Chat log from the meeting on 2025-08-26

From OpenSimulator

Revision as of 12:01, 26 August 2025 by Tampa (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
[11:02 AM] Andrew Hellershanks: Hello, everyone.
[11:03 AM] Joe Magarac: Allo.
[11:03 AM] Gavin.Hird @grid.xmir.org:8002: Hi JNoe
[11:03 AM] Gavin.Hird @grid.xmir.org:8002: Joe
[11:03 AM] Andrew Hellershanks: Hello, Joe.
[11:04 AM] Vincent.Sylvester @hg.zetaworlds.com: Had a bit commit this week, huge, best commit ever, they say, best commit of the year, that's what they say
[11:04 AM] Andrew Hellershanks: :)
[11:05 AM] Ubit Umarov: who had ?
[11:05 AM] Andrew Hellershanks: What was it for?
[11:05 AM] Gavin.Hird @grid.xmir.org:8002: Imagine the commit messages if that person made them
[11:05 AM] Vincent.Sylvester @hg.zetaworlds.com: Imagine the code...
[11:05 AM] Gavin.Hird @grid.xmir.org:8002: :-)
[11:05 AM] Ubit Umarov: what are you talking about? lol
[11:05 AM] Gavin.Hird @grid.xmir.org:8002: weaving
[11:05 AM] Vincent.Sylvester @hg.zetaworlds.com: The inventory thing
[11:06 AM] Gavin.Hird @grid.xmir.org:8002: maybe he is good at making threaded code
[11:06 AM] Ubit Umarov: my commit ?
[11:06 AM] Vincent.Sylvester @hg.zetaworlds.com: ;P
[11:07 AM] Ubit Umarov: well no idea, but talking about commits and last week code changes
[11:07 AM] Andrew Hellershanks: There were a few commits this past week.
[11:08 AM] Ubit Umarov: yes i made a little fix on inventory, adding a missind urlencode on folder name
[11:08 AM] Ubit Umarov: in fact reported by Some Tampa person
[11:09 AM] Ubit Umarov: also had to force Yengine scripts reset and recompile, sadly losing state, Due to changes a few weeks ago
[11:09 AM] Ubit Umarov: and that was it
[11:10 AM] Vincent.Sylvester @hg.zetaworlds.com: I'm surprised the inventory thing had not come up earlier, it's been like that for a long time ever since melanie and diva came up with that mess basically
[11:10 AM] Ubit Umarov: err forced that by updateing the compiler and serializer versions
[11:11 AM] Ubit Umarov: no one uses silly chars in inventory
[11:11 AM] Vincent.Sylvester @hg.zetaworlds.com: I found over 200 entries with more than one leading space so people must have used them over time
[11:11 AM] Ubit Umarov: in fact the rules should be more restrictive, like no space
[11:12 AM] Gavin.Hird @grid.xmir.org:8002: yeah, lets get back to MS DO naming
[11:12 AM] Gavin.Hird @grid.xmir.org:8002: DOS
[11:12 AM] Vincent.Sylvester @hg.zetaworlds.com: Enforce camelCase lol
[11:12 AM] Ubit Umarov: well in many things we never left those >(
[11:13 AM] Ubit Umarov: and unless you are chinese or similar, no reason
[11:13 AM] Ubit Umarov: well even those type en
[11:14 AM] Ubit Umarov: anyway.. the urlencode is back there and some chars do wrok again
[11:14 AM] Vincent.Sylvester @hg.zetaworlds.com: Oh there was one more piece of news that happened worth mentioning
[11:15 AM] Vincent.Sylvester @hg.zetaworlds.com: About a week ago I ran updates on machinery running ubuntu 24 noble and that broke libgdiplus again
[11:15 AM] Gavin.Hird @grid.xmir.org:8002: anyone tried Debian 13 yet?
[11:15 AM] Vincent.Sylvester @hg.zetaworlds.com: I have the iso sitting around haven't gotten around to installing it
[11:16 AM] Andrew Hellershanks: Gavin, I didn't know it was out.
[11:16 AM] Vincent.Sylvester @hg.zetaworlds.com: It might be impacted as well
[11:16 AM] Gavin.Hird @grid.xmir.org:8002: about a week or so
[11:16 AM] Vincent.Sylvester @hg.zetaworlds.com: I am not exactly sure how it broke, but the only broken part seems to be font rendering
[11:17 AM] Ubit Umarov: think linux distros are great on breaking soffware
[11:17 AM] Ubit Umarov: well now win updates trash ssds
[11:17 AM] Vincent.Sylvester @hg.zetaworlds.com: Somehow it's unable to load any font that isn't of fixed glyph type, so svg fonts and such more modern ones just break displaying boxes instead
[11:17 AM] Gavin.Hird @grid.xmir.org:8002: MS want everything stored in their cloud
[11:17 AM] Gavin.Hird @grid.xmir.org:8002: network boot even
[11:17 AM] Vincent.Sylvester @hg.zetaworlds.com: The upside is that you can fetch the old version from 22 jammy and install that along with the libtiff5 dependency and it works again
[11:18 AM] Vincent.Sylvester @hg.zetaworlds.com: You just have to fetch the .deb and install them downgrading the installed version
[11:18 AM] Vincent.Sylvester @hg.zetaworlds.com: Leave it to Canonical to fuck up a compile
[11:18 AM] Vincent.Sylvester @hg.zetaworlds.com: Not their first blunder
[11:19 AM] Vincent.Sylvester @hg.zetaworlds.com: Seems that the problem has been around for months now and there is an open bug report so maybe it'll get fixed at some point, even blind chickens find a kernel once in a while
[11:19 AM] Andrew Hellershanks: hehe
[11:20 AM] Vincent.Sylvester @hg.zetaworlds.com: It's not really that feasible, but it might at some point become necessary to ship libgdiplus along with binaries
[11:20 AM] Vincent.Sylvester @hg.zetaworlds.com: The problem is the dependencies and various system configs between distros make that rather complex
[11:20 AM] Vincent.Sylvester @hg.zetaworlds.com: It depends on pango, cairo, libtiff and more
[11:21 AM] Andrew Hellershanks: If you grabbed an older copy of libgdiplus and tiff5 It raises the question as to which is faulty.which
[11:21 AM] Gavin.Hird @grid.xmir.org:8002: does it have anything to do with more and more Wayland code creeping into Linux distros?
[11:22 AM] Vincent.Sylvester @hg.zetaworlds.com: Well font rendering is supposed to be cairo and pango, but I only noticed around the time libtiff was moved to 6, so it's possible that it is just a bad compile not having setup the dependencies properly or something
[11:22 AM] Vincent.Sylvester @hg.zetaworlds.com: From the issue ticket that exists it seems that possibly it was a problem with pango and code was removed that should have stayed or fixed
[11:22 AM] Ubit Umarov: are they using libgdi from wine?
[11:22 AM] Vincent.Sylvester @hg.zetaworlds.com: That's also not clear
[11:23 AM] Vincent.Sylvester @hg.zetaworlds.com: The package info lists mono as maintainer, but they kinda gave it to winehq, but seemingly not officially either
[11:23 AM] Ubit Umarov: the wine one did got a few changes in may
[11:23 AM] Ubit Umarov: and jun
[11:23 AM] Vincent.Sylvester @hg.zetaworlds.com: It seems they took it and announcing as such, but mono team may have never handed over the connections with the distros
[11:23 AM] Vincent.Sylvester @hg.zetaworlds.com: I also can't find any builds from winehq, just the code
[11:23 AM] Ubit Umarov: mono team is ms team
[11:24 AM] Ubit Umarov: no changes in 6 years or so
[11:24 AM] Vincent.Sylvester @hg.zetaworlds.com: It might be that they aren't allowed to hand over control over it, leaving the distros to do that
[11:24 AM] Ubit Umarov: 4 years
[11:25 AM] Ubit Umarov: https://github.com/mono/libgdiplus
[11:25 AM] Vincent.Sylvester @hg.zetaworlds.com: If nothing happens with the distro version of it and the issue ticket remains stale I may just email the winehq maintainer of it to ask if they have any clue as to what's going on
[11:25 AM] Ubit Umarov: https://gitlab.winehq.org/mono/libgdiplus.git had changes recently as i said
[11:26 AM] Vincent.Sylvester @hg.zetaworlds.com: Canonical and others may have to say "ok we are gonna grab libgdiplus from these guys now, bye ms", wouldn't surprise me
[11:27 AM] Vincent.Sylvester @hg.zetaworlds.com: The only other option is fetching source, setting up runners and rolling our own version of it and the dependencies for different distros. A lot of work, but not impossible
[11:27 AM] Ubit Umarov: "Make system_fonts initialization thread-safe."
[11:28 AM] Vincent.Sylvester @hg.zetaworlds.com: At first I thought it was only the custom fonts through privatefontcollection that broke due to the addmemory way of loading them, but even system wide fonts were broken if they were newer ones with svg scaling or whatever that is
[11:29 AM] Vincent.Sylvester @hg.zetaworlds.com: The old stuff like courier new works fine
[11:29 AM] Vincent.Sylvester @hg.zetaworlds.com: But the ubuntu one, at least the modern version of that wouldn't load
[11:29 AM] Ubit Umarov: no idea what source they use
[11:30 AM] Ubit Umarov: but they keep breaking things
[11:30 AM] Ubit Umarov: dotnet from them is broken a lot of time
[11:31 AM] Vincent.Sylvester @hg.zetaworlds.com: Compiling libgdiplus is not too difficult, I used to have a environment for that back when it had that massive memory leak
[11:31 AM] Ubit Umarov: and how are your impostors doing Joe?
[11:31 AM] Ubit Umarov: and dayturn ?
[11:32 AM] Joe Magarac: Impostors? Working on the server side. I have to have my own server to go scan regions and distribute info.
[11:32 AM] Joe Magarac: It
[11:32 AM] Ubit Umarov: outch
[11:32 AM] Joe Magarac: is at animats.info but not doing much yet.
[11:32 AM] Gavin.Hird @grid.xmir.org:8002: I jsut worked on the build environment last week to make sure I can actually build the Windows version
[11:33 AM] Gavin.Hird @grid.xmir.org:8002: https://github.com/mono/libgdiplus has the following message This repository was archived by the owner on Mar 7, 2025. It is now read-only.
[11:33 AM] Ubit Umarov: yeah ms wants libgdiplus dead and buried
[11:33 AM] Gavin.Hird @grid.xmir.org:8002: you bet
[11:33 AM] Vincent.Sylvester @hg.zetaworlds.com: Same with System.Drawing for some reason
[11:34 AM] Ubit Umarov: it does linux... theirs things do not
[11:35 AM] Ubit Umarov: but their excuse is to refuso to mantain libgd adn its dependencies they cant control
[11:35 AM] Ubit Umarov: like cairo and pango
[11:35 AM] Gavin.Hird @grid.xmir.org:8002: Have you gusy seen the AI generated environments Google has demonstrated that is fully navigatable?
[11:35 AM] Ubit Umarov: so they tell you to use things from google
[11:35 AM] Vincent.Sylvester @hg.zetaworlds.com: Poor billion dollar company can't afford some developers
[11:35 AM] Gavin.Hird @grid.xmir.org:8002: can be changed too in the most recent demos
[11:36 AM] Gavin.Hird @grid.xmir.org:8002: looks like a total SL killer too me
[11:36 AM] Gavin.Hird @grid.xmir.org:8002: to me*
[11:36 AM] Vincent.Sylvester @hg.zetaworlds.com: lol
[11:36 AM] Ubit Umarov: that actually ends up having its own form of libgdi
[11:39 AM] Vincent.Sylvester @hg.zetaworlds.com: https://bugs.launchpad.net/ubuntu/+source/libgdiplus/+bug/2069473 For anyone that wants to also complain to Canonical about breaking things :)
[11:39 AM] Gavin.Hird @grid.xmir.org:8002: https://genie3.ai
[11:40 AM] Vincent.Sylvester @hg.zetaworlds.com: How is that meant to kill SL, you can't grief others with it so why bother logging in xD
[11:41 AM] Gavin.Hird @grid.xmir.org:8002: I am sure you can fill it with all kinds in a couple iterations
[11:41 AM] Gavin.Hird @grid.xmir.org:8002: have athe AI hallucinate up something
[11:41 AM] Vincent.Sylvester @hg.zetaworlds.com: LSD seems more energy efficient
[11:42 AM] Gavin.Hird @grid.xmir.org:8002: true, but nobody cares about that
[11:42 AM] Gavin.Hird @grid.xmir.org:8002: except some net-zero nuts
[11:42 AM] Ubit Umarov: couple iterations, lsd.. errr that is a bit out this meeting theme
[11:43 AM] Ubit Umarov: :)
[11:43 AM] Gavin.Hird @grid.xmir.org:8002: it is not like SL is particularly energy efficient, yeah?
[11:43 AM] Vincent.Sylvester @hg.zetaworlds.com: Well we are well and truly screwed on that front already so even that wouldn't help
[11:43 AM] Ubit Umarov: this fs limited to 30fps is using 30W on cpu here
[11:44 AM] Vincent.Sylvester @hg.zetaworlds.com: They want the holodeck
[11:44 AM] Ubit Umarov: pff plus 130W on gpu
[11:44 AM] Gavin.Hird @grid.xmir.org:8002: I was referring to the SL backend
[11:45 AM] Ubit Umarov: ohh ll pays for that, not on my bills
[11:45 AM] Ubit Umarov: :)
[11:45 AM] Gavin.Hird @grid.xmir.org:8002: :-)
[11:45 AM] Vincent.Sylvester @hg.zetaworlds.com: Isn't that just mostly AWS at this point?
[11:45 AM] Vincent.Sylvester @hg.zetaworlds.com: Not sure about energy efficient, but certainly not cost effective
[11:46 AM] Vincent.Sylvester @hg.zetaworlds.com: Speaking of machine learning things though, still looking for someone that has some knowledge in that for a project I been on and off on for a while now
[11:47 AM] Vincent.Sylvester @hg.zetaworlds.com: Just throwing that out there
[11:47 AM] Gavin.Hird @grid.xmir.org:8002: even the Rsapberry PI has machine learning add on hardware these days
[11:47 AM] Vincent.Sylvester @hg.zetaworlds.com: facepalm
[11:48 AM] Gavin.Hird @grid.xmir.org:8002: there must be tons of info out there
[11:48 AM] Vincent.Sylvester @hg.zetaworlds.com: ai is becoming like DOOM, trying to run on a damn toaster next... that's how you get pyromaniac toasters wanting to burn the world
[11:49 AM] Ubit Umarov: duhh a few wasted silicon on doing math with low precision floats... and they call it ai
[11:49 AM] Gavin.Hird @grid.xmir.org:8002: you make the AI HW the toaster
[11:49 AM] Ubit Umarov: even a stm32 controler has ai now ( ARM based )
[11:50 AM] Ubit Umarov: "ai"
[11:50 AM] Vincent.Sylvester @hg.zetaworlds.com: People used to fold proteins to heat their houses, now it's ai generating smut
[11:50 AM] Gavin.Hird @grid.xmir.org:8002: :-)
[11:50 AM] Gavin.Hird @grid.xmir.org:8002: yes "ai"
[11:50 AM] Ubit Umarov: every thing has AI now
[11:51 AM] Vincent.Sylvester @hg.zetaworlds.com: more like actually incompetent
[11:51 AM] Ubit Umarov: or made by AI
[11:51 AM] Ubit Umarov: nonsense publicity
[11:51 AM] Vincent.Sylvester @hg.zetaworlds.com: delete prod database
[11:52 AM] Ubit Umarov: ai designed toilet paper.. ofc
[11:52 AM] Vincent.Sylvester @hg.zetaworlds.com: Half way through the roll morphs into sandpaper or worse
[11:53 AM] Ubit Umarov: err don't say that... some readers my rush looking where to buy that
[11:53 AM] Ubit Umarov: errr hmm nm :p
[11:53 AM] Vincent.Sylvester @hg.zetaworlds.com: xD
[11:54 AM] Gavin.Hird @grid.xmir.org:8002: haha
[11:54 AM] Gavin.Hird @grid.xmir.org:8002: that and pizza glue
[11:58 AM] Ubit Umarov: well we are close to meeting end time, any other issue?
[11:58 AM] Andrew Hellershanks: That's usually my question. :)
[11:59 AM] Gavin.Hird @grid.xmir.org:8002: nope
[12:00 PM] Joe Magarac: Thanks all! Off to next meeting.
[12:00 PM] Andrew Hellershanks: With nothing further for today I'll call the meeting to a close.
[12:00 PM] Andrew Hellershanks: Thank you all for coming. See you again next week.
[12:00 PM] Ubit Umarov: ok rl calls, cya you all next week, if not before
General
About This Wiki