Chat log from the meeting on 2019-12-17

From OpenSimulator

Revision as of 12:46, 17 December 2019 by Sheera Khan (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

[11:11] Kayaker Magic: I have a question about region caching: It only works for objects in the region? Why not terrain patches?
[11:11] Ubit Umarov: ask ll kayaker
[11:11] Ubit Umarov: :)
[11:11] Andrew.Hellershanks @hg.osgrid.org: It is 10 past now. Looks like it will be just us today. As mentioned last week this will be the last meeting of 2019 due to the holidays. The next meeting will be on January 7 of next year.
[11:11] Leighton.Marjoram @opensimfest.com: makes note:
[11:12] Ubit Umarov: i only activated something viewers had for years and we ignored
[11:13] Ubit Umarov: the way they cache things guess land patchs don't matter much on their 256m regions
[11:13] Kayaker Magic: Gavin: Do you have a feel for what it is caching? Whole objects? All assets in the region? Textures?
[11:13] Ubit Umarov: see the cache is raw lludp packets
[11:14] Ubit Umarov: the object opes
[11:14] Ubit Umarov: just the contents of compressed Updates packets ( and not normal update packets )
[11:15] Ubit Umarov: all the decode and work to convert to rendering things will happen later
[11:15] Ubit Umarov: just plain network trafic saving
[11:15] Kayaker Magic: and terrain is not sent with similar network traffic?
[11:16] Ubit Umarov: don't think they care about the 256 land patchs their regions have
[11:16] Gavin.Hird @grid.xmir.org:8002: when you say terrain patches, do you mean the terrain geometry for a region patch (4x4m)?
[11:16] Ubit Umarov: no
[11:16] Ubit Umarov: terrain is 16x16m units == patches
[11:16] Ubit Umarov: each compressed using DCTs
[11:17] Ubit Umarov: a simplified jpeg
[11:17] Gavin.Hird @grid.xmir.org:8002: I was asking about the geometry
[11:17] Ubit Umarov: viewer side its same
[11:17] Ubit Umarov: viewers have a list of 16x16m things
[11:18] Ubit Umarov: each counts as a object on the viewer rendering stats
[11:18] Gavin.Hird @grid.xmir.org:8002: the terrain expands to a pretty large structure, so if you were to cache that for every region passed through it would fill the cache pretty fast
[11:18] Ubit Umarov: ( they do apply tesselation on top of that..
[11:19] Ubit Umarov: not that big
[11:19] Kayaker Magic: Not on todays hard drives...
[11:19] Ubit Umarov: they get the packet and decode it to 16x16 heights
[11:20] Gavin.Hird @grid.xmir.org:8002: it was designed for when 1 TB drive cost $3 millon
[11:20] Ubit Umarov: that is the basic object
[11:20] Ubit Umarov: later they do tesselation blabla bla
[11:20] Ubit Umarov: but basic is 16x16m heights
[11:20] Ubit Umarov: and as i said that is a object for it
[11:21] Ubit Umarov: so viewer objects cache is not as effective as it could be
[11:21] Gavin.Hird @grid.xmir.org:8002: LL is completely rewriting the cache, so they are well aware of it
[11:22] Ubit Umarov: it could be a more prepared object like in rendering mesh format or so
[11:22] Ubit Umarov: i did hear about textures and mesh cache
[11:22] Ubit Umarov: not objects
[11:22] Ubit Umarov: seen a few viewers of their meetings with tpvs
[11:23] Ubit Umarov: like tpvs are not allowed to do releases for SL from 20th to jan 1st
[11:23] Kayaker Magic: Well, a model like the current region cache could add the terrain packets on the way by...
[11:23] Ubit Umarov: the dcts are a bit heavy
[11:23] Kayaker Magic: Heavy?
[11:23] Ubit Umarov: i plane to cache those region side one day also
[11:23] Ubit Umarov: plan
[11:24] Kayaker Magic: Difficult algorithm to decode?
[11:24] Ubit Umarov: see fft?
[11:24] Ubit Umarov: its half a fft
[11:24] Ubit Umarov: only cosines are used
[11:24] Ubit Umarov: ( like jpeg )
[11:25] Kayaker Magic: But if you cache the packets, the terrain code to decode them again later already exists.
[11:25] Ubit Umarov: and bad
[11:25] Kayaker Magic: Of course, that is all viewer side code.....
[11:25] Ubit Umarov: does generate a lot of noise on terrain
[11:25] Kayaker Magic: The current system is noisy.
[11:26] Ubit Umarov: any sharp terrain is ugly
[11:26] Kayaker Magic: right.
[11:26] Ubit Umarov: creates lots of noise outside
[11:26] Ubit Umarov: nothing new..
[11:26] Ubit Umarov: is side effect of window
[11:26] Kayaker Magic: Known problem: The terrain you see is not quite where the server thinks it is.
[11:27] Ubit Umarov: on ffts we always reduce that by apply a window funtion
[11:27] Ubit Umarov: like hanning etc etc
[11:27] Kayaker Magic: Unless your terrain is very smooth to start with.
[11:27] Ubit Umarov: yeah
[11:27] Ubit Umarov: but noise is very ugly indead
[11:28] Ubit Umarov: terrain must be smooth
[11:29] Ubit Umarov: you can see it on the base of the peeks i made out there
[11:29] Ubit Umarov: and how far they do extend
[11:29] Kayaker Magic: So the viewer is looking at packets as they go by, and some of them are stored in the region cache, but not all of them.
[11:29] Kayaker Magic: Why bother to differentiate?
[11:30] Ubit Umarov: patches are not stored
[11:30] Ubit Umarov: objects, some there is no sense to cache.. they change
[11:31] Ubit Umarov: ( and their position is part of the things stored in cache)
[11:31] Ubit Umarov: so even a position change invalidates that entry
[11:31] Ubit Umarov: again typical SL half thinking
[11:32] Ubit Umarov: but the network side saving is well visible
[11:33] Ubit Umarov: we do see things rez faster
[11:34] Kayaker Magic: I was just imagining terrain patches acting like objects, and imagining them breaking apart and moving in independent 16x16m chunks.
[11:35] Ubit Umarov: no they don't
[11:35] Kayaker Magic: Just in my imagination!
[11:35] Ubit Umarov: bc on top of them there are the terrain textures
[11:35] Ubit Umarov: and those even take in consideration the region global position
[11:36] Ubit Umarov: as you can see on the new warp3d maps
[11:36] Kayaker Magic: They use something like Perlin noise to blend textures together at altitude...
[11:37] Ubit Umarov: thats less that global position
[11:37] Ubit Umarov: so terrain on 2 nearby regions with similar settings does transition smothly
[11:38] Ubit Umarov: we actual have those numbers to help that to happen
[11:38] Kayaker Magic: But the textures are different for each viewer!
[11:38] Ubit Umarov: no they are not
[11:38] Kayaker Magic: They are different every time you log out and back in again!
[11:38] Ubit Umarov: they are assets
[11:38] Ubit Umarov: then you have something wrong somewhere :p
[11:39] Ubit Umarov: terrain textures are normal assets
[11:39] Ubit Umarov: viewers are told about them on region handshake and regionInfo
[11:39] Andrew.Hellershanks @hg.osgrid.org: I thought that was a known issue. Looking at the same region with different viewers at the same time shows a slightly different look to the land.
[11:40] Kayaker Magic: Yes, but as altitude changes, two textures are blended together at the transistion, that blending changes every time you log in.
[11:40] Gavin.Hird @grid.xmir.org:8002: it does not look exactly the same for any given viewer
[11:40] Ubit Umarov: seen that with some dayturn version only
[11:40] Ubit Umarov: but now its closer to fs
[11:40] Ubit Umarov: the math must be as SL does
[11:40] Ubit Umarov: at least the same on all viewers
[11:40] Ubit Umarov: or results will be dif of course
[11:41] Ubit Umarov: our maps where totally wrong on that
[11:41] Ubit Umarov: a lot better on warp3d now
[11:42] Ubit Umarov: that means how global position is used, how the 4 textures do blend with height etc etc
[11:43] Ubit Umarov: ( and yes some perk noise is added )
[11:43] Ubit Umarov: see warp3d source for details :p
[11:44] Ubit Umarov: i mean our module
[11:44] Kayaker Magic: "the code is the documentation"
[11:44] Ubit Umarov: of course :p
[11:44] Ubit Umarov: if the cpu can read it why can't you??
[11:44] Andrew.Hellershanks @hg.osgrid.org: :)
[11:45] Gavin.Hird @grid.xmir.org:8002: the cpu cannot execute a might do statement
[11:45] Andrew.Hellershanks @hg.osgrid.org: There is 15 minutes remaining before the top of the hour. What did everything think about this years OSCC?
[11:46] Andrew.Hellershanks @hg.osgrid.org: I was interested to hear about the SceneGate viewer and Echo Voice.
[11:46] Leighton.Marjoram @opensimfest.com: Its always nice to hear what people are doing with Opensim, particularly from an education point of view.
[11:46] Ubit Umarov: a bit less All move out of opensim to this great other thing
[11:46] Ubit Umarov: but still there
[11:46] Leighton.Marjoram @opensimfest.com: There seemed to be a fair bit of stuff not really related to Opensim but its was good.
[11:47] Ubit Umarov: and fun the one.. we all should move.. the thing is now dead... but we should move
[11:47] Kayaker Magic: Gavin: will their work on Echo Voice be usable by all other viewers?
[11:47] Ubit Umarov: you mean mumble
[11:47] Kayaker Magic: (opensim viewers)
[11:47] Gavin.Hird @grid.xmir.org:8002: I only noticed that after having bitched about viewer security for an hour, they provided a http link for the viewer download
[11:47] Ubit Umarov: :)
[11:47] Gavin.Hird @grid.xmir.org:8002: no
[11:48] Leighton.Marjoram @opensimfest.com: :)
[11:48] Gavin.Hird @grid.xmir.org:8002: it will not be usefull at all with age old libs, and no Linux and Mac support
[11:48] Ubit Umarov: i was promised the region side updated mumble module last year
[11:48] Ubit Umarov: now it was renamed and placed on that what was the name thing
[11:49] Andrew Hellershanks: They talked about replacing the SLVoice module with something that would talk to the actual voice system. It sounds like it would be possible to use Echo Voice with other viewers.
[11:49] Gavin.Hird @grid.xmir.org:8002: not if there is no code that supports it
[11:49] Andrew Hellershanks: Linux and Mac support are on their ToDo list. They are also updating the code to use modern, readily available libraries.
[11:49] Ubit Umarov: they just did some automation of the user side module change
[11:49] Bill Blight: Echo voice is just the old mumble renamed
[11:49] Gavin.Hird @grid.xmir.org:8002: Linux and Mac support was on their to-do list for everyting they presented
[11:50] Ubit Umarov: on old mumble usage we would need to run a batch file to change the bin
[11:50] Gavin.Hird @grid.xmir.org:8002: does anyone of them even own a Mac?
[11:50] Ubit Umarov: now finaly they added a way for the viewer to use the right one
[11:50] Andrew Hellershanks: Bill, it might be at the moment but won't be for long.
[11:50] Bill Blight: scenegate is based on almost 2 year old code
[11:50] Ubit Umarov: But some wrong ideas on how to do that
[11:50] Bill Blight: We will see Andrew
[11:50] Ubit Umarov: details
[11:51] Ubit Umarov: they did promissed to talk more with us
[11:51] Ubit Umarov: lets see
[11:51] Gavin.Hird @grid.xmir.org:8002: nobody have seen their source which they stressed was so important was open source
[11:51] Ubit Umarov: well lisa actually did shown be a bug
[11:51] Bill Blight: much violation of the GPL going on there
[11:51] Ubit Umarov: we had for ages and don't remember any reporting
[11:52] Ubit Umarov: "request teleport" uff
[11:52] Gavin.Hird @grid.xmir.org:8002: it is the same talk they did back in 2016 when Lisa took ownership of a Kokua repository that had not seen an update for 8 years
[11:52] Andrew Hellershanks: It remains to be seen what will happen with SceneGate. There have been several starts at OS specific viewers and they all pretty much died out. If the people behind SceneGate can do what they talked about we may finally have an OS specific viewer that would open the doors for other changes in the OS code base.
[11:52] Bill Blight: LOL
[11:53] Bill Blight: I was part of that group for a while, ummm yeah , no comment
[11:54] Ubit Umarov: ok i did got a spam of tp offers :)
[11:54] Ubit Umarov: so yeah now works
[11:54] Ubit Umarov: selby could tell
[11:54] Andrew Hellershanks: Ubit, What were you testing?
[11:54] Ubit Umarov: he is IMA
[11:55] Ubit Umarov: that you received was not working andrew
[11:55] Ubit Umarov: "request teleport"
[11:55] Kayaker Magic: I got a request
[11:55] Leighton.Marjoram @opensimfest.com: me too
[11:55] Andrew Hellershanks: Since when has that been broken?
[11:55] Ubit Umarov: for some reason we where not dispatching the special im
[11:55] Ubit Umarov: some told me since ever
[11:56] Ubit Umarov: and i found no code at all doing it
[11:56] Andrew Hellershanks: ok.
[11:56] Ubit Umarov: easy fix
[11:56] Ubit Umarov: done it during oscc :p
[11:57] Gavin.Hird @grid.xmir.org:8002: LL sent out a 5 page primer on how to covert viewer repositories from HG to git
[11:57] Gavin.Hird @grid.xmir.org:8002: so this will be fun
[11:57] Andrew Hellershanks: Most times I see offers of TPs going out but not requests for a TP.
[11:57] Gavin.Hird @grid.xmir.org:8002: completely new workflow
[11:57] Ubit Umarov: terry users also reported a issue on objects position on rez
[11:57] Ubit Umarov: fixed now
[11:57] Andrew Hellershanks: LL is moving to git?
[11:57] Ubit Umarov: yes
[11:57] Gavin.Hird @grid.xmir.org:8002: Bitbucket is dripping HG support
[11:57] Ubit Umarov: all must move to git
[11:57] Gavin.Hird @grid.xmir.org:8002: so we have no choice
[11:58] Ubit Umarov: all at bitbucket
[11:58] Kayaker Magic: Bill has installed the latest version here with changes done during OSCC?
[11:58] Ubit Umarov: bill made a automatic tool
[11:58] Andrew Hellershanks: Did they say why they want to drop HG support?
[11:58] Gavin.Hird @grid.xmir.org:8002: from feb you cannot create HG repositories and come July they will remove all of them
[11:58] Bill Blight: yes this region auto updates on git commits
[11:58] Ubit Umarov: this region is automaticly updated to master
[11:58] Kayaker Magic: Cool
[11:58] Gavin.Hird @grid.xmir.org:8002: so all the history not converted will be gone
[11:59] Ubit Umarov: so no.. im not doing any commit now
[11:59] Ubit Umarov: or Bill would need to rush prevent the thing from killing us :)
[11:59] Kayaker Magic: I'm confused: what does HG stand for in this conversation?
[11:59] Gavin.Hird @grid.xmir.org:8002: they also said they will completely stop to fix any issues or bugs related to Windows 7
[11:59] Andrew Hellershanks: We are just at the top of the hour. For anyone who came late this is the last meeting of 2019 due to Christmas and New Years. The next meeting will be on January 7 of next year.
[12:00] Ubit Umarov: LL and tpvs use HG ( mercurial) in place of GIT
[12:00] Andrew Hellershanks: Kayaker, HG refers to the Mercurial version control system.
[12:00] Ubit Umarov: similar system
[12:00] Gavin.Hird @grid.xmir.org:8002: because their legal does not even allow them to have Win 7 installed after official support is dropped in a few days
[12:00] Kayaker Magic: I wondered how you could use git to Hyper Grid.
[12:00] Ubit Umarov: *facepalm*
[12:01] Andrew Hellershanks: Kayaker, magic. ;)
[12:01] Gavin.Hird @grid.xmir.org:8002: what makes it possible to Hyper Grid will from now on be in git
[12:01] Kayaker Magic: Collisions in acronym space....
[12:01] Andrew Hellershanks: The reference to Bitbucket put it in context.
[12:01] Ubit Umarov: kay remember the chemical symbol of mercury ?
[12:02] Kayaker Magic: I see how they used that name....
[12:02] Ubit Umarov: yes it is Hg ;)
[12:02] Andrew Hellershanks: hm... that means I will be able to remove Mercurial from my computer in the new year.
[12:03] Ubit Umarov: ll even published instructions or how they did it to help tvps
[12:03] Ubit Umarov: i was told
[12:03] Gavin.Hird @grid.xmir.org:8002: 5 pages
[12:04] Ubit Umarov: well i was on a meeting where Oz told that also
[12:05] Ubit Umarov: so you liked oscc?
[12:05] Ubit Umarov: i kinda liked the first 2 sections
[12:05] Ubit Umarov: :)
[12:05] Kayaker Magic: It is always fun.
[12:06] Kayaker Magic: I've noticed the education people were taking over for a while, now they say OpenSim Education is dead. ;0
[12:06] Andrew Hellershanks: I liked it. I always find some panels worth attending to hear what people are doing with OpenSim or what new things are in the works from various groups of people.
[12:06] Ubit Umarov: did not expect animesh to be such sucess
[12:06] Ubit Umarov: its a old thing here at osg now
[12:06] Gavin.Hird @grid.xmir.org:8002: it is still pretty unexplored
[12:07] Ubit Umarov: when i seen the region on ubode and did sit on the floor
[12:07] Gavin.Hird @grid.xmir.org:8002: it is more difficult to make than just rigging something with avastar
[12:07] Andrew Hellershanks: I haven't seen the panel about OS being dead. I need to find that in the video of the recorded sessions.
[12:07] Ubit Umarov: felt the need for the bodyguards Bill gave me
[12:07] Ubit Umarov: so did wear them :)
[12:08] Ubit Umarov: those we have there protecting our door also :)
[12:08] Andrew Hellershanks: I was happy that the OSCC event was run using 0.9.1 this year.
[12:08] Ubit Umarov: and amasing didn't crash once
[12:08] Leighton.Marjoram @opensimfest.com: me too Andrew
[12:08] Gavin.Hird @grid.xmir.org:8002: I bought a few sets of FP animesh scripts in SL that also worked ok here
[12:09] Andrew Hellershanks: I've got three more items on my ToDo list as a result of the keynote and VIP Q&A sessions.
[12:09] Ubit Umarov: yeap ppl doing fun things with them at sl
[12:10] Kayaker Magic: I want to find the time to play with animesh...
[12:10] Kayaker Magic: so many projects, so little time....
[12:10] Ubit Umarov: :)
[12:11] Andrew Hellershanks: I need to find the time to finish some old inworld projects. Been too busy with RL projects.
[12:11] Ubit Umarov: but oscc had less ppl this here no?
[12:11] Ubit Umarov: counted 60 something 70 at most on keynotes
[12:12] Ubit Umarov: expo 3 basicly empty
[12:12] Andrew Hellershanks: It felt that way to me. I counted the number of people in the four keynote regions at one point and it seemed less than past years. One region only had one user and the the other less than half dozens.
[12:12] Gavin.Hird @grid.xmir.org:8002: as I said before the timing is not good
[12:12] Gavin.Hird @grid.xmir.org:8002: December is busy everywhere
[12:13] Andrew Hellershanks: It has been in December before but not as late as this.
[12:13] Leighton.Marjoram @opensimfest.com: Its a bad time, they start advertising 10 days before and dont really reach out beyond people who have already been before.
[12:13] Gavin.Hird @grid.xmir.org:8002: that too
[12:13] Kayaker Magic: I always went to sit in the least full region to give the poor servers a break.
[12:13] Gavin.Hird @grid.xmir.org:8002: I got an email invite with program 2 days before
[12:13] Andrew Hellershanks: I was seeing a large number of familiar names in the audience and fewer whose names were not.
[12:14] Ubit Umarov: servers didn't felt anything
[12:14] Kayaker Magic: lol
[12:14] Ubit Umarov: sitting people is zero load
[12:14] Leighton.Marjoram @opensimfest.com: IT could be so much more but it feels that people are exhausted and trying to keep up with the ever shifting technology landscape.
[12:15] Andrew Hellershanks: Might be a carry over from past years. Conference staff would ask people to sit after they arrived at a keynote region.
[12:15] Leighton.Marjoram @opensimfest.com: VR/AR is no where near ready and WebXR is probably the one to watch but still no where near ready.
[12:15] Ubit Umarov: we should to that anywhere
[12:15] Kayaker Magic: They did ask people to sit a few times.
[12:15] Ubit Umarov: hate lbsa ppl now
[12:15] Leighton.Marjoram @opensimfest.com: Virtual worlds still have a lot to offer
[12:15] Ubit Umarov: love to stand there
[12:16] Ubit Umarov: inc osgrid boss shhh don't tell :p
[12:16] Andrew Hellershanks: One thing that wasn't announced properly was their load test before the conference. I didn't hear about it until it was over.
[12:16] Gavin.Hird @grid.xmir.org:8002: SL now has a competition for best names
[12:16] Gavin.Hird @grid.xmir.org:8002: when you are again allowed a full name at a price
[12:16] Ubit Umarov: yeah i was seening 50 ppl at lbsa when gavin told me it had happened
[12:17] Gavin.Hird @grid.xmir.org:8002: I only knew because of a tweet I happened to see
[12:17] Ubit Umarov: they are selling name chaging to premium users
[12:17] Andrew Hellershanks: The attendance for the load test wasn't that high from what I heard.
[12:17] Gavin.Hird @grid.xmir.org:8002: yep
[12:17] Kayaker Magic: I went and threw snowballs at everyone.
[12:17] Gavin.Hird @grid.xmir.org:8002: you sure did :-))
[12:17] Ubit Umarov: no idea why they made those changes to names
[12:17] Andrew Hellershanks: I don't use twitter so mentioning the load test there isn't good. They should have sent out a notice on the OS mailing lists.
[12:17] Ubit Umarov: in past
[12:17] Gavin.Hird @grid.xmir.org:8002: they felt a bit wet
[12:18] Gavin.Hird @grid.xmir.org:8002: you should let a few of them knock people over
[12:18] Ubit Umarov: they don't know or ignore OS mail lists andrew
[12:18] Andrew Hellershanks: Let's wrap up this last meeting of the year.
[12:19] Kayaker Magic: Oh! Ubit: anything to say about the LocalTeleport OOSL function?
[12:20] Ubit Umarov: either that i don't know that one?
[12:20] Kayaker Magic: I saw it appear in the Wiki,
[12:20] Ubit Umarov: mb you mean objectTeleport
[12:20] Ubit Umarov: thats the fun one
[12:21] Kayaker Magic: lhttp://opensimulator.org/wiki/OsLocalTeleportAgent
[12:21] Andrew Hellershanks: I need to update the Wiki with the OS version number when some of the newest functions appeared.
[12:21] Kayaker Magic: You didn't write osLocalTeleportAgent? Who did?
[12:21] Ubit Umarov: ok i remember that now
[12:21] Ubit Umarov: it does tp setting avatar lookat etc
[12:22] Ubit Umarov: if viewers allow
[12:22] Ubit Umarov: and velocity etc
[12:22] Ubit Umarov: well its what it tells on wiki
[12:23] Ubit Umarov: made it ages ago :)
[12:23] Kayaker Magic: I'm interested in using it to solve the problem of where the avatar lands when standing up.
[12:23] Ubit Umarov: well some viewer settings my get in the way
[12:23] Ubit Umarov: don't remember details
[12:23] Kayaker Magic: Since osLocalTeleportAgent has threat level NONE, it can be used by anyone.
[12:24] Ubit Umarov: and does a fast tp
[12:24] Andrew Hellershanks: Kayaker, you want the avatar to land lying down??
[12:24] Ubit Umarov: its just setting postion etc
[12:24] Ubit Umarov: but viewers are the master
[12:24] Ubit Umarov: on several avatar things
[12:24] Kayaker Magic: When I stand up from this chair, I land on top of the bar.\
[12:24] Ubit Umarov: so some may not work
[12:25] Ubit Umarov: lookat is a case i think
[12:25] Kayaker Magic: I want to write furniture scripts that force the avatar to stand up BEHIND the chair in cases like this
[12:25] Andrew Hellershanks: Kayaker, I think that is because it moves you forward so you don't wind up on top of the thing you were sitting on.
[12:25] Gavin.Hird @grid.xmir.org:8002: I have made some changes to a script or the parameters for that to happen on some chairs
[12:25] Kayaker Magic: Yes, and I'd like thee to be a way to adjust that default behavior
[12:26] Andrew Hellershanks: That would be a bit odd to wind up behind a chair you were sitting in when you stand.
[12:26] Kayaker Magic: Landing on top of the bar is also a bit odd
[12:26] Gavin.Hird @grid.xmir.org:8002: not necessarily
[12:26] Andrew Hellershanks: True.
[12:26] Andrew Hellershanks: That would also work when sitting on stools without backs to them.
[12:26] Gavin.Hird @grid.xmir.org:8002: place a chair close to a window looking out and when you stand up you fall out the window
[12:26] Ubit Umarov: we should need a few more parameters and a OSSL to set them
[12:27] Kayaker Magic: I submitted a mantis asking for a way to set those paramters!
[12:27] Ubit Umarov: OSSLSitTarget i guess
[12:27] Andrew Hellershanks: Gavin, that sounds like something fun to do for unsuspecting users. :)
[12:27] Gavin.Hird @grid.xmir.org:8002: right
[12:27] Gavin.Hird @grid.xmir.org:8002: it is
[12:27] Gavin.Hird @grid.xmir.org:8002: but
[12:27] Gavin.Hird @grid.xmir.org:8002: it becomes tedious fast
[12:27] Ubit Umarov: and more parameters to add to all dbs oars etc yeack
[12:27] Ubit Umarov: pain :P
[12:35] Gavin.Hird @grid.xmir.org:8002: I got all the regions converted except those running on PI
[12:36] Gavin.Hird @grid.xmir.org:8002: but they use the same robust
[12:36] Ubit Umarov: odd that error
[12:36] Gavin.Hird @grid.xmir.org:8002: it was a very oddball script error
[12:37] Gavin.Hird @grid.xmir.org:8002: but it is not there any more, so will probably never find out
[12:37] Ubit Umarov: :)
[12:38] Gavin.Hird @grid.xmir.org:8002: something with a faulty script got loaded on the region, which triggered a rollback, and then the region simply stopped loading the rest
[12:38] Ubit Umarov: dbs should not rollback
[12:38] Gavin.Hird @grid.xmir.org:8002: which left one region completely empty and one with a less than 50 items
[12:39] Gavin.Hird @grid.xmir.org:8002: no, not there

Personal tools
General
About This Wiki