Open Source Meeting/2007-11-29

From Second Life Wiki
Jump to navigation Jump to search

< Open Source Meeting


Open source meeting - Thursday, 2pm PT.

Teleport to the Linden Open Source Project headquarters.


Agenda

  • Please add items here
  • ...

Transcript

  • [14:04] Dzonatas Sol: No items on the Agenda.. not even Open Discussion... hmmm.. wait some more see if others show up?
  • [14:05] Michelle2 Zenovka: I saw an update to 1.18.5.3 avaiable, is the source out yet?
  • [14:05] Harrison Partch: I never have been introduced to a Linden!
  • [14:05] Harrison Partch: Hello sir.
  • [14:05] Q Linden: hey there
  • [14:06] Harrison Partch: I have luckily found Michelle2's apt repository for 64bit Linux builds.
  • [14:07] Rob Linden: hi folks
  • [14:07] Michelle2 Zenovka: Hi Rob, have you committed 1.18.5.3 yet to svn
  • [14:07] Dzonatas Sol: Hello
  • [14:08] Rob Linden: Michelle2: will do later today
  • [14:08] Harrison Partch: Can I ask a foolish question?
  • [14:08] Rob Linden: go for it
  • [14:08] Harrison Partch: It's about the viewer.
  • [14:08] Harrison Partch: I hang out in Luskwood when I come to SL
  • [14:08] Harrison Partch: and it is often very laggy
  • [14:08] Harrison Partch: and my question is,
  • [14:09] Harrison Partch: why does the text box that I type into slow down?
  • [14:09] Michelle2 Zenovka: it all runs in the same thread
  • [14:09] Harrison Partch: Surely "accept keypress" need NOT be in the same thread as the graphics loop
  • [14:09] Gigs Taggart: Surely :)
  • [14:10] Harrison Partch: So I always wondered.
  • [14:10] Michelle2 Zenovka: everything basicly runs in sequence in the main loop so if one bit hangs it all slows down
  • [14:10] Harrison Partch: Why not have one thread updating most of the screen and one doing the chat widnow?
  • [14:10] Harrison Partch: window?
  • [14:10] Q Linden: you don't ever want more than one thread updating graphics. that way lies madness
  • [14:10] Rob Linden: see https://wiki.secondlife.com/wiki/Viewer_App_Cleanup
  • [14:10] Gigs Taggart: do it
  • [14:11] Harrison Partch: Madness? Why, if they are partioning the screen between them?
  • [14:11] Q Linden: single graphics context
  • [14:11] Q Linden: it's not separate windows
  • [14:11] Gigs Taggart: you can still separate input though
  • [14:12] Gigs Taggart: right now it can actually miss keypresses
  • [14:12] Harrison Partch: Right, hardware GPU.
  • [14:12] Harrison Partch: OK that sort of makes sense.
  • [14:12] Harrison Partch: I'm so used to thinking about CPU graphics only.
  • [14:13] Gigs Taggart: thraxsis you been working out? :)
  • [14:13] Michelle2 Zenovka: Do we have a release schedule for the 1.18.5 branch, things seem to be moving nicely there now i got the swap storm under control, lol
  • [14:13] Rob Linden: at this point, with 1.18.5 out, we're probably going to be pushing 1.18.6 rc0 pretty quickly
  • [14:13] Thraxis Epsilon: Just hitting the steroid sliders
  • [14:14] Michelle2 Zenovka: windlight?
  • [14:14] Seg Baphomet: HAY GUYS
  • [14:14] Gigs Taggart: hi
  • [14:14] Rob Linden: Michelle2: no, viewer auth and one or two other things
  • [14:14] Tao Takashi: Hi
  • [14:15] Harrison Partch: f bubbles?
  • [14:15] Michelle2 Zenovka: ok cool, i need to make some more noice but windlight will not compile on gcc 4.2
  • [14:15] Gigs Taggart: some of my patches will liekly need serious porting to windlight :(
  • [14:15] Harrison Partch: are those analemma bubbles?
  • [14:16] Seg Baphomet: Windlight is win but it doesn't even work with fglrx. ;P
  • [14:16] Seg Baphomet: I need to see if I can get an older fglrx or something.
  • [14:16] Tao Takashi: on this machine windlight SL does not even start, instant crash ;-)
  • [14:16] Michelle2 Zenovka: neither does my ati card :-)
  • [14:16] Rob Linden: so, I have something I want to ask about:
  • [14:16] Seg Baphomet: The latest ones are apparently a new codebase and completely screw up shaders on older cards.
  • [14:16] Dzonatas Sol: I'll need a new CPU and graphics card to run Windlight and voice
  • [14:17] Rob Linden: there's been a lot of talk about "fixed internally"
  • [14:17] Rob Linden: I spent some time doing a detailed status update to https://jira.secondlife.com/browse/SVC-421
  • [14:18] Rob Linden: I'm guessing people are going to like that level of info, but it's probably not realistic to give that detail for every single issue
  • [14:18] Seg Baphomet: Neat why does my camera keep swinging around to the back.
  • [14:18] Seg Baphomet: Seriously, WTF
  • [14:19] Gigs Taggart: use my camera patch
  • [14:19] Rob Linden: so I'm wondering what is a good middle ground between having every issue updated like SVC-421 and the status quo
  • [14:19] Rob Linden: thoughts?
  • [14:19] Seg Baphomet: I don't see any invisiprims.
  • [14:19] Gigs Taggart: Rob: I think LL needs to go through and clean up ones that have indeed been rolled out so we can have some confidence the ones that are there now really are still in the pipeline
  • [14:20] Seg Baphomet: Given that we don't have access to LL's SCM, you're going to have to give us status reports.
  • [14:20] Seg Baphomet: We can't just look in the SCM and see WTF is going on.
  • [14:20] Gigs Taggart: Thraxsis and I have closed hundreds of them over the last two days, that makes LL's job easier
  • [14:20] Rob Linden: well, be careful when you tell us what we "have" to do ;-)
  • [14:20] Gigs Taggart: I said "I think"
  • [14:21] Rob Linden: was talking to Seg
  • [14:21] Gigs Taggart: oh
  • [14:21] Seg Baphomet: Well without status reports, we have no other way of knowing if anything got merged until it turns up in a release.
  • [14:21] Mm Alder: Rob: The problem is the appearance that LL has the problem fixed, but they forgot to release it.
  • [14:21] Seg Baphomet: Which can take months.
  • [14:21] Rob Linden: Gigs: yes, we've had a conversation this morning about how to update that
  • [14:22] Michelle2 Zenovka: COuld be good if there is some kind of estimate what version a patch may appear in.
  • [14:22] Rob Linden: ok.....I'm not asking "if", I'm asking "how"
  • [14:22] Harrison Partch: Nicholaz blogged about exactly this issue - the two jiras
  • [14:22] Seg Baphomet: Is there any plans within LL to open up the SCM, or are we going to go on like this forever?
  • [14:22] Thraxis Epsilon: Mm Alder, it's been a bit more then that. Fixes are being released to the grid and not being moved from "Fixed Internally" to "Fixed"
  • [14:22] Rob Linden: Seg: yes, but let's stay on this topic
  • [14:22] Dzonatas Sol: I like to suggest, but I'm not sure I completely understand the current issue
  • [14:23] Gigs Taggart: Rob: well we de-chaffed if for you as much as we could, I'm just saying LL should complete that task, so we can be sure that the remaining ones are indeed still pending.
  • [14:23] Rob Linden: Michelle2: we don't even know that
  • [14:23] Gigs Taggart: that would probably be sufficient for now
  • [14:23] Rob Linden: k....mandatory reading: my comment on SVC-421
  • [14:23] Seg Baphomet: Yeah. OpenJPEG 1.2 went out with 1.18.3 but the bug never got updated.
  • [14:23] Q Linden: rob, is there anything we could do to allow a comment on internal Jira to be automatically cross-posted to an external jira?
  • [14:23] Q Linden: specific comments, I mean
  • [14:23] Seg Baphomet: And the lossy patch did get merged. Which was kinda expected to happen at the same time.
  • [14:24] Rob Linden: Status update: it appears as though the fix for this issue hasn't made it out into the wild yet. Hard to say when it will make it out at this point, but likely two or three releases down the road. The exact timing is hard to project. Here's a horrificially detailed response in case you're interested in why we're handwaving here. The problem is that there are one or two showstopper bugs left in the branch where this issue was checked in (branch: task-inv-perms-5, svn rev 73978). That branch last went to QA on November 16, and failed the test pass on November 27. The branch has been kicked back to development to fix the last remaining known showstoppers before being merged into the release branch. Once merged into the release, it eventually makes its way into the final product. See: https://wiki.secondlife.com/wiki/Source_branches
  • [14:24] Seg Baphomet: DIDN'T get merged, rather.
  • [14:24] Rob Linden: (that was my update to SVC-421)
  • [14:24] Rob Linden: my point in giving you that status was to show you how tough it is to project when something will be released
  • [14:24] Gigs Taggart: yep, all I'm saying is that no comment is necessary, just make sure the status really means what it says by staying on top of ones that really shouldn't be in that workflow status anymore.
  • [14:24] Harrison Partch: The mythical furry-fortnight.
  • [14:25] Rob Linden: does everyone agree with Gigs here?
  • [14:25] Mm Alder: Rob: is QA the only reason why Fixed Internally doesent get released?
  • [14:25] Seg Baphomet: Well my point is that sometimes stuff doesn't seem to actually make it into a branch at all. And we really have no way of knowing unless you tell us.
  • [14:25] Seg Baphomet: For example, the lossy patch. :)
  • [14:26] Rob Linden: ok, that can be another case study
  • [14:26] Thraxis Epsilon: It would be a big help keeping the Fixed Internal status under control... Figure in 2 days we took it from 222 down to around 111
  • [14:26] Michelle2 Zenovka: When somethin gets merged is it possible to have a merged at xyz branch, sounds like manual work though
  • [14:26] Rob Linden: Seg: what's the issue number on that one?
  • [14:26] Thraxis Epsilon: I was just looking at that one earlier today
  • [14:27] Thraxis Epsilon: Was actually reported as released to the grid twice I think
  • [14:27] Gigs Taggart: heh, we've audited all of them :)
  • [14:27] Gigs Taggart: between us
  • [14:27] Q Linden: re missing patches, we've been revising our internal processes in the last couple of months, and I think it'll be a lot more rare that stuff gets committed but never released
  • [14:28] Q Linden: But the flip side of that is that stuff is taking longer
  • [14:28] Thraxis Epsilon: VWR-2404
  • [14:28] Dzonatas Sol: So one concern is (a) avoid the need to write detailed tracking comments for 'Fixed Internally's and (b) being able to keep the pjira status field current with the internal progress
  • [14:28] Gigs Taggart: Rob: WEB-401 might help ;)
  • [14:28] Seg Baphomet: VWR-1475, now merged by Tofu. :)
  • [14:30] Dzonatas Sol: Kinda sounds like there needs to be a way to link jira systems together
  • [14:30] Mm Alder: On another topic: I've been seeing lots of source releases that do not build cleanly. Has that been cleared up?
  • [14:30] Seg Baphomet: Errr, the split bug tracker just needs to go away.
  • [14:30] Dzonatas Sol: or an external program to automate cross-updates
  • [14:31] Gigs Taggart: Mm: I can field that one. The code that's released goes through a meat grinder compared to the code LL works from. :)
  • [14:31] Rob Linden: split bug tracker: WEB-89, I believe
  • [14:31] Seg Baphomet: I know the Red Hat bug tracker, they can just mark stuff "private" for security issues and RHEL stuff that clients don't want public.
  • [14:31] Rob Linden: please make a detailed case there
  • [14:32] Rob Linden: please....this issue actually has pretty wide attention right now, and some well-thought-out comments by you all would be helpful there
  • [14:32] Seg Baphomet: The Fedora project shares the same bugzilla with basically everything Red Hat does.
  • [14:32] Gigs Taggart: yeah
  • [14:32] Gigs Taggart: Other than exploits in the bug tracker itself, I see no reason against a merge
  • [14:32] Gigs Taggart: well, except for the extreme slow performance of PJIRA :)
  • [14:32] Rob Linden: ok, regarding VWR-1475, that got checked into maintenance-4 r74571
  • [14:33] Gigs Taggart: Rob: can you update the source brances page with these branch numbers?
  • [14:33] Seg Baphomet: VWR-2383 can probably be closed.
  • [14:33] Rob Linden: that might be a more maintainable way
  • [14:34] Seg Baphomet: I haven't confirmed that debuginfo works though.
  • [14:34] Squirrel Wood: /ao off
  • [14:34] Seg Baphomet: No one's reported crashes lately it seems. :)
  • [14:34] Rob Linden: ok, how is this for a process
  • [14:35] Rob Linden: we make sure that any time something gets marked "fixed internally" has a branch and a rev on it
  • [14:35] Squirrel Wood: I'd say that would help a lot
  • [14:35] Rob Linden: that branch may not mean anything to most people, but....
  • [14:36] Rob Linden: we then endeavor to keep the list of branches up to date on the wiki for at least the branches referred to in PJIRA
  • [14:36] Michelle2 Zenovka: Well that does give people an indication of where the patch is and can be watched as the braches head towards release
  • [14:36] Harrison Partch: I am obsessed with realtime voxel raytracing.
  • [14:36] Rob Linden: I don't think that we can do that for ALL branches, because there are branches in there which are still confidential work
  • [14:37] Gigs Taggart: you mean the NSA branch that has the backdoor code?
  • [14:37] Q Linden: dammit, gigs
  • [14:37] Rob Linden: yeah, that one Gigs :)
  • [14:37] Gigs Taggart: :P
  • [14:37] Harrison Partch: Backdoor code? Only for adult regions I hope.
  • [14:37] Gigs Taggart: hehe
  • [14:37] Harrison Partch:  ;}
  • [14:38] Michelle2 Zenovka: Even if we know things are in branch X and we can see the branch map that shows where X is this is helpful.
  • [14:38] Rob Linden: should I add new fields for LL Fixed in Branch and LL Revision?
  • [14:38] Gibson Willis: about branches...
  • [14:38] Gigs Taggart: I do question the wisdom of there being any "confidential branches" in an open source project O_o
  • [14:39] Gibson Willis: I think that's a great idea Rob
  • [14:39] Michelle2 Zenovka: Id vote for that Rob
  • [14:39] Rob Linden: Gigs: simulator isn't open source
  • [14:39] Tao Takashi: I just wanted to raise that issue, Gigs ;-)
  • [14:39] Gigs Taggart: the branch structure of the simulator is confidential?
  • [14:39] Gigs Taggart: :P
  • [14:39] Gibson Willis: but i also think it would be very beneficial if there were more frequent merges to the externally-facing svn
  • [14:39] Harrison Partch: There IS an open source simulator but it is different.
  • [14:39] Dzonatas Sol: Does the extra field cause more work to keep it up to date?
  • [14:39] Tao Takashi: but at some point this will go away
  • [14:40] Dzonatas Sol: (more work than what it is worth)
  • [14:40] Gibson Willis: on the bug fix or feature level
  • [14:40] Squirrel Wood: link the branch info to the branch tree in some way to make people even happier?
  • [14:40] Seg Baphomet: The SCM needs to be merged as well. Eventually.
  • [14:40] Rob Linden: I don't think it's more than its worth if we get community help on this
  • [14:40] Seg Baphomet: Open sourcing the sim will probably make that easier.
  • [14:40] Rob Linden: Seg: yup
  • [14:40] Rob Linden: as will the work that Sardonyx is doing
  • [14:41] Harrison Partch: The simulator will be opensourced? will that be separate from the OpenSim project?
  • [14:41] Gibson Willis: is that likely to happen anytime soon?
  • [14:41] Tao Takashi: it would be great if in the end all that code could end up in some foundation to manage it and it gets more like open source projects started by a OS community
  • [14:42] Gigs Taggart: we need to ditch proprietary dependancies first :(
  • [14:42] Michelle2 Zenovka: Harrison, Open sim is an effot to prodce a simulator from scratch, whist one day its hoped the LL sim will be opened, part of a long term plan
  • [14:42] Harrison Partch: Right, k.
  • [14:42] Tao Takashi: Harrison there is also the SL Grid Architecture group which works on an open standard for the grid
  • [14:42] Tao Takashi: this might incoporate opensim
  • [14:43] Rob Linden: Tao: I think the steps we would need to take to spin off a foundation are all that would likely to be needed, nto the actual foundation itself
  • [14:43] Harrison Partch: Is the LL sim not in C# like OpenSim?
  • [14:43] Tao Takashi: and there might be open source components available as a result
  • [14:43] Gibson Willis: any reason that the viewer svn couldn't be made entirely public? with all commits available?
  • [14:43] Rob Linden: Gibson: code is too tangled up still
  • [14:43] Liana Linden: Harrison: fyi https://wiki.secondlife.com/wiki/Architecture_Working_Group
  • [14:43] Harrison Partch: Thx
  • [14:43] Gigs Taggart: liana!
  • [14:43] Tao Takashi: but I think in the end the source code should be owned by some foundation and not LL ;)
  • [14:43] Rob Linden: architecture is too tangled up
  • [14:43] Tao Takashi: in the far future at least
  • [14:43] Rob Linden: Tao: who's going to pay for that?
  • [14:44] Gibson Willis: but you are able to have a publicly facing svn that one can compile the viewer from
  • [14:44] Tao Takashi: it would be uncool if LL goes down for some reason and the SL would be gone, too
  • [14:44] Thraxis Epsilon: Well if it is released as GPL... it's not really owned by anyone ;)
  • [14:44] Gibson Willis: how big a step is it from there to having the branches be public as well?
  • [14:44] Liana Linden: Gigs! ;-)
  • [14:44] Gigs Taggart: that's not true, as long as there's a contrib agreement
  • [14:44] Seg Baphomet: We'd be stuck with GPLv2 only... :)
  • [14:44] Gigs Taggart: I don't see LL needing to let go of overship of code to embrace the open source community
  • [14:44] Rob Linden: Copyright still applies, even with the gpl
  • [14:44] Thraxis Epsilon: I just mean that with GPL it could continue to be developed
  • [14:45] Rob Linden: agree with Gigs. it's just going to take time
  • [14:45] Gigs Taggart: rob: people are getting cynical and discouraged now :(
  • [14:45] Q Linden: don't forget that someone has to run the servers too
  • [14:45] Tao Takashi: well, who runs servers is a completely different thing
  • [14:45] Seg Baphomet: Copyright still applies, but given the GPL the only problem would be not being able to upgrade to GPLv3. :)
  • [14:45] Tao Takashi: isn't the goal anyway that everybody theoretically can run servers?
  • [14:46] Gigs Taggart: eventually
  • [14:46] Harrison Partch: They also server who only stand and run servers.
  • [14:46] Q Linden: oh dear
  • [14:46] Rob Linden: Gigs: have we backslid?
  • [14:46] Harrison Partch: They also serve...
  • [14:46] Gigs Taggart: Rob: I think so
  • [14:46] Rob Linden: how?
  • [14:46] Gigs Taggart: Rob: the patch queue has grown a lot in the last month or two
  • [14:46] Seg Baphomet: Nicholaz?
  • [14:47] Gigs Taggart: Rob: also we really need to commit to getting rid of proprietary deps, I know that has cost me business
  • [14:47] Harrison Partch: Tao, the Opensim server runs now, at least it did months ago.
  • [14:47] Seg Baphomet: BTW, OpenJPEG is in Fedora now. :)
  • [14:47] Gigs Taggart: Rob: people come to me, ask for custom viewer work, when they find out they are going to need a proprietary license for any realistic use, they back out
  • [14:47] Tao Takashi: I know that it runs.. and the more independant projects are done the less the source code ownership is maybe an issue
  • [14:47] Mm Alder: Rob: I'm getting discouraged because it's just too hard to build the vanilla source release with VS2005. Bug it's good news from BOS that cmake is on it's way.....eventually.
  • [14:48] Seg Baphomet: I just need to get xmlrpc-epi in and then I can get slviewer itself in. :)
  • [14:48] Q Linden: cmake should help there, yes
  • [14:48] Dzonatas Sol: has a C++ version of OpenJPEG now =)
  • [14:48] Michelle2 Zenovka: Debian is not moving so fast, :-(
  • [14:48] Seg Baphomet: That's debian for you? ;)
  • [14:48] Thraxis Epsilon: The c-ares issue with the last couple releases did make compiling with 2005 not so much fun.
  • [14:48] Q Linden: so patch queue getting longer ... that's also true internally, related to the changed processes we've been talking about
  • [14:49] Tao Takashi: btw, any news on wiki speed? this discourages me a little to do any work on it..
  • [14:49] Squirrel Wood: The Second Life 1.18.5 Viewer is now available.
  • [14:49] Rob Linden: I think having Linux distros in the mix will help get us to remove proprietary deps
  • [14:49] Gibson Willis: Rob: do you have client and server code on your internal branches?
  • [14:49] Gigs Taggart: yes
  • [14:49] Gibson Willis: still not seeing why the viewer branches couldn't be made public?
  • [14:49] Gigs Taggart: but LL does need to commit at some point and say "we are dropping this proprietary option"
  • [14:49] Rob Linden: Gibson: it's really seriously tangled.
  • [14:49] Seg Baphomet: I'm just hoping the trademark thing isn't a showstopper. No one's said anything about it so far.
  • [14:49] Gibson Willis: how do you manage to even do merges?
  • [14:49] Dzonatas Sol: Gibson, "monolithic" is the word
  • [14:50] Q Linden: gibson, we are trying to detangle now
  • [14:50] Gigs Taggart: until the open source dependancy is the one that the user base is actually using, it'll always be a second class citizen
  • [14:50] Rob Linden: Seg: you saw Liana's mail, right?
  • [14:50] Liana Linden: Seg: I sent an update to the list.
  • [14:50] Harrison Partch: Which one is Nicholaz?
  • [14:50] Liana Linden: We're working on the TM question.
  • [14:51] Michelle2 Zenovka: Gigs, i'm starting to get a user base only running opensource vieiwers
  • [14:51] Liana Linden: / everyone types faster than me. :-P
  • [14:51] Gigs Taggart: Q: might it hit an asymptotic wall sometime, with the patches? Eventually the amount of porting work that builds up will snowball :)
  • [14:51] Michelle2 Zenovka: no prop deps
  • [14:51] Rob Linden: I don't think the trademark thing is going to be a showstopper. we'll figure something out there
  • [14:51] Seg Baphomet: Well there's also possible issues from the Fedora side.
  • [14:51] Michelle2 Zenovka: Wost comes to worst we do a "mozilla" with the trademarks
  • [14:51] Seg Baphomet: Spot is really the one to talk to about that.
  • [14:51] Harrison Partch: you mean like iceweasel?
  • [14:51] Thraxis Epsilon: I lost alot of faith in Fedora when they killed their old version support with little notice
  • [14:52] Michelle2 Zenovka: exactly
  • [14:52] Q Linden: gigs, once we manage to split the repositories up, we'll be able to be faster because merges will be easier
  • [14:52] Harrison Partch: ultimate linux distro = Sidux
  • [14:52] Q Linden: but we have to do that VERY carefully
  • [14:52] Harrison Partch: Debian unstable done right
  • [14:52] Seg Baphomet: No distro wars right now, please.
  • [14:53] Harrison Partch: Sorry.
  • [14:53] Gigs Taggart: yeah that's a good way to waste hours accomplishing nothing :)
  • [14:53] Harrison Partch: We have meetings for that.
  • [14:53] Harrison Partch: Sorry
  • [14:53] Gigs Taggart: hehe
  • [14:53] Rob Linden: ok, back to the "fixed internally" issue....
  • [14:54] Rob Linden: I think we've got something that will help keep really tech savvy people up to date
  • [14:54] Thraxis Epsilon: Should "Fixed Internally" be tagged with "In Progress" vs "Resolved" also?
  • [14:54] Rob Linden: ...but will be utterly baffling to everyone else
  • [14:54] Dzonatas Sol: Rob, I think the fixed version field can also be used for branches
  • [14:54] Gigs Taggart: Rob: that's good enough, we can explain it to all buy one of them.
  • [14:54] Gigs Taggart: all but one rather
  • [14:54] Mm Alder: Rob: There's somebody else listening? :-)
  • [14:54] Gigs Taggart: the "edge case" can just get over it
  • [14:55] Gigs Taggart: can I hijack the topic? sljirastats.com is coming along nicely, what stats would you all like to see?
  • [14:56] Gigs Taggart: fixed internally will obviously be tracked, as will pending patches
  • [14:56] Rob Linden: yes, let's talk about that
  • [14:56] Thraxis Epsilon: Open tickets by submitted month
  • [14:56] Rob Linden: weekly or monthly trendlines on those would be helpful
  • [14:56] Tao Takashi: has to go, cya later! :)
  • [14:56] Thraxis Epsilon: open or unresolved
  • [14:57] Rob Linden: Gigs: how much do config changes on JIRA hurt?
  • [14:57] Gigs Taggart: rob: new customfield shouldn't break it, but they won't be imported until I change the DB
  • [14:57] Rob Linden: ...because, for example, I'm thinking of turning "fixed internally" into a state rather than a resolution
  • [14:57] Gigs Taggart: hrm
  • [14:57] Gigs Taggart: what's a state?
  • [14:57] Gigs Taggart: like resolved?
  • [14:57] Rob Linden: "open", "resolved"
  • [14:57] Gigs Taggart: retroactive?
  • [14:57] Liana Linden: a workflow action state
  • [14:58] Seg Baphomet: That would make more sense, to me yeah.
  • [14:58] Gigs Taggart: I don't know just do it and I'll sort out any breakage
  • [14:58] Rob Linden: we can make a bulk change
  • [14:58] Gigs Taggart: it's early enough in the game right now
  • [14:58] Seg Baphomet: How about "need more info" being a state instead of a resolution? :)
  • [14:58] Michelle2 Zenovka: Yes thats a confusing one
  • [14:58] Rob Linden: hmmm.....that's interesting, too, Seg
  • [14:58] Seg Baphomet: Which is the way bugzilla does it.
  • [14:59] Gigs Taggart: add notabug while you are in there
  • [14:59] Thraxis Epsilon: Cannot Reproduce would make more sense as a State also
  • [14:59] Gigs Taggart: we really need a non-bug resolution
  • [14:59] Seg Baphomet: Yes.
  • [14:59] Rob Linden: "Cannot repro" is more of a state
  • [14:59] Rob Linden: or I'm sorry, is more of a resolution
  • [14:59] Seg Baphomet: You can have both.
  • [15:00] Rob Linden: and they said jira was complicated before
  • [15:00] Seg Baphomet: State: can't repo Resolution: closed
  • [15:00] Gigs Taggart: yeah don't make it both
  • [15:00] Gigs Taggart: hehe
  • [15:00] Michelle2 Zenovka: yes its kind of a dead end, but shoudn't a big really hit needs more info first then later move to that resolution of cannot repro
  • [15:00] Gigs Taggart: no
  • [15:00] Gigs Taggart: needs more info means we can't even try to reproduce it
  • [15:00] Liana Linden: I'm going to have to take off. See you next week.
  • [15:01] Rob Linden: I think "needs more info" is a resolution, come to think of it
  • [15:01] Michelle2 Zenovka: ok i see
  • [15:01] Seg Baphomet: Can't Repo doesn't mean the bug doesn't exist.
  • [15:01] Seg Baphomet: Just means its harder for LL to debug.
  • [15:01] Gibson Willis: take care
  • [15:01] Rob Linden: basically, anything where the next state is likely "closed" and where the info on why it was closed needs to be retained
  • [15:01] Gigs Taggart: Aric would disagree there Seg, even though I've made the same argument to him :)
  • [15:01] Thraxis Epsilon: That's why it shouldn't be "resolved"
  • [15:01] Q Linden: / bye liana
  • [15:01] Dzonatas Sol: "open," "implemented", and "closed"
  • [15:02] Rob Linden: k....I've got to run, too
  • [15:02] Seg Baphomet: In Fedora's bugzilla "neesmoreinfo" means we're waiting on the reporter to supply more information. The bug is still open.
  • [15:02] Dzonatas Sol: thanks for all being here
  • [15:02] Rob Linden: thanks everyone for coming ....good conversation