Chat log from the meeting on 2013-10-15

From OpenSimulator

Revision as of 21:54, 24 October 2015 by Fritigern (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
[11:03] Andrew Hellershanks is online.
[11:04] Teravus Ousley spots a hyphenated last name developer
[11:04] Nebadon Izumi: lol
[11:04] Simulator Version v0.5 shouts: OpenSim 0.8.0 Dev          daf44cc: 2013-10-12 20:47:10 -0400 (Unix/Mono)
[11:04] Nebadon Izumi: hello Justin
[11:05] Andrew Hellershanks: Your mic is on, Nebadon
[11:05] Teravus Ousley: hello Mr CC
[11:05] Nebadon Izumi: really
[11:05] Richardus Raymaker: oh nebadon. found the problem with disapearing prims and mesh. it where the singularity release
[11:05] Andrew Hellershanks: RL is calling. bbiab.
[11:05] Teravus Ousley: mic is now off
[11:05] Justin Clark-Casey: hi nebadon, teravus, folks
[11:05] Dahlia Trimble is online.
[11:05] Nebadon Izumi: ya something must be very broken in Singularity
[11:05] Justin Clark-Casey: I was wondering why I was hearing water
[11:05] Nebadon Izumi: there was no indication on my end mic was on
[11:06] Nebadon Izumi: water?
[11:06] Andrew Hellershanks: Nebadon, It is off now.
[11:06] Teravus Ousley: fountain sounds, traveling over the voice channel
[11:06] Richardus Raymaker: i have 1 probleem seen, simborder crossings sometimes shoot you back to the begin of the sim. today ons tandalone i shoot to other place and not moved back
[11:06] Nebadon Izumi: wow really?
[11:06] Nebadon Izumi: my speakers are not even turned on
[11:06] Nebadon Izumi: i dont hear anything
[11:06] Nebadon Izumi: its totally quiet in here
[11:07] Teravus Ousley: :) Maybe you've got your sound card set to record the wave output
[11:07] Richardus Raymaker: hi justin.
[11:07] Nebadon Izumi: oh
[11:07] Nebadon Izumi: you know what
[11:07] Andrew Hellershanks: wow. This is a very different look for justin.
[11:07] Nebadon Izumi: i have a fiber optic cable looping back on itself
[11:07] Nebadon Izumi: for fraps
[11:07] Richardus Raymaker: yes
[11:07] fine.Dreadlow hello to all
[11:08] Richardus Raymaker: lol nebadon. thats not the loopback people need :)
[11:08] Justin Clark-Casey: I was uing this avatar to test something and now I can't be bothered to change it
[11:08] Justin Clark-Casey: though if Nebadon wants me to change I will
[11:08] Justin Clark-Casey: it is his av after all
[11:08] Nebadon Izumi: I didnt make that, SLB made that
[11:08] Richardus Raymaker: as long not both wear the same avatar
[11:08] fine.Dreadlow wow fibre optic for home will be nice :D
[11:08] Nebadon Izumi: and no thats fine its a freebie at my sandbox
[11:08] Teravus Ousley is still barefoot suited with a red highlight :)
[11:08] Nebadon Izumi: oh its Audio Fiber Optic
[11:08] Nebadon Izumi: not network haha
[11:09] Nebadon Izumi: my soundcard has SPDIF in/out
[11:09] Nebadon Izumi: and i have a fiber cable connecting out to in so i can capture
[11:09] fine.Dreadlow lool ah i allready have it :D
[11:09] Teravus Ousley: well.. Welcome everyone to the weekly OpenSimulator developer group/meeting.
[11:09] fine.Dreadlow hello Dahlia
[11:09] Justin Clark-Casey: been playing around with getting a gamepad working with oculus street view
[11:10] fine.Dreadlow oh great i didn't know about :D
[11:10] Nebadon Izumi: nice, how is that going Justin?
[11:10] Justin Clark-Casey: cool to be able to virtually drive through the streets
[11:10] Andrew Hellershanks: I was having a problem with offline messaging not working. I finally fixed the problem with a minor change to the OpenSim code and the PHP file.
[11:10] Teravus Ousley: I'm glad that you resolved, it Andrew
[11:10] Justin Clark-Casey: anyway,I will stop being offtopic
[11:10] logger sewell: hey everyone
[11:11] Justin Clark-Casey: Hi logger
[11:11] Dahlia Trimble: hi
[11:11] Richardus Raymaker: hi logger
[11:11] Pathfinder.Lester howdy folks
[11:11] Teravus Ousley thinks it would be interesting to ride around town with an augmented reality set that displays interest items
[11:11] Richardus Raymaker: hi pathfinder
[11:11] Andrew Hellershanks: The offline PHP file was being called with a path. That is sort of non-standard way of passing parameters to a PHP file. If the webserver was configured a certain way you can't pass the data using a path.
[11:12] Richardus Raymaker: mac donalds, burger king. icecream bar starbucks. not sure if call that intressting
[11:12] Richardus Raymaker: :)
[11:12] Andrew Hellershanks: I changed the offline files to just use the normal file.php?blah type format for passing data.
[11:12] Nebadon Izumi: I took fleep to my Sundbyerg Model which is a real world representation of a street there, with bunch of NPCs and stuff
[11:12] Nebadon Izumi: she was wearing the Occulus Rift, she was totally blown away
[11:12] Pathfinder.Lester nice
[11:12] Teravus Ousley: :)
[11:12] Justin Clark-Casey: it's cool, though the low res screen is annoying now
[11:12] Andrew Hellershanks: The changes were trivial once I understood what why the offline stuff wasn't working.
[11:12] Justin Clark-Casey: I will probably end up shelling out for the hi-res consumer version
[11:13] Teravus Ousley thinks he will shell out for the hi-rez version also
[11:13] Justin Clark-Casey: hi dahlia, pathfinder
[11:13] Dahlia Trimble: hi
[11:13] Dahlia Trimble: you bought a rift justin?
[11:13] Richardus Raymaker: hi dahlia
[11:13] Teravus Ousley: So, since last meeting we had a Minor Gitatstrophy that was resolved with a server side git reset --hard
[11:14] Nebadon Izumi: I am afraid to buy a Rift, because i get motion sick pretty easily
[11:14] Nebadon Izumi: everyone i hear who has one seems to feel ill after using it
[11:14] Andrew Hellershanks still has to file a mantis with the code changes for the offline messaging fix.
[11:14] Justin Clark-Casey: dhalia: yeah, had one for a while
[11:14] Dahlia Trimble: do you get sick wearing it?
[11:14] Justin Clark-Casey: simulation sickness is quite a big factor, though you do get more used to it over time
[11:14] Teravus Ousley: If you encounter a message like 'mismatched heads', you may need to re-clone or git pull -f
[11:15] Teravus Ousley: It wasn't that way for very long, so I doubt many people are affected
[11:15] Dahlia Trimble: dont think I'll be getting a rift but a google glass is kinda interesting
[11:15] Justin Clark-Casey nods
[11:15] Teravus Ousley: Also, there's a new database connector for Postgres
[11:16] Justin Clark-Casey: who is committing those postgres patches?
[11:16] Teravus Ousley: Thanks to fernando oliveirax
[11:16] Dahlia Trimble: I did the one that "messed up" the repo :) *hides*
[11:16] Justin Clark-Casey: I knew it was your fault, Dahlia :)
[11:16] Teravus Ousley: I pushed the last two.
[11:17] Teravus Ousley: I had to hack the author info because the patches didn't include them
[11:17] Richardus Raymaker: weird question, because i never heared about a version 0.7.7 and 9.7.8 or 0.7.9 and still someone is talking about the numbers
[11:17] Justin Clark-Casey: because I'm paranoid, I did double check the dll binary and re-added it to record the version number
[11:17] Justin Clark-Casey: I'm paranoid about binary stuff that comes in from non-core ppl
[11:17] Andrew Hellershanks sighs
[11:17] Robert Adams will now add variable sized terrain patches for Postgres
[11:17] Andrew Hellershanks: I'm needed again. bbiab
[11:18] Teravus Ousley: It's good to be paranoid, and it's good to have a paranoid person on the team :D
[11:18] Justin Clark-Casey: richardus: I made an autocratic decision to bump to 0.8, partly because the plan is to update the min mono/.net versions
[11:18] Dahlia Trimble: I didnt see the binary in there or I would have been paranoid too
[11:18] Richardus Raymaker: Yes. justin. but thats why the numbers are so strange
[11:18] fine.Dreadlow i've not found a reference for postgres installation.. is there any? or is it future stuff?
[11:18] Justin Clark-Casey: richardus: how are they strange?
[11:18] Nebadon Izumi: ya that is something we need to talk about soon, the minimum mono version
[11:19] Richardus Raymaker: not heared about a versionnumbered 0.7.8
[11:19] Justin Clark-Casey: some time ago we had a discussion about going to .net 4 and the result was to do it after the next release
[11:19] Justin Clark-Casey: that will make minimum mono 2.8, realistically 2.10
[11:19] Teravus Ousley: fine.Dreadlow: postres is a new connector, that you can configure in your StandaloneCommon.ini/Gridcommon.ini
[11:19] Justin Clark-Casey: but ever major linux distro is on at least 2.10 now
[11:19] Robert Adams: the mono version is gated by what the Linux distributions are shipping with
[11:20] Justin Clark-Casey: richardus: we don't need to monotonically increase the version numbers
[11:20] Andrew Hellershanks returns
[11:20] Dahlia Trimble: mono version should be the minimum that is needed to support the feature set
[11:20] Robert Adams: we are technically able to goto the latest version... but we don't want to make installation on common Linux distributions difficult
[11:20] Teravus Ousley thinks that once we move to 4.0, we get cool things like ConcurrentDictionary :D
[11:20] Justin Clark-Casey: yes, so going to 2.8 as a min is now reasonable - that's all we need to get to .NET 4.0
[11:20] Richardus Raymaker: Justin. its just something i read on forum. and i think he use wrong version numbers
[11:21] Justin Clark-Casey: yes, and being on .NET 4.0 will make the modInvoke stuff work better
[11:21] Richardus Raymaker: mono 2.4 is really old
[11:21] Justin Clark-Casey: richardus: he?
[11:21] Justin Clark-Casey: oh, someone else?
[11:21] Richardus Raymaker: its a modified 0.7.6 version. but i read the following line.
[11:21] Richardus Raymaker: "- 0.7.6 contains all patches from 0.7.7 , 0.7.8 and 0.7.9 the current dev version number is 0.8.0 and will be the following version when its stable enough.
[11:21] Justin Clark-Casey: ppl often seem to get mixed up with version numbers for some reason
[11:21] fine.Dreadlow my oldest server have mono 2.10 and this is really old :D
[11:21] Justin Clark-Casey: richardus: well, that's wrong
[11:22] Justin Clark-Casey: fine: the hold up was Debian, but they are now 2.10
[11:22] Richardus Raymaker: thats what i mean, so  my feleing is right
[11:22] fine.Dreadlow ah k
[11:22] Nebadon Izumi: who wrote that and where?
[11:23] Richardus Raymaker: you have the link. then i keep it in the middle for now
[11:23] Nebadon Izumi: ya definitely not right, but oh well not a big deal
[11:23] Teravus Ousley: New public grid people, if you have not already sent your grid info to HyperGrid Business, it might be good to do so. She tries to keep track of stats and stuff
[11:24] Richardus Raymaker: ok
[11:24] fine.Dreadlow i'll do cause its my third day with hypergrid and its not fully working for now ;)
[11:24] Teravus Ousley:
[11:24] Nebadon Izumi: Justin I have to say i think launcing mono 3.2.3 with the --server option definitely has some impact
[11:24] Justin Clark-Casey: nebadon: good or bad?
[11:24] Nebadon Izumi: ive not had any problems really
[11:24] Nebadon Izumi: good
[11:24] Nebadon Izumi: the region seems to start faster
[11:24] Nebadon Izumi: and its definitely using more threads
[11:24] Richardus Raymaker: cool. need to remember that switch
[11:24] Justin Clark-Casey: they said that it changes some threadpool characteristiics
[11:25] Richardus Raymaker: did you wrote that on the forum ?
[11:25] fine.Dreadlow i like the newest version much its great
[11:25] Nebadon Izumi: yes its supposedly more aggressive, and it is definitely using more threads
[11:25] Nebadon Izumi: smxy had to increase the limit
[11:25] Nebadon Izumi: when he started using --server regions wouldnt start
[11:25] Nebadon Izumi: because the thread limit on his kernel was 1024
[11:25] Dahlia Trimble: --server kinda concerns me considering all the treadpool abuse we have
[11:25] Nebadon Izumi: where as my server was set for 65535 or something
[11:25] Nebadon Izumi: out of the box
[11:26] Nebadon Izumi: once he upped it it ran fine
[11:26] Robert Adams suspects there is thread starving under heavy load
[11:26] Nebadon Izumi: well so far Dahlia i have only seen improvement
[11:26] Nebadon Izumi: which is good thing in my book
[11:27] Dahlia Trimble: you only get as many threads as there are logical cores. any more is just more layers of management overhead
[11:27] Dahlia Trimble: it may start faster but context switching is probably slower
[11:27] Nebadon Izumi: we are running it here right now
[11:27] Teravus Ousley has a hexicore processor and can do up to two hardware threads per processor for a total of 12.
[11:28] Nebadon Izumi: last couple weeks
[11:28] Dahlia Trimble: need good empirical data
[11:28] Andress Renault is online.
[11:28] fine.Dreadlow thats why i've builded more then one virtual machine on esx
[11:28] Nebadon Izumi: we also had to increase the nursery size for sgen
[11:28] Nebadon Izumi: to 64m
[11:28] Nebadon Izumi: that seems to be working well also
[11:28] Robert Adams: it's probably another discussion who's resolution is "it depends"
[11:28] Nebadon Izumi: i have added these options to my mono 3.2.3 compile instructions
[11:29] Nebadon Izumi: here >
[11:30] Richardus Raymaker: i want to play with virtual machines to. but i need to test it with virtualbox. the only change i now make. i run it on ssd
[11:30] Justin Clark-Casey: robert adams: On another note, are we ready to switch over to bulletsim as the default phys engine now?
[11:30] Nebadon Izumi: we have been testing on a varitey of different hardware too
[11:30] Nebadon Izumi: probably about 6 different servers
[11:30] fine.Dreadlow yes thats a great idea :) ssd's are faster
[11:30] Richardus Raymaker: nebadon, did you explained the new settings for mno on osgrid forum ?
[11:31] Teravus Ousley wonders if people have time to do stuff.. If so, it might be helpful to take a look at and verify/close issues
[11:31] Nebadon Izumi: Justin I think we are close, I am not sure its 100% ready, yet though however again we are running it here on this region, and most of my regions do run it now too, works pretty good
[11:31] Robert Adams: justin: yes, switch over is a good thing
[11:31] Justin Clark-Casey: is it one of those things where switching would make it easier to get to suitable ready?
[11:32] Nebadon Izumi: ya I think if Robert is comfortable since it will impact him the most in terms of support
[11:32] Justin Clark-Casey: ok, sounds like we are good to switch then
[11:32] Robert Adams: switching over could generate bug reports, but then things will get fixed
[11:32] Nebadon Izumi: then i am definitely +1
[11:32] Justin Clark-Casey: top of the dev cycle is the time to do it
[11:32] Teravus Ousley: Yep +1 for bollux switch :)
[11:32] Justin Clark-Casey: you're thinking of something else, Ter ;)
[11:32] Andress Renault is offline.
[11:32] Nebadon Izumi: lol
[11:33] Nebadon Izumi: we need to keep in mind too that melanie is probably going to be doing some physics stuff here soon
[11:33] Nebadon Izumi: so might get bumpy for a bit
[11:33] Nebadon Izumi: though heres is more api level stuff i think
[11:33] Nebadon Izumi: heres=hers*
[11:33] Nebadon Izumi: than physics engine changes
[11:33] Richardus Raymaker: sorry lost track, switching what over ? mono ?
[11:33] Nebadon Izumi: border crossing stuff and such
[11:34] fine.Dreadlow :D
[11:34] Nebadon Izumi: Bulletsim Richardus
[11:34] Nebadon Izumi: making it the default engine
[11:34] Teravus Ousley: as the default fizix engine
[11:34] Richardus Raymaker: yes, its great to have it default
[11:35] Nebadon Izumi: Has anyone gotten a chance to test the Postgre stuff at all yet?
[11:35] Richardus Raymaker: oh robert, is the movement when you walk fixt in newer releases ? because in the 0.7.6 final i get a bit fast/normal/fats normal world movement
[11:35] Nebadon Izumi: I have not had the time, i dont know Postgre at all, first step is learn it for me
[11:36] fine.Dreadlow i wannt cause i had no reference what to configure on opensim
[11:36] Teravus Ousley has a mostly wind0ws shop
[11:36] Justin Clark-Casey: Someone wrote that it's faster but would want to see more anecdotes
[11:36] Justin Clark-Casey: I've never used postgresql
[11:36] Nebadon Izumi: ya me either, im hoping in a few weeks i'll have some time to give it a whirl
[11:36] Dahlia Trimble: ive never used it either
[11:37] Richardus Raymaker: same here. so take a while before i can test it on standalone. but i like to test it
[11:37] Robert Adams: Richardus: i would have to check... movement was fixed but not sure what release it is in
[11:37] Teravus Ousley: apparently there's a windows installer now
[11:37] Teravus Ousley:
[11:38] Teravus Ousley: the last time I checked, which was many years ago, it was *nix only
[11:38] fine.Dreadlow ubuntu rules in my case :D
[11:38] Andress Renault is online.
[11:38] Richardus Raymaker: i know. i need to test a dev version
[11:38] Andrew Hellershanks: I forgot what the differences used to be between PostgresQL and MySQL. Claims were MySQL used to be faster but PostgresQL had some other things going for it if you were using it more commercially.
[11:38] Andrew Hellershanks: That was ages ago. The two have probably gotten closer to each other in features/behaviour since then.
[11:39] Richardus Raymaker: i heared talking about postgress is much betetr for big data and use less resources
[11:39] Teravus Ousley: MySQL uses one dialect of SQL
[11:39] Justin Clark-Casey: There seem to fanatics on both sides. I'd be interested to see the differences - db isn't critical during normal sim ops but speed plays a bigger role on oar/iar updates
[11:39] Teravus Ousley: ... Postgres uses a form of PL-SQL
[11:40] Andress Renault is online.
[11:40] Teravus Ousley: which is apparently similar to Oracle's dialect
[11:40] Cuteulala Artis is offline.
[11:40] Andrew Hellershanks: ok
[11:41] Freaky Tech: they got clother each other, have pretty recent MySQL and PostgeSQL; both have transactions these days; that is what commercial guys like a lot
[11:41] Teravus Ousley: (And I'm talking about the Oracle database server.. not the Oracle company)
[11:41] Andrew Hellershanks: Teravus, thought you were. Transactions are good to have.
[11:42] fine.Dreadlow yeah i wanna now what exactly runs faster on psql
[11:42] Teravus Ousley: well, there are some people who swear by Postgres who were made very happy
[11:42] Andrew Hellershanks: With MysQL it still seems that you only get transactions with certain DB engines.
[11:42] Freaky Tech: transactions on MySQL is InnoDB only ; MyISAM does not have transactions
[11:43] Justin Clark-Casey: the only point that uses a transaction right now is in the deduplicating asset store
[11:43] Justin Clark-Casey: afair
[11:43] Freaky Tech: replication needs transactions badly
[11:45] Freaky Tech: and that is also doing live db backup
[11:46] Teravus Ousley: Questions, comments, blog posts of note On OpenSim of late?
[11:46] Nebadon Izumi: nothing here at the moment :)
[11:46] Andrew Hellershanks: I discovered a few new features were supported in the scripting engine. Sound queueing is working now. Don't know when that got added.
[11:47] Justin Clark-Casey: I did that about 3 months ago
[11:47] Andress Renault is offline.
[11:47] Teravus Ousley: :)
[11:47] Nebadon Izumi: i'll be in Washington DC in about 8 days showing off OpenSim :)
[11:47] fine.Dreadlow yeah and a couple things more :) i've tested it too
[11:47] Andrew Hellershanks: ok, cool. ty. I have a device that plays music that works now because of that.
[11:47] Kayaker Magic: Seems to me that the list of supported function in the Wiki is behind the LSL implimentation now.
[11:47] Justin Clark-Casey: nebadon: who to?
[11:47] Justin Clark-Casey: kayaker: yeah, wiki does not get much love
[11:47] Nebadon Izumi:
[11:47] Nebadon Izumi: here Justin
[11:48] Andrew Hellershanks: I used the on_rez passed parameter for the first time recently.
[11:48] Teravus Ousley: ah, cool
[11:48] Andress Renault is online.
[11:48] Andress Renault is offline.
[11:48] Richardus Raymaker: Andrew, so thats fixt ? because i have mantis open. but not tested
[11:48] Justin Clark-Casey: you're physically going there. With Crista?
[11:48] Kayaker Magic: I'll do a pass over that list of LSL functions and see if I can update it.
[11:48] Andrew Hellershanks: Richardus, I didn't know there is a mantis open about it.
[11:48] Nebadon Izumi: Crista cant make it, but yes
[11:48] Justin Clark-Casey: kayaker: cool
[11:49] Nebadon Izumi: i will phsyically be there showing it off on a big screen
[11:49] Nebadon Izumi: at a booth
[11:49] Justin Clark-Casey: nebadon: she is listed as a speaker?
[11:49] Richardus Raymaker:
[11:49] Nebadon Izumi: yes she was, but ended up having schedule conflict she could not resolve
[11:49] Justin Clark-Casey: richardus: yes, this is an issue where opensim should scan for such problems at script compilation time rather than failing later on
[11:50] Justin Clark-Casey: there is at least one other related issues in the mantis db
[11:50] Andrew Hellershanks is offline.
[11:50] Richardus Raymaker: that mantis need a better search
[11:50] Andrew Hellershanks is online.
[11:50] Justin Clark-Casey: nebadon: hey, well good luck!
[11:50] Richardus Raymaker: thats the whole problem you most end with makeing new onbe
[11:50] Nebadon Izumi: thanks, im excited
[11:50] Teravus Ousley notes that there are lots of bugs open in OpenSimulator, I'd love for people to take a crack at fixing them :D. it's a challenge. :)
[11:50] Nebadon Izumi: use google richardus :)
[11:51] Justin Clark-Casey: teravus: I'm not sure how one encourages that
[11:51] Nebadon Izumi: Richardus ie : mono
[11:51] Nebadon Izumi: that would search mantis for "mono"
[11:51] Richardus Raymaker: teravus Am starting to work more with opensim again at the work side. so then you get to code and can check things
[11:51] Nebadon Izumi: paste that whole thing into google search bo
[11:51] Teravus Ousley: :) code.. Tell people we want them to help :)
[11:51] Nebadon Izumi: box*, it also works very well for the main wiki too
[11:51] Richardus Raymaker: yeah that google function i always forget
[11:52] Andrew Hellershanks: Richardus, 6697 is about a different issue. I used the valid syntax for on_rez declaration. Always have even when I didn't use the parameter.
[11:52] Andrew Hellershanks: Teravus, I crashed my viewer again. You would like to see people take a crack at fixing what?
[11:52] Justin Clark-Casey: I fix loads of bugs but that doesn't seem to encourage others so much :)
[11:52] Teravus Ousley: bugs :)
[11:53] Teravus Ousley: We as developers, need to mention that we'd love people to take a crack at solving some of the bugs and posting patches
[11:53] Andrew Hellershanks: Some bugs can be buried deep in the internals of OS. I tend to stay on the outside edges.
[11:53] Richardus Raymaker: Well, if i can close a mantis i do it..
[11:53] Justin Clark-Casey: tbh, I think that has little impact but I'm not sure what does
[11:54] Justin Clark-Casey: or it may just be that there aren't enough motivated people to look at these problems
[11:54] Teravus Ousley nods
[11:54] Teravus Ousley: If we don't tell people, then they won't know though..
[11:54] Andrew Hellershanks: I fixed a few things where I wanted some features or wanted a feature working properly.
[11:54] Dahlia Trimble: yay motivation :)
[11:54] Kayaker Magic: I'm retiring from RL work in 1.5 years, plan to get modivated to help after that...
[11:54] Justin Clark-Casey: yes, and then when someone does fix something their patch gets left open for ages ;)
[11:54] Andrew Hellershanks: hehe
[11:54] Teravus Ousley: They might think that we're not interested
[11:55] Andrew Hellershanks: justin, the fix was in my source tree. That was the imporant part for me. ;)
[11:55] Justin Clark-Casey: I think it can be quite a daunting codebase as well
[11:55] Teravus Ousley nods
[11:55] Cuteulala Artis is online.
[11:55] Andrew Hellershanks: yeah. THe problem is some things can be interaction between viewer and OS.
[11:56] Teravus Ousley: I think there are some things that core devs don't want to touch because it's safer :)
[11:56] Justin Clark-Casey nods
[11:56] Justin Clark-Casey: Things like script functions tend to be a lot easier
[11:56] Andrew Hellershanks: If it ain't broke...
[11:56] fine.Dreadlow lol Teravus
[11:56] Justin Clark-Casey: I think that can be a good place for ppl to start
[11:56] Richardus Raymaker: oh fix that with a botton thats saying "dont p[ress me"
[11:56] Andrew Hellershanks: Sometimes even if it is broke, safer not to break it even more. :)
[11:56] Teravus Ousley: I also think that patch git workflows that are daunting as well for patchers and core devs alike
[11:57] Justin Clark-Casey: teravus: yes. The modern way seems to be github pull requests
[11:57] Teravus Ousley: <strikethrough>extra that</strikethrough>
[11:57] Andrew Hellershanks: git takes some getting used to. Its ok until you hit a code conflict then it can be a pita to resolve.
[11:57] Kayaker Magic: I seem to recall a page in the wiki about how to get started helping with opensim, but I cannot find that page now... does it exist?
[11:57] Teravus Ousley: well, I agree with pull requests, it's just a challenge when they're based off of a local branch
[11:57] Andrew Hellershanks: Kayaker, it should be on the developer section.
[11:58] Justin Clark-Casey: There is but I never ever get time to write more
[11:58] Teravus Ousley: .. extra 'stuff' gets merged if we're not careful
[11:58] Andress Renault is online.
[11:58] Justin Clark-Casey: ha, I wrote that back in july 2008
[11:58] Andrew Hellershanks: I run git diff before doing commits
[11:58] Andrew Hellershanks: Check what changes may go in.
[11:58] Justin Clark-Casey: teravus: yes
[11:58] Kayaker Magic: Thanks Justin, that is the page I remember!
[11:58] Andrew Hellershanks: The biggest headache I have is that my machine automatically fixes whitespace and that causes issues for other people working on code.
[11:59] Justin Clark-Casey: kayaker: I't s aurprisingly time-consuming to write that kind of stuff
[11:59] Freaky Tech: the best way running a git copy of opensim is to clone first and then doing the fixes
[11:59] Justin Clark-Casey: with such a complex project
[11:59] Andrew Hellershanks: Freaky, that is the only way. You can't fix the code if you don't have a copy of it first. :)
[11:59] Teravus Ousley: Andrew, mine does too, in my directory it converts to CRLF and when committing, it stores them as LF
[11:59] Freaky Tech: exactly anything else is just nightmare
[11:59] Andrew Hellershanks: Biggest project I worked on previously was only 50,000 lines or so. OS is around 8 times that.
[11:59] fine.Dreadlow :D
[12:00] Justin Clark-Casey: andrew: yeah, it's not small
[12:00] Robert Adams: in my hacking of varregion, I'm going to write a series of blog entries explaining my decisions... I'm not being gentle with the existing code
[12:00] Justin Clark-Casey: robert: are you going to merge your branch back into master soon or continue working on the branch?
[12:00] Teravus Ousley was gentle with MegaRegions :)
[12:00] Richardus Raymaker: SOund s like robert is going to play with dynamite :)
[12:00] Andrew Hellershanks: Its also in C# which I barely know. Fortunately, its close enough to C and I have enough (limited) knowledge of OOP that I have been able to fix some things.
[12:01] Dahlia Trimble: not being gentle with which code?
[12:01] Andress Renault is offline.
[12:01] Andrew Hellershanks: Monodevelop has helped me track down things in the code at times. Trying to find things with just find/grep often turned up too many results that the real thing I was looking for got lost in the noise.
[12:01] Robert Adams: I'm going to merge it eventually... need to have it at least to the point that the legacy region size works
[12:01] Dahlia Trimble: lol
[12:01] Justin Clark-Casey: robert: that woudl be good :)
[12:02] Teravus Ousley: Robert have you seen my DeMoronize function? :D
[12:02] Andrew Hellershanks notices he needs to commit an updated DLL for osprofile.
[12:02] Justin Clark-Casey: robert: are you working with viewer devs as well?
[12:02] Robert Adams: yes... what the heck does that routine do?
[12:02] Robert Adams: looks like it was for fixing some old cruft that might have been in the database?
[12:02] Nebadon Izumi: lol demoronize
[12:02] Andrew Hellershanks: I should pass on a script to Robert. I find fireworks objects move a lot faster than in SL.
[12:03] Teravus Ousley: It rebases to Constants.Regionsize
[12:03] Teravus Ousley: .. And I think it's name might be moronize
[12:03] Teravus Ousley: .. not demoronize
[12:03] Andrew Hellershanks: oh, how do core groups, search, profile compare in functionality to the addon external modules?
[12:03] Justin Clark-Casey: LLClientView.LLHeightFieldMoronize() ?
[12:04] Teravus Ousley: That's the one :)
[12:04] Robert Adams: Andrew: I'd like that... there is a lot of tuning that can be done to make it more SL-like
[12:04] Justin Clark-Casey: groups is better than XmlRpc - I don't know about the others
[12:04] Justin Clark-Casey: groups should be on by default in my opinion
[12:04] Andrew Hellershanks: Robert, now that I'm not in crisis mode fixing somethings in a grid, I'll have a bit more time to talk with you about that.
[12:04] Robert Adams: there's a function like that??
[12:04] Simulator Version v0.5 shouts: OpenSim 0.8.0 Dev          daf44cc: 2013-10-12 20:47:10 -0400 (Unix/Mono)
[12:04] Teravus Ousley: In english, It takes a <256m float array and sticks it in a 256m float array
[12:04] fine.Dreadlow i have some troubles with groups and friendship on hypergrid
[12:05] Teravus Ousley nods
[12:05] Teravus Ousley wrote it :)
[12:05] Andrew Hellershanks: justin, core groups functionality is the same as flotsam groups?
[12:05] Justin Clark-Casey: andrew: is it better afaik, but I have never used it
[12:05] Freaky Tech: seems same with my hypergrid avi just standing around
[12:05] Robert Adams: I've replaced most of the instances of passing around bare float arrays with a class that includes the regions sized and a small represenation of the heightmap
[12:05] Joe Radik is having with groups on new grid.
[12:06] Joe Radik: having problems with
[12:06] Justin Clark-Casey: core groups or xmlrpc?
[12:06] Andrew Hellershanks: Justin, ok. I'll have to give it a test. I can write a PHP script to help with migrating the group notice attachment data.
[12:06] Justin Clark-Casey: andrew: isn't that what diva was talking about some time ago?
[12:06] Robert Adams: c# is an OO language, for diety's sake.... shouldn't be passing around bare data structures
[12:06] Justin Clark-Casey: I think somebody else made a script to migrate groups code from xmlrpc to core groups db entries
[12:06] Teravus Ousley: :)
[12:06] Andrew Hellershanks: Justin, there was a basic conversion script but it didn't handle migration of notice attachment data.
[12:07] Andrew Hellershanks: It was mainly just a set of SQL statements.
[12:07] Justin Clark-Casey: right
[12:07] Andrew Hellershanks: It needs a bit more sophistication to handle notice data.
[12:07] Teravus Ousley: :) Cool, sounds good
[12:07] Dahlia Trimble: bye all :)
[12:07] Joe Radik: core groups
[12:08] Joe Radik: first time setting up grid, though.
[12:08] Andrew Hellershanks: I'll have to get core groups set up on my standalone first.
[12:08] Justin Clark-Casey: joe: ok, unfortunately I have not used it so I can't help much
[12:08] Teravus Ousley: Yeah, it's officially 7 minutes over.. so I before I go, I again want to point people to mantis and look for things that you might be able to fix and post a patch :)
[12:08] Justin Clark-Casey: bye dahlia
[12:08] Richardus Raymaker: bye dahlia
[12:08] Joe Radik: I'm just getting started. Currently running recent core with wifi.
[12:08] Justin Clark-Casey: I need to go myself
[12:08] Richardus Raymaker: soon start to setup grid to
[12:08] Teravus Ousley:
[12:08] Andrew Hellershanks: Joe, that's always fun. OS is a complicated beast to set up. Once you've done it a few times it gets a bit easier but it is so easy to miss some little detail when first working with OS.
[12:08] Justin Clark-Casey: joe: you've used opensim before though?
[12:09] Justin Clark-Casey: or am I confusing you with someone else
[12:09] Richardus Raymaker: after i figured out if VM have big impact
[12:09] Joe Radik: running sim on osgrid for a while and D2 for home "grid"
[12:09] Andrew Hellershanks: I have some bugs and things to deal with on some non-OS projects before I could go through the mantis for OS.
[12:09] Joe Radik: Decided to move to core instead of D2
[12:10] Justin Clark-Casey: ok, I need to go. bye folks
[12:10] Dahlia Trimble is offline.
[12:10] Andrew Hellershanks still hasn't put in the bigger HD's for his compuer.
Personal tools
About This Wiki