Before everyone else jumps on Lars here, I'm going to try and make a bit of sense of this situation as I see it (puts on his Sumanah hat). It's primarily a misunderstanding.
a) Lars may not be a native english speaker, and is here representing his community's decisions, please do not read too much into his tone. b) The Swedish community clearly did not fully understand that experimental status of LQT, that it is not deemed ready for production use. A review of events as they seem to me: #1 So during the 1.17 upgrade LQT brroke for the Swedish community, they filed bugs and went "it happens" thinking this was fully tested production mw.org supported extension. #2 Then it broke in 1.18 again and they went (and I'm extrapolating here) "ugh, this isn't a one time thing, there are big problems here we don't want to babysit this extension when so many of our users are unhappy every time it breaks, leaving us with non-working talk pages." #3 So they talked it over and decided to go back to regular talk pages. #4 They let us know through mw.org mailing list that they weren't about to use what they view as such a poorly supported extension. Since I love analogies, they feel like they went to the store and bought a game (Rage for the PC), brought it home and it was buggy as hell, they submitted reports about the problems. Then after the next patch it was still buggy. This time they went back to the store and got a refund, they didn't take the time to file meticulous bug reports (that is non-trivial to do for regular users and admins imo). They really don't care anymore about whatever bugs it may have, they've moved on. Of course the misunderstanding here is that LQT is production ready and would operate more-or-less bugfree (in terms of show stoppers and major problems). This just isn't the case. So it's unfortunate that the Swedish community has been operating under this assumption, but they clearly made the right decision for their users if they need something stable right now and aren't ready to deal with big problems right now during upgrades. Lars I hope this clears thing up for you, and you can go back to your community with a bit of enthusiasm restored, we care about LQT, it's getting better, and maybe one day we'll have a "production" ready version that you can try out again and re-evaluate with your users. But that day is not today. >From our perspective, perhaps more can be done to clearly indicate the status of extensions and who they appropriate for. - Finlay On Tue, Oct 11, 2011 at 8:01 PM, John Du Hart <compwhi...@gmail.com> wrote: > Wow. I'm in a state of complete shock for the lack of care here, #1 for the > fact that you (or apparently the whole swedish Wikipedia community which I > find very hard to believe) can't put in the 5 minutes needed entering a bug > report for something that may effect many others, and (#2) you instead go > the route of discussion to end up at a resolution to abandon the system in > whole. > >> LiquidThreads has again crashed after the upgrade to MediaWiki 1.18. > > Something will always break. There are no guarantees that stuff will always > work, expect it to be wonky right after deployment (it *IS* pre-release > software) > >> We don't see how LiquidThreads could ever become a reliable system > > It's in the midst being rewritten > http://www.mediawiki.org/wiki/LiquidThreads_3.0 > > Honestly I'm completely blown away by the fact that the remote thought of > letting this go unreported for over a *WEEK* without being reported to the > technical staff was at any point an acceptable decision, and your response > one week later being "nah we're just not going to use this anymore". > Honestly, how do you all decide not to report an issue as bad sounding > as "crashed" > (I'm assuming a DB error of some sort). I'm pretty sure if you reported this > when it happened it would most likely be resolved by now. > > Above all, you have the right to decide that you don't want LQT, but to do > so saying that there's some bug and refusing to report it because "have no > interest in this bug getting fixed" is absolutely infuriating to me and > probably other developers. > > (Note: I speak for myself and my own opinions) > > > On Tue, Oct 11, 2011 at 7:38 PM, Lars Aronsson <l...@aronsson.se> wrote: > >> On 10/12/2011 01:18 AM, K. Peachey wrote: >> > On Wed, Oct 12, 2011 at 9:13 AM, Lars Aronsson<l...@aronsson.se> wrote: >> >> The Swedish Wikisource community has decided not to file any >> >> bug report for the fact that LiquidThreads has again crashed >> >> after the upgrade to MediaWiki 1.18. >> > >> > So you aren't going to file a bug so everyone else has to suffer and >> > possibly find this bug themselves when it could be fixed if someone >> > filed it? >> >> Correct. We have no interest in this bug getting fixed. The matter >> doesn't exist anymore. It is a non-topic. This is the difference >> from last time this happened. Thanks for understanding. >> >> >> -- >> Lars Aronsson (l...@aronsson.se) >> Aronsson Datateknik - http://aronsson.se >> >> >> >> _______________________________________________ >> Wikitech-l mailing list >> Wikitech-l@lists.wikimedia.org >> https://lists.wikimedia.org/mailman/listinfo/wikitech-l >> > > > > -- > John > _______________________________________________ > Wikitech-l mailing list > Wikitech-l@lists.wikimedia.org > https://lists.wikimedia.org/mailman/listinfo/wikitech-l > _______________________________________________ Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l
