Open Source Meeting/2010-05-06

From Second Life Wiki
Jump to: navigation, search

Agenda Thursday, 6 May 2010

  1. Weekly Snowglobe update - Merov Linden
  2. Code-level QA (maybe nit-picking, dunno):
    1. This code was recently changed to return mStatus <= LLSpeaker::STATUS_VOICE_ACTIVE || mStatus == LLSpeaker::STATUS_MUTED;. This depends on the declaration order of the enum values. Someone changing/adding those values might not be aware of that. How to handle this?
      • do nothing: everyone changing the .h must have read the .cpp, first
      • add comment at enum declaration
      • use only == and list all applicable values explicitly
      • ...?
    2. Doesn't our Coding Standard require braces here?
  3. Unscheduled bugs triage: would be good to go with the community through those and decide for some what we should take immediately (2.0 or 1.4) or later.
  4. VWR-2085 Current shortcut for hiding UI can't be used under Linux - also see On the Road to Viewer 2.1 for a mention of this feature


[13:41] Boroondas Gupte: hi Techwolf, hi WolfPup
[13:41] WolfPup Lowenhar: hey boroondas
[13:41] Techwolf Lupindo: Hi boroondas.
[13:41] Techwolf Lupindo: Wolfpup, I dno't do winblows. :-)
[13:42] WolfPup Lowenhar: ok
[13:42] WolfPup Lowenhar: figured i would come to the meeting dressed a little differently today
[13:43] Techwolf Lupindo: /me smiels
[13:43] Boroondas Gupte: I came as (Loading...), it seems.
[13:44] WolfPup Lowenhar: this one is fun for a 'normal' size AV
[13:44] WolfPup's Google Translator: this one is fun for a 'normal' size AV
[13:44] WolfPup Lowenhar: so who want 'roasted' :p
[13:46] Morgaine Dinova: Hi Bor, Wolf
[13:46] Boroondas Gupte: hi Morgaine
[13:46] Morgaine Dinova: How goes?
[13:46] Boroondas Gupte: Fine, thanks. Yourself?
[13:46] Morgaine Dinova: Not too bad, could do with some food.
[13:46] Morgaine Dinova: No UNIX today ... :-(
[13:46] Morgaine Dinova: UXIG
[13:46] WolfPup Lowenhar: hmmm me wonders what scorched kitty fur smells like :p
[13:47] Morgaine Dinova: We don't burn. Dilithium
[13:48] Morgaine Dinova: I'll go see if Armin or someone is logged in on OSgrid
[13:49] WolfPup Lowenhar: i like the flight mode on this one :p
[13:50] WolfPup Lowenhar: hey ardy
[13:51] Ardy Lay: Hi
[13:56] Morgaine Dinova: Hi Ardy, Fred
[13:56] Fred Rookstown: Hi.
[13:56] Fred Rookstown: LL still ignoring you on VWRAP?
[13:57] Boroondas Gupte: hi Jonathan
[13:57] Jonathan Yap: Hi Boroondas & everyone
[13:58] Morgaine Dinova: VWRAP is its own boss. LL can ignore all they like, it's not going to make any difference. It has momentum of its own :-)
[13:58] Fred Rookstown: Might as well add in support for Voxel terrain, I'm adding it to opensim right now.
[13:58] Morgaine Dinova: It's more closely aligned with Opensim and IBM actually, than with LL.
[13:58] Techwolf Lupindo: Morgain, any more vwrap client you want me to add to the overlay?
[13:59] Ardy Lay: Hi Merov
[13:59] Morgaine Dinova: Heh, I wish there were others. Actually we don't have even one VWRAP client. Snowglobe as an early implementation of OGP in it, which is not VWRAP.
[13:59] Techwolf Lupindo: HI merov
[13:59] Merov Linden: hi guys
[13:59] Jonathan Yap: Here is a product using the new Socks option in SG:
[13:59] Morgaine Dinova: Hiya Merov :-)
[14:00] Techwolf Lupindo: sim lag
[14:00] Fred Rookstown: What else is new
[14:00] Fred Rookstown: :P
[14:00] WolfPup Lowenhar: hey merov
[14:00] Merov Linden: OK, it's 2pm, time to start
[14:00] Michelle2 Zenovka: Too many meetings too little time, even if i'm in 2 at once
[14:01] WolfPup Lowenhar: hey open
[14:01] Opensource Obscure: hello everybody : )
[14:01] Merov Linden: hi
[14:01] Fred Rookstown: Hi.
[14:01] WolfPup Lowenhar: sorry opnesource my flight wings might get ing the way
[14:01] Opensource Obscure: ^^
[14:01] Merov Linden: so, we have an interesting agenda today with some very specific questions
[14:01] Merov Linden:
[14:02] Michelle2 Zenovka: I'm listening to a very interesting talk about how my socks5 snowglobe code is being deployed in the enterprise with SLE and firewalled access ;-)
[14:02] Jonathan Yap: I just came from that talk, interesting.
[14:02] Merov Linden: makes me think Michelle2: when are we going to get that ported to SG2.0? *cough*
[14:02] Merov Linden: :)
[14:03] Michelle2 Zenovka: yes we really need to now
[14:03] Merov Linden: you bet!
[14:03] Jonathan Yap: That was brought up at the talk
[14:03] Merov Linden: so, first item was a general update on what's brewing in Merov's land
[14:03] Michelle2 Zenovka: Jonathan, who do you think did the snowglobe 1.4 builds and set the proxies up ;-)
[14:03] WolfPup Lowenhar: im running on a custome built version of SG V2 now
[14:04] Merov Linden: 1.x: nothing really, now that the tree is open and building again, I'm expecting some commits from committers (wink wink nudge nudge...)
[14:04] Jonathan Yap: Sorry, getting it into Vwr 2 was brought up at that seminar
[14:04] Michelle2 Zenovka: my work is backing up, sorry ;-(
[14:04] Fred Rookstown: Committers aren't really enthused about committing thanks to the lab plugging their ears and going LA LA LA I CAN'T HEAR YOU in regards to TPV feedback.
[14:04] Merov Linden: on 1.4 though, we now have the new wwater culling code from Aleric and I was wondering if anyone tested it
[14:05] Merov Linden: is that working for you guys?
[14:05] Ardy Lay: Merov, seems to be working quite well, yes.
[14:05] Boroondas Gupte: I tested it back when it was created
[14:05] Merov Linden: it's a rather big patch so test in all sorts of situation is important before we consider porting to 2.x
[14:06] Merov Linden: Aleric left the bug open though so I'm wondering if he's done or not (unclear from JIRA)
[14:06] Boroondas Gupte: I think for 1.4 it's done
[14:06] Techwolf Lupindo: /me retuns
[14:06] Merov Linden: I should add some comment to it to get that moving
[14:07] Techwolf Lupindo: Talk? Darm, I keep missing all the good stuff out there.
[14:07] Merov Linden: k, on 2.x, the *big* news internally is that we created viewer-public
[14:07] Fred Rookstown: When are we going to see floating text fixed in viewer 2.0?
[14:07] Merov Linden: I've been merging with viewer-external on my repo and struggling with some export issues (some new files in there)
[14:07] Fred Rookstown: (Darling's threatening to come here if it's not fixed)
[14:08] Merov Linden: those export issues should be resolved today
[14:08] Merov Linden: expect a big commit on viewer-external
[14:08] Merov Linden: that will makes us in sync with the most recent 2.0.2 internal effort
[14:09] Boroondas Gupte: will viewer-public be merged right into viewer-external?
[14:09] Merov Linden: from there, I'll turn export on automatic and we'll see viewer-external move at the same pace as viewer-public
[14:09] WolfPup Lowenhar: wil that be a compileate up date of all SG V@ source files?
[14:09] WolfPup's Google Translator: would that be a compileate up date of all SG V @ source files?
[14:09] Merov Linden: i.e. : we'll export everything in near real time
[14:09] WolfPup Lowenhar: dang translator
[14:09] WolfPup's Google Translator: dang translator
[14:10] Boroondas Gupte: shouldn't we have separate external branches for different internal ones?
[14:10] Merov Linden: yes, that will be the complete V2 main code
[14:10] Jonathan Yap: Wasn't that issue raised yesterday at an OH Boroondas...anybody know what I am talking about?
[14:10] WolfPup Lowenhar: now im getting everything in german
[14:10] Merov Linden: in line with what Q and Howard described some months ago
[14:10] WolfPup's Google Translator: now im getting everything in German
[14:11] Merov Linden: Boroondas: we may indeed have some other external branches for some projects
[14:11] Techwolf Lupindo: wolfpup, if you are on snowglobe, just and make sure you turned off "Translate Chat(powered by Goodgle)"
[14:11] Merov Linden: but that will be the exception rather than the rule
[14:12] WolfPup Lowenhar: i have a translator hud
[14:12] Merov Linden: the idea is that by default, viewer dev push to viewer-public as often as possible
[14:12] WolfPup Lowenhar: not useing the one built into the viewer
[14:12] Boroondas Gupte: Then just detach the object, WolfPup
[14:13] WolfPup Lowenhar: it is ok after i talk a couple of lines it will correct and has
[14:13] Merov Linden: another thing I've been doing on 2.x is going through the whole list of 1.x patches and see what needs to be ported to 2.x
[14:13] Boroondas Gupte: :-)
[14:13] Techwolf Lupindo: /me smiles
[14:13] Merov Linden: I have a spreadsheet that I think I should make public soon so that we can share that burden
[14:13] Merov Linden: of porting
[14:13] Merov Linden: some ports are easy, some are not as you can imagine...
[14:14] Merov Linden: e.g. Socks5 *cough again*
[14:14] Merov Linden: ;)
[14:14] Michelle2 Zenovka: I've got chris nagging my alt about this ;-)
[14:14] Techwolf Lupindo: I did some cmake work yesterday on snow 1.3 based code. I learned more about cmake and also relised just how spagtiee the code from LL is. :-/
[14:14] WolfPup Lowenhar: i have not even though of trying to work it 1.x cause of the inventory diferences between 1.x ans v2
[14:14] Boroondas Gupte: VWR-12984 could be a hard one, too
[14:14] JIRA-helper:
[#VWR-12984] 1.23: Water flickers and disappears in patches
[14:15] Merov Linden: yeah, that's the one I was refering to at the beginning of this chat
[14:15] Merov Linden: Aleric's big patch
[14:15] WolfPup Lowenhar: that one has been around for a long time
[14:15] Techwolf Lupindo: I'me afraid Aleric Inglewood is the only one that can port that one. :-/
[14:15] Merov Linden: it's a pretty hot one though for all those res with sky boxes
[14:16] Merov Linden: Techwolf: well, that needs to be looked into, that part of the rendering hasn't seen lots of changes AFAIK
[14:16] Boroondas Gupte: 90 votes, 25 watchers
[14:16] Techwolf Lupindo: Hmm....
[14:16] Merov Linden: 90/25: that's *hot*
[14:17] Merov Linden: k, well, I will not work on it immediately (busy with viewer-public hook up and cleaning of 1.x patch ports)
[14:17] Ardy Lay: Seems unlikely that Aleric will spend the time porting it, based on his expressed opinion of Viewer 2.0 and how it's development was handled so don't hold your breath.
[14:17] Merov Linden: but I'll happily consider patches from anyone around :)
[14:18] Merov Linden: well, Aleric was at the OH Tuesday and seemed to be in good spirit
[14:19] Michelle2 Zenovka: Aleric is happy with things last i heard but he has prommised some other people some time on a project too
[14:19] Merov Linden: He's busy with a non Snowglobe project though but I haven't gave up on him
[14:19] Merov Linden: Michelle2, yes, that's what he told me
[14:20] Ardy Lay: We shall see...
[14:20] Merov Linden: ok, let;s move down the agenda
[14:20] Merov Linden: we have a code style question
[14:21] Merov Linden: from Boroondas I think
[14:21] Boroondas Gupte: yeah
[14:21] Merov Linden: what about that style of code: return mStatus <= LLSpeaker::STATUS_VOICE_ACTIVE || mStatus == LLSpeaker::STATUS_MUTED;
[14:21] Merov Linden: eeek!
[14:22] Merov Linden: agree with Boroondas: we need "("
[14:22] Techwolf Lupindo: arg...I just tried to run snowglobe 2 svn, error due to a b0rked filepath....another bug for me to work on....
[14:22] Merov Linden: the absence of which pose problem for some linux compilers IIRC
[14:22] Merov Linden: Techwolf did a clean up of those some months ago
[14:22] Techwolf Lupindo: /me ears perk up
[14:23] Boroondas Gupte: um, my question isn't about brackets, here
[14:23] Merov Linden: so, yeah, it's annoying to see them sprouting again through V2.0 LL written code
[14:23] Merov Linden: /me checks the question again
[14:23] Boroondas Gupte: rather that this depends on the enum order to stay intact
[14:23] Boroondas Gupte: because of the <=
[14:24] Boroondas Gupte: change the order in the enum and that method will give different results
[14:24] Merov Linden: o yes, you're right
[14:24] Boroondas Gupte: which might not be obvious to someone changing the enum
[14:25] Merov Linden: I haven't looked at the enum declaration: is tere something there that caution devs about the ordering of the values?
[14:25] Boroondas Gupte: no
[14:25] Merov Linden: aaagh... well that's asking for trouble then :(
[14:25] Merov Linden: looks like that test needs to be rewritten
[14:26] Merov Linden: may be some semantically meaningful method added to the class wrapping the enum
[14:26] Merov Linden: so that the semantic is surfaced for devs modifying the enum
[14:27] Merov Linden: Boroondas: could you log that as a bug in JIRA so we track it and fix it?
[14:27] Boroondas Gupte: will do
[14:28] Merov Linden: that will be something pick up by viewer-public I'm sure. It's clear there's a disaster waiting there...
[14:28] Boroondas Gupte: general question: should issues like this block merging the respective changeset from viewer-public?
[14:28] Boroondas Gupte: (into snowglobe)
[14:28] Merov Linden: I would say no: we should log them and fix
[14:29] Merov Linden: the only "block merging" issues would be loss of functionality, serious breakage
[14:29] Boroondas Gupte: ok
[14:29] Opensource Obscure: sorry if I interrumpt, I have to afk for a few time. I just added an item to the agenda, but I'm not sure I can get back in time, so here is some more context about that...I read on the blog that 'hide UI' feature (currently broken in Viewer 2.0) will be fixed in 2.1 .. it would be cool if at the same time the shortcut for that feature got moved to something that Linux users can actually use. that feature can *only* be used through a shortcut :) btw, it's a 1-word fix, awaiting since 2007 :s
[14:29] Techwolf Lupindo: eeekk...I caught up with the issues here, I also see some () need to be added too.
[14:30] Merov Linden: but, great catch Boroondas, I didn't see it when merging
[14:31] Merov Linden: OO: make sure that Linux issue is mentioned in the relevant JIRA
[14:31] Merov Linden: if none, log one please
[14:31] Opensource Obscure: ok thanks!
[14:32] Boroondas Gupte: VWR-2085 is assigned to soft
[14:32] JIRA-helper:
[#VWR-2085] Ctrl-Alt-F1 Hide/Show UI Does Not Work Correctly Under Linux
[14:32] Opensource Obscure: and linked to
[14:32] Merov Linden: k, moving down the agenda: I'd like to do some triaging of "unscheduled" bugs
[14:32] JIRA-helper: [#VWR-17009] Suppressing UI works just partially.
[14:33] Jonathan Yap: /me is curious what an "unscheduled bug" is
[14:33] Boroondas Gupte: we skipped item 2.2 (should be a quick one)
[14:34] Boroondas Gupte: (or I can file a jira about that one, too)
[14:34] Merov Linden: isn't 2.2 about the braces?
[14:34] Boroondas Gupte: yeah
[14:34] Boroondas Gupte: curly ones, this time
[14:35] WolfPup Lowenhar: i see what boroon is talking about
[14:36] Merov Linden: ah right, I see that: after the "if" you mean
[14:36] WolfPup Lowenhar: there are no brackest around the linse of code that follow the if statement
[14:36] Boroondas Gupte: yep
[14:36] Merov Linden: yes, braces... darn!
[14:36] Boroondas Gupte: (maybe I should spell stuff out on the agenda instead of just linking)
[14:37] WolfPup Lowenhar: i would think that ould normaly cause the test to fail
[14:37] Merov Linden: keep us on our toes Boroondas :)
[14:37] Boroondas Gupte: :-P
[14:37] Merov Linden: yeap, another one to log
[14:37] Ardy Lay: The link is fine, I think.
[14:37] Techwolf Lupindo: /me jots down three jira entries he needs to make...
[14:38] Merov Linden: k, "unscheduled bugs": those are bugs that have no "Fix version" set meaning they are not scheduled for being fixed in any of the upcoming version
[14:38] Merov Linden: more generally, that means that we haven't looked at and triaged them
[14:39] WolfPup Lowenhar: boroondas when you sugestes tyo me on tuesday about moving the time set-up code to inside the check avout making the date stamp i relized it would make processing fastor for those that woould have it off
[14:39] Boroondas Gupte: yeah, that's the idea, WolfPup
[14:40] Boroondas Gupte: might also save memory
[14:40] Merov Linden: first one in the triage list: SNOW-640
[14:40] JIRA-helper:
[#SNOW-640] Images in landmarks/search/classifieds displaying using many different aspect ratios
[14:40] WolfPup Lowenhar: now if i can just get the code to compile in VS2008 with out crashing the compiler
[14:41] Merov Linden: This is one I moved from EXT to SNOW this morning while triaging EXT bugs
[14:41] Merov Linden: reported by our own Trilobyte :)
[14:42] Merov Linden: this one is "easy" technically to fix but we couldn't get to an agreement of what is the expected best solution
[14:42] Techwolf Lupindo: LL sould picked one aspect ration and stuck with it.
[14:42] Merov Linden: for all possible UI?
[14:43] Merov Linden: that doesn't seem realistic to me
[14:43] Boroondas Gupte: one ratio, not one size
[14:43] Boroondas Gupte: size could still change
[14:43] Techwolf Lupindo: Or narrow it to two of them. Right now, *checks*
[14:43] Merov Linden: I understood
[14:43] Techwolf Lupindo: There are FIVE different apect ratios used.
[14:44] Merov Linden: well, one ratio means lots of UI reviewing
[14:44] WolfPup Lowenhar: /me wonders if he should create a meta issue concering compileing issues
[14:45] Merov Linden: an algo that would decide between scaling and cropping or mix/match both to accomodate all possible situation would give more flexibility to all UI designers
[14:45] Ardy Lay: That should be part of the framework already.
[14:46] Merov Linden: the framework?
[14:46] Boroondas Gupte: unexpected (from the texture creator's view) cropping might be almost as bad as wrong aspect ratio
[14:46] Ardy Lay: Yes. the ... you called it "scafolding".
[14:47] Merov Linden: well, the framework then right now is simply scaling to fill up the space, it doesn't maintain the original aspect ratio
[14:48] Merov Linden: it's an OK solution as long as the compression in one of the dimension is not too bad
[14:48] Techwolf Lupindo: I think it should be narrow down to 4:3 and 16:9. There are some places that use the same textue but dispaly it at differen ratios. Very fusterating for the shop owner advertizing there shop.
[14:48] Ardy Lay: #Textures
[14:49] Boroondas Gupte: yes ... when the UI displays the same texture at several places those HAVE to be at the same ratio. That should be a design requirement.
[14:49] Ardy Lay: WAY to many varients.
[14:49] Merov Linden: well, one can easily scale with same factor in both dimensions, leaving black or background in the unfilled area
[14:50] Merov Linden: but I think I see you with using a limited set of known ratios throughout
[14:50] Techwolf Lupindo: The problem is that textures are in powers of 2, not aspect ratios.
[14:50] Merov Linden: +1 Ardy
[14:51] Boroondas Gupte: Yeah, and the wanted display aspect ratio isn't saved with the asset.
[14:51] Techwolf Lupindo: 512x512. 1024x512. Can't go in between.
[14:51] Ardy Lay: Open textures in SG 1.4.0 in the texture preview floater.
[14:52] Ardy Lay: Unconstrained plus 7 ratios available to preview in.
[14:53] Merov Linden: that was an Aimee's patch
[14:53] Merov Linden: it's in 2.0 BTW...
[14:53] Merov Linden: but it's a different issue
[14:53] WolfPup Lowenhar: that it is
[14:53] Ardy Lay: She did that partly because of this mess so we could preview what would happen to our textures.
[14:53] Boroondas Gupte: Merov, have you also had a look at  ? IIRC, I listed a lot of screen measurements, there.
[14:53] Merov Linden: it's really silly not to save the original dimension in the asset...
[14:54] Techwolf Lupindo: I get "PERMISSION VIOLATION"
[14:54] Merov Linden: Thanks Boroondas
[14:55] WolfPup Lowenhar: same here
[14:55] Boroondas Gupte: well, standard way of handling this is having pixel aspect ratio and display aspect ratio the same, and only convert to power-of-two sizes when required (i.e. just before pushing the texture into OpenGL)
[14:56] Boroondas Gupte: but of course, the aspect ratio could be added as meta-data, too
[14:56] WolfPup Lowenhar: what about those of us that have a 16:9 screen but run in a 4:3 window most of the time?
[14:56] Merov Linden: Boroondas: I agree but I'm afraid that "power of 2" assumption has percolated everywhere
[14:56] Ardy Lay: The metadata stuffed into the image asset is here, I think.
[14:57] Merov Linden: ugly I agree and should be taked out...
[14:57] Ardy Lay: Hmm... no. Not what I was looking for.
[14:58] Merov Linden: k, as far as 640 is concerned, sould that be a "fix in 2.0" issue?
[14:59] Merov Linden: I'd say we should fix that in 2.1, especially considering the implication if done corrrectly...
[14:59] Boroondas Gupte: we need to cooperate with upstream and TPVs on that one ... only one viewer chaning the ratios (back), doesn't help anything.
[14:59] Ardy Lay: Here maybe:
[15:00] Ardy Lay: Yeah, original size is in there.
[15:00] Merov Linden: Boroondas: we can't really impose all viewers to walk in sync
[15:00] Merov Linden: the idea is more to show better experience and have them pull the patches or, better, rebase their viewers
[15:01] WolfPup Lowenhar: dang i need to go have a nother OH to goto on aditi
[15:01] Ardy Lay: I would say parse out the metadata if it exists, some old images won't have it, then calculate the aspect ratio from that?
[15:01] Boroondas Gupte: well, the experiance will depend on user created textures
[15:01] Techwolf Lupindo: aditi oh is distrated by tech problems
[15:01] Boroondas Gupte: for those to change, we need some kind of critical mass
[15:02] Techwolf Lupindo: I lot of viewers are switching to snowglobe for there base code to work off of.
[15:02] WolfPup Lowenhar: ill check the log for here later
[15:02] Merov Linden: hmmm... we don't even know which change we are promoting,,,
[15:02] Merov Linden: yet
[15:02] Merov Linden: changing the meta data and using them is big for sure
[15:03] Boroondas Gupte: that's why we need cooperation: we have to pick a change that has good chances to be adopted by others
[15:03] Merov Linden: true
[15:03] Merov Linden: those "others" though are all on @opensource-dev
[15:04] Techwolf Lupindo: With http texture in the pipeline, it sould be easy to add someting for aspect ratio, either by pixil size or metadata.
[15:04] Merov Linden: so I think you'r making a case to bring that one to discussion there
[15:04] Merov Linden: which is a good thing I agree
[15:04] Boroondas Gupte: yeah, sounds good
[15:05] Merov Linden: Techwolf: true, we'll still have a legacy of old asset though
[15:05] Merov Linden: what do we do for those? display ugly and tell owners to change them?
[15:05] Boroondas Gupte: though we should have Esbee etc in on that, too, I guess
[15:06] Boroondas Gupte: well, easiest would be to revert to aspect ratios as used in 1.x, as most content will still be made for those
[15:06] Merov Linden: Boroondas: I think Esbee lurks on opensource-dev, anyoway, I do chat with her on a regular basis now
[15:06] Boroondas Gupte: cool
[15:06] Merov Linden: /me realizes he needs to get in admin mode in JIRA and add new versions...
[15:07] Merov Linden: I think that one should be a 2.1 feature
[15:07] Ardy Lay: I understand many images that are metatagged will have been manipulated to force the desired presentation in Viewer 1.23.5
[15:07] Merov Linden: as a general rule, we said that Snowglobe 2.0 would be "Viewer 2.0 + all Snowglobe 1.x specific patches"
[15:08] Ardy Lay: Changing that behavior will be asking for screaming and you are getting it from people trying out Viewer 2.0
[15:08] Ardy Lay: Will adding logic to use the metatag cause more or less screaming?
[15:08] Merov Linden: Ardy: I think you're right, that's what someone referred to as "non squared pixels compensation" this morning
[15:08] Merov Linden: a crazy workaround really...
[15:09] Techwolf Lupindo: Note that aspect ratios here are just for the UI. Every busness in busness will keep up with the changes. It not like there is old content that needs to be keeped. If something is abanded, it disapears off the UI.
[15:09] Ardy Lay: Would be awesome if image asset owners could be allowed to "re-metatag" their image assets without reuploading them.
[15:10] Boroondas Gupte: indeed
[15:10] Ardy Lay: Many many parcels have really old and still applicable images associated with them.
[15:10] Merov Linden: this is one which is easy to fix conceptually (use square pixels, scale only isotropically, don't use power of 2 sizes in asset) but it's made complicated only because of tech legacy and debt...
[15:11] Boroondas Gupte: and legacy content
[15:11] Merov Linden: yup
[15:11] Techwolf Lupindo: There is no aspect ratio in world. Tecture on prims are streatched to fit. So the creater adjust the prim to get the disired effect.
[15:11] Ardy Lay: Tech, we are not debating that. ;-)
[15:12] Boroondas Gupte: yeah, but profile pictures, e.g.
[15:12] Techwolf Lupindo: :-)
[15:12] Boroondas Gupte: right now you have to decide whether you want it to look right in 1.x or 2.x
[15:12] Boroondas Gupte: both doesn't go
[15:12] Techwolf Lupindo: I see many profile pics of the wrong aspect ratio
[15:12] Merov Linden: ok, I see it's already 3pm
[15:12] Ardy Lay: But yes, lots of old code, old images and old residents to bend when changin stuff. :-)
[15:12] Merov Linden: we've looked only to one bug but, boy, that opened a lot of cans of worms...
[15:13] Boroondas Gupte: let's go fishing ;-)
[15:13] Ardy Lay: Hehe
[15:13] Merov Linden: :D
[15:13] Ardy Lay: Watch out for "phishing" via forum PMs.
[15:13] Boroondas Gupte: dunno if canned worms can be used for that ...
[15:13] Merov Linden: OK, I need to go but I think we should continue on triaging those newcomers
[15:14] Merov Linden: I'll be editing JIRA to add new release/versions so we can assign them correctly
[15:14] Ardy Lay: Merov, what the heck is this? secondlife:///app/chat/0/Test message.
[15:14] Boroondas Gupte: Merov, can you post the permalink next time? just copying the url from the browser's adress bar doesn't work for jira searches.
[15:14] Jonathan Yap: Test
[15:14] Boroondas Gupte: Test
[15:15] Ardy Lay: Hehe, Got ya.
[15:15] Boroondas Gupte: lol
[15:15] Jonathan Yap: lol
[15:15] Ardy Lay: That's being used against people now.
[15:15] Techwolf Lupindo: Test message.
[15:15] Techwolf Lupindo: Test message.
[15:15] Ardy Lay: Making them say stuff.
[15:15] Techwolf Lupindo: Hmm....
[15:15] Jonathan Yap: It only put out "Test" for me, should I file a jira?
[15:15] Merov Linden: eeek!...
[15:15] Ardy Lay: Consider yourselves cautioned.
[15:15] Techwolf Lupindo: !!!....
[15:15] Ardy Lay: Depends on your viewer version.
[15:16] Merov Linden: Test
[15:16] Techwolf Lupindo: ....secondlife:///app/chat/4/hud commnad.
[15:16] Boroondas Gupte: secondlife:///app/chat/0/I%20agree%20to%20everything%20Boroondas%20says.
[15:16] Ardy Lay: Hehe
[15:16] Ardy Lay: I%20agree%20to%20everything%20Boroondas%20says.
[15:16] Ardy Lay: Not so good, SG 1.4.0 here.
[15:16] Jonathan Yap: I%20agree%20to%20everything%20Boroondas%20says.
[15:17] Merov Linden: does it passes though SG's chat translation ? :)
[15:17] Jonathan Yap: I%20agree%20to%20everything%20Boroondas%20says.
[15:17] Ardy Lay: Not using translations at the moment.
[15:17] Merov Linden: I%20agree%20to%20everything%20Boroondas%20says.
[15:17] Ardy Lay: I%20agree%20to%20everything%20Boroondas%20says.
[15:17] Techwolf Lupindo: If one know thay have a hud, craft a url to send a command to that hud, just get the victom to clik on the link.
[15:17] Jonathan Yap: secondlife:///app/chat/0/Bonjour
[15:17] Jonathan Yap: Bonjour
[15:18] Jonathan Yap: It passes through the translator
[15:18] Merov Linden: Salut les p'tits loups
[15:18] Ardy Lay: Kinda scary. I have been getting warnings through admin channels to watch out for abuse via this method.
[15:18] Merov Linden: not bad...
[15:19] Merov Linden: That should be reported for sure... too tempting to use
[15:19] Michelle2 Zenovka: Its useful for group invites from an object, gives you a link to bring up the group windows
[15:19] Ardy Lay: The chat function isn't needed for that.
[15:20] Michelle2 Zenovka: no its not chat but a different operation via the secondlife:: URI
[15:20] Ardy Lay: There are many uses for other APP URI strings but what is the chat one good for?
[15:20] Boroondas Gupte: well, linking to content is ok. The issue here is that the urls don't point to the actual content, but to RPCs
[15:21] Ardy Lay: HTML based UI elements were the goal but I don't know how well used that is.
[15:22] Ardy Lay: I use the teleport URI in some devices.
[15:22] Merov Linden: I'll ping the "security" folks to let them know though I guess they already do
[15:22] Ardy Lay: A friend has a group invite device that uses a URI.
[15:23] Merov Linden: see you Tuesday folks!
[15:23] Ardy Lay: See you, Merov
[15:23] Boroondas Gupte: /me waits for someone embedding secondlife://app URIs in src attributes.
[15:23] Merov Linden: Thanks for coming
[15:24] Michelle2 Zenovka: *NOTE: Text editors are considered sources of trusted URLs
// in order to make avatar profile links in chat history work.
// While a malicious resident could chat an app SLURL, the
// receiving resident will see it and must affirmatively
// click on it.
// *TODO: Make this trust model more refined. JC
[15:24] Ardy Lay: I once went to Ahern and dropped an app uri into chat an half the mob teleported to another infohub. That was my amusement for the evening.
[15:24] Boroondas Gupte: lol
[15:24] Techwolf Lupindo: /me lol
[15:24] Techwolf Lupindo: Drop that here now. :-)
[15:25] Ardy Lay: They look like this secondlife:///app/teleport/Lusk/215/170/62/
[15:26] Boroondas Gupte: well, those at least say what they do
[15:26] Boroondas Gupte: but the chat ones aren't translated to human readable

Generated with SLog Wikifier

Post-meeting/Follow-up activity

  • SNOW-641: code depending on enum value declaration order (patches attached for 3 different potential resolutions)
  • SNOW-642: if-statement without braces (violates coding standard) (trivial patch attached)