OpenSimulator - User contributions [en]

Office hours

2008-08-26T20:03:25Z

ChrisD: /* Logs */

OpenSim weekly office hours are held in an OpenSim world somewhere on the Internet. The purpose of these office hours is to created a focussed weekly discussion on the state of OpenSim, and things that need attention. It is intended for people wishing to become contributors to the project, so if you show up, you'll probably walk away with something on your todo list.

The office hours are held on Tuesdays at 19:00 UTC. This corresponds to:
* 11:00 PST - San Francisco (U.S.A. - California)
* 14:00 - New York (U.S.A. - New York)
* 19:00 UST - Oslo (Norway)
* 20:00 - Italy
* 02:00 Wed - Perth (Australia) (sorry adam)
* 15:00 AST - Atlantic Canada

There is useful [http://www.timeanddate.com/ Time'n'date] tool for timezone aware planning, with [http://www.timeanddate.com/worldclock/converter.html Time Zone Converter] & [http://www.timeanddate.com/worldclock/meeting.html Meeting Planner] (last thing helps to plan meeting for up 4 tz places)

The office hours will take place at the sim "Wright Plaza" on osgrid, which was called "Sang" until last Friday.

To get to this sim, make sure you have created an account with the normal first, last and password and set the home location to "Wright Plaza". This is done at http://www.osgrid.org.

Then startup the client with a -loginuri of "-loginuri http://osgrid.org:8002" and you should arrive on "Wright Plaza".

=== Previous Agenda Items ===

Cfk would like to suggest that we give some thought to the following bugs.
#) Remoting exceptions with adjacent non-running sims.
#) Prim unlinking.
#) Enabling the script editor from the edit prim dialog box.
#) jpeg/ode dll and glibc-2.4 exceptions and stability.
#) GridServer.exe locks up in r2090 and works fine in r2060 on OSGrid.
#) Enabling avatar 'sit' on a prim.
#) More progress on avatar customization consistency. (Avatar changes when crossing region boundaries).

== Logs ==
* [[Chat log from the meeting on 2008-08-26]]
* [[Chat log from the meeting on 2008-08-19]]
* [[Chat log from the meeting on 2008-08-12]]
* [[Chat log from the meeting on 2008-08-05]]
* [[Chat log from the meeting on 2008-07-22]]
* [[Chat log from the meeting on 2008-07-15]]
* [[Chat log from the meeting on 2008-07-08]]
* [[Chat log from the meeting on 2008-07-01]]
* [[Chat log from the meeting on 2008-06-17]]
* [[Chat log from the meeting on 2008-06-10]]
* [[Chat log from the meeting on 2008-04-15]]
* [[Chat log from the meeting on 2008-04-08]]
* [[Chat log from the meeting on 2008-03-11]]
* [[Chat log from the meeting on 2008-03-04]]
* [[Chat log from the meeting on 2008-02-17]]
* [[Chat log from the meeting on 2008-02-12]]
* [[Chat log from the meeting on 2008-02-05]]
* [[Chat log from the meeting on 2008-01-29]]
* [[Chat log from the meeting on 2008-01-09]]
* [[Chat log from the meeting on 2007-12-18]]
* [[Chat log from the meeting on 2007-12-11]]
* Chat log from the meeting on 2007-12-4 (meeting cancelled)
* [[Chat log from the meeting on 2007-11-27]]
* Chat log from the meeting on 2007-11-20 (missing)
* [[Chat log from the meeting on 2007-11-13]]
* [[Chat log from the meeting on 2007-11-06]]
* [[Chat log from the meeting on 2007-10-30]]
* [[Chat log from the meeting on 2007-10-23]]
* [[Chat log from the meeting on 2007-10-16]]
* [[Chat log from the meeting on 2007-10-09]]

[[Category:Development]]
[[Category:Office Hours]]

Chat log from the meeting on 2008-08-26

2008-08-26T20:02:10Z

ChrisD: New page: [11:07] Charles Krinkeb: Do I assume there are no Lindens here right now so we skip the interop discussion until later? [11:07] BlueWall Slade: tada! [11:08] Whump Linden...

[11:07] Charles Krinkeb: Do I assume there are no Lindens here right now so we skip the interop discussion until later?
 [11:07] BlueWall Slade: tada!
 [11:08] Whump Linden: Howdy.
 [11:08] Justin Clark-Casey: Hi Whump
 [11:08] Teravus Ousley: heh, Mr Bill is here.
 [11:08] Teravus Ousley: :D
 [11:09] Charles Krinkeb: Morning, Whump. Glad to see you. We were just about to get started and I believe discussing the OGP gridnaut project is the first item on the agenda.
 [11:09] Charles Krinkeb: I believe that Teravus has commited most of the OGP patch to svn. Perhaps Teravus could describe what he did for you and others, Whump.
 [11:10] Whump Linden: Yes, Teravus, I yield the floor.
 [11:11] Teravus Ousley: hmm. Ok. Essentially, I reimplemented the patch as a region module. Very little code was copy and pasted. Though, the original patch was used as a reference.
 [11:11] Teravus Ousley: What that means for OGP, is it will benefit from not requiring to keep and maintain a separate patch.
 [11:12] egion going down
 [11:12] Charles Krinkeb: So, this would be svn past r5970 where r5970 was also just released last weekend as 0.5.9, right, Terauvs?
 [11:12] Justin Clark-Casey: uhh
 [11:12] region going down
 [11:12] Justin Clark-Casey: why is the region going down
 [11:12] Whump Linden: yes, running svn -up is much easier than downloading and running a patch
 [11:12] Teravus Ousley: It probably isn't.. it's a neighbor
 [11:12] Charles Krinkeb: dont think it is.
 [11:12] Teravus Ousley: blue messages get sent to child agents too..
 [11:12] Charles Krinkeb: maybe someone has figured out how to send messages cross region.
 [11:13] Justin Clark-Casey: mmm, I presume that's a bug?
 [11:13] Teravus Ousley: I'd say that's probably a bug, yes
 [11:13] Charles Krinkeb: One mans bug is another mans feature.
 [11:13] Justin Clark-Casey: I think this is a bug for all men
 [11:13] Justin Clark-Casey: and women
 [11:13] Justin Clark-Casey: and dragons
 [11:13] Charles Krinkeb: I'm neutral as long as the region doesnt crash.
 [11:13] Teravus Ousley: Well, to turn on OGP, you must pick a spot that isn't already taken on http://wiki.secondlife.com/wiki/Open_Grid_Public_Beta/Map_Locations#Teravus_Ousley
 [11:14] Teravus Ousley: that's your 'gridspace' in OGP
 [11:14] Teravus Ousley: your /Regions/*.xml file contains your region X, and Y.
 [11:14] Teravus Ousley: don't overlap on another region.. or you'll get cache problems in the client. :D
 [11:14] Charles Krinkeb: Will we be still working out details on X versus X*256 for a while?
 [11:15] Whump Linden: That was disucssed at AWG this morning.
 [11:15] Whump Linden: https://wiki.secondlife.com/wiki/AW_Groupies/Chat_Logs/AWGroupies-2008-08-26
 [11:15] Teravus Ousley: no.. I think it's covered right now. OpenSimulator will be configured at the moment the same way that.. say it would be in standalone or grid mode. The gridspaces mean essentially the same thing.
 [11:16] Teravus Ousley: it's just an internal representation that's different, if you know what I mean.
 [11:16] Neas Bade: hey all
 [11:16] Justin Clark-Casey: Hi Neas
 [11:16] Teravus Ousley: Hey Sean
 [11:16] Charles Krinkeb: Are there any action items for OGP that need to be addressed this week on the OpenSim side?
 [11:17] Teravus Ousley: One bug with teleporting sending you to 0,0,2 if a parcel teleport location isn't set
 [11:17] Whump Linden: yes, I'll open a copy of that on JIRA
 [11:17] Justin Clark-Casey: actually how good is teleporting generally at the moment?
 [11:17] Charles Krinkeb: Morning, SDague. We are just finishing up on the first agenda item, OGP.
 [11:17] Justin Clark-Casey: oh sorry, don't let me interrupt the ogp discussion
 [11:18] Teravus Ousley: Additionally, after you pick a non-conflicting X and Y, you'll need to set the following OpenSim.ini options
 [11:18] Charles Krinkeb: no problem, Justin. Basically, Core developers or Lindens have the floor at any time they wish.
 [11:18] Whump Linden: and we have two high priority bugs on our side, one's fixing the sim_host, and the other is the content type on responses to caps
 [11:18] Whump Linden: at the moment, OpenSim has workarounds for both of those
 [11:19] Teravus Ousley: [OpenGridProtocol]
 [11:19] Charles Krinkeb: Great. Whump: "What would you suggest we should be thinking about for 2-4 weeks down the road?"
 [11:19] Teravus Ousley: ogp_enabled=true
 [11:19] Whump Linden: so we will be releasing another viewer binary later this week
 [11:19] Whump Linden: Charles: getting OGP working in Grid mode
 [11:20] Teravus Ousley heard that the protocol spec 3 is comming out very soon also
 [11:20] Charles Krinkeb: Got it. Sounds like a good thought. I will carry that banner around the field a bit.
 [11:20] Whump Linden: yes, Zero's reviewing it today
 [11:20] Teravus Ousley: Ooh, Whump. Technically you can do OGP in grid mode, though, only the region knows about you.
 [11:20] Charles Krinkeb: Will I see you and Zero next week at VW08? I will be in the OpenSim/3di booth.
 [11:21] Teravus Ousley: and the Grid servers 'shrug' at your identity when they get asked about you :D
 [11:21] Whump Linden: Tess and Hamilton will be at VW08
 [11:21] Charles Krinkeb: Cool. I'll look for them.
 [11:21] Whump Linden: and they'll be at the LL booth part of the conference
 [11:22] Teravus Ousley: Though.. you can cross borders and do all of the normal stuff in grid mode with OGP.. no inventory or appearance though.
 [11:22] Whump Linden: so you're somewhat of a ghost?
 [11:22] Whump Linden: in grid mode
 [11:23] Whump Linden: will other agents be aware of your presense?
 [11:23] Teravus Ousley: yep, essentially a visitor.. nobody knows about you but the region you're in.. but the essential state information gets sent in border crossings/teleports. Yes.. they are.
 [11:23] Whump Linden: okay
 [11:24] Whump Linden: well, I can be available to set up another region and hook it up to the OSGrid
 [11:24] Neas Bade: charles, is there an agenda? I had a question, but it's unrelated to the current topic, and didn't want to push anyone off the stage
 [11:24] Charles Krinkeb: Ok. Trying to move on, but it is fair to talk about anything previously covered throughout the meeting. I believe we have a new release for OpenSim and it is 0.5.9. It will get up on the web site in the next day or so. I dont think the OGP is in 0.5.9 but is post-0.5.9 for commit.
 [11:25] Whump Linden: okay, I'm done
 [11:25] Neas Bade: Charles, is that going on the opensim project on the forge?
 [11:25] Teravus Ousley: right, OGP was after 0.5.9 on purpose.
 [11:25] Neas Bade: I saw adam set that up, and it will make a good place for file releases
 [11:25] Salahzar Raptor: yep: can you tell us the url of the agenda?
 [11:25] Charles Krinkeb: Our first agenda item was OGP. The second agenda item is "What things do the developers need from the testers and other participants this week?"
 [11:25] Justin Clark-Casey: is there any way to identify visitors via ogp to osgrid
 [11:26] Charles Krinkeb: Our agenda always has three items. 1) OGP 2) What developers need. 3) What testers need.
 [11:26] Justin Clark-Casey: ?
 [11:26] Teravus Ousley: at the moment, they come in with their linden name..
 [11:26] Charles Krinkeb: Go ahead, SDague.
 [11:26] Justin Clark-Casey: any other identifying features? birthmark, tattoos?
 [11:26] Teravus Ousley: heh
 [11:26] Neas Bade: doh, client crash
 [11:27] Justin Clark-Casey: I'm just wondering if we can try and stop spurious information requests to the varoius grid services
 [11:27] Teravus Ousley: If you look up their profile, they probably won't have one.
 [11:27] Justin Clark-Casey: since that would make it easier to identify real bugs
 [11:27] Whump Linden: / justin: should have "_EXTERNAL" appended to their last name.
 [11:27] You: somtimes they are clouds :)
 [11:27] Neas Bade: heh, missing terr head
 [11:27] Neas Bade: man, that looks crazy
 [11:27] Justin Clark-Casey: whump: thanks, that's a good point
 [11:27] Teravus Ousley: hmm, I didn't add _EXTERNAL to the current version.
 [11:27] Teravus Ousley: Would you like me to add _EXTERNAL?
 [11:27] Neas Bade: ter, wouldn't it be better to just put something in profile?
 [11:28] Teravus Ousley: I do.
 [11:28] Neas Bade: ok
 [11:28] Justin Clark-Casey: yes, perhaps a bool on the client view?
 [11:28] Whump Linden: / Teravus: yes, I'll Mantis it
 [11:28] Teravus Ousley: It says something like 'OGP User' and 'About says, "I'm testing the OpenGridProtocol"
 [11:28] Justin Clark-Casey: oh okay.... so there's some information in profile that can be used?
 [11:28] Justin Clark-Casey: hmm, okay
 [11:28] Neas Bade: so, I heard that in the latest message templates the protocol supports click position on surfaces
 [11:28] Justin Clark-Casey: it's alright - I just want to eliminate issues which might be leading to the huge upsurge in paranormal (ghost) activity lately
 [11:28] Neas Bade: does anyone know more about that?
 [11:29] Neas Bade: and how we might enable that in opensim
 [11:29] Teravus Ousley: hmm. OGP users are currently treated as regular agents as far as the simulator is concerned.
 [11:29] Whump Linden: / okay, I need to run, folks are in #gridnauts if you have more questions
 [11:29] Justin Clark-Casey: yeah, that's what I thought
 [11:29] Whump Linden: Thanks Charles.
 [11:29] Teravus Ousley: Thanks :D
 [11:30] Charles Krinkeb: thanks, Whump. Looks like things are moving along.
 [11:30] Whump Linden: And big thanks to Teravus, he jumped in and was all out of bubble gum.
 [11:31] Charles Krinkeb: Ok, next agenda item. "What are the things the core developers want or need from the testers?"
 [11:31] Teravus Ousley: I'd like to thank Zha.. also .. but he's not here.
 [11:32] Justin Clark-Casey: are there still lots of ghosts around?
 [11:32] Shenlei Winkler: yes
 [11:32] Shenlei Winkler: there are
 [11:32] Justin Clark-Casey: I came on wp the other day and there were a bunchj of people who showed up in the users list, but were unresponsive
 [11:32] Justin Clark-Casey: which is almost worse than straightforward ghosts
 [11:32] Charles Krinkeb: Last week, we said that Mantis with confirming notes on both Windows and Linux and differing configurations or at least additional observers indicating they see the same problem would help the developers.
 [11:32] Neas Bade: yeh, I'm thinking of diving into the scene data structures and cleaning some of those up
 [11:33] You: Currently there are ghosts in Ferrmi for just about every avatar at this meeting.
 [11:33] Neas Bade: it may at least help us track down some of these ghosts
 [11:33] Neas Bade: so, no one have any idea on the surface touch stuff?
 [11:33] Justin Clark-Casey: the trouble is that I think there are multiple ghost sources
 [11:33] Charles Krinkeb: We are opening more Mantis then we are closing, so we need some help from the testers to add confirming notes to help the developers.
 [11:33] Justin Clark-Casey: firstly there were those that were around already, and then there are the new ones
 [11:33] Teravus Ousley: well remember. Ghosts are caused by a full object update on an avatar that shouldn't get one.
 [11:33] Justin Clark-Casey: Neas: That would be excellent stuff
 [11:33] Justin Clark-Casey: Neas: perhaps even splitting up the massive Scene class in some way?
 [11:34] Neas Bade: yeh, that was my thought
 [11:34] Justin Clark-Casey: actually what would be really useful for testers to do
 [11:34] Justin Clark-Casey: is see if they can isolate the most simple way in which a ghost can be replicated...
 [11:34] Neas Bade: after I head on colided with it trying to do nhibernate region store stuff right before vacation I started thinking that it would be better to put the nh work on the back burner and putter on cleaning up the scene stuff
 [11:35] Justin Clark-Casey: Neas: I don't suppose you can tell me what the real driving purpose behind InnerScene is?
 [11:35] Shenlei Winkler: Can you be more specific, Justin?
 [11:35] Shenlei Winkler: We get ghosts everytime a user logs out
 [11:35] Neas Bade: nope, that predates me :)
 [11:35] Charles Krinkeb: What can testers do to help developers that we are not doing?
 [11:36] Justin Clark-Casey: Shenlei: Hmm, I've seen log outs occur without ghosts... but perhaps its different if someone logs into the region
 [11:36] Justin Clark-Casey: Shenlei: do you see the avatar actually disappear?
 [11:36] Shenlei Winkler: nope
 [11:36] Shenlei Winkler: to all intents and purposes, it is that user
 [11:36] Neas Bade: yeh, I think we used to have a sweeper that cleaned up child agents
 [11:36] Neas Bade: but at some point that stopped being in there
 [11:36] Justin Clark-Casey: Interesting, I saw it happen properly on Wp the other day so it's not completely consistent...
 [11:36] Shenlei Winkler: but since we know exactly who is in the region, we also know that the avatar representation is a ghost
 [11:36] Shenlei Winkler: and, you get a green dot on the minimap
 [11:37] Teravus Ousley: right.. got lost somewhere in the LLUDP rewrite/refactoring.
 [11:37] Justin Clark-Casey: yeah, possibly
 [11:37] Justin Clark-Casey: but it would be good to pin it down
 [11:37] Shenlei Winkler: well, we can work on that
 [11:37] Shenlei Winkler: the version on our internal region is 5988
 [11:38] Justin Clark-Casey: if we could get a simple series of replication steps (even if it's straight login and then logout), that would help
 [11:38] Melanie Milland: we should not really depend on a sweeper
 [11:38] Justin Clark-Casey: Shenlei: and I guess I could even come see it for myself soon :)
 [11:38] Melanie Milland: good to have one
 [11:38] Shenlei Winkler: this afternoon :)
 [11:38] Melanie Milland: but if that's more than a last line of defense, then somethings else stinks
 [11:38] Justin Clark-Casey: well, it looks like things sohuld timeout in the LLClientView code and clean them up, in principle
 [11:38] Teravus Ousley: Melanie: we have to.. the scene doesn't know when a client is dead or not
 [11:39] Neas Bade: yeh, this is the ping sweep
 [11:39] Melanie Milland: :/
 [11:39] Neas Bade: though we should really have 1 reaper, instead of the one per client view the way they used to be
 [11:39] Charles Krinkeb: Last of my three "Office Hour" general agenda questions: "What do the testers need from the developers to make their life easier?"
 [11:39] Melanie Milland thinks: "Borken protocol indeed"
 [11:39] Neas Bade: we always need to allow for a hard crashed client
 [11:39] Shenlei Winkler: well...can I say thank you to everyone ...because life in OpenSim is way easier
 [11:39] Justin Clark-Casey: well you know, without unit tests we don't know what to eliminate very well...
 [11:39] Neas Bade: when means no graceful logout
 [11:40] Justin Clark-Casey: or at least, look where to divert more attention
 [11:40] Shenlei Winkler: I was looking at my notes from Spirit and things are just so much smoother
 [11:40] You: The gohsts on Fermi are for avatars at this meeting and I doubt if they arrived at WP via Fermi.
 [11:40] Justin Clark-Casey: Chris D: where is Fermi?
 [11:40] Shenlei Winkler: Other than that, vehicle stuff would be great
 [11:40] Justin Clark-Casey: Shenlei: thanks :)
 [11:40] Teravus Ousley: Thanks for doing that! It has been quite illuminating and helpful :D
 [11:40] Charles Krinkeb: SouthEast
 [11:40] Shenlei Winkler: cool :)
 [11:40] Shenlei Winkler: Melanie, thank yu for avatar persistence ;)
 [11:40] Justin Clark-Casey: to me, that sounds like the old child avatar issue...
 [11:41] Justin Clark-Casey: er child agent
 [11:41] Neas Bade: yes, that's the child agent issue
 [11:41] You: Yes, it has been like that for quite a while
 [11:41] Neas Bade: anyway, with any luck, I may make it easier to figure out these issues. We'll see
 [11:41] Justin Clark-Casey: right, regarding inventory I'm hoping 1903 will be patched up soon
 [11:42] Neas Bade: the net effect however is that the nh stuff is probably on indefinite hold, unless someone else has cycles to dig in.
 [11:42] Justin Clark-Casey: so it might be worth holding off on reporting any more problems with it until then
 [11:42] Neas Bade only has so many hours in a week
 [11:42] Teravus Ousley: haha, yes indeed.. time is a limited resource
 [11:42] Shenlei Winkler: Justin - do you think the inventory issue might be contributing to the off texture loading anomaly I've been noticing?
 [11:42] Shenlei Winkler: Sometimes texture don't load.
 [11:42] Teravus Ousley: hehe, I suppose the ultimiate limited resource.. but that's a topic for another meeting :D
 [11:43] Justin Clark-Casey: Shenlei: No, they will be unrelated
 [11:43] Neas Bade: yep :)
 [11:43] Shenlei Winkler: Pity
 [11:43] Justin Clark-Casey: are all the non-loading textures large ones?
 [11:43] Shenlei Winkler: no
 [11:43] Shenlei Winkler: in fact, some are sculpti textures, which are tiny
 [11:43] UUID Speaker: Neas Bade, your key is f5ce1c54-dd24-441f-bb0e-99d34e0f35fd
 [11:43] Charles Krinkeb: I have a question: "Do I need to change my three agenda items in any way or is this the best way to move forward for now?"
 [11:43] Teddy Ruxpin: .
 [11:43] Teravus Ousley: .
 [11:43] Teddy Ruxpin: Weird
 [11:43] Justin Clark-Casey: ..
 [11:44] Charles Krinkeb: ?
 [11:44] Teddy Ruxpin: My av started typing on its own
 [11:44] Teravus Ousley: Ö¿Ö
 [11:44] Neas Bade: I do really want to figure out this surface click thing as well though :)
 [11:44] Justin Clark-Casey: shenlei: well, another thing I can look at son :)
 [11:44] Justin Clark-Casey: soon
 [11:44] Charles Krinkeb: That happens to all of us from time to time, Teddy.
 [11:44] Shenlei Winkler grins
 [11:44] Justin Clark-Casey: neas: I notice Whump took off when you asked - a coincidence I'm sure ;)
 [11:45] Neas Bade: http://wiki.secondlife.com/wiki/Release_Notes/Second_Life_Server/1.24
 [11:45] Charles Krinkeb: co-incidence
 [11:45] Neas Bade: it's LSL documented there
 [11:45] Neas Bade: :)
 [11:45] Neas Bade: no one likes answering my questions ;)
 [11:46] Neas Bade: but, seriously, the ability to do in world apps goes way up once we have that
 [11:46] Justin Clark-Casey: yeah, definitely
 [11:46] Charles Krinkeb: Chris: do you mind copying the chat log and putting it up on the wiki today? Do you know how to do that?
 [11:46] Teravus Ousley: Ok, since whump is requesting _External in the name. I'll set it as an option in the OpenSim.ini which defaults to _External
 [11:46] Shenlei Winkler: I was kind of curious to hear what he had to say, myself
 [11:46] Melanie Milland: external in the name for what?
 [11:47] Teravus Ousley: OGP name, gets modified so you can know that it's an external avatar
 [11:47] Neas Bade: it seems like :OGP or something would be btter
 [11:47] Salahzar Raptor: if the agenda has ended.. can I put a question?
 [11:47] Neas Bade: fire away
 [11:48] Salahzar Raptor: yes the question is about the XEngine..
 [11:48] Salahzar Raptor: I've read that you are considering to have it as default and that's might be a good thing
 [11:48] Salahzar Raptor: since now there is a lot of confusion in would-be helpers
 [11:48] Salahzar Raptor: since lsl-coding is duplicated in two places :(
 [11:48] Salahzar Raptor: is that project still on?
 [11:49] Justin Clark-Casey: we're hoping to recieve patches which reduce the duplication soon
 [11:49] Charles Krinkeb: Until we resolve the duplicated file, please patch the original LSL_BuiltIn_Commands.cs and the LSL_Api.cs.
 [11:49] Justin Clark-Casey: kurt is apparantly working on them - it's a pity he isn't here since we really need a status update, I think
 [11:49] You: Charles: Sorry got called away. Yes no problem with the chat log.
 [11:49] Charles Krinkeb: Thank you kindly, Chris.
 [11:50] Charles Krinkeb: Your help is greatly appreciated. I am a one-armed paper hanger sometimes with this project.
 [11:50] Salahzar Raptor: Yes we are working that way, but it might become increasingly complex to have them aligned properly
 [11:50] Salahzar Raptor: since they are not only copies but also some refactoring has been done..
 [11:50] Salahzar Raptor: so it behaves like a "fork"
 [11:51] Charles Krinkeb: We do now know when the duplication will be resolved. Until it is, it is best to patch both LSL_BuiltIn_Commands.cs and LSL_Api.cs if possible.
 [11:51] Charles Krinkeb: "do not know"
 [11:51] Justin Clark-Casey: if it doesn't get resolved soon, I shall resolve it in a way that makes everybody cry
 [11:51] Teravus Ousley: haha
 [11:51] Teravus Ousley: sounds like something Adam would do :D
 [11:51] Junta Kohime: lol
 [11:51] Charles Krinkeb: The duplication of LSL_BuiltIn_Commands.cs to another file was a mistake in the first place and we are still working to resolve it.
 [11:52] Justin Clark-Casey: yeah, it's a good tactic sometimes if you have skin of steel
 [11:52] Neas Bade: heh
 [11:52] Teravus Ousley: hehe, if your hair was black.. you could hide in the shadows..
 [11:52] Salahzar Raptor: Ok so I will go on on checking the differences between the two files
 [11:52] Charles Krinkeb: As the guy that committed the patch that duplicated LSL_BuiltIn_Commands.cs in the first place, the whole thing is a stone around my neck until it gets resolved.
 [11:52] Justin Clark-Casey: :)
 [11:52] Neas Bade: oh man, justin surpassed ter for #1 commit spot
 [11:53] Teravus Ousley: yep
 [11:53] Justin Clark-Casey: not for long though, I'm sure
 [11:53] Justin Clark-Casey: Teravus: finished with your big deadline?
 [11:53] Teravus Ousley: nah. I'm still stuck with work. Just until Sept 1st though likely.
 [11:53] Charles Krinkeb: Ok. I declare the formal meeting over and any questions or statements are fair.

LLUDP Dissector

2008-08-01T08:32:16Z

ChrisD: /* on Windows */

== LLUDP protocol dissector ==
On this page you will find the Lua code for a wireshark protocol dissector that can parse the message_template.msg file and use that information to decode all the message fields from the Linden UDP protocol.

== Installing ==
* Requires wireshark with Lua 5.1.x support. [http://wiki.wireshark.org/Lua See this page for getting wireshark to support lua]

=== on Linux ===
* Copy all four source files into ~/.wireshark
* If you need to run wireshark as the root user or using sudo then you will need to edit the scripts into one file by replacing the dofile("script.lua") calls with the contents of file between the quotes.
* The other method is to add your user account to the correct group (on Gentoo it is group "wireshark") that will allow your non-root user to capture packets.

=== on Windows ===
* Copy all four source files into your user profiles directory

'''Vista*'''

C:\Users\<username>\AppData\Roaming\Wireshark

'''XP/2000'''

C:\Documents and Settings\<username>\Application Data\Wireshark

* Edit C:\Program Files\Wireshark\init.lua and change ''disable_lua'' to ''false'' (default is true)

* *Note: I have only tested this on Windows XP

== Description of source files ==
* "init.lua" -- simple script that loads the "lludp.lua" script.
* "lludp.lua" -- contains the code that decodes each packet header and decompresses zero-encoded packets. This file uses wireshark only functions for accessing packet bytes and building a tree of information from each packet.
* "llmessage.lua" -- contains the message_template.msg file parser the decodes the tokens from the lexer into an tree of tables containing all details about each message/block/variable from the template file. This file only has pure lua code.
* "lexer.lua" -- contains the template file lexer. This lexer knows how to tokenize the template file into the follow tokens: IDENTIFIER, NUMBER, COMMENT, EOL. The stream of tokens produced by this lexer is parsed by the "llmessage.lua" file. This file only has pure lua code.

=== File "init.lua" ===
<source lang="lua">

-- register http to handle tcp ports 8000-8010 and 9000
do
local tcp_port_table = DissectorTable.get("tcp.port")
local http_dissector = tcp_port_table:get_dissector(80)
for port = 8000,8010 do
tcp_port_table:add(port,http_dissector)
end
tcp_port_table:add(9000,http_dissector)
end

-- Load lludp protocol dissector.
dofile("lludp.lua")

</source>

=== File "lludp.lua" ===
<source lang="lua">
--[[
BSD-Licensed:
Copyright (c) 2008, Robert G. Jakabosky <bobby@sharedrealm.com>
All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

* Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
* Neither the name of the SharedRealm nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

]]

dofile("llmessage.lua")

-- cache globals to local for speed.
local str_format=string.format

-- test if ByteArray only has printable ASCII character and ends with '\0'
local allowed_special = {
[0] = true, -- null
[9] = true, -- tab
[10] = true, -- new line
[13] = true, -- carriage return
}
local function is_string(bytes)
local c
local max = bytes:len() - 1
for i=0,max do
c = bytes:get_index(i)
if c >= 127 then -- not ascii character
return false
elseif c < 32 then -- control character range
if not allowed_special[c] then
-- control characters between NULL and Space
return false
elseif c == 0 and i < max then
-- null byte only allowed at end.
return false
end
end
end
return true
end

-- lludp protocol example
-- declare our protocol
local lludp_proto = Proto("lludp","LLUDP","LindenLabs UDP Protocol")

-- setup preferences
lludp_proto.prefs["template_file"] =
Pref.string("Message template file", "message_template.msg", "Message template file")
lludp_proto.prefs["udp_port_start"] =
Pref.string("UDP port range start", "13000", "First UDP port to decode as this protocol")
lludp_proto.prefs["udp_port_end"] =
Pref.string("UDP port range end", "13050", "Last UDP port to decode as this protocol")
-- current preferences settings.
local current_settings = {
template_file = "",
udp_port_start = -1,
udp_port_end = -1,
}
-- current list of parsed messages.
local message_details = nil

-- setup protocol fields.
lludp_proto.fields = {}
local fds = lludp_proto.fields
fds.flags = ProtoField.new("Flags", "lludp.flags", "FT_UINT8", nil, "BASE_HEX", "0xFF")
fds.flags_zero = ProtoField.new("Zero", "lludp.flags.zero", "FT_UINT8", nil, "BASE_HEX", "0x80")
fds.flags_reliable = ProtoField.new("Reliable", "lludp.flags.rel", "FT_UINT8", nil, "BASE_HEX", "0x40")
fds.flags_resent = ProtoField.new("Resent", "lludp.flags.res", "FT_UINT8", nil, "BASE_HEX", "0x20")
fds.flags_ack = ProtoField.new("Ack", "lludp.flags.ack", "FT_UINT8", nil, "BASE_HEX", "0x10")
fds.sequence = ProtoField.new("Sequence", "lludp.sequence", "FT_UINT32", nil, "BASE_DEC")
fds.extra_len = ProtoField.new("Extra length", "lludp.extra_len", "FT_UINT8", nil, "BASE_DEC")
fds.extra_bytes = ProtoField.new("Extra header", "lludp.extra_bytes", "FT_BYTES", nil, "BASE_HEX")
fds.msg_id = ProtoField.new("Message ID", "lludp.msg.id", "FT_UINT32", nil, "BASE_HEX")
fds.msg_name = ProtoField.new("Message name", "lludp.msg.name", "FT_STRINGZ", nil)
fds.msg = ProtoField.new("Message body", "lludp.msg", "FT_BYTES", nil, "BASE_HEX")
fds.acks_count = ProtoField.new("Acks count", "lludp.acks_count", "FT_UINT8", nil, "BASE_DEC")
fds.acks = ProtoField.new("Acks", "lludp.acks", "FT_UINT32", nil, "BASE_DEC")
fds.block_count = ProtoField.new("Block count", "lludp.block_count", "FT_UINT8", nil, "BASE_DEC")
fds.block = ProtoField.new("Block", "lludp.block", "FT_BYTES", nil, "BASE_HEX")
fds.var_fixed = ProtoField.new("Fixed blob", "lludp.var.fixed", "FT_BYTES", nil, "BASE_HEX")
fds.var_variable = ProtoField.new("Variable blob", "lludp.var.variable", "FT_BYTES", nil, "BASE_HEX")
fds.var_string = ProtoField.new("String", "lludp.var.string", "FT_STRINGZ", nil)
fds.var_u8 = ProtoField.new("U8", "lludp.var.u8", "FT_UINT8", nil, "BASE_DEC")
fds.var_u16 = ProtoField.new("U16", "lludp.var.u16", "FT_UINT16", nil, "BASE_DEC")
fds.var_u32 = ProtoField.new("U32", "lludp.var.u32", "FT_UINT32", nil, "BASE_DEC")
fds.var_u64 = ProtoField.new("U64", "lludp.var.u64", "FT_UINT64", nil, "BASE_DEC")
fds.var_s8 = ProtoField.new("S8", "lludp.var.s8", "FT_INT8", nil, "BASE_DEC")
fds.var_s16 = ProtoField.new("S16", "lludp.var.s16", "FT_INT16", nil, "BASE_DEC")
fds.var_s32 = ProtoField.new("S32", "lludp.var.s32", "FT_INT32", nil, "BASE_DEC")
fds.var_s64 = ProtoField.new("S64", "lludp.var.s64", "FT_INT64", nil, "BASE_DEC")
fds.var_f32 = ProtoField.new("F32", "lludp.var.f32", "FT_FLOAT", nil, "BASE_DEC")
fds.var_f64 = ProtoField.new("F64", "lludp.var.f64", "FT_DOUBLE", nil, "BASE_DEC")
fds.var_llvector3 = ProtoField.new("LLVector3", "lludp.var.llvector3", "FT_BYTES", nil, "BASE_HEX")
fds.var_llvector3d = ProtoField.new("LLVector3d", "lludp.var.llvector3d", "FT_BYTES", nil, "BASE_HEX")
fds.var_llvector4 = ProtoField.new("LLVector4", "lludp.var.llvector4", "FT_BYTES", nil, "BASE_HEX")
fds.var_llquaternion = ProtoField.new("LLQuaternion", "lludp.var.llquaternion", "FT_BYTES", nil, "BASE_HEX")
fds.var_lluid = ProtoField.new("LLUID", "lludp.var.lluid", "FT_BYTES", nil, "BASE_HEX")
fds.var_bool = ProtoField.new("BOOL", "lludp.var.bool", "FT_UINT8", nil, "BASE_DEC")
fds.var_ipaddr = ProtoField.new("IPADDR", "lludp.var.ipaddr", "FT_IPv4", nil, "BASE_DEC")
fds.var_ipport = ProtoField.new("IPPORT", "lludp.var.ipport", "FT_UINT16", nil, "BASE_DEC")
-- variable type handlers.
local variable_handlers = {
Fixed = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_fixed, rang)
if len <= 4 then
ti:set_text(str_format("%s: 0x%08x",var.name, rang:uint()))
else
ti:set_text(str_format("%s: length=%d, Blob:%s", var.name, len, tostring(rang)))
end
end,
Variable = function(block_tree, buffer, offset, len, var)
local is_data = false
-- try to guess if this field is text.
if var.name:find("Data") then
-- this is a data find.
is_data = true
end
local str_rang = buffer(offset + var.count_length, len - var.count_length)
local bytes = str_rang:bytes()
if not is_data and is_string(bytes) then
local str = str_rang:string()
local ti = block_tree:add(fds.var_string, buffer(offset, len), str)
ti:set_text(str_format("%s: %s", var.name, str))
else
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_variable, rang)
if len <= 4 then
ti:set_text(str_format("%s: 0x%08x",var.name, rang:uint()))
else
ti:set_text(str_format("%s: length=%d, Blob:%s", var.name, len, tostring(rang)))
end
end
end,
U8 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_u8, rang)
ti:set_text(str_format("%s: %d", var.name, rang:le_uint()))
end,
U16 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_u16, rang)
ti:set_text(str_format("%s: %d", var.name, rang:le_uint()))
end,
U32 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_u32, rang)
ti:set_text(str_format("%s: %d", var.name, rang:le_uint()))
end,
U64 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_u64, rang)
ti:set_text(str_format("%s: 0x%s",var.name, tostring(rang)))
end,
S8 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_s8, rang)
local num = rang:le_uint()
if num > 127 then num = num - 256 end
ti:set_text(str_format("%s: %d",var.name, num))
end,
S16 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_s16, rang)
local num = rang:le_uint()
if num > 32768 then num = num - 65536 end
ti:set_text(str_format("%s: %d",var.name, num))
end,
S32 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_s32, rang)
local num = rang:le_uint()
if num > 2147483648 then num = num - 4294967296 end
ti:set_text(str_format("%s: %d",var.name, num))
end,
S64 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_s64, rang)
ti:set_text(str_format("%s: 0x%s",var.name, tostring(rang)))
end,
F32 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_f32, rang)
ti:set_text(str_format("%s: %f", var.name, rang:le_float()))
end,
F64 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_f64, rang)
ti:set_text(str_format("%s: %f", var.name, rang:le_float()))
end,
LLVector3 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_llvector3, rang)
-- parse LLVector3
local x,y,z
x = buffer(offset + 0,4):le_float()
y = buffer(offset + 4,4):le_float()
z = buffer(offset + 8,4):le_float()
-- display
ti:set_text(str_format("%s: <%f,%f,%f>", var.name, x, y, z))
end,
LLVector3d = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_llvector3d, rang)
-- parse LLVector3d
local x,y,z
x = buffer(offset + 0,8):le_float()
y = buffer(offset + 8,8):le_float()
z = buffer(offset + 16,8):le_float()
-- display
ti:set_text(str_format("%s: <%f,%f,%f>", var.name, x, y, z))
end,
LLVector4 = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_llvector4, rang)
-- parse LLVector4
local x,y,z,s
x = buffer(offset + 0,4):le_float()
y = buffer(offset + 4,4):le_float()
z = buffer(offset + 8,4):le_float()
s = buffer(offset + 12,4):le_float()
-- display
ti:set_text(str_format("%s: <%f,%f,%f,%f>", var.name, x, y, z, s))
end,
LLQuaternion = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_llquaternion, rang)
-- parse LLQuaternion
local x,y,z,w
x = buffer(offset + 0,4):le_float()
y = buffer(offset + 4,4):le_float()
z = buffer(offset + 8,4):le_float()
-- calculate W
w = 1 - (x * x) - (y * y) - (z * z)
if w > 0 then
w = math.sqrt(w)
else
w = 0
end
-- display
ti:set_text(str_format("%s: <%f,%f,%f,%f>", var.name, x, y, z, w))
end,
LLUUID = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_lluid, rang)
local str = tostring(rang)
str = str:sub(1,8) .. '-' ..
str:sub(9,12) .. '-' .. str:sub(13,16) .. '-' ..
str:sub(17,20) .. '-' .. str:sub(21)
ti:set_text(str_format("%s: %s", var.name, str))
end,
BOOL = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add_le(fds.var_bool, rang)
local val = "false"
if rang:le_uint() > 0 then
val = "true"
end
ti:set_text(str_format("%s: %s", var.name, val))
end,
IPADDR = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add(fds.var_ipaddr, rang)
ti:set_text(str_format("%s: %s", var.name, tostring(rang:ipv4())))
end,
IPPORT = function(block_tree, buffer, offset, len, var)
local rang = buffer(offset, len)
local ti = block_tree:add(fds.var_ipport, rang)
ti:set_text(str_format("%s: %d", var.name, rang:uint()))
end,
}

-- un-register lludp to handle udp port range
local function unregister_udp_port_range(start_port, end_port)
if not start_port or start_port <= 0 or not end_port or end_port <= 0 then
return
end
udp_port_table = DissectorTable.get("udp.port")
for port = start_port,end_port do
udp_port_table:remove(port,lludp_proto)
end
end

-- register lludp to handle udp port range
local function register_udp_port_range(start_port, end_port)
if not start_port or start_port <= 0 or not end_port or end_port <= 0 then
return
end
udp_port_table = DissectorTable.get("udp.port")
for port = start_port,end_port do
udp_port_table:add(port,lludp_proto)
end
end

-- handle preferences changes.
function lludp_proto.init(arg1, arg2)
local old_start, old_end
local new_start, new_end
-- check if preferences have changed.
for pref_name,old_v in pairs(current_settings) do
local new_v = lludp_proto.prefs[pref_name]
if new_v ~= old_v then
if pref_name == "template_file" then
-- load & parse message_template.msg file.
local file = new_v
if file and file:len() > 0 then
local new_details = parse_template(file)
if new_details then
message_details = new_details
end
end
elseif pref_name == "udp_port_start" then
old_start = old_v
new_start = new_v
elseif pref_name == "udp_port_end" then
old_end = old_v
new_end = new_v
end
-- save new value.
current_settings[pref_name] = new_v
end
end
-- un-register old port range
if old_start and old_end then
unregister_udp_port_range(tonumber(old_start), tonumber(old_end))
end
-- register new port range.
if new_start and new_end then
register_udp_port_range(tonumber(new_start), tonumber(new_end))
end
end

-- parse flag bits.
local FLAG_ZER = 4
local FLAG_REL = 3
local FLAG_RES = 2
local FLAG_ACK = 1
local flag_names = {"ACK", "RES", "REL", "ZER"}
local bits_lookup = {
{},
{1},
{2},
{2,1},
{3},
{3,1},
{3,2},
{3,2,1},
{4},
{4,1},
{4,2},
{4,2,1},
{4,3},
{4,3,1},
{4,3,2},
{4,3,2,1},
}
local function parse_flags(flags)
flags = (flags / 16) + 1
local bit_list = bits_lookup[flags]
local bits = {}
local names = ""
for _, bit in ipairs(bit_list) do
bits[bit] = true
if names:len() > 0 then
names = names .. ", "
end
names = names .. flag_names[bit]
end
return bits, names
end

local function grow_buff(buff, size)
local old_size = buff:len()
if old_size > size then return end
-- buffer needs to grow
buff:set_size(size)
-- fill new space with zeros
for i = old_size,size-1 do
buff:set_index(i, 0)
end
end

local function zero_decode(zero_buf)
local out_buf = ByteArray.new()
local zero_off = 0
local zero_len = zero_buf:len()
local out_size = 0
local out_off = 0
local b
-- pre-allocate
grow_buff(out_buf, zero_len)
out_size = zero_len
-- zero expand
repeat
b = zero_buf:get_index(zero_off)
if b == 0 then
-- get zero count
local count = zero_buf:get_index(zero_off + 1)
if count == 0 then count = 255 end
out_off = out_off + count
-- fill zeros
if out_off > out_size then
out_size = out_off + 128
grow_buff(out_buf, out_size)
end
zero_off = zero_off + 2
else
if out_off >= out_size then
out_size = out_off + (zero_len - zero_off) + 4
grow_buff(out_buf, out_size)
end
-- copy non-zero bytes.
out_buf:set_index(out_off,b)
zero_off = zero_off + 1
out_off = out_off + 1
end
until zero_off == zero_len
-- truncate to real size.
out_buf:set_size(out_off)

return out_buf:tvb("Decompressed Data")
end

local function parse_msg_id(buff)
local b = buff:get_index(0)
local msg_id = b
local msg_id_len = 1
if b == 255 then
b = buff:get_index(1)
msg_id = msg_id * 256 + b
msg_id_len = 2
if b == 255 then
b = buff:get_index(2)
msg_id = msg_id * 256 + b
b = buff:get_index(3)
msg_id = msg_id * 256 + b
msg_id_len = 4
end
end
return msg_id, msg_id_len
end

-- get message name.
local function get_msg_name(msg_id)
-- check that we have message details
if message_details == nil then
return str_format("0x%08x", msg_id)
end
-- find message name from id.
local msg = message_details.msgs[msg_id]
-- Invalid message id
if msg == nil then
return str_format("0x%08x", msg_id)
end
return msg.name
end

-- calculate length a block.
local function get_block_length(msg_buffer, start_offset, block)
-- check if bock is fixed length.
if block.fixed_length then
return block.min_length
end
-- parse block's variables to calculate total block length.
local offset = start_offset
local rang
for _,var in ipairs(block) do
local len = 0
if var.has_count then
-- variable with length bytes.
len = var.count_length
--print(var.name, offset, ", len:", len)
rang = msg_buffer(offset, len)
len = len + rang:le_uint()
--print(var.name, var.count_length, ", total:", len)
else
-- fixed length variable
len = var.length
--print(var.name, ", total:", len)
end
offset = offset + len
end
return (offset - start_offset)
end

-- build block tree
local function build_block_tree(msg_buffer, block_tree, start_offset, block)
local offset = start_offset
local rang
-- parse block's variables
for _,var in ipairs(block) do
local len = 0
if var.has_count then
-- variable with length bytes.
len = var.count_length
rang = msg_buffer(offset, len)
len = len + rang:le_uint()
else
-- fixed length variable
len = var.length
end
-- get variable's type field.
local handler = variable_handlers[var.type]
-- parse variable.
if handler then
handler(block_tree, msg_buffer, offset, len, var)
end
offset = offset + len
end
return (offset - start_offset)
end

-- buid message tree
local function build_msg_tree(msg_buffer, msg_tree, msg_id)
local offset = 0
local rang
-- check that we have message details
if message_details == nil then
msg_tree:set_text(str_format("Message Id: 0x%08x", msg_id))
return nil
end
-- find message name from id.
local msg = message_details.msgs[msg_id]
-- Invalid message id
if msg == nil then
msg = str_format("Invalid message id: 0x%08x", msg_id)
msg_tree:add_expert_info(PI_MALFORMED, PI_ERROR, msg)
msg_tree:set_text(msg)
return nil
end
-- skip message id bytes.
offset = msg.id_length
-- set message name.
msg_tree:set_text(msg.name .. ":")
-- proccess message blocks
for _,block in ipairs(msg) do
local count = block.count
if count == nil then
-- parse count byte.
rang = msg_buffer(offset,1)
count = rang:uint()
msg_tree:add(fds.block_count,rang)
offset = offset + 1
end
-- print("block name: ", block.name, count)
for n=1,count do
local block_len = get_block_length(msg_buffer, offset, block)
-- parse block
rang = msg_buffer(offset, block_len)
local block_tree = msg_tree:add(fds.block,rang)
if count > 1 then
block_tree:set_text(str_format("%s: %d of %d",block.name,n,count))
else
block_tree:set_text(block.name)
end
-- parse block variables.
build_block_tree(msg_buffer, block_tree, offset, block)
offset = offset + block_len
end
end
return msg.name
end

-- packet dissector
function lludp_proto.dissector(buffer,pinfo,tree)
local rang,offset
pinfo.cols.protocol = "LLUDP"
local lludp_tree = tree:add(lludp_proto,buffer(),"Linden UDP Protocol")
-- Flags byte.
offset = 0
rang = buffer(offset,1)
local flags = rang:uint()
local flags_bits, flags_list = parse_flags(flags)
flags_tree = lludp_tree:add(fds.flags, rang)
flags_tree:set_text("Flags: " .. str_format('0x%02X (%s)', flags, flags_list))
flags_tree:add(fds.flags_zero, rang)
flags_tree:add(fds.flags_reliable, rang)
flags_tree:add(fds.flags_resent, rang)
flags_tree:add(fds.flags_ack, rang)
offset = offset + 1
-- Sequence number 4 bytes.
rang = buffer(offset,4)
local sequence = rang:uint()
lludp_tree:add(fds.sequence, rang)
offset = offset + 4
-- Extra header length.
rang = buffer(offset,1)
local extra_length = rang:uint()
lludp_tree:add(fds.extra_len,rang)
offset = offset + 1
-- Extra header data.
if extra_length > 0 then
rang = buffer(offset, extra_length)
lludp_tree:add(fds.extra_bytes, rang)
offset = offset + extra_length
end
-- Appended Acks. count
local acks_bytes = 0
local acks_count = 0
if flags_bits[FLAG_ACK] then
rang = buffer(buffer:len() - 1, 1)
acks_count = rang:uint()
acks_bytes = (acks_count * 4) + 1
end
-- Zero Decode
local msg_len = (buffer:len() - acks_bytes) - offset
if flags_bits[FLAG_ZER] then
msg_buffer=zero_decode(buffer(offset,msg_len):bytes())
msg_len = msg_buffer:len()
offset = 0
else
msg_buffer = buffer(offset, msg_len):tvb()
offset = 0
end
-- Message ID
local msg_id, msg_id_len = -1, 4
if msg_id_len > msg_len then
msg_id_len = msg_len
end
msg_id, msg_id_len = parse_msg_id(msg_buffer(offset, msg_id_len):bytes())
rang = msg_buffer(offset, msg_id_len)
lludp_tree:add(fds.msg_id, rang)
local msg_name = get_msg_name(msg_id)
if msg_name == nil then
msg_name = str_format("0x%08x", msg_id)
else
lludp_tree:add(fds.msg_name, rang, msg_name)
end
-- Message body.
rang = msg_buffer(offset, msg_len)
local msg_tree = lludp_tree:add(fds.msg, rang)
build_msg_tree(msg_buffer, msg_tree, msg_id)
-- Appended Acks. list.
if flags_bits[FLAG_ACK] then
local acks_off = buffer:len()
rang = buffer(acks_off - 1, 1)
acks_off = acks_off - acks_bytes
local acks_tree = lludp_tree:add(fds.acks_count, rang)
for i = 1,acks_count do
rang = buffer(acks_off,4)
acks_tree:add(fds.acks, rang)
acks_off = acks_off + 4
end
end
-- Info column
pinfo.cols.info = str_format('[%s] Seq=%u Type=%s', flags_list, sequence, msg_name)
end

-- register lludp to handle udp ports 9000-9003
register_udp_port_range(9000,9003)
</source>

=== File "llmessage.lua" ===
<source lang="lua">
--[[
BSD-Licensed:
Copyright (c) 2008, Robert G. Jakabosky <bobby@sharedrealm.com>
All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

* Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
* Neither the name of the SharedRealm nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

]]

dofile("lexer.lua")

local lexer
local cur_token = nil
local cur_token_str = nil

local function get_token(skip_tokens)
repeat
token = lexer.get_token()
if token ~= nil then
cur_token = token[1]
cur_token_str = token[2]
end
until token == nil or not skip_tokens[cur_token]
return token
end

local function run_parser(parser)
local state = parser.init()
if parser.skip_tokens == nil then parser.skip_tokens = {} end
while get_token(parser.skip_tokens) do
-- check what the parser is expecting next.
if state.expect then
-- check expected type
if state.expect ~= cur_token then
error(string.format("state.expected token '%s' instead of '%s'",
TokenNames[state.expect], TokenNames[cur_token]))
end
-- reset expect field
state.expect = nil
end
if state.expect_str then
-- check expected string
if state.expect_str ~= cur_token_str then
error(string.format("state.expected token '%s' instead of '%s'",
state.expect_str, cur_token_str))
end
-- reset expect_str field
state.expect_str = nil
end
-- get handler function for current token.
local f = parser[cur_token]
if f ~= nil then
local ret = f(state)
if ret then
-- praser finished.
return ret
end
elseif parser.unhandled_error then
error(string.format("unhandled token '%s' when paring '%s'\n", cur_token_str, parser.name))
end
end
return parser.eof(state)
end

-- Known variable types and there fixed length.
-- length == -1, requires a number after the type that is the length of count field
-- length == -2, requires a number after the type that is the fixed variable length.
VariableTypes = {
Null = 0,
Fixed = -2,
Variable = -1,
U8 = 1,
U16 = 2,
U32 = 4,
U64 = 8,
S8 = 1,
S16 = 2,
S32 = 4,
S64 = 8,
F32 = 4,
F64 = 8,
LLVector3 = 12,
LLVector3d = 24,
LLVector4 = 16,
LLQuaternion = 12,
LLUUID = 16,
BOOL = 1,
IPADDR = 4,
IPPORT = 2,
}

--
-- Variable parser
--
local variable_parser = {
name = "variable",
unhandled_error = false,
skip_tokens = {[Token.EOL] = true},
init = function()
return {
name = "<MISSING VARIABLE NAME>",
type = "Null",
has_count = false,
length = 0,
expect = Token.IDENTIFIER,
expect_field = "name",
required = 2
}
end,
[Token.IDENTIFIER] = function(state)
if state.expect_field == "name" then
state.name = cur_token_str
state.expect = Token.IDENTIFIER
state.expect_field = "type"
state.required = state.required - 1
elseif state.expect_field == "type" then
state.type = cur_token_str
state.length = VariableTypes[state.type]
if state.length == nil then
error("Unknown variable type: " .. cur_token_str)
elseif state.length == -1 or state.length == -2 then
state.expect = Token.NUMBER
else
state.required = state.required - 1
end
else
error(string.format("unhandled variable identifier: %s\n",cur_token_str))
end
return nil
end,
[Token.NUMBER] = function(state)
if state.expect_field == "type" then
if state.length == -1 then
-- variable field length uses embedded count field
state.has_count = true
state.count_length = tonumber(cur_token_str)
state.length = nil
elseif state.length == -2 then
-- fixed field length
state.length = tonumber(cur_token_str)
end
state.required = state.required - 1
else
error(string.format("unhandled variable number: %s\n",cur_token_str))
end
return nil
end,
["{"] = function(state)
error("sub block not allowed in variable block")
end,
["}"] = function(state)
if state.required > 0 then
error("missing " .. state.required .. " fields")
end
-- clean state.
state.required = nil
state.expect = nil
state.expect_field = nil
return state
end,
eof = function(state)
error("missing '}' token at end of variable: " .. state.name)
end,
}

-- Block Quantities
local BlockQuantity = {
Single = 1,
Variable = -1,
Multiple = -2,
}

--
-- Block parser
--
local block_parser = {
name = "block",
unhandled_error = false,
skip_tokens = {[Token.EOL] = true},
init = function()
return {
name = "<MISSING BLOCK NAME>",
quantity = "Single",
count = 0,
min_length = 0,
fixed_length = true,
expect = Token.IDENTIFIER,
expect_field = "name",
required = 2
}
end,
[Token.IDENTIFIER] = function(state)
if state.expect_field == "name" then
state.name = cur_token_str
state.expect = Token.IDENTIFIER
state.expect_field = "quantity"
state.required = state.required - 1
elseif state.expect_field == "quantity" then
state.quantity = cur_token_str
state.count = BlockQuantity[cur_token_str]
if state.count == nil then
error("Unknown block quantity: " .. cur_token_str)
elseif state.count == -2 then
state.expect_field = "count"
state.expect = Token.NUMBER
else
if state.count == -1 then
state.has_count = true
state.count_length = 1
state.count = nil
end
state.required = state.required - 1
end
else
error(string.format("unhandled block identifier: %s\n",cur_token_str))
end
return nil
end,
[Token.NUMBER] = function(state)
if state.expect_field == "count" then
state.count = tonumber(cur_token_str)
state.required = state.required - 1
else
error(string.format("unhandled block number: %s\n",cur_token_str))
end
return nil
end,
["{"] = function(state)
local variable = run_parser(variable_parser)
table.insert(state,variable)
-- add length of fixed length variables to minimal length of block.
if variable.has_count then
state.min_length = state.min_length + variable.count_length
state.fixed_length = false
else
state.min_length = state.min_length + variable.length
end
end,
["}"] = function(state)
if state.required > 0 then
error("missing " .. state.required .. " fields")
end
-- clean state.
state.required = nil
state.expect = nil
state.expect_field = nil
return state
end,
eof = function(state)
error("missing '}' token at end of block: " .. state.name)
end,
}

--
-- Message parser
--
local message_parser = {
name = "message",
unhandled_error = false,
skip_tokens = {[Token.EOL] = true},
init = function()
-- create state
return {
name = "<MISSING MESSAGE NAME>",
expect = Token.IDENTIFIER,
expect_field = "name",
fixed_length = true,
min_length = 0,
required = 5
}
end,
[Token.IDENTIFIER] = function(state)
if state.expect_field == "name" then
state.name = cur_token_str
state.expect = Token.IDENTIFIER
state.expect_field = "frequency"
state.required = state.required - 1
elseif state.expect_field == "frequency" then
state.frequency = cur_token_str
state.expect = Token.NUMBER
state.required = state.required - 1
elseif state.expect_field == "trust" then
state.trust = cur_token_str
state.expect = Token.IDENTIFIER
state.expect_field = "compression"
state.required = state.required - 1
elseif state.expect_field == "compression" then
state.compression = cur_token_str
state.required = state.required - 1
else
error(string.format("unhandled message identifier: %s\n",cur_token_str))
end
return nil
end,
[Token.NUMBER] = function(state)
if state.expect_field == "frequency" then
state.number = tonumber(cur_token_str)
state.expect = Token.IDENTIFIER
state.expect_field = "trust"
state.required = state.required - 1
-- create true message id from frequency and message number
local freq = state.frequency
if freq == "High" then
-- High is already correct.
state.id = state.number
state.id_length = 1
elseif freq == "Medium" then
state.id = tonumber("0xFF" .. string.format("%02X", state.number))
state.id_length = 2
elseif freq == "Low" then
state.id = tonumber("0xFFFF" .. string.format("%04X", state.number))
state.id_length = 4
else
-- Fixed is already correct.
state.id = state.number
state.id_length = 4
end
else
error(string.format("unhandled message number: %s\n",cur_token_str))
end
return nil
end,
["{"] = function(state)
local block = run_parser(block_parser)
table.insert(state,block)
-- add min length of block to minimal length of message
local min_length = block.min_length
if block.has_count then
-- add one byte for the block count
min_length = min_length + 1
state.fixed_length = false
else
-- if block is not fixed length then message can't be fixed length.
if not block.fixed_length then
state.fixed_length = false
end
min_length = min_length * block.count
end
state.min_length = state.min_length + min_length
end,
["}"] = function(state)
if state.required > 0 then
error("missing " .. state.required .. " fields")
end
-- clean state.
state.required = nil
state.expect = nil
state.expect_field = nil
return state
end,
eof = function(state)
error("missing '}' token at end of message: " .. state.name)
end,
}

--
-- Template file parser
--
local template_parser = {
name = "message_template",
unhandled_error = false,
init = function()
return {
version = 0,
msg_count = 0,
msgs = {}
}
end,
[Token.IDENTIFIER] = function(state)
-- handle version
if cur_token_str == "version" then
state.expect = Token.NUMBER
state.last_ident = cur_token_str
else
error(string.format("unknown template identifier: %s\n",cur_token_str))
end
return nil
end,
[Token.NUMBER] = function(state)
-- handle version number
if state.last_ident == "version" then
state.last_ident = nil
state.version = tonumber(cur_token_str)
-- check version number
if state.version ~= 2 then
error("invalid verion: " .. state.version)
end
else
error(string.format("unhandled template number: %s\n",cur_token_str))
end
return nil
end,
["{"] = function(state)
local message = run_parser(message_parser)
state.msg_count = state.msg_count + 1
state.msgs[message.id] = message
end,
["}"] = function(state)
error(string.format("unhandled '%s' token",cur_token_str))
end,
eof = function(state)
return state
end,
}

function parse_template(file)
-- create lexer
local status, ret = pcall(get_lexer,file)
if not status then
io.write(string.format("Failed parse file into tokens: %s\n", ret))
return nil
end
lexer = ret
-- parse template file
local status, ret = pcall(run_parser,template_parser)
if not status then
io.write(string.format("Failed parsing on line %d: %s\n", lexer.get_line_number(), ret))
return nil
end
io.write("finished parsing: " .. file .. "\n")
-- return list of messages parsed from file.
return ret
end

--parse_template("message_template.msg")
--print_tokens("message_template.msg")

</source>

=== File "lexer.lua" ===
<source lang="lua">
--[[
BSD-Licensed:
Copyright (c) 2008, Robert G. Jakabosky <bobby@sharedrealm.com>
All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

* Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
* Neither the name of the SharedRealm nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

]]

-- Token types.
Token = {
NONE = -1,
IDENTIFIER = -2,
NUMBER = -3,
COMMENT = -6,
EOL = -7,
["{"] = "{",
["}"] = "}",
}
TokenNames = {
[-1] = "NONE",
[-2] = "IDENTIFIER",
[-3] = "NUMBER",
[-6] = "COMMENT",
[-7] = "EOL",
["{"] = "{",
["}"] = "}",
}

-- parse line into array of tokens.
local function default_parse_tokens(line)
local tokens = {}
local comment = nil
-- check for a comment on this line.
local idx = line:find("//")
if idx ~= nil then
comment = {Token.COMMENT, line:sub(idx)}
-- remove comment from line
line = line:sub(1,idx - 1)
end

-- split line into tokens using white-space as token delimitator
for tok in line:gmatch("%s?([^%s]+)") do
local tok_type = Token.NONE
-- check for number
if tonumber(tok) ~= nil then
tok_type = Token.NUMBER
elseif Token[tok] then
-- token is same as type
tok_type = Token[tok]
else
-- token is an identifier
tok_type = Token.IDENTIFIER
end
table.insert(tokens,{tok_type,tok})
end
-- insert comment token.
if comment ~= nil then
table.insert(tokens,comment)
end
-- add token to mark the end of this line
table.insert(tokens,{Token.EOL, ""})
return tokens
end

function get_lexer(file, parse_tokens)
-- use the default line tokenizer if one is not provided
if parse_tokens == nil then parse_tokens = default_parse_tokens end
-- next/current line code
local line_num = 0
local line = nil
local next_line = io.lines(file)
-- parse line tokens code
local get_next_token = nil
local next_tokens = function ()
local f, tokens, idx
repeat
line_num = line_num + 1
line = next_line()
if line == nil then return nil end
tokens = parse_tokens(line)
until tokens ~= nil
-- create get_next_toekn function from table iterator
f, tokens, idx = ipairs(tokens)
get_next_token = function()
idx, token = f(tokens, idx)
return token
end
return tokens
end

-- get first group of tokens
if next_tokens() == nil then
-- error reading file or empty file
return nil
end
-- build lexer table.
local lexer = {
get_token = function ()
local token
repeat
token = get_next_token()
if token == nil then
-- get next group of tokens
if next_tokens() == nil then
-- end of file.
return nil
end
end
until token ~= nil
return token
end,
get_line_number = function() return line_num end
}
return lexer
end

function print_tokens(file)
local lexer = get_lexer(file)
local num = -1
while true do
local tok = lexer.get_token()
if tok == nil then
break
end
if num ~= lexer.get_line_number() then
num = lexer.get_line_number()
io.write("\n")
io.write(string.format("%d: ",num))
end
io.write(string.format("%s ",tok[2]))
end
io.write("\n")
end

</source>

Office hours

2008-07-22T21:26:25Z

ChrisD: /* Logs */

OpenSim weekly office hours are held in an OpenSim world somewhere on the Internet. The purpose of these office hours is to created a focussed weekly discussion on the state of OpenSim, and things that need attention. It is intended for people wishing to become contributors to the project, so if you show up, you'll probably walk away with something on your todo list.

The office hours are held on Tuesdays at 19:00 UTC. This corresponds to:
* 11:00 PST - San Francisco (U.S.A. - California)
* 14:00 - New York (U.S.A. - New York)
* 19:00 UST - Oslo (Norway)
* 20:00 - Italy
* 02:00 Wed - Perth (Australia) (sorry adam)
* 15:00 AST - Atlantic Canada

There is useful [http://www.timeanddate.com/ Time'n'date] tool for timezone aware planning, with [http://www.timeanddate.com/worldclock/converter.html Time Zone Converter] & [http://www.timeanddate.com/worldclock/meeting.html Meeting Planner] (last thing helps to plan meeting for up 4 tz places)

The office hours will take place at the sim "Wright Plaza" on osgrid, which was called "Sang" until last Friday.

To get to this sim, make sure you have created an account with the normal first, last and password and set the home location to "Wright Plaza". This is done at http://www.osgrid.org.

Then startup the client with a -loginuri of "-loginuri http://osgrid.org:8002" and you should arrive on "Wright Plaza".

=== Previous Agenda Items ===

Cfk would like to suggest that we give some thought to the following bugs.
#) Remoting exceptions with adjacent non-running sims.
#) Prim unlinking.
#) Enabling the script editor from the edit prim dialog box.
#) jpeg/ode dll and glibc-2.4 exceptions and stability.
#) GridServer.exe locks up in r2090 and works fine in r2060 on OSGrid.
#) Enabling avatar 'sit' on a prim.
#) More progress on avatar customization consistency. (Avatar changes when crossing region boundaries).

== Logs ==
* [[Chat log from the meeting on 2008-07-22]]
* [[Chat log from the meeting on 2008-07-15]]
* [[Chat log from the meeting on 2008-07-08]]
* [[Chat log from the meeting on 2008-07-01]]
* [[Chat log from the meeting on 2008-06-17]]
* [[Chat log from the meeting on 2008-06-10]]
* [[Chat log from the meeting on 2008-04-15]]
* [[Chat log from the meeting on 2008-04-08]]
* [[Chat log from the meeting on 2008-03-11]]
* [[Chat log from the meeting on 2008-03-04]]
* [[Chat log from the meeting on 2008-02-17]]
* [[Chat log from the meeting on 2008-02-12]]
* [[Chat log from the meeting on 2008-02-05]]
* [[Chat log from the meeting on 2008-01-29]]
* [[Chat log from the meeting on 2008-01-09]]
* [[Chat log from the meeting on 2007-12-18]]
* [[Chat log from the meeting on 2007-12-11]]
* Chat log from the meeting on 2007-12-4 (meeting cancelled)
* [[Chat log from the meeting on 2007-11-27]]
* Chat log from the meeting on 2007-11-20 (missing)
* [[Chat log from the meeting on 2007-11-13]]
* [[Chat log from the meeting on 2007-11-06]]
* [[Chat log from the meeting on 2007-10-30]]
* [[Chat log from the meeting on 2007-10-23]]
* [[Chat log from the meeting on 2007-10-16]]
* [[Chat log from the meeting on 2007-10-09]]

[[Category:Development]]
[[Category:Office Hours]]

Chat log from the meeting on 2008-07-22

2008-07-22T21:21:57Z

ChrisD:

[10:58] Bill Humphries: Yes.
 [10:58] Justin Clark-Casey: hey shenlei!
 [10:58] Bill Humphries: Hey Dahlia, Charles.
 [10:58] Justin Clark-Casey: hi dahlia, adam, bill, samuel
 [10:58] Dahlia Trimble: Hi :)
 [10:58] Justin Clark-Casey: nice to see you Charles
 [10:58] Adam Frisby: Hey guys
 [10:59] Dahlia Trimble quickly does a vanity rebake
 [10:59] You: morning.
 [10:59] You: Nice dress, Shenlei
 [10:59] Shenlei Winkler blinks....
 [11:00] Shenlei Winkler: hi guys
 [11:00] Shenlei Winkler: Thanks Charles
 [11:00] Dahlia Trimble: Hi :)
 [11:00] You: you guys and gals are shaming me out of my "ruth state"
 [11:00] Shenlei Winkler: Hi Justin, it's nice to see you again
 [11:00] Strawberry Fride: hi everyone :)
 [11:00] Shenlei Winkler grins ruthlessly
 [11:00] Shenlei Winkler: good
 [11:00] Dahlia Trimble: lol
 [11:00] Justin Clark-Casey: well, me in my poor grey/black ruthed state
 [11:00] Bill Humphries: Ah, rebaked.
 [11:00] Justin Clark-Casey: graphics card issues
 [11:00] Shenlei Winkler: I lost some of my hair in inventory somewhere. I'm sulking.
 [11:00] Justin Clark-Casey: hey dahlia, strawberry
 [11:00] You: being a nerd, I just barely get my pants and shirt on in the morning.
 [11:01] Strawberry Fride: most of you haven't rezzed in any form for me yet - getting disembodied voices :)
 [11:01] Shenlei Winkler: they make GR animals for that you know
 [11:01] Justin Clark-Casey: interesting
 [11:01] Dahlia Trimble: I cant see Shenlei's dress, she looks like a cloud :(
 [11:01] Shenlei Winkler: Match the gorilla to the gorilla
 [11:01] Justin Clark-Casey: I can see her dress, though there's a lot of grey....
 [11:01] Shenlei Winkler: I'm moving content over...or trying
 [11:01] Bill Humphries: She's wearing the mother of all crinolines with that dress.
 [11:01] Charles Krinkeb: Shall we start or wait another minute?
 [11:02] Shenlei Winkler: Ballgowns are the epitome of virtual worlds, Bill
 [11:02] Bill Humphries: No argument there.
 [11:02] Charles Krinkeb: Maybe I need a ball gown sometime so I can keep up.
 [11:02] Justin Clark-Casey: interesting, my graphics card actually appears to be playing ball today
 [11:02] Strawberry Fride: can still only see 4 avatars - though plenty more dots on minimap
 [11:03] Justin Clark-Casey: yes, lots more people than that here sf
 [11:03] Strawberry Fride: I guessed as much :)
 [11:03] Bill Humphries: Well, at some point, maybe Hya will open a branch out here.
 [11:03] Dahlia Trimble: everyone is grey but Bluewall and Strawberry
 [11:03] Dahlia Trimble: and Bill
 [11:03] Justin Clark-Casey: yeah, we do seem to have avatar appearance issues still.....
 [11:04] Strawberry Fride: I can see Dahlia (with clothes), Chris D (also with clothes) and a grey Monk and a grey Adam :)
 [11:04] Charles Krinkeb: I see lots of "AgentHeightWidth" and "MapItemRequest" packets so the sim is loading a bit.
 [11:04] Bill Humphries: maybe hitting rebake?
 [11:04] Shenlei Winkler: Persistence is getting better
 [11:04] Chris D was worried he might be naked
 [11:04] Charles Krinkeb: Adam? Are you here or a zombie?
 [11:04] Shenlei Winkler: I teleported today and did not lose my hair
 [11:04] Strawberry Fride: my hair on my sim is remarkably resilient
 [11:05] Charles Krinkeb: great, Shenlei
 [11:05] Strawberry Fride: especially since we added that missing method to SQL Server. No wonder it wasn't working so well :)
 [11:05] Strawberry Fride: nothing in the Agents table at all before Sunday
 [11:05] Charles Krinkeb: Lets try to get started. I think the first item of business is to chat with our LL colleagues and discuss interop issues.
 [11:05] Justin Clark-Casey: mssql tends to lag in opensim
 [11:05] Justin Clark-Casey: strawberry: good stuff
 [11:06] Justin Clark-Casey: hey mic
 [11:06] Strawberry Fride: mssql not so much any more - whizzy these days
 [11:06] Justin Clark-Casey: hi neas
 [11:06] Justin Clark-Casey: I mean in terms of functionality
 [11:06] Bill Humphries: okay, Tess and Greg are coming, if you have another item you can do first
 [11:06] Strawberry Fride: :)
 [11:06] Charles Krinkeb: Morning, Neas, Mic.
 [11:06] Mic Bowman: morning all
 [11:07] Neas Bade: hey all
 [11:07] Monk Zymurgy: g'morning
 [11:07] Charles Krinkeb: Ready to get started?
 [11:07] Neas Bade: charles, can you turn on build for me?
 [11:07] Neas Bade: I want to generate the sit bug test case
 [11:08] Adam Frisby: That was weeird
 [11:08] Adam Frisby: Anyway
 [11:09] Adam Frisby: I have a slight announcement for these hours - can someone add a note on the agenda for me?
 [11:09] Charles Krinkeb: Neas. You are an estate manager just like me and I just rezzed a box.
 [11:09] Dahlia Trimble: slight?
 [11:09] Neas Bade: yeh, no dice
 [11:09] Adam Frisby: Oh, I totally need estate manager privilidges. >_>
 [11:09] Strawberry Fride: oo - now I see people! flew into adjacent region and back :) Shenlei - fab outfit!
 [11:10] Neas Bade: oh, never mind
 [11:10] Shenlei Winkler grins
 [11:10] Neas Bade: my build button is just greyed out
 [11:10] Shenlei Winkler: Thanks
 [11:10] Neas Bade: ok, I'm going to go build this a little ways away
 [11:10] Mic Bowman: brb
 [11:10] Neas Bade: back shortly
 [11:10] Charles Krinkeb: Adam, Neas. The estate stuff appears to be a bit fubar lately.
 [11:11] Zha Ewry: Hah. Shen is a ruth to me
 [11:11] Zha Ewry: What a fate
 [11:11] Shenlei Winkler: Darling, I can haz hair
 [11:11] Shenlei Winkler smiles
 [11:11] Adam Frisby: Hehe
 [11:11] Zha Ewry: No
 [11:11] Zha Ewry: You can haz particle cloud
 [11:11] Strawberry Fride: lol
 [11:11] Charles Krinkeb: Lets get going. Sorry about the estate stuff, Neas. I'll have Nebadon look at it a little later. I dont want to loose our hour.
 [11:11] G2 Proto: :) hello all
 [11:11] Shenlei Winkler: well, there are worse things...+
 [11:11] Primitive: Script running
 [11:12] Charles Krinkeb: Lets let Adam go first. Adam, you have the floor.
 [11:12] Adam Frisby: Alright, I have shinies.
 [11:12] Adam Frisby: Zha, you're going to appreciate this.
 [11:12] Dahlia Trimble: shinies? o.O
 [11:12] Charles Krinkeb: Somebody hit you in the eye?
 [11:12] Adam Frisby: Myself and Mike just finished setting up forge.opensimulator.org -- it's a install of GForge, which is Source-forge like software. It's for people who are working on opensim related projects, to get hosting and SVN access for it
 [11:13] Adam Frisby: Things like say the Perl Gridservers, the ASP.NET gridservers, Zha's interop patches, etc can all be hosted there
 [11:13] Adam Frisby: We can make projects, you get mailing lists, bug trackers, etc all included plus SVN hosting (looks like we have Hg and Git too)
 [11:13] BlueWall Slade: awesome!
 [11:13] Hiro Protagonist: +1 Adam
 [11:13] Hiro Protagonist: Hi All :D
 [11:13] Zha Ewry: Cool
 [11:13] Zha Ewry: Better still, when the pacth isn't ugly as sin
 [11:13] Hiro Protagonist: we need to get Jeroen's 'moo' on there too
 [11:14] Charles Krinkeb: How do we handle keeping the patches in sync with the opensim svn?
 [11:14] Dahlia Trimble: good question Charles
 [11:14] Justin Clark-Casey: adam: nice
 [11:14] Adam Frisby: Well, doing full opensim clones probably isnt the intent of hosting it there, but I imagine just producing .patches based on SVN revisions might be the way to go.
 [11:14] Hiro Protagonist: Hiya WHump :)
 [11:14] Charles Krinkeb: Hiro. Can you figure out how to set Neas so he can build and add Adam to estate managers. I cant tell if it is broken or not.
 [11:14] Adam Frisby: In any case, it should help - I'm going to throw up a lot of the internal DTL code onto it so you can play with some of our internal stuff. Anyone else is free to register projects, etc.
 [11:15] Adam Frisby: We've got a 15 user limit right now, but that will be removed in a day or two when we get the proper OSS license reg'd.
 [11:15] Hiro Protagonist: sure thing Charles one sec
 [11:15] Bill Humphries: Sorry about that, I lost my presense.
 [11:15] Charles Krinkeb: No problem, Bill. Anything else, Adam? Any questions for Adam?
 [11:15] Justin Clark-Casey: didn't appear to be directly a server issue
 [11:15] G2 Proto: its very cool thanks Adam
 [11:15] Bill Humphries: What was the topic?
 [11:16] Hiro Protagonist: Ckrinke, need to relog as plaza builder
 [11:16] Charles Krinkeb: [11:12] Adam Frisby: Myself and Mike just finished setting up forge.opensimulator.org -- it's a install of GForge, which is Source-forge like software. It's for people who are working on opensim related projects, to get hosting and SVN access for it
 [11:16] Tao Takashi: somehow I seem to be underwater and I cannot see myself ...
 [11:16] Charles Krinkeb: k, Hiro
 [11:16] Hiro Protagonist: also the estate managers list for this region is impacted; I'll have to purge a few non-participants
 [11:16] Charles Krinkeb: [11:13] Adam Frisby: Things like say the Perl Gridservers, the ASP.NET gridservers, Zha's interop patches, etc can all be hosted there
 [11:16] Dahlia Trimble: Thanks Adam :)
 [11:16] Justin Clark-Casey: we still have these login issues I see...
 [11:16] Charles Krinkeb: Anything else, Adam? Any questions for Adam?
 [11:17] Bill Humphries: okay, thanks for catching me back up
 [11:17] Justin Clark-Casey: adam: are there any criteria for getting hosting?
 [11:17] Neas Bade: btw, this is the log off doom. It has funny effects if people log in later as to what you look like. Fun bugs in the sit code
 [11:17] Neas Bade: btw, Adam, +1 for setting up gforge
 [11:17] Justin Clark-Casey: apart from that you have code related to opensim?
 [11:18] Justin Clark-Casey: don't log off is you're on the log of doom?
 [11:18] Charles Krinkeb: Lets give Bill the floor and Adam can chime in if there is more, or anyone can ask questions about anything we previously talked about.
 [11:18] Adam Frisby: Justin: no criteria other than it's somewhat opensim related.
 [11:18] Adam Frisby: Projects have to be approved by a site admin (opensim devs)
 [11:18] Justin Clark-Casey: cool :)
 [11:18] Neas Bade: Adam, I thought there was a free version of gforge, not just the commercial version. Or did they stop doing that?
 [11:19] Plaza Builder: Ok Neas
 [11:19] Adam Frisby: Neas: The free version is crippled.
 [11:19] Adam Frisby: Majorly, so we're applying for the full F/OSS one.
 [11:19] Plaza Builder: if you select 'Advanced->View Admin Options' it should enable your estate powers
 [11:20] Dahlia Trimble: who is Plaza Builder today?
 [11:20] Plaza Builder: Hiro, but only temproarily
 [11:20] Neas Bade: Plaza, no, it's working for me
 [11:20] Plaza Builder: cool
 [11:20] Neas Bade: I didn't realize the build button would stay greyed out
 [11:20] Neas Bade: that's something we should change
 [11:20] Plaza Builder: Ahhhh
 [11:20] Neas Bade: but it's a minor bug
 [11:20] Plaza Builder: yup
 [11:20] Neas Bade: sorry, my confusion
 [11:20] Plaza Builder: Charles, what all was it you asked me to do? Neas, and...
 [11:21] Bill Humphries: Okay, Tess and Zha are here, so I can go ahead if you all are ready.
 [11:21] Charles Krinkeb: Adam as estate manager. Gryc might have to go off the list.
 [11:21] Charles Krinkeb: Please go ahead, Bill.
 [11:21] Plaza Builder: Gryc is off, as are Babble and Pablo
 [11:21] Plaza Builder: they can be readded later if need be
 [11:21] Charles Krinkeb: k, Hiro
 [11:21] Bill Humphries: Okay, so next week, the 31st, we're kicking off the Open Grid Protocol Beta Program.
 [11:21] Adam Frisby: Charles: we should be able to make it so there's multiple admins.
 [11:22] Adam Frisby: (FYI: Mike just got the F/OSS license then, he's installing now)
 [11:22] Tao Takashi: now this looks better. Hello :)
 [11:22] Justin Clark-Casey: Hi Tao
 [11:22] Bill Humphries: As you may know, Zha Ewry developed a patch for OpenSim that enables the protocols for teleport.
 [11:22] G2 Proto: hey Tao
 [11:22] Plaza Builder: Ok Adam you are set sir
 [11:22] Plaza Builder: brb w/Hiro
 [11:22] Zha Ewry: FFor some epsilon, around the phrase enables
 [11:23] Bill Humphries: At the moment, we need to do a little more work to get Zha's changes in sync with our agent domain.
 [11:23] Tao Takashi: I should have tried to login to here earlier to be here in time ...
 [11:23] Dahlia Trimble: is the current interop patch on mantis up to date?
 [11:23] Zha Ewry: I should have a new on to offer in a day or so
 [11:24] Zha Ewry: Among other things, catching up to the current Linden Beta grid
 [11:24] Zha Ewry: and.. somewhat better integrated, less horrible
 [11:24] Bill Humphries: But what I'm hoping to come out of here with is a plan for people participating in the beta to either be able to build OpenSim out of top of trunk with OGP support, or a well-defined way to get it patched with OGP support.
 [11:24] G2 Proto: great!
 [11:25] Justin Clark-Casey: Zha, is it still the case that applying the patch disrupts the current region login ops?
 [11:25] Hiro Protagonist: lol
 [11:25] Adam Frisby: >_> <_<
 [11:25] Hiro Protagonist: Adam
 [11:25] Strawberry Fride: :)
 [11:25] Zha Ewry: Less so, but yes
 [11:25] Charles Krinkeb: <harumph>
 [11:25] Zha Ewry: Mostly in the handles
 [11:25] Bill Humphries: I saw the blue card, LOL.
 [11:25] Dahlia Trimble: lol
 [11:25] Zha Ewry: Once we sort out the current prtocol update, I want to look at that
 [11:26] Justin Clark-Casey: I mean, the best thing right now might be just for people to apply the patch
 [11:26] Zha Ewry: I think, and this is problematic, for the longer term story
 [11:26] Justin Clark-Casey: or for some kind soul to supply a build with it pre-applied
 [11:26] Bill Humphries: So that would be a blocker, then, alright.
 [11:26] Hiro Protagonist: Justin, you are in color mate ;)
 [11:26] Dahlia Trimble: I have a build with it applied
 [11:26] Tao Takashi: btw, it would be great if the protocol docs would be updated or discussion about changes would go about some mailing list so I can keep up with my agent domain implementation. not sure if that's possible
 [11:26] Zha Ewry: that you won't be able to login, via the normal path, and teleport, because you won't be on the agent domain correctly
 [11:26] Charles Krinkeb: So, Bill. Is your understanding that the interop is teleporting to an OpenSim standalone region from the betagrid and not to a region attached to OSGrid?
 [11:26] Justin Clark-Casey: hiro: I know - it's wierd - there's nothing different on my machine :)
 [11:27] Bill Humphries: Charles, at the moment, yes. But I think teleporting to a grid enabled sim would be a great next step in the beta.
 [11:28] Hiro Protagonist: hrmmm
 [11:28] Zha Ewry: I clearly want to get this synced with the full grid version
 [11:28] Hiro Protagonist: I guess I'll wait for phase 2 then
 [11:28] Zha Ewry: Part of that is going to be sorting out handles
 [11:28] Hiro Protagonist: I dont have the resources to set up a standalone
 [11:28] Neas Bade: so, that means proper agent domain handoff, right?
 [11:28] Zha Ewry: (my one attempt to patch that way, was a nightmare)
 [11:28] Neas Bade: because I don't think we want osgrid accounts to be dependant on SL accounts
 [11:29] Bill Humphries: Neas, this is just a beta.
 [11:29] Neas Bade: most people running grids want seperate account controls
 [11:29] Neas Bade: Bill, yeh, I understand that :)
 [11:29] Tao Takashi: I would more call it alpha ;-)
 [11:29] Justin Clark-Casey: agent domain handoff is going to be considerably more complicated I imagine
 [11:29] Neas Bade: but if we're trying for real grid interop, it seems like doing it early would be a good thing
 [11:29] Charles Krinkeb: For, now it is fair to say that a common username could be used so we dont have to open the openid/liveid/authorization bit yet.
 [11:29] Zha Ewry: I think there is a very legitimate concern, in not handling the AD list
 [11:30] Zha Ewry: (bit)
 [11:30] Zha Ewry: too many SL style meetings, my typoing dies
 [11:30] Neas Bade: :)
 [11:30] Zha Ewry: There are two, issues, I think, one is the agent domain code, the other, woudl be the client handling the hand off
 [11:31] Charles Krinkeb: Well, I dont want to derail any of the efforts, so I'll stand in the background a bit, and OSGrid is happy to support this effort when appropriate for LL's plans.
 [11:32] Charles Krinkeb: Although one has to understand that the notion of "interop" implies grid<-->grid, both betagrid<->OpenSimGrid *and* OpenSimGrid<->OpenSimGrid
 [11:32] Bill Humphries: Okay, so for the short term, if we could host the patch on adam's gforge, until we can work through the issues with merging that would be good. And I can see if we can provide a precompiled mono .exe as well.
 [11:33] Adam Frisby: Go ahead - if you make the project right now, I can approve it.
 [11:33] Dahlia Trimble: couldnt interop also include grid<->standalone?
 [11:33] Adam Frisby: (I'll also add the other opensim admins to the site admins if they care to register accounts. hehe.)
 [11:33] Tao Takashi: I hope the patch will then also be published
 [11:33] Neas Bade: Dahlia, I think so
 [11:33] Charles Krinkeb: We can run a mono .exe on the OSGrid gridserver as long as both LL and the core developers agree.
 [11:33] Tess Linden: Interop means that the code implements against the protocol, and any code taht runs against the protocol should interoperate with each other
 [11:33] Bill Humphries: Yes, and that's where, given the time left before the beta starts, where we'll start from.
 [11:33] Digi Fly: hi
 [11:34] Hiro Protagonist: I think it should also support standalone as well; however, it's important to note, that there is almost no representation of that use case on osgrid
 [11:34] Zha Ewry: Well, baring opermisions and all
 [11:34] Justin Clark-Casey: tao: you mean the zha's existing work? that is already out there
 [11:34] Neas Bade: the issue on the opensim side is that we've got to address moving between standalone and grid instances relatively shortly, especially because there is no URL bar in the client
 [11:34] Justin Clark-Casey: hiro: you mean local assets?
 [11:34] Tess Linden: we've got code thats working against a protocol that will allow you to TP between virtual worlds, we're talking about making it easy to iterate on
 [11:34] Bill Humphries: A patch that will enable OGP on a standalone grid. With the caveats about no inventory, no baked textures coming over.
 [11:34] Neas Bade: and if there is no ability to do agent domain handoff, then that is a real issue
 [11:34] Charles Krinkeb: Ok. This may be an important point then. Are you saying that interop has nothing to do with moving from grid to grid, or sim to sim, but merely that certain code communicates with each other in two internet seperated sites?
 [11:34] Tess Linden: Neas: Suzy put in a url bar in the client before she left for vacation
 [11:34] Hiro Protagonist: Justin: I mean, coming to osgrid to look for people running standalone is, well, a little misdirected
 [11:35] Tess Linden: it remembers the last 10 region_urls youve typed in
 [11:35] Tao Takashi: Justin: I also mean the changes which are happening now
 [11:35] Justin Clark-Casey: hiro: true :)
 [11:35] Neas Bade: Tess, ok, so when does that release in a GA browser? :P)
 [11:35] Zha Ewry: Interop is a series of steps, and teleport is only one part
 [11:35] Tao Takashi: I have Zha's patch running here on a standalone server
 [11:35] Neas Bade: oops sorr for the P, I just meant a smilely
 [11:35] Charles Krinkeb: k. I best start correcting a mis-conception that exists in our users.
 [11:35] Neas Bade: Tess, those urls include grid?
 [11:36] Zha Ewry: We need to make sure it's all the places where we ahve urls
 [11:36] Tess Linden: region_urls identify the region you're rezzing into, could be on any grid
 [11:36] Zha Ewry: including AD, Region, and login
 [11:36] Tao Takashi: and it would be nicer of course to have a branch instead of a patch which you have to apply
 [11:36] paulie Flomar waves
 [11:36] Zha Ewry: We really want to make it easy to hop to any of the permuations
 [11:36] Zha Ewry: and. ideally, without a logout/login
 [11:36] Dahlia Trimble: Hi Paulie
 [11:36] Tess Linden: sorry CHarles, whats the misconception?
 [11:36] paulie Flomar: Howdeh, everyone. :)
 [11:37] BlueWall Slade: stand alone seems a good, safe first step
 [11:37] Neas Bade: ok, so when do we get to see that in a release client? As it would be really nice to not need to have all these icon links for different OS instances
 [11:37] paulie Flomar: thx, d.
 [11:37] Digi Fly: stand alone still the required firts step.
 [11:37] Justin Clark-Casey: hey paulie
 [11:37] Tess Linden: thats what the Beta is for, so we can all try it out and work out all the kinks to a ready state
 [11:37] Tao Takashi: will this beta client also have some cmdline option to change the agent domain?
 [11:38] paulie Flomar: Howdeh, JCC.
 [11:38] Neas Bade: right, but the beta client won't do old auth, right?
 [11:38] Charles Krinkeb: The prevalent misconception is that the interop implies a full handoff from betagrid<->OpenSimGrid, both ways, in an arms-length manner. I may have to soften the perceptions in the users a bit.
 [11:38] Zha Ewry: I think its in the branch
 [11:38] Dahlia Trimble: maybe we need to develop some criteria for including the interop patch into core?
 [11:38] Zha Ewry: (the various switches, to select all the bits, if not, it needs to be)
 [11:38] Tao Takashi: interop4 I guess
 [11:38] Tess Linden: +1 Dahlia
 [11:39] Charles Krinkeb: No problem. I just dont want the users to assume it is further ahead then it is. That adds additional pressure on everyone.
 [11:39] Zha Ewry: I think handles, grid , whitelisting, good controls, and cleaner code, would begin to do it
 [11:39] Charles Krinkeb: +1, Zha. Thats about where my thinking starts.
 [11:39] Bill Humphries: Zha by grid you mean teleporting into a sim in a grid, or a handoff?
 [11:39] Tess Linden: Charles: its not a full handoff because your identity lives with the agent domain, and the agent domain is the second life one, but the opensim one doesn't implement against the protocol yet
 [11:39] Zha Ewry: Being able to run in grid mode
 [11:40] Bill Humphries: Zha: +1
 [11:40] Zha Ewry: Full handoff, I think is another good, but, somewhat seperable
 [11:40] Zha Ewry: issue
 [11:40] Justin Clark-Casey: I think we're really aiming for modules for the protocol stuff if possible
 [11:40] Tess Linden: do we need all those things for the Beta?
 [11:40] Zha Ewry: We need to be able to do it, so we can have both the various unified and seperable cases
 [11:40] Zha Ewry: I think we need all of those to exit beta ;-)
 [11:41] TARDIS Wiki-Lith: No wikipedia URL found
 [11:41] Tao Takashi: these are the times of perpetual betas though
 [11:41] Tess Linden: oh yes, there is so much to do during the Beta, I don't think anybody is arguing that
 [11:41] Zha Ewry: I personaly have my own list of horrible things inside my code to fix too
 [11:42] Charles Krinkeb: Its ok. One step at a time is just fine. We will move forward faster that way. I just need to make sure the perceptions are not out of line with reality over the next few months.
 [11:42] Zha Ewry: (there's also some discussion we need to have about how we signal how the grids cooperate, so w eknow if we do/don't need a AD handoff
 [11:42] Tess Linden: but am I hearing that we don't want to apply the patch to trunk until the Beta is done and we're actually releasing the code?
 [11:42] Dahlia Trimble: I would hope sooner Tess
 [11:42] Bill Humphries: okay, so I'd propose the plan would be to start a branch with standalone OGP support and add the features Zha mentioned above to that branch.
 [11:42] Zha Ewry: When the OpenSim core, thinks its appropriate
 [11:42] Justin Clark-Casey: there really is a philosphical question as to whether it would ever be applied directly to trunk
 [11:42] Justin Clark-Casey: we may well want to keep protocol code in separate modules
 [11:43] Neas Bade: no, I think you are hearing that AD handoff and having grids be able to connect but not be in the same AD is really important
 [11:43] Tao Takashi: Justin: wouldn't this be more a question of refactoring the code so these things can be plugins?
 [11:43] Justin Clark-Casey: Tao: yep, that's certainly a factor
 [11:43] Zha Ewry: A bit of both, I think tao
 [11:43] Neas Bade: and something short of that, in which OpenSim is largely just extra land is a lot less interesting to a lot of opensim folks
 [11:44] Dahlia Trimble: so we may want to work towards putting the hooks into core so interop can work as a module?
 [11:44] TARDIS Wiki-Lith: Ask a question with '???' to get a Wikipedia answer
(maybe). Touch me to get the Wikipedia article used to provide the answer.
 [11:44] Charles Krinkeb: We also have the issue of OpenSimGrid<->OpenSimGrid interop, so there may be more then one implementation of a similar notion.
 [11:44] TARDIS Wiki-Lith: Ask a question with '???' to get a Wikipedia answer
(maybe). Touch me to get the Wikipedia article used to provide the answer.
 [11:44] Justin Clark-Casey: in which case those plugins may be bundled with a separate OpenSim core kernel rather than being part of that core kernel itself
 [11:44] Justin Clark-Casey: Dahlia: +1
 [11:44] Charles Krinkeb: stop typing three question marks
 [11:44] TARDIS Wiki-Lith: Ask a question with '???' to get a Wikipedia answer

(maybe). Touch me to get the Wikipedia article used to provide the answer.
 [11:44] Dahlia Trimble: lol
 [11:45] Hiro Protagonist: I think kinic is actually working on the wikilith
 [11:45] Hiro Protagonist: *knio
 [11:45] Zha Ewry: I'd be glad to work with anyoen who wants to
 [11:45] Hiro Protagonist: pffft
 [11:45] Hiro Protagonist: KINOC
 [11:45] Hiro Protagonist: lol
 [11:45] Zha Ewry: dive into the modularity bits
 [11:45] Tao Takashi: having it on a branch should definitiely make it easier to collaborate
 [11:45] paulie Flomar: It was me playin with the wikilith. I'll cut it out. :)
 [11:45] Daxxon Kinoc: not i
 [11:46] Hiro Protagonist: the thing that concerns me about a branch is that the code there will rapidly go stale
 [11:46] Charles Krinkeb: Thanks for clarifying, Zha, Tess and Bill. I was operating under the misconception that this was a full handoff.
 [11:46] Neas Bade: Tao, we shot down the branch idea in core. The opensim code moves too fast and no one really wants to maintain a branch.
 [11:46] Dahlia Trimble: I dont see how a branch would have any advantages over a patch set
 [11:46] Zha Ewry: So far, I've been able to keep on branch wihtout too much pain
 [11:46] Neas Bade: branches in svn is the evil
 [11:46] Zha Ewry: and. I'm willing to comit to it
 [11:46] Zha Ewry: i mean trunk
 [11:46] Justin Clark-Casey: depends if other people also want to work on that code I think
 [11:47] Adam Frisby: Zha: well there's an advantage there to having the seperate project on the forge then - since you can maintain that and release your in-dev code
 [11:47] Zha Ewry: I've got the stuff in placet o build and test daily
 [11:47] Justin Clark-Casey: patches aren't very scalable
 [11:47] Zha Ewry: One thing I'd think we can do
 [11:47] Zha Ewry: is break it into two parts
 [11:47] Zha Ewry: the hooks
 [11:47] Zha Ewry: and the core functoin
 [11:47] Zha Ewry: at which point, we can get 90% of it in trunk
 [11:47] Dahlia Trimble: that would help
 [11:47] Tao Takashi: well, I usually prefer branches to sending patches forth and back.
 [11:47] Zha Ewry: and only have the patch out there
 [11:47] Justin Clark-Casey: zha: that would be cool
 [11:47] Hiro Protagonist: one thing that concerns me is taht I know that a lot of the packet and stack code is being worked over as we speak
 [11:48] Zha Ewry: I don't touch that
 [11:48] Bill Humphries: and the core function becomes a module, Zha?
 [11:48] Hiro Protagonist: it may well be that the very parts you're depending on being stable will be moving the most quickly
 [11:48] Tao Takashi: If Zha works alone on it then ok, but if somebody should join I would see some problem.
 [11:48] Justin Clark-Casey: hiro: I think that'slargely orthogonal Hiro
 [11:48] Zha Ewry: This is 90%, in parsing/responsinf to the AD
 [11:48] Hiro Protagonist bows to more knowlegable heads
 [11:48] Hiro Protagonist: just makin' sure folks
 [11:48] Zha Ewry: and.. calling back into the guts of the OpenSim world
 [11:49] Zha Ewry: (and I'm trying to focus on keepiing it seperate that way)
 [11:50] Zha Ewry: (There is some really tricky re-facoring that might imply, at some point, but it would be nice to havbe nayway, seperate out the login, fromt he state updates behind it)
 [11:50] Charles Krinkeb: Adam. I would be most interested in your opinion and guidance.
 [11:50] Justin Clark-Casey: Zha: I think we're using you as cheap labour to improve our modularity ;)
 [11:50] Adam Frisby: Haha
 [11:51] Dahlia Trimble: lol
 [11:51] Bill Humphries: okay, I think I'm back.
 [11:51] Zha Ewry: That's fine with me
 [11:51] Dahlia Trimble: wb :)
 [11:51] Adam Frisby: Well, I've talked with Zha about this in person - but I think Zha's onto the right track with having some seperate patches for improving OpenSim's modularity, and the interop functionality.
 [11:51] Zha Ewry: When I get the core patch done, I'm willign to sort out other thignds I've noticed. I don't miind making thing bvetter
 [11:51] Charles Krinkeb: Oh Great Existential Wizard, Adam. What would you counsel us on the best way to move forward on the interop patch with maximum happiness and minimal frustration to all.
 [11:52] Zha Ewry: That involves good beer and decent fajitas, I think.
 [11:52] Adam Frisby: Dont look at me, I'm usually the one breaking things. >_>
 [11:52] Tess Linden: Neas I understand your concern that the patch to OpenSim doesn't create an opensim agent domain, but just because we've added coded to make the region domain work against the protcol doesn't prevent anybody else writing code to add an agent domain that works against the protocol too
 [11:52] G2 Proto: lol
 [11:52] Bill Humphries: So we have an offsite in Austin where we can get good beer and fajita's to has this out? :)
 [11:52] Adam Frisby: Tess: I think that does actually raise a valid point here -- is there a OSS project to build a agent domain?
 [11:53] Adam Frisby: A reference one that is
 [11:53] Neas Bade: no, but it prevents the way we can do grid jump today with just the region <-> client protocol
 [11:53] Adam Frisby: If not - that strikes me as a major deficiency that all parties would be best trying to fix.
 [11:53] Neas Bade: which would be a step backwards, and something that would be good not to loose
 [11:53] Justin Clark-Casey: neas: can we really do that today though? The raw jump might be possible but we would hjave asset and inventory probs, potentially
 [11:53] Charles Krinkeb: I have RL intruding and must be passive. Is someone going to get the chat log on the wiki.
 [11:54] Neas Bade: if you are changing domains of control, you are leaving that behind
 [11:54] Neas Bade: but that's ok
 [11:54] Zha Ewry: I think, looking intot he code, that, the current OpenSim code gets pretty close to being an easy basis for an agent domain, in some of the UGAS code
 [11:54] Neas Bade: it's like moving from myspace -> facebook
 [11:54] Justin Clark-Casey: the trouble is I don't think you can tell the client to suddenly reload its inventory
 [11:54] Tess Linden: that just tells me that there isn't a big push for it. We've been working hard to meet these deadlines to show that interoperability works, and its amazing, but people are pushing back because they want more? why not take steps at a time and work together?
 [11:55] Zha Ewry: I think the client, often gets in the way of this
 [11:55] Justin Clark-Casey: I think one issue is that we want really to be protocol agnostic
 [11:55] Zha Ewry: because it's so hard to do simple hacks, on it end, which would make it easier for people to test and flow between parts of the developing envrionemnt
 [11:56] Justin Clark-Casey: so potentially an opensim server could be on some protocol other than ogp, when/if there is one
 [11:56] Tess Linden: if we can get the code patched to trunk then we can all be working on the same code base to improve interoperability, and we can discuss details at this office hours, at the AWG office hour, or Zero's
 [11:56] Justin Clark-Casey: and opensim itself wasn't really meant ot be a full vw implementation, really just kernel of a general vw server
 [11:56] Charles Krinkeb: I need to go. Sorry folks. Please carry on without me.
 [11:57] Bill Humphries: Okay, thanks Charles.
 [11:57] Hiro Protagonist: See ya soo Charles
 [11:57] Dahlia Trimble: tc Charles :)
 [11:57] Neas Bade: later charles
 [11:57] Justin Clark-Casey: bye charles
 [11:57] Charles Krinkeb: My heart is with this, but I have to pretend to work.
 [11:57] Dahlia Trimble: lol
 [11:57] Daxxon Kinoc: happy RL'ing ...
 [11:57] Charles Krinkeb: <sigh>
 [11:57] Adam Frisby: Zha: just quickly heading back 5 minutes -
 [11:58] Zha Ewry: yes?
 [11:58] Adam Frisby: Adding AD support to the opensim userserver, what needs to be done?
 [11:58] Adam Frisby: Is there some specs someone could spend a couple of days nailing out?
 [11:58] Zha Ewry: I htink, and this was a quick glance
 [11:58] Adam Frisby: And do we have a test scenario we can use?
 [11:58] Zha Ewry: you'd need to be willign to handle about four calls
 [11:58] Zha Ewry: and.. Tao, ought to be able to drive the test cases
 [11:59] Zha Ewry: Bascially, you need to tyake the place_avatar, fromt he client, in the modded form, and pass it on to the regions, as a rez or rez/de-rez pair
 [11:59] Bill Humphries: Yes, Tao, Infinity and Lochlainn have been working on a test harness as well.
 [11:59] Tess Linden: Justin: we've iterated on the OGP protocol because we think this is the common language that will define interoperability between virtual worlds. We should discuss more if there are specific pieces that a group of people don't agree with that people would want another protocol for
 [12:00] Zha Ewry: I'd be game for a discussion on how to do a sprint on that
 [12:00] Justin Clark-Casey: Tess: I'm not sure OpenSim really wants to choose protocols
 [12:00] Bill Humphries: http://wiki.secondlife.com/wiki/SLGOGP_Teleport_Strawman
 [12:00] Justin Clark-Casey: which is why separation of the stuff into modules works for most people on the project
 [12:00] Hiro Protagonist: OpenSim wants it's users to choose protocols
 [12:00] Zha Ewry: I think Opensim, is strctured well enough to allow ipeople to pick and chose sets
 [12:01] Dahlia Trimble is going to take a rl break then head to Zero's ... bye all :)
 [12:01] Hiro Protagonist: bye Dahlia
 [12:01] Zha Ewry: I've plenty of use cases where I don't want or need a full interop set, just some parts
 [12:01] Zha Ewry: Oh. Yikes.time.
 [12:01] Justin Clark-Casey: bye Dahlia
 [12:02] Zha Ewry: I think keepign in mind, that OpenSim is a base, which can have mutiple towers is important
 [12:02] Justin Clark-Casey: absolutely. Also there are many people who want to use opensim for standalone applications without any grid concerns
 [12:02] Justin Clark-Casey: zha: +1
 [12:02] Adam Frisby is going to head to Zero's hours. (maybe we should look at extending the OpenSim ones since we keep running into Zeros)
 [12:02] Adam Frisby: Seeya over there
 [12:03] Justin Clark-Casey needs to eat
 [12:03] Zha Ewry: yeah, I have an hour on Tuesday when I can work. Can't have that
 [12:03] Bill Humphries: Justin, OpenSim doesn't need to be hard coded to use one protocol, I agree.
 [12:04] Tess Linden: sorry, pushed a funny button
 [12:05] Bill Humphries: But, as the AWG and Linden Lab are working on a candidate for an interopt protocol, I would hate to get blocked on moving forward.
 [12:05] Hiro Protagonist: step away from the 'big X' LoL
 [12:05] Justin Clark-Casey: yes. Which is why the module solution will hopefully be a good way forward
 [12:05] Strawberry Fride: I'm gonna head off - thanks all, see you later :)
 [12:05] Digi Fly: bye strawberry
 [12:05] Hiro Protagonist: bye :)
 [12:06] Shenlei Winkler: night, Strawberry
 [12:06] Justin Clark-Casey: so that OpenSim can be bundled with various ogp modules sitting on top potentially (assuming that something doesn't crop up later on in connection with the fundamental architecture of vws)
 [12:06] paulie Flomar: I'm off, too. Good meeting. :)
 [12:06] Bill Humphries: okay
 [12:06] Justin Clark-Casey: bye paulie
 [12:06] paulie Flomar: .me waves
 [12:07] Bill Humphries: So what's the intermediate step?
 [12:07] Tess Linden: I think this has been a very constructive and open conversation, and we should continue to work together to create protocols that'll make the metaverse work, whether its standalone or interoperable
 [12:07] Justin Clark-Casey: yes, I do agree
 [12:07] Bill Humphries: I've heard concerns about developing that as a branch, and as a patch. But we will need a way to develop that will get us to a OGP module.
 [12:08] Hiro Protagonist: I dont question the utility of using standalone first as an intermediary step - it's just that few (if any) beyond the devs in this group run standalone
 [12:08] Tess Linden: I think the next step is Zha will refactor code to work this way, and after that would it be good enough to merge into trunk?
 [12:08] Justin Clark-Casey: I would think if Zha does the work to put in modular hooks
 [12:08] Hiro Protagonist: we're grid operators/region operators who operate grid attatched regions
 [12:08] Bill Humphries: How difficult would it be to take Zha's contribution and modularize it?
 [12:08] Justin Clark-Casey: the modular parts can be put into trunk and then the module itself shipped either seperately or in an opensim bundle
 [12:08] Tess Linden: Hiro: I didnt know that.. good to know
 [12:08] Justin Clark-Casey: there's probably a reasonable amount of work involved since I don't think we currently have modularity in that area
 [12:09] Bill Humphries: Hiro, yes, I understand that. And as we work out the kinks in standalone, then we'll be better prepared to develop grid mode.
 [12:09] Justin Clark-Casey: I mean,in the mean time perhaps someone would want to just supply an opensim build with Zha's patch applied
 [12:09] Hiro Protagonist: Understood :)
 [12:09] Justin Clark-Casey: if you want to widen the testing participation as much as possible...
 [12:09] Bill Humphries: thanks, Hiro
 [12:10] BlueWall Slade: clean patches against recent svn's would be ok?
 [12:10] Bill Humphries: Yes, broad participation, so we can work out bugs in code and the protocoal.
 [12:10] Tess Linden: Justin: would it be too much to ask for that to be done during the Beta so we can get the code in earlier?
 [12:10] Justin Clark-Casey: sorry, for what to be done?
 [12:11] Tess Linden: for the refactor for modularity
 [12:11] Justin Clark-Casey: I think as long as people put in the manpower then those patches can come in any time
 [12:11] Tess Linden: if we keep the code away from the trunk for too long then maintenance on it will be expensive
 [12:11] Hiro Protagonist: afaik, we already have much spport for modules in place - in fact, probably all
 [12:11] Justin Clark-Casey: yes
 [12:12] Justin Clark-Casey: so I guess the way forward is for Zha or other interested parties to submit patches containing the modularity hooks and as much code as possible that isn't ogp specific
 [12:12] Justin Clark-Casey: those can be tiny little patches at the work gets done, in principle
 [12:13] Bill Humphries: okay, so the consensus is OGP lives in a module
 [12:14] Justin Clark-Casey: for now at least - who knows what can happen in the future in principle. I mean, if it becomes HTTP then things could be different
 [12:14] Justin Clark-Casey: but I think we're concerned with concensus building in the OpenSim community
 [12:14] Carlos Roundel: hola
 [12:15] Hiro Protagonist: Hola Carlos
 [12:15] Carlos Roundel: in my region higth avatars density
 [12:15] Carlos Roundel: :DD
 [12:15] Bill Humphries: I understand, I've worked with groups that work by consensus building, and it's work, but worth the effort.
 [12:15] Justin Clark-Casey: Bill: Cool - we're definitely much the same here
 [12:16] Justin Clark-Casey: hard work but very much necessary to our project setup, I would say
 [12:17] Bill Humphries: and if you have questions about OGP, you're very welcome to join in on the process
 [12:17] Justin Clark-Casey: I would like too - unfortunately I'm horribly busy for the next couple of months
 [12:17] Axaes Xandal: very funny
 [12:18] Axaes Xandal: hehe
 [12:18] Justin Clark-Casey: I think OpenSim very much progresses by coalitions of the willing - though there are strong opinions in core about keeping the project 'neutral' right now
 [12:18] Bill Humphries: are there other developers who would be interested in participating?
 [12:18] Hiro Protagonist: unfortunately, JustinCC is the last dev here
 [12:19] Justin Clark-Casey: the last puppy in the shop
 [12:19] Hiro Protagonist: heh
 [12:19] Shenlei Winkler: awwww
 [12:19] Bill Humphries: okay, that's a question I can follow up on later.
 [12:19] Justin Clark-Casey: but I know Adam and Dahlia are on the OGP list
 [12:19] Tess Linden: thanks for having me :)
 [12:19] Tess Linden: gotta run
 [12:19] Justin Clark-Casey: bye Tess
 [12:19] G2 Proto: cya Tess
 [12:19] Tess Linden: bye Justin, bye G2
 [12:19] Bill Humphries: Yes, Tess is presenting at Zero's office hours.
 [12:19] Tess Linden: bye everybody
 [12:19] Hiro Protagonist: Thanks for coming Tess
 [12:19] Bill Humphries: Thanks Tess.
 [12:19] Axaes Xandal: bye
 [12:20] G2 Proto: I'll look into the oGP list and see if we can help
 [12:20] Hiro Protagonist: every little bit helps G2
 [12:20] Hiro Protagonist: but what they really need is opensim dev representation in the ogp group
 [12:20] Hiro Protagonist: core devs
 [12:20] Bill Humphries: Great, and I'll follow up there. But I think the final takeaway is short term patch, longer term module?
 [12:21] Justin Clark-Casey: There are people in there - I'm in there though I don't have a lot of time
 [12:21] Justin Clark-Casey: bill: in my opinion, that is definitely the way to go to move the process forward
 [12:21] Hiro Protagonist: this may be heresy, but...
 [12:21] G2 Proto: hehe cool ok well I'll get involved I am mostly web mashups but I have others on my team who could help with client stuff possibly
 [12:21] Bill Humphries: Yes, as mentioned ablve Adam and Dahlia.
 [12:21] Bill Humphries: and otehrs
 [12:21] Justin Clark-Casey: The problem is, there aren't always people with time :)
 [12:21] Hiro Protagonist: I think that you guys need to quit having office hours about this - get our core group together with your engineers, and hammer this thing out
 [12:21] Justin Clark-Casey: Us not being a corporate entity and all ;)
 [12:21] Hiro Protagonist: just my 0.02$L
 [12:22] Bill Humphries: Hiro, I've pitched the idea internally about having a sprint.
 [12:22] G2 Proto: I'm willing to commit to help represent OpenSim there regularly
 [12:22] Justin Clark-Casey: hiro: I think the situation needs to evolve a bit first
 [12:22] G2 Proto: OGP that is
 [12:22] Hiro Protagonist: G2: you arent OpenSIm core
 [12:22] Hiro Protagonist: no offense
 [12:22] G2 Proto: i see
 [12:22] Hiro Protagonist: but hat's whats needfull
 [12:23] Justin Clark-Casey: I reckon Dahlia and Adam will do an excellent job there though
 [12:23] Hiro Protagonist: Justin: I'm sure it will :)
 [12:23] UUID Speaker: Axaes Xandal, your key is 29540840-7c2d-4038-9089-0d429611644c
 [12:23] Rip Oxbar: hello folks
 [12:23] G2 Proto: hey Rip
 [12:23] Axaes Xandal: Hi Rip
 [12:23] Justin Clark-Casey: hi rip. You're merely 83 minutes late :)
 [12:23] Monk Zymurgy: hi axaes...has anyone made a chatlog for nebadon?
 [12:23] Bill Humphries: Justin, by evolve, what do you mean, a more fleshed out spec or acceptance?
 [12:23] Axaes Xandal: Hi Monk
 [12:24] Rip Oxbar: Charlie u here with us
 [12:24] Hiro Protagonist: Charles is not
 [12:24] Rip Oxbar: heck
 [12:24] Axaes Xandal: Next time y'all come to a meeting you should get properly dressed:--)
 [12:24] Justin Clark-Casey: I think more fleshing out really, seeing how things develop
 [12:25] Axaes Xandal: Looks like an army of clones here
 [12:25] Rip Oxbar: well i would but my clothes didnt DL
 [12:25] Digi Fly: rebake.
 [12:25] Axaes Xandal: hehe
 [12:25] Bill Humphries: hit rebake
 [12:25] Justin Clark-Casey: I think there is a lot of concern with seeing grid to grid, though I personally think getting opensim properly hooked up to thenew domain stuff is enough work
 [12:25] Justin Clark-Casey: I have to imagine that's much much easier than doing grid to grid
 [12:26] Rip Oxbar: ok everythings worn
 [12:26] Rip Oxbar: its just not rezzing :)
 [12:26] Hiro Protagonist: you are rezzed here Rip
 [12:27] Rip Oxbar: not here :(
 [12:27] Justin Clark-Casey: anyhow, must dash
 [12:27] Hiro Protagonist: TC Justin
 [12:27] Rip Oxbar: it maybe be our connection
 [12:27] Hiro Protagonist: catch you in the channels
 [12:27] Axaes Xandal: bb Justin
 [12:27] You: bye Justin
 [12:27] Justin Clark-Casey: see you hiro, chris
 [12:27] Digi Fly: bye justin
 [12:27] Justin Clark-Casey: see you folks
 [12:27] Bill Humphries: kk, thanks for your time Justin
 [12:28] Rip Oxbar: i heard Microsoft is going to put a opensim package together...is that true?
 [12:28] Hiro Protagonist: LOL
 [12:28] Digi Fly: ow. that will kick SL back.
 [12:28] Bill Humphries: I don't know. One of their developer evangelists is interested in OpenSim.
 [12:28] Hiro Protagonist: I seriously doubt any such thing is happening
 [12:29] Hiro Protagonist: who knows though
 [12:29] Rip Oxbar: Hang on ill get you a article
 [12:29] Axaes Xandal: Is the smudgy-terrain-bug know already?
 [12:29] Axaes Xandal: it seems to be spreading
 [12:30] Rip Oxbar: http://secondlife.reuters.com/stories/2008/07/18/microsoft-eyes-integration-between-opensim-and-windows-live-id/
 [12:30] Bill Humphries: okay, I'm the last person in the metaverse who should express an opinon about that.
 [12:30] Hiro Protagonist: Heh
 [12:30] Hiro Protagonist: I have to discount that article without even reading it
 [12:30] Hiro Protagonist: the guy that writes those articles is notorious for not checking his facts
 [12:30] Bill Humphries: alright, I'm going to dash
 [12:30] Hiro Protagonist: he even got Zero's real name wrong recently
 [12:31] Rip Oxbar: Ahh ok
 [12:31] Hiro Protagonist: Take care Bill
 [12:31] Bill Humphries: see you all next week here
 [12:31] Hiro Protagonist: Thanks for coming Sir
 [12:31] Rip Oxbar: cu later bill
 [12:31] You: bye Bill, thanks
 [12:31] Bill Humphries: Office Hours tomorrow on Prep for the Beta
 [12:31] G2 Proto: that article was twisted
 [12:31] Bill Humphries: I'll send the reminder out to the gridnauts list.
 [12:31] Rip Oxbar: Beta when?
 [12:31] Rip Oxbar: here?
 [12:31] Hiro Protagonist sighs
 [12:31] G2 Proto: G2 my company is working on some asp.net mods for admin for open sim
 [12:31] Bill Humphries: http://wiki.secondlife.com/wiki/Open_Grid_Public_Beta
 [12:32] G2 Proto: yet Eric claimed it was Microosft
 [12:32] Rip Oxbar: ok thanks bill
 [12:32] G2 Proto: they are working on no OpenSim package only with us to work with the community
 [12:32] Hiro Protagonist: Eric is a gossip factory
 [12:32] Rip Oxbar: ive been so busy in rl i havnt been keeping up im sorry to say
 [12:32] G2 Proto: yes he really went on a tangent
 [12:32] Hiro Protagonist: you can even quote me on that
 [12:32] Rip Oxbar: haha
 [12:32] G2 Proto: lol
 [12:32] G2 Proto: we wont be doing nay more onterviews
 [12:33] Hiro Protagonist: we have adopted the posture that we will release a press release to him if we have anything for him to report
 [12:33] Hiro Protagonist: (OSGrid, that is)
 [12:33] Rip Oxbar: haha
 [12:33] G2 Proto: we certainly do not plan to compete with Sl which is what is said, neither G2 nor MS
 [12:33] Rip Oxbar: so hes that bad then huh
 [12:33] G2 Proto: just offer a login option is all for opensim
 [12:33] Hiro Protagonist: yeah, fraid so
 [12:34] Hiro Protagonist: G2, dont let him slow ya down
 [12:34] Rip Oxbar: well it would be nice if true though
 [12:34] You: Have t o go now cya
 [12:34] Hiro Protagonist: TC Chris
 [12:34] G2 Proto: I won't but he clearly went for the more suggestive story and could get the MS Dev in trouble
 [12:34] Axaes Xandal: bye Chris
 [12:34] G2 Proto: but don't get me worng
 [12:34] Hiro Protagonist: G2: indeed.
 [12:34] G2 Proto: MS loves OpenSim
 [12:35] Hiro Protagonist: who doesnt?
 [12:35] G2 Proto: they are just trying to find out how to support it
 [12:35] Hiro Protagonist: and they are welcome to do what they will with it of course
 [12:35] Hiro Protagonist: including submit patcehs
 [12:35] G2 Proto: even Steve ballmer is seeing a video from g2 about recreating the MS campus
 [12:35] You: I have a log of office hour pus a couple of screenshots .
 [12:35] G2 Proto: in OpenSim internally
 [12:35] Rip Oxbar: i still dont like L$ being considered for and economy though
 [12:35] You: Bye
 [12:35] G2 Proto: vya Chris
 [12:35] G2 Proto: cya
 [12:35] Hiro Protagonist: Cool Chrios - there's a place on the wiki for the log
 [12:35] Hiro Protagonist: if you would post it that would rock
 [12:35] Hiro Protagonist: *Chris
 [12:36] You: Will do
 [12:36] G2 Proto: they will use OpenSIm internally to teach refactoring too
 [12:36] Digi Fly: bye chris
 [12:36] Hiro Protagonist: tHANKS mAN :d

Office hours

2008-07-22T21:03:47Z

ChrisD: /* Logs */

OpenSim weekly office hours are held in an OpenSim world somewhere on the Internet. The purpose of these office hours is to created a focussed weekly discussion on the state of OpenSim, and things that need attention. It is intended for people wishing to become contributors to the project, so if you show up, you'll probably walk away with something on your todo list.

The office hours are held on Tuesdays at 19:00 UTC. This corresponds to:
* 11:00 PST - San Francisco (U.S.A. - California)
* 14:00 - New York (U.S.A. - New York)
* 19:00 UST - Oslo (Norway)
* 20:00 - Italy
* 02:00 Wed - Perth (Australia) (sorry adam)
* 15:00 AST - Atlantic Canada

There is useful [http://www.timeanddate.com/ Time'n'date] tool for timezone aware planning, with [http://www.timeanddate.com/worldclock/converter.html Time Zone Converter] & [http://www.timeanddate.com/worldclock/meeting.html Meeting Planner] (last thing helps to plan meeting for up 4 tz places)

The office hours will take place at the sim "Wright Plaza" on osgrid, which was called "Sang" until last Friday.

To get to this sim, make sure you have created an account with the normal first, last and password and set the home location to "Wright Plaza". This is done at http://www.osgrid.org.

Then startup the client with a -loginuri of "-loginuri http://osgrid.org:8002" and you should arrive on "Wright Plaza".

=== Previous Agenda Items ===

Cfk would like to suggest that we give some thought to the following bugs.
#) Remoting exceptions with adjacent non-running sims.
#) Prim unlinking.
#) Enabling the script editor from the edit prim dialog box.
#) jpeg/ode dll and glibc-2.4 exceptions and stability.
#) GridServer.exe locks up in r2090 and works fine in r2060 on OSGrid.
#) Enabling avatar 'sit' on a prim.
#) More progress on avatar customization consistency. (Avatar changes when crossing region boundaries).

== Logs ==
* [[Chat log from the meeting on 2008-07-22]]
* [[Chat log from the meeting on 2008-07-21]]
* [[Chat log from the meeting on 2008-07-15]]
* [[Chat log from the meeting on 2008-07-08]]
* [[Chat log from the meeting on 2008-07-01]]
* [[Chat log from the meeting on 2008-06-17]]
* [[Chat log from the meeting on 2008-06-10]]
* [[Chat log from the meeting on 2008-04-15]]
* [[Chat log from the meeting on 2008-04-08]]
* [[Chat log from the meeting on 2008-03-11]]
* [[Chat log from the meeting on 2008-03-04]]
* [[Chat log from the meeting on 2008-02-17]]
* [[Chat log from the meeting on 2008-02-12]]
* [[Chat log from the meeting on 2008-02-05]]
* [[Chat log from the meeting on 2008-01-29]]
* [[Chat log from the meeting on 2008-01-09]]
* [[Chat log from the meeting on 2007-12-18]]
* [[Chat log from the meeting on 2007-12-11]]
* Chat log from the meeting on 2007-12-4 (meeting cancelled)
* [[Chat log from the meeting on 2007-11-27]]
* Chat log from the meeting on 2007-11-20 (missing)
* [[Chat log from the meeting on 2007-11-13]]
* [[Chat log from the meeting on 2007-11-06]]
* [[Chat log from the meeting on 2007-10-30]]
* [[Chat log from the meeting on 2007-10-23]]
* [[Chat log from the meeting on 2007-10-16]]
* [[Chat log from the meeting on 2007-10-09]]

[[Category:Development]]
[[Category:Office Hours]]

Chat log from the meeting on 2008-07-22

2008-07-22T21:02:00Z

ChrisD: New page: [11:01] Shenlei Winkler: I'm moving content over...or trying [11:01] Bill Humphries: She's wearing the mother of all crinolines with that dress. [11:01] Charles Krinkeb: ...

[11:01] Shenlei Winkler: I'm moving content over...or trying
 [11:01] Bill Humphries: She's wearing the mother of all crinolines with that dress.
 [11:01] Charles Krinkeb: Shall we start or wait another minute?
 [11:02] Shenlei Winkler: Ballgowns are the epitome of virtual worlds, Bill
 [11:02] Bill Humphries: No argument there.
 [11:02] Charles Krinkeb: Maybe I need a ball gown sometime so I can keep up.
 [11:02] Justin Clark-Casey: interesting, my graphics card actually appears to be playing ball today
 [11:02] Strawberry Fride: can still only see 4 avatars - though plenty more dots on minimap
 [11:03] Justin Clark-Casey: yes, lots more people than that here sf
 [11:03] Strawberry Fride: I guessed as much :)
 [11:03] Bill Humphries: Well, at some point, maybe Hya will open a branch out here.
 [11:03] Dahlia Trimble: everyone is grey but Bluewall and Strawberry
 [11:03] Dahlia Trimble: and Bill
 [11:03] Justin Clark-Casey: yeah, we do seem to have avatar appearance issues still.....
 [11:04] Strawberry Fride: I can see Dahlia (with clothes), Chris D (also with clothes) and a grey Monk and a grey Adam :)
 [11:04] Charles Krinkeb: I see lots of "AgentHeightWidth" and "MapItemRequest" packets so the sim is loading a bit.
 [11:04] Bill Humphries: maybe hitting rebake?
 [11:04] Shenlei Winkler: Persistence is getting better
 [11:04] Chris D was worried he might be naked
 [11:04] Charles Krinkeb: Adam? Are you here or a zombie?
 [11:04] Shenlei Winkler: I teleported today and did not lose my hair
 [11:04] Strawberry Fride: my hair on my sim is remarkably resilient
 [11:05] Charles Krinkeb: great, Shenlei
 [11:05] Strawberry Fride: especially since we added that missing method to SQL Server. No wonder it wasn't working so well :)
 [11:05] Strawberry Fride: nothing in the Agents table at all before Sunday
 [11:05] Charles Krinkeb: Lets try to get started. I think the first item of business is to chat with our LL colleagues and discuss interop issues.
 [11:05] Justin Clark-Casey: mssql tends to lag in opensim
 [11:05] Justin Clark-Casey: strawberry: good stuff
 [11:06] Justin Clark-Casey: hey mic
 [11:06] Strawberry Fride: mssql not so much any more - whizzy these days
 [11:06] Justin Clark-Casey: hi neas
 [11:06] Justin Clark-Casey: I mean in terms of functionality
 [11:06] Bill Humphries: okay, Tess and Greg are coming, if you have another item you can do first
 [11:06] Strawberry Fride: :)
 [11:06] Charles Krinkeb: Morning, Neas, Mic.
 [11:06] Mic Bowman: morning all
 [11:07] Neas Bade: hey all
 [11:07] Monk Zymurgy: g'morning
 [11:07] Charles Krinkeb: Ready to get started?
 [11:07] Neas Bade: charles, can you turn on build for me?
 [11:07] Neas Bade: I want to generate the sit bug test case
 [11:08] Adam Frisby: That was weeird
 [11:08] Adam Frisby: Anyway
 [11:09] Adam Frisby: I have a slight announcement for these hours - can someone add a note on the agenda for me?
 [11:09] Charles Krinkeb: Neas. You are an estate manager just like me and I just rezzed a box.
 [11:09] Dahlia Trimble: slight?
 [11:09] Neas Bade: yeh, no dice
 [11:09] Adam Frisby: Oh, I totally need estate manager privilidges. >_>
 [11:09] Strawberry Fride: oo - now I see people! flew into adjacent region and back :) Shenlei - fab outfit!
 [11:10] Neas Bade: oh, never mind
 [11:10] Shenlei Winkler grins
 [11:10] Neas Bade: my build button is just greyed out
 [11:10] Shenlei Winkler: Thanks
 [11:10] Neas Bade: ok, I'm going to go build this a little ways away
 [11:10] Mic Bowman: brb
 [11:10] Neas Bade: back shortly
 [11:10] Charles Krinkeb: Adam, Neas. The estate stuff appears to be a bit fubar lately.
 [11:11] Zha Ewry: Hah. Shen is a ruth to me
 [11:11] Zha Ewry: What a fate
 [11:11] Shenlei Winkler: Darling, I can haz hair
 [11:11] Shenlei Winkler smiles
 [11:11] Adam Frisby: Hehe
 [11:11] Zha Ewry: No
 [11:11] Zha Ewry: You can haz particle cloud
 [11:11] Strawberry Fride: lol
 [11:11] Charles Krinkeb: Lets get going. Sorry about the estate stuff, Neas. I'll have Nebadon look at it a little later. I dont want to loose our hour.
 [11:11] G2 Proto: :) hello all
 [11:11] Shenlei Winkler: well, there are worse things...+
 [11:11] Primitive: Script running
 [11:12] Charles Krinkeb: Lets let Adam go first. Adam, you have the floor.
 [11:12] Adam Frisby: Alright, I have shinies.
 [11:12] Adam Frisby: Zha, you're going to appreciate this.
 [11:12] Dahlia Trimble: shinies? o.O
 [11:12] Charles Krinkeb: Somebody hit you in the eye?
 [11:12] Adam Frisby: Myself and Mike just finished setting up forge.opensimulator.org -- it's a install of GForge, which is Source-forge like software. It's for people who are working on opensim related projects, to get hosting and SVN access for it
 [11:13] Adam Frisby: Things like say the Perl Gridservers, the ASP.NET gridservers, Zha's interop patches, etc can all be hosted there
 [11:13] Adam Frisby: We can make projects, you get mailing lists, bug trackers, etc all included plus SVN hosting (looks like we have Hg and Git too)
 [11:13] BlueWall Slade: awesome!
 [11:13] Hiro Protagonist: +1 Adam
 [11:13] Hiro Protagonist: Hi All :D
 [11:13] Zha Ewry: Cool
 [11:13] Zha Ewry: Better still, when the pacth isn't ugly as sin
 [11:13] Hiro Protagonist: we need to get Jeroen's 'moo' on there too
 [11:14] Charles Krinkeb: How do we handle keeping the patches in sync with the opensim svn?
 [11:14] Dahlia Trimble: good question Charles
 [11:14] Justin Clark-Casey: adam: nice
 [11:14] Adam Frisby: Well, doing full opensim clones probably isnt the intent of hosting it there, but I imagine just producing .patches based on SVN revisions might be the way to go.
 [11:14] Hiro Protagonist: Hiya WHump :)
 [11:14] Charles Krinkeb: Hiro. Can you figure out how to set Neas so he can build and add Adam to estate managers. I cant tell if it is broken or not.
 [11:14] Adam Frisby: In any case, it should help - I'm going to throw up a lot of the internal DTL code onto it so you can play with some of our internal stuff. Anyone else is free to register projects, etc.
 [11:15] Adam Frisby: We've got a 15 user limit right now, but that will be removed in a day or two when we get the proper OSS license reg'd.
 [11:15] Hiro Protagonist: sure thing Charles one sec
 [11:15] Bill Humphries: Sorry about that, I lost my presense.
 [11:15] Charles Krinkeb: No problem, Bill. Anything else, Adam? Any questions for Adam?
 [11:15] Justin Clark-Casey: didn't appear to be directly a server issue
 [11:15] G2 Proto: its very cool thanks Adam
 [11:15] Bill Humphries: What was the topic?
 [11:16] Hiro Protagonist: Ckrinke, need to relog as plaza builder
 [11:16] Charles Krinkeb: [11:12] Adam Frisby: Myself and Mike just finished setting up forge.opensimulator.org -- it's a install of GForge, which is Source-forge like software. It's for people who are working on opensim related projects, to get hosting and SVN access for it
 [11:16] Tao Takashi: somehow I seem to be underwater and I cannot see myself ...
 [11:16] Charles Krinkeb: k, Hiro
 [11:16] Hiro Protagonist: also the estate managers list for this region is impacted; I'll have to purge a few non-participants
 [11:16] Charles Krinkeb: [11:13] Adam Frisby: Things like say the Perl Gridservers, the ASP.NET gridservers, Zha's interop patches, etc can all be hosted there
 [11:16] Dahlia Trimble: Thanks Adam :)
 [11:16] Justin Clark-Casey: we still have these login issues I see...
 [11:16] Charles Krinkeb: Anything else, Adam? Any questions for Adam?
 [11:17] Bill Humphries: okay, thanks for catching me back up
 [11:17] Justin Clark-Casey: adam: are there any criteria for getting hosting?
 [11:17] Neas Bade: btw, this is the log off doom. It has funny effects if people log in later as to what you look like. Fun bugs in the sit code
 [11:17] Neas Bade: btw, Adam, +1 for setting up gforge
 [11:17] Justin Clark-Casey: apart from that you have code related to opensim?
 [11:18] Justin Clark-Casey: don't log off is you're on the log of doom?
 [11:18] Charles Krinkeb: Lets give Bill the floor and Adam can chime in if there is more, or anyone can ask questions about anything we previously talked about.
 [11:18] Adam Frisby: Justin: no criteria other than it's somewhat opensim related.
 [11:18] Adam Frisby: Projects have to be approved by a site admin (opensim devs)
 [11:18] Justin Clark-Casey: cool :)
 [11:18] Neas Bade: Adam, I thought there was a free version of gforge, not just the commercial version. Or did they stop doing that?
 [11:19] Plaza Builder: Ok Neas
 [11:19] Adam Frisby: Neas: The free version is crippled.
 [11:19] Adam Frisby: Majorly, so we're applying for the full F/OSS one.
 [11:19] Plaza Builder: if you select 'Advanced->View Admin Options' it should enable your estate powers
 [11:20] Dahlia Trimble: who is Plaza Builder today?
 [11:20] Plaza Builder: Hiro, but only temproarily
 [11:20] Neas Bade: Plaza, no, it's working for me
 [11:20] Plaza Builder: cool
 [11:20] Neas Bade: I didn't realize the build button would stay greyed out
 [11:20] Neas Bade: that's something we should change
 [11:20] Plaza Builder: Ahhhh
 [11:20] Neas Bade: but it's a minor bug
 [11:20] Plaza Builder: yup
 [11:20] Neas Bade: sorry, my confusion
 [11:20] Plaza Builder: Charles, what all was it you asked me to do? Neas, and...
 [11:21] Bill Humphries: Okay, Tess and Zha are here, so I can go ahead if you all are ready.
 [11:21] Charles Krinkeb: Adam as estate manager. Gryc might have to go off the list.
 [11:21] Charles Krinkeb: Please go ahead, Bill.
 [11:21] Plaza Builder: Gryc is off, as are Babble and Pablo
 [11:21] Plaza Builder: they can be readded later if need be
 [11:21] Charles Krinkeb: k, Hiro
 [11:21] Bill Humphries: Okay, so next week, the 31st, we're kicking off the Open Grid Protocol Beta Program.
 [11:21] Adam Frisby: Charles: we should be able to make it so there's multiple admins.
 [11:22] Adam Frisby: (FYI: Mike just got the F/OSS license then, he's installing now)
 [11:22] Tao Takashi: now this looks better. Hello :)
 [11:22] Justin Clark-Casey: Hi Tao
 [11:22] Bill Humphries: As you may know, Zha Ewry developed a patch for OpenSim that enables the protocols for teleport.
 [11:22] G2 Proto: hey Tao
 [11:22] Plaza Builder: Ok Adam you are set sir
 [11:22] Plaza Builder: brb w/Hiro
 [11:22] Zha Ewry: FFor some epsilon, around the phrase enables
 [11:23] Bill Humphries: At the moment, we need to do a little more work to get Zha's changes in sync with our agent domain.
 [11:23] Tao Takashi: I should have tried to login to here earlier to be here in time ...
 [11:23] Dahlia Trimble: is the current interop patch on mantis up to date?
 [11:23] Zha Ewry: I should have a new on to offer in a day or so
 [11:24] Zha Ewry: Among other things, catching up to the current Linden Beta grid
 [11:24] Zha Ewry: and.. somewhat better integrated, less horrible
 [11:24] Bill Humphries: But what I'm hoping to come out of here with is a plan for people participating in the beta to either be able to build OpenSim out of top of trunk with OGP support, or a well-defined way to get it patched with OGP support.
 [11:24] G2 Proto: great!
 [11:25] Justin Clark-Casey: Zha, is it still the case that applying the patch disrupts the current region login ops?
 [11:25] Hiro Protagonist: lol
 [11:25] Adam Frisby: >_> <_<
 [11:25] Hiro Protagonist: Adam
 [11:25] Strawberry Fride: :)
 [11:25] Zha Ewry: Less so, but yes
 [11:25] Charles Krinkeb: <harumph>
 [11:25] Zha Ewry: Mostly in the handles
 [11:25] Bill Humphries: I saw the blue card, LOL.
 [11:25] Dahlia Trimble: lol
 [11:25] Zha Ewry: Once we sort out the current prtocol update, I want to look at that
 [11:26] Justin Clark-Casey: I mean, the best thing right now might be just for people to apply the patch
 [11:26] Zha Ewry: I think, and this is problematic, for the longer term story
 [11:26] Justin Clark-Casey: or for some kind soul to supply a build with it pre-applied
 [11:26] Bill Humphries: So that would be a blocker, then, alright.
 [11:26] Hiro Protagonist: Justin, you are in color mate ;)
 [11:26] Dahlia Trimble: I have a build with it applied
 [11:26] Tao Takashi: btw, it would be great if the protocol docs would be updated or discussion about changes would go about some mailing list so I can keep up with my agent domain implementation. not sure if that's possible
 [11:26] Zha Ewry: that you won't be able to login, via the normal path, and teleport, because you won't be on the agent domain correctly
 [11:26] Charles Krinkeb: So, Bill. Is your understanding that the interop is teleporting to an OpenSim standalone region from the betagrid and not to a region attached to OSGrid?
 [11:26] Justin Clark-Casey: hiro: I know - it's wierd - there's nothing different on my machine :)
 [11:27] Bill Humphries: Charles, at the moment, yes. But I think teleporting to a grid enabled sim would be a great next step in the beta.
 [11:28] Hiro Protagonist: hrmmm
 [11:28] Zha Ewry: I clearly want to get this synced with the full grid version
 [11:28] Hiro Protagonist: I guess I'll wait for phase 2 then
 [11:28] Zha Ewry: Part of that is going to be sorting out handles
 [11:28] Hiro Protagonist: I dont have the resources to set up a standalone
 [11:28] Neas Bade: so, that means proper agent domain handoff, right?
 [11:28] Zha Ewry: (my one attempt to patch that way, was a nightmare)
 [11:28] Neas Bade: because I don't think we want osgrid accounts to be dependant on SL accounts
 [11:29] Bill Humphries: Neas, this is just a beta.
 [11:29] Neas Bade: most people running grids want seperate account controls
 [11:29] Neas Bade: Bill, yeh, I understand that :)
 [11:29] Tao Takashi: I would more call it alpha ;-)
 [11:29] Justin Clark-Casey: agent domain handoff is going to be considerably more complicated I imagine
 [11:29] Neas Bade: but if we're trying for real grid interop, it seems like doing it early would be a good thing
 [11:29] Charles Krinkeb: For, now it is fair to say that a common username could be used so we dont have to open the openid/liveid/authorization bit yet.
 [11:29] Zha Ewry: I think there is a very legitimate concern, in not handling the AD list
 [11:30] Zha Ewry: (bit)
 [11:30] Zha Ewry: too many SL style meetings, my typoing dies
 [11:30] Neas Bade: :)
 [11:30] Zha Ewry: There are two, issues, I think, one is the agent domain code, the other, woudl be the client handling the hand off
 [11:31] Charles Krinkeb: Well, I dont want to derail any of the efforts, so I'll stand in the background a bit, and OSGrid is happy to support this effort when appropriate for LL's plans.
 [11:32] Charles Krinkeb: Although one has to understand that the notion of "interop" implies grid<-->grid, both betagrid<->OpenSimGrid *and* OpenSimGrid<->OpenSimGrid
 [11:32] Bill Humphries: Okay, so for the short term, if we could host the patch on adam's gforge, until we can work through the issues with merging that would be good. And I can see if we can provide a precompiled mono .exe as well.
 [11:33] Adam Frisby: Go ahead - if you make the project right now, I can approve it.
 [11:33] Dahlia Trimble: couldnt interop also include grid<->standalone?
 [11:33] Adam Frisby: (I'll also add the other opensim admins to the site admins if they care to register accounts. hehe.)
 [11:33] Tao Takashi: I hope the patch will then also be published
 [11:33] Neas Bade: Dahlia, I think so
 [11:33] Charles Krinkeb: We can run a mono .exe on the OSGrid gridserver as long as both LL and the core developers agree.
 [11:33] Tess Linden: Interop means that the code implements against the protocol, and any code taht runs against the protocol should interoperate with each other
 [11:33] Bill Humphries: Yes, and that's where, given the time left before the beta starts, where we'll start from.
 [11:33] Digi Fly: hi
 [11:34] Hiro Protagonist: I think it should also support standalone as well; however, it's important to note, that there is almost no representation of that use case on osgrid
 [11:34] Zha Ewry: Well, baring opermisions and all
 [11:34] Justin Clark-Casey: tao: you mean the zha's existing work? that is already out there
 [11:34] Neas Bade: the issue on the opensim side is that we've got to address moving between standalone and grid instances relatively shortly, especially because there is no URL bar in the client
 [11:34] Justin Clark-Casey: hiro: you mean local assets?
 [11:34] Tess Linden: we've got code thats working against a protocol that will allow you to TP between virtual worlds, we're talking about making it easy to iterate on
 [11:34] Bill Humphries: A patch that will enable OGP on a standalone grid. With the caveats about no inventory, no baked textures coming over.
 [11:34] Neas Bade: and if there is no ability to do agent domain handoff, then that is a real issue
 [11:34] Charles Krinkeb: Ok. This may be an important point then. Are you saying that interop has nothing to do with moving from grid to grid, or sim to sim, but merely that certain code communicates with each other in two internet seperated sites?
 [11:34] Tess Linden: Neas: Suzy put in a url bar in the client before she left for vacation
 [11:34] Hiro Protagonist: Justin: I mean, coming to osgrid to look for people running standalone is, well, a little misdirected
 [11:35] Tess Linden: it remembers the last 10 region_urls youve typed in
 [11:35] Tao Takashi: Justin: I also mean the changes which are happening now
 [11:35] Justin Clark-Casey: hiro: true :)
 [11:35] Neas Bade: Tess, ok, so when does that release in a GA browser? :P)
 [11:35] Zha Ewry: Interop is a series of steps, and teleport is only one part
 [11:35] Tao Takashi: I have Zha's patch running here on a standalone server
 [11:35] Neas Bade: oops sorr for the P, I just meant a smilely
 [11:35] Charles Krinkeb: k. I best start correcting a mis-conception that exists in our users.
 [11:35] Neas Bade: Tess, those urls include grid?
 [11:36] Zha Ewry: We need to make sure it's all the places where we ahve urls
 [11:36] Tess Linden: region_urls identify the region you're rezzing into, could be on any grid
 [11:36] Zha Ewry: including AD, Region, and login
 [11:36] Tao Takashi: and it would be nicer of course to have a branch instead of a patch which you have to apply
 [11:36] paulie Flomar waves
 [11:36] Zha Ewry: We really want to make it easy to hop to any of the permuations
 [11:36] Zha Ewry: and. ideally, without a logout/login
 [11:36] Dahlia Trimble: Hi Paulie
 [11:36] Tess Linden: sorry CHarles, whats the misconception?
 [11:36] paulie Flomar: Howdeh, everyone. :)
 [11:37] BlueWall Slade: stand alone seems a good, safe first step
 [11:37] Neas Bade: ok, so when do we get to see that in a release client? As it would be really nice to not need to have all these icon links for different OS instances
 [11:37] paulie Flomar: thx, d.
 [11:37] Digi Fly: stand alone still the required firts step.
 [11:37] Justin Clark-Casey: hey paulie
 [11:37] Tess Linden: thats what the Beta is for, so we can all try it out and work out all the kinks to a ready state
 [11:37] Tao Takashi: will this beta client also have some cmdline option to change the agent domain?
 [11:38] paulie Flomar: Howdeh, JCC.
 [11:38] Neas Bade: right, but the beta client won't do old auth, right?
 [11:38] Charles Krinkeb: The prevalent misconception is that the interop implies a full handoff from betagrid<->OpenSimGrid, both ways, in an arms-length manner. I may have to soften the perceptions in the users a bit.
 [11:38] Zha Ewry: I think its in the branch
 [11:38] Dahlia Trimble: maybe we need to develop some criteria for including the interop patch into core?
 [11:38] Zha Ewry: (the various switches, to select all the bits, if not, it needs to be)
 [11:38] Tao Takashi: interop4 I guess
 [11:38] Tess Linden: +1 Dahlia
 [11:39] Charles Krinkeb: No problem. I just dont want the users to assume it is further ahead then it is. That adds additional pressure on everyone.
 [11:39] Zha Ewry: I think handles, grid , whitelisting, good controls, and cleaner code, would begin to do it
 [11:39] Charles Krinkeb: +1, Zha. Thats about where my thinking starts.
 [11:39] Bill Humphries: Zha by grid you mean teleporting into a sim in a grid, or a handoff?
 [11:39] Tess Linden: Charles: its not a full handoff because your identity lives with the agent domain, and the agent domain is the second life one, but the opensim one doesn't implement against the protocol yet
 [11:39] Zha Ewry: Being able to run in grid mode
 [11:40] Bill Humphries: Zha: +1
 [11:40] Zha Ewry: Full handoff, I think is another good, but, somewhat seperable
 [11:40] Zha Ewry: issue
 [11:40] Justin Clark-Casey: I think we're really aiming for modules for the protocol stuff if possible
 [11:40] Tess Linden: do we need all those things for the Beta?
 [11:40] Zha Ewry: We need to be able to do it, so we can have both the various unified and seperable cases
 [11:40] Zha Ewry: I think we need all of those to exit beta ;-)
 [11:41] TARDIS Wiki-Lith: No wikipedia URL found
 [11:41] Tao Takashi: these are the times of perpetual betas though
 [11:41] Tess Linden: oh yes, there is so much to do during the Beta, I don't think anybody is arguing that
 [11:41] Zha Ewry: I personaly have my own list of horrible things inside my code to fix too
 [11:42] Charles Krinkeb: Its ok. One step at a time is just fine. We will move forward faster that way. I just need to make sure the perceptions are not out of line with reality over the next few months.
 [11:42] Zha Ewry: (there's also some discussion we need to have about how we signal how the grids cooperate, so w eknow if we do/don't need a AD handoff
 [11:42] Tess Linden: but am I hearing that we don't want to apply the patch to trunk until the Beta is done and we're actually releasing the code?
 [11:42] Dahlia Trimble: I would hope sooner Tess
 [11:42] Bill Humphries: okay, so I'd propose the plan would be to start a branch with standalone OGP support and add the features Zha mentioned above to that branch.
 [11:42] Zha Ewry: When the OpenSim core, thinks its appropriate
 [11:42] Justin Clark-Casey: there really is a philosphical question as to whether it would ever be applied directly to trunk
 [11:42] Justin Clark-Casey: we may well want to keep protocol code in separate modules
 [11:43] Neas Bade: no, I think you are hearing that AD handoff and having grids be able to connect but not be in the same AD is really important
 [11:43] Tao Takashi: Justin: wouldn't this be more a question of refactoring the code so these things can be plugins?
 [11:43] Justin Clark-Casey: Tao: yep, that's certainly a factor
 [11:43] Zha Ewry: A bit of both, I think tao
 [11:43] Neas Bade: and something short of that, in which OpenSim is largely just extra land is a lot less interesting to a lot of opensim folks
 [11:44] Dahlia Trimble: so we may want to work towards putting the hooks into core so interop can work as a module?
 [11:44] TARDIS Wiki-Lith: Ask a question with '???' to get a Wikipedia answer

(maybe). Touch me to get the Wikipedia article used

to provide the answer.

 [11:44] Charles Krinkeb: We also have the issue of OpenSimGrid<->OpenSimGrid interop, so there may be more then one implementation of a similar notion.
 [11:44] TARDIS Wiki-Lith: Ask a question with '???' to get a Wikipedia answer

(maybe). Touch me to get the Wikipedia article used

to provide the answer.

 [11:44] Justin Clark-Casey: in which case those plugins may be bundled with a separate OpenSim core kernel rather than being part of that core kernel itself
 [11:44] Justin Clark-Casey: Dahlia: +1
 [11:44] Charles Krinkeb: stop typing three question marks
 [11:44] TARDIS Wiki-Lith: Ask a question with '???' to get a Wikipedia answer

(maybe). Touch me to get the Wikipedia article used

to provide the answer.

 [11:44] Dahlia Trimble: lol
 [11:45] Hiro Protagonist: I think kinic is actually working on the wikilith
 [11:45] Hiro Protagonist: *knio
 [11:45] Zha Ewry: I'd be glad to work with anyoen who wants to
 [11:45] Hiro Protagonist: pffft
 [11:45] Hiro Protagonist: KINOC
 [11:45] Hiro Protagonist: lol
 [11:45] Zha Ewry: dive into the modularity bits
 [11:45] Tao Takashi: having it on a branch should definitiely make it easier to collaborate
 [11:45] paulie Flomar: It was me playin with the wikilith. I'll cut it out. :)
 [11:45] Daxxon Kinoc: not i
 [11:46] Hiro Protagonist: the thing that concerns me about a branch is that the code there will rapidly go stale
 [11:46] Charles Krinkeb: Thanks for clarifying, Zha, Tess and Bill. I was operating under the misconception that this was a full handoff.
 [11:46] Neas Bade: Tao, we shot down the branch idea in core. The opensim code moves too fast and no one really wants to maintain a branch.
 [11:46] Dahlia Trimble: I dont see how a branch would have any advantages over a patch set
 [11:46] Zha Ewry: So far, I've been able to keep on branch wihtout too much pain
 [11:46] Neas Bade: branches in svn is the evil
 [11:46] Zha Ewry: and. I'm willing to comit to it
 [11:46] Zha Ewry: i mean trunk
 [11:46] Justin Clark-Casey: depends if other people also want to work on that code I think
 [11:47] Adam Frisby: Zha: well there's an advantage there to having the seperate project on the forge then - since you can maintain that and release your in-dev code
 [11:47] Zha Ewry: I've got the stuff in placet o build and test daily
 [11:47] Justin Clark-Casey: patches aren't very scalable
 [11:47] Zha Ewry: One thing I'd think we can do
 [11:47] Zha Ewry: is break it into two parts
 [11:47] Zha Ewry: the hooks
 [11:47] Zha Ewry: and the core functoin
 [11:47] Zha Ewry: at which point, we can get 90% of it in trunk
 [11:47] Dahlia Trimble: that would help
 [11:47] Tao Takashi: well, I usually prefer branches to sending patches forth and back.
 [11:47] Zha Ewry: and only have the patch out there
 [11:47] Justin Clark-Casey: zha: that would be cool
 [11:47] Hiro Protagonist: one thing that concerns me is taht I know that a lot of the packet and stack code is being worked over as we speak
 [11:48] Zha Ewry: I don't touch that
 [11:48] Bill Humphries: and the core function becomes a module, Zha?
 [11:48] Hiro Protagonist: it may well be that the very parts you're depending on being stable will be moving the most quickly
 [11:48] Tao Takashi: If Zha works alone on it then ok, but if somebody should join I would see some problem.
 [11:48] Justin Clark-Casey: hiro: I think that'slargely orthogonal Hiro
 [11:48] Zha Ewry: This is 90%, in parsing/responsinf to the AD
 [11:48] Hiro Protagonist bows to more knowlegable heads
 [11:48] Hiro Protagonist: just makin' sure folks
 [11:48] Zha Ewry: and.. calling back into the guts of the OpenSim world
 [11:49] Zha Ewry: (and I'm trying to focus on keepiing it seperate that way)
 [11:50] Zha Ewry: (There is some really tricky re-facoring that might imply, at some point, but it would be nice to havbe nayway, seperate out the login, fromt he state updates behind it)
 [11:50] Charles Krinkeb: Adam. I would be most interested in your opinion and guidance.
 [11:50] Justin Clark-Casey: Zha: I think we're using you as cheap labour to improve our modularity ;)
 [11:50] Adam Frisby: Haha
 [11:51] Dahlia Trimble: lol
 [11:51] Bill Humphries: okay, I think I'm back.
 [11:51] Zha Ewry: That's fine with me
 [11:51] Dahlia Trimble: wb :)
 [11:51] Adam Frisby: Well, I've talked with Zha about this in person - but I think Zha's onto the right track with having some seperate patches for improving OpenSim's modularity, and the interop functionality.
 [11:51] Zha Ewry: When I get the core patch done, I'm willign to sort out other thignds I've noticed. I don't miind making thing bvetter
 [11:51] Charles Krinkeb: Oh Great Existential Wizard, Adam. What would you counsel us on the best way to move forward on the interop patch with maximum happiness and minimal frustration to all.
 [11:52] Zha Ewry: That involves good beer and decent fajitas, I think.
 [11:52] Adam Frisby: Dont look at me, I'm usually the one breaking things. >_>
 [11:52] Tess Linden: Neas I understand your concern that the patch to OpenSim doesn't create an opensim agent domain, but just because we've added coded to make the region domain work against the protcol doesn't prevent anybody else writing code to add an agent domain that works against the protocol too
 [11:52] G2 Proto: lol
 [11:52] Bill Humphries: So we have an offsite in Austin where we can get good beer and fajita's to has this out? :)
 [11:52] Adam Frisby: Tess: I think that does actually raise a valid point here -- is there a OSS project to build a agent domain?
 [11:53] Adam Frisby: A reference one that is
 [11:53] Neas Bade: no, but it prevents the way we can do grid jump today with just the region <-> client protocol
 [11:53] Adam Frisby: If not - that strikes me as a major deficiency that all parties would be best trying to fix.
 [11:53] Neas Bade: which would be a step backwards, and something that would be good not to loose
 [11:53] Justin Clark-Casey: neas: can we really do that today though? The raw jump might be possible but we would hjave asset and inventory probs, potentially
 [11:53] Charles Krinkeb: I have RL intruding and must be passive. Is someone going to get the chat log on the wiki.
 [11:54] Neas Bade: if you are changing domains of control, you are leaving that behind
 [11:54] Neas Bade: but that's ok
 [11:54] Zha Ewry: I think, looking intot he code, that, the current OpenSim code gets pretty close to being an easy basis for an agent domain, in some of the UGAS code
 [11:54] Neas Bade: it's like moving from myspace -> facebook
 [11:54] Justin Clark-Casey: the trouble is I don't think you can tell the client to suddenly reload its inventory
 [11:54] Tess Linden: that just tells me that there isn't a big push for it. We've been working hard to meet these deadlines to show that interoperability works, and its amazing, but people are pushing back because they want more? why not take steps at a time and work together?
 [11:55] Zha Ewry: I think the client, often gets in the way of this
 [11:55] Justin Clark-Casey: I think one issue is that we want really to be protocol agnostic
 [11:55] Zha Ewry: because it's so hard to do simple hacks, on it end, which would make it easier for people to test and flow between parts of the developing envrionemnt
 [11:56] Justin Clark-Casey: so potentially an opensim server could be on some protocol other than ogp, when/if there is one
 [11:56] Tess Linden: if we can get the code patched to trunk then we can all be working on the same code base to improve interoperability, and we can discuss details at this office hours, at the AWG office hour, or Zero's
 [11:56] Justin Clark-Casey: and opensim itself wasn't really meant ot be a full vw implementation, really just kernel of a general vw server
 [11:56] Charles Krinkeb: I need to go. Sorry folks. Please carry on without me.
 [11:57] Bill Humphries: Okay, thanks Charles.
 [11:57] Hiro Protagonist: See ya soo Charles
 [11:57] Dahlia Trimble: tc Charles :)
 [11:57] Neas Bade: later charles
 [11:57] Justin Clark-Casey: bye charles
 [11:57] Charles Krinkeb: My heart is with this, but I have to pretend to work.
 [11:57] Dahlia Trimble: lol
 [11:57] Daxxon Kinoc: happy RL'ing ...
 [11:57] Charles Krinkeb: <sigh>
 [11:57] Adam Frisby: Zha: just quickly heading back 5 minutes -
 [11:58] Zha Ewry: yes?
 [11:58] Adam Frisby: Adding AD support to the opensim userserver, what needs to be done?
 [11:58] Adam Frisby: Is there some specs someone could spend a couple of days nailing out?
 [11:58] Zha Ewry: I htink, and this was a quick glance
 [11:58] Adam Frisby: And do we have a test scenario we can use?
 [11:58] Zha Ewry: you'd need to be willign to handle about four calls
 [11:58] Zha Ewry: and.. Tao, ought to be able to drive the test cases
 [11:59] Zha Ewry: Bascially, you need to tyake the place_avatar, fromt he client, in the modded form, and pass it on to the regions, as a rez or rez/de-rez pair
 [11:59] Bill Humphries: Yes, Tao, Infinity and Lochlainn have been working on a test harness as well.
 [11:59] Tess Linden: Justin: we've iterated on the OGP protocol because we think this is the common language that will define interoperability between virtual worlds. We should discuss more if there are specific pieces that a group of people don't agree with that people would want another protocol for
 [12:00] Zha Ewry: I'd be game for a discussion on how to do a sprint on that
 [12:00] Justin Clark-Casey: Tess: I'm not sure OpenSim really wants to choose protocols
 [12:00] Bill Humphries: http://wiki.secondlife.com/wiki/SLGOGP_Teleport_Strawman
 [12:00] Justin Clark-Casey: which is why separation of the stuff into modules works for most people on the project
 [12:00] Hiro Protagonist: OpenSim wants it's users to choose protocols
 [12:00] Zha Ewry: I think Opensim, is strctured well enough to allow ipeople to pick and chose sets
 [12:01] Dahlia Trimble is going to take a rl break then head to Zero's ... bye all :)
 [12:01] Hiro Protagonist: bye Dahlia
 [12:01] Zha Ewry: I've plenty of use cases where I don't want or need a full interop set, just some parts
 [12:01] Zha Ewry: Oh. Yikes.time.
 [12:01] Justin Clark-Casey: bye Dahlia
 [12:02] Zha Ewry: I think keepign in mind, that OpenSim is a base, which can have mutiple towers is important
 [12:02] Justin Clark-Casey: absolutely. Also there are many people who want to use opensim for standalone applications without any grid concerns
 [12:02] Justin Clark-Casey: zha: +1
 [12:02] Adam Frisby is going to head to Zero's hours. (maybe we should look at extending the OpenSim ones since we keep running into Zeros)
 [12:02] Adam Frisby: Seeya over there
 [12:03] Justin Clark-Casey needs to eat
 [12:03] Zha Ewry: yeah, I have an hour on Tuesday when I can work. Can't have that
 [12:03] Bill Humphries: Justin, OpenSim doesn't need to be hard coded to use one protocol, I agree.
 [12:04] Tess Linden: sorry, pushed a funny button
 [12:05] Bill Humphries: But, as the AWG and Linden Lab are working on a candidate for an interopt protocol, I would hate to get blocked on moving forward.
 [12:05] Hiro Protagonist: step away from the 'big X' LoL
 [12:05] Justin Clark-Casey: yes. Which is why the module solution will hopefully be a good way forward
 [12:05] Strawberry Fride: I'm gonna head off - thanks all, see you later :)
 [12:05] Digi Fly: bye strawberry
 [12:05] Hiro Protagonist: bye :)
 [12:06] Shenlei Winkler: night, Strawberry
 [12:06] Justin Clark-Casey: so that OpenSim can be bundled with various ogp modules sitting on top potentially (assuming that something doesn't crop up later on in connection with the fundamental architecture of vws)
 [12:06] paulie Flomar: I'm off, too. Good meeting. :)
 [12:06] Bill Humphries: okay
 [12:06] Justin Clark-Casey: bye paulie
 [12:06] paulie Flomar: .me waves
 [12:07] Bill Humphries: So what's the intermediate step?
 [12:07] Tess Linden: I think this has been a very constructive and open conversation, and we should continue to work together to create protocols that'll make the metaverse work, whether its standalone or interoperable
 [12:07] Justin Clark-Casey: yes, I do agree
 [12:07] Bill Humphries: I've heard concerns about developing that as a branch, and as a patch. But we will need a way to develop that will get us to a OGP module.
 [12:08] Hiro Protagonist: I dont question the utility of using standalone first as an intermediary step - it's just that few (if any) beyond the devs in this group run standalone
 [12:08] Tess Linden: I think the next step is Zha will refactor code to work this way, and after that would it be good enough to merge into trunk?
 [12:08] Justin Clark-Casey: I would think if Zha does the work to put in modular hooks
 [12:08] Hiro Protagonist: we're grid operators/region operators who operate grid attatched regions
 [12:08] Bill Humphries: How difficult would it be to take Zha's contribution and modularize it?
 [12:08] Justin Clark-Casey: the modular parts can be put into trunk and then the module itself shipped either seperately or in an opensim bundle
 [12:08] Tess Linden: Hiro: I didnt know that.. good to know
 [12:08] Justin Clark-Casey: there's probably a reasonable amount of work involved since I don't think we currently have modularity in that area
 [12:09] Bill Humphries: Hiro, yes, I understand that. And as we work out the kinks in standalone, then we'll be better prepared to develop grid mode.
 [12:09] Justin Clark-Casey: I mean,in the mean time perhaps someone would want to just supply an opensim build with Zha's patch applied
 [12:09] Hiro Protagonist: Understood :)
 [12:09] Justin Clark-Casey: if you want to widen the testing participation as much as possible...
 [12:09] Bill Humphries: thanks, Hiro
 [12:10] BlueWall Slade: clean patches against recent svn's would be ok?
 [12:10] Bill Humphries: Yes, broad participation, so we can work out bugs in code and the protocoal.
 [12:10] Tess Linden: Justin: would it be too much to ask for that to be done during the Beta so we can get the code in earlier?
 [12:10] Justin Clark-Casey: sorry, for what to be done?
 [12:11] Tess Linden: for the refactor for modularity
 [12:11] Justin Clark-Casey: I think as long as people put in the manpower then those patches can come in any time
 [12:11] Tess Linden: if we keep the code away from the trunk for too long then maintenance on it will be expensive
 [12:11] Hiro Protagonist: afaik, we already have much spport for modules in place - in fact, probably all
 [12:11] Justin Clark-Casey: yes
 [12:12] Justin Clark-Casey: so I guess the way forward is for Zha or other interested parties to submit patches containing the modularity hooks and as much code as possible that isn't ogp specific
 [12:12] Justin Clark-Casey: those can be tiny little patches at the work gets done, in principle
 [12:13] Bill Humphries: okay, so the consensus is OGP lives in a module
 [12:14] Justin Clark-Casey: for now at least - who knows what can happen in the future in principle. I mean, if it becomes HTTP then things could be different
 [12:14] Justin Clark-Casey: but I think we're concerned with concensus building in the OpenSim community
 [12:14] Carlos Roundel: hola
 [12:15] Hiro Protagonist: Hola Carlos
 [12:15] Carlos Roundel: in my region higth avatars density
 [12:15] Carlos Roundel: :DD
 [12:15] Bill Humphries: I understand, I've worked with groups that work by consensus building, and it's work, but worth the effort.
 [12:15] Justin Clark-Casey: Bill: Cool - we're definitely much the same here
 [12:16] Justin Clark-Casey: hard work but very much necessary to our project setup, I would say
 [12:17] Bill Humphries: and if you have questions about OGP, you're very welcome to join in on the process
 [12:17] Justin Clark-Casey: I would like too - unfortunately I'm horribly busy for the next couple of months
 [12:17] Axaes Xandal: very funny
 [12:18] Axaes Xandal: hehe
 [12:18] Justin Clark-Casey: I think OpenSim very much progresses by coalitions of the willing - though there are strong opinions in core about keeping the project 'neutral' right now
 [12:18] Bill Humphries: are there other developers who would be interested in participating?
 [12:18] Hiro Protagonist: unfortunately, JustinCC is the last dev here
 [12:19] Justin Clark-Casey: the last puppy in the shop
 [12:19] Hiro Protagonist: heh
 [12:19] Shenlei Winkler: awwww
 [12:19] Bill Humphries: okay, that's a question I can follow up on later.
 [12:19] Justin Clark-Casey: but I know Adam and Dahlia are on the OGP list
 [12:19] Tess Linden: thanks for having me :)
 [12:19] Tess Linden: gotta run
 [12:19] Justin Clark-Casey: bye Tess
 [12:19] G2 Proto: cya Tess
 [12:19] Tess Linden: bye Justin, bye G2
 [12:19] Bill Humphries: Yes, Tess is presenting at Zero's office hours.
 [12:19] Tess Linden: bye everybody
 [12:19] Hiro Protagonist: Thanks for coming Tess
 [12:19] Bill Humphries: Thanks Tess.
 [12:19] Axaes Xandal: bye
 [12:20] G2 Proto: I'll look into the oGP list and see if we can help
 [12:20] Hiro Protagonist: every little bit helps G2
 [12:20] Hiro Protagonist: but what they really need is opensim dev representation in the ogp group
 [12:20] Hiro Protagonist: core devs
 [12:20] Bill Humphries: Great, and I'll follow up there. But I think the final takeaway is short term patch, longer term module?
 [12:21] Justin Clark-Casey: There are people in there - I'm in there though I don't have a lot of time
 [12:21] Justin Clark-Casey: bill: in my opinion, that is definitely the way to go to move the process forward
 [12:21] Hiro Protagonist: this may be heresy, but...
 [12:21] G2 Proto: hehe cool ok well I'll get involved I am mostly web mashups but I have others on my team who could help with client stuff possibly
 [12:21] Bill Humphries: Yes, as mentioned ablve Adam and Dahlia.
 [12:21] Bill Humphries: and otehrs
 [12:21] Justin Clark-Casey: The problem is, there aren't always people with time :)
 [12:21] Hiro Protagonist: I think that you guys need to quit having office hours about this - get our core group together with your engineers, and hammer this thing out
 [12:21] Justin Clark-Casey: Us not being a corporate entity and all ;)
 [12:21] Hiro Protagonist: just my 0.02$L
 [12:22] Bill Humphries: Hiro, I've pitched the idea internally about having a sprint.
 [12:22] G2 Proto: I'm willing to commit to help represent OpenSim there regularly
 [12:22] Justin Clark-Casey: hiro: I think the situation needs to evolve a bit first
 [12:22] G2 Proto: OGP that is
 [12:22] Hiro Protagonist: G2: you arent OpenSIm core
 [12:22] Hiro Protagonist: no offense
 [12:22] G2 Proto: i see
 [12:22] Hiro Protagonist: but hat's whats needfull
 [12:23] Justin Clark-Casey: I reckon Dahlia and Adam will do an excellent job there though
 [12:23] Hiro Protagonist: Justin: I'm sure it will :)
 [12:23] UUID Speaker: Axaes Xandal, your key is 29540840-7c2d-4038-9089-0d429611644c
 [12:23] Rip Oxbar: hello folks
 [12:23] G2 Proto: hey Rip
 [12:23] Axaes Xandal: Hi Rip
 [12:23] Justin Clark-Casey: hi rip. You're merely 83 minutes late :)
 [12:23] Monk Zymurgy: hi axaes...has anyone made a chatlog for nebadon?
 [12:23] Bill Humphries: Justin, by evolve, what do you mean, a more fleshed out spec or acceptance?
 [12:23] Axaes Xandal: Hi Monk
 [12:24] Rip Oxbar: Charlie u here with us
 [12:24] Hiro Protagonist: Charles is not
 [12:24] Rip Oxbar: heck
 [12:24] Axaes Xandal: Next time y'all come to a meeting you should get properly dressed:--)
 [12:24] Justin Clark-Casey: I think more fleshing out really, seeing how things develop
 [12:25] Axaes Xandal: Looks like an army of clones here
 [12:25] Rip Oxbar: well i would but my clothes didnt DL
 [12:25] Digi Fly: rebake.
 [12:25] Axaes Xandal: hehe
 [12:25] Bill Humphries: hit rebake
 [12:25] Justin Clark-Casey: I think there is a lot of concern with seeing grid to grid, though I personally think getting opensim properly hooked up to thenew domain stuff is enough work
 [12:25] Justin Clark-Casey: I have to imagine that's much much easier than doing grid to grid
 [12:26] Rip Oxbar: ok everythings worn
 [12:26] Rip Oxbar: its just not rezzing :)
 [12:26] Hiro Protagonist: you are rezzed here Rip
 [12:27] Rip Oxbar: not here :(
 [12:27] Justin Clark-Casey: anyhow, must dash
 [12:27] Hiro Protagonist: TC Justin
 [12:27] Rip Oxbar: it maybe be our connection
 [12:27] Hiro Protagonist: catch you in the channels
 [12:27] Axaes Xandal: bb Justin
 [12:27] You: bye Justin
 [12:27] Justin Clark-Casey: see you hiro, chris
 [12:27] Digi Fly: bye justin
 [12:27] Justin Clark-Casey: see you folks
 [12:27] Bill Humphries: kk, thanks for your time Justin
 [12:28] Rip Oxbar: i heard Microsoft is going to put a opensim package together...is that true?
 [12:28] Hiro Protagonist: LOL
 [12:28] Digi Fly: ow. that will kick SL back.
 [12:28] Bill Humphries: I don't know. One of their developer evangelists is interested in OpenSim.
 [12:28] Hiro Protagonist: I seriously doubt any such thing is happening
 [12:29] Hiro Protagonist: who knows though
 [12:29] Rip Oxbar: Hang on ill get you a article
 [12:29] Axaes Xandal: Is the smudgy-terrain-bug know already?
 [12:29] Axaes Xandal: it seems to be spreading
 [12:30] Rip Oxbar: http://secondlife.reuters.com/stories/2008/07/18/microsoft-eyes-integration-between-opensim-and-windows-live-id/
 [12:30] Bill Humphries: okay, I'm the last person in the metaverse who should express an opinon about that.
 [12:30] Hiro Protagonist: Heh
 [12:30] Hiro Protagonist: I have to discount that article without even reading it
 [12:30] Hiro Protagonist: the guy that writes those articles is notorious for not checking his facts
 [12:30] Bill Humphries: alright, I'm going to dash
 [12:30] Hiro Protagonist: he even got Zero's real name wrong recently
 [12:31] Rip Oxbar: Ahh ok
 [12:31] Hiro Protagonist: Take care Bill
 [12:31] Bill Humphries: see you all next week here
 [12:31] Hiro Protagonist: Thanks for coming Sir
 [12:31] Rip Oxbar: cu later bill
 [12:31] You: bye Bill, thanks
 [12:31] Bill Humphries: Office Hours tomorrow on Prep for the Beta
 [12:31] G2 Proto: that article was twisted
 [12:31] Bill Humphries: I'll send the reminder out to the gridnauts list.
 [12:31] Rip Oxbar: Beta when?
 [12:31] Rip Oxbar: here?
 [12:31] Hiro Protagonist sighs
 [12:31] G2 Proto: G2 my company is working on some asp.net mods for admin for open sim
 [12:31] Bill Humphries: http://wiki.secondlife.com/wiki/Open_Grid_Public_Beta
 [12:32] G2 Proto: yet Eric claimed it was Microosft
 [12:32] Rip Oxbar: ok thanks bill
 [12:32] G2 Proto: they are working on no OpenSim package only with us to work with the community
 [12:32] Hiro Protagonist: Eric is a gossip factory
 [12:32] Rip Oxbar: ive been so busy in rl i havnt been keeping up im sorry to say
 [12:32] G2 Proto: yes he really went on a tangent
 [12:32] Hiro Protagonist: you can even quote me on that
 [12:32] Rip Oxbar: haha
 [12:32] G2 Proto: lol
 [12:32] G2 Proto: we wont be doing nay more onterviews
 [12:33] Hiro Protagonist: we have adopted the posture that we will release a press release to him if we have anything for him to report
 [12:33] Hiro Protagonist: (OSGrid, that is)
 [12:33] Rip Oxbar: haha
 [12:33] G2 Proto: we certainly do not plan to compete with Sl which is what is said, neither G2 nor MS
 [12:33] Rip Oxbar: so hes that bad then huh
 [12:33] G2 Proto: just offer a login option is all for opensim
 [12:33] Hiro Protagonist: yeah, fraid so
 [12:34] Hiro Protagonist: G2, dont let him slow ya down
 [12:34] Rip Oxbar: well it would be nice if true though
 [12:34] You: Have t o go now cya
 [12:34] Hiro Protagonist: TC Chris
 [12:34] G2 Proto: I won't but he clearly went for the more suggestive story and could get the MS Dev in trouble
 [12:34] Axaes Xandal: bye Chris
 [12:34] G2 Proto: but don't get me worng
 [12:34] Hiro Protagonist: G2: indeed.
 [12:34] G2 Proto: MS loves OpenSim
 [12:35] Hiro Protagonist: who doesnt?
 [12:35] G2 Proto: they are just trying to find out how to support it
 [12:35] Hiro Protagonist: and they are welcome to do what they will with it of course
 [12:35] Hiro Protagonist: including submit patcehs
 [12:35] G2 Proto: even Steve ballmer is seeing a video from g2 about recreating the MS campus
 [12:35] You: I have a log of office hour pus a couple of screenshots .
 [12:35] G2 Proto: in OpenSim internally
 [12:35] Rip Oxbar: i still dont like L$ being considered for and economy though
 [12:35] You: Bye
 [12:35] G2 Proto: vya Chris
 [12:35] G2 Proto: cya
 [12:35] Hiro Protagonist: Cool Chrios - there's a place on the wiki for the log
 [12:35] Hiro Protagonist: if you would post it that would rock
 [12:35] Hiro Protagonist: *Chris
 [12:36] You: Will do
 [12:36] G2 Proto: they will use OpenSIm internally to teach refactoring too
 [12:36] Digi Fly: bye chris
 [12:36] Hiro Protagonist: tHANKS mAN :d