IRC office hours/Office hours 2013-07-10

Time: 17:00-18:20 UTC (22:30-23:50 IST)
Channel: #wikimedia-office
Timestamps are in IST (UTC+5:30).

[22:30] <siebrand> Hello, Welcome to the office hour of the Wikimedia Language Engineering Team.
[22:30] <siebrand> My name is Siebrand Mazeland and I am Product Manager Language Engineering at the Wikimedia Foundation.
[22:31] <siebrand> The other members of our team present today are:
[22:31] <siebrand> Niklas Laxström / Nikerabbit
[22:31] <siebrand> Pau Giner / pginer
[22:31] <siebrand> Kartik Mistry / kart_
[22:31] <siebrand> Amir Aharoni / aharoni
[22:31] <siebrand> Runa Bhattacharjee / arrbee
[22:31] <siebrand> Our last office hour was held on June 12, 2013. The logs are at:
[22:31] <siebrand> http://meta.wikimedia.org/wiki/IRC_office_hours/Office_hours_2013-06-12
[22:31] <siebrand> Over the past few weeks we have enabled UniversalLanguageSelector on all Wikimedia wikis.
[22:32] <siebrand> To us it appears that the feedback has been mixed and tending less towards positive.
[22:32] <siebrand> We have found that people in our communities can judge very harshly over the work we do.
[22:33] <siebrand> We have found out empirically, that issues in for example bugzilla that contain strong negative judgements, are harder for us to look at objectively and deal with immediately.
[22:33] <siebrand> Harsh criticism, incomplete or unactionable bug reports and "pushing your personal highest priority" on multiple issues is not effective, though.
[22:33] <siebrand> We are people, too.
[22:33] <siebrand> We acknowledge that change is sometimes hard to deal with.
[22:33] <siebrand> We acknowledge that the software that we develop as a team is not free of bugs.
[22:33] <siebrand> We commit to finding solutions, and creating them as soon as possible.
[22:33] <siebrand> We acknowledge that we need your help with that;
[22:33] <siebrand> even more so, you are who we are making this software for,
[22:34] <siebrand> as well as the hundreds of millions of monthly readers of Wikimedia web sites that we hardly ever hear from.
[22:34] <siebrand> When this office hour is over (18:00 UTC), we will deploy a fix for the largest performance issue that was identified in ULS.
[22:34] <siebrand> For those interested https://bugzilla.wikimedia.org/50836 / https://gerrit.wikimedia.org/r/#/c/72967/
[22:34] <siebrand> This makes the web fonts feature in ULS perform about 4 times faster.
[22:34] <siebrand> (according to our measurements on a page on beta labs)
[22:35] <siebrand> It's active already at http://en.wikipedia.beta.wmflabs.org and an example page is http://en.wikipedia.beta.wmflabs.org/wiki/Barack_Obama
[22:35] <siebrand> We are looking into further optimisations.
[22:35] <siebrand> If you want to help, you can.
[22:35] <siebrand> The code is at https://github.com/wikimedia/jquery.webfonts
[22:35] <siebrand> https://github.com/wikimedia/jquery.webfonts/issues?state=open contains two performance related reports.
[22:35] <siebrand> Another thing to mention is that VisualEditor and the input methods of ULS really don't like each other.
[22:36] <siebrand> We are in contact with the Visual Editor team about this.
[22:36] <siebrand> It looks like architectural issues in VE may cause it to take a few months to iron this issue out.
[22:36] <siebrand> I understand you may want to ask us for details on this, but we do not have them yet.
[22:36] <siebrand> As soon as we do, we will share them on the relevant bugzilla issues.
[22:36] <siebrand> (looking up bugzilla issues you can CC yourself to to keep updated on that)
[22:37] <siebrand> James_F may be able to provide those sooner.
[22:37] <siebrand> Right. That's enough from me, I guess.
[22:37] <siebrand> What should we discuss?
[22:37] <matanya> siebrand: i would like to ask about 2GM ram required
[22:37] <siebrand> matanya: excuse me?
[22:38] <matanya> the extenstion page recommends: https://www.mediawiki.org/wiki/Extension:UniversalLanguageSelector
[22:38] <matanya> "Recommended: 2 GB RAM on your PC to load all pages."
[22:38] <siebrand> matanya: I'm not aware of that. I'm fairly certain it's not something anyone from our team added.
[22:38] <siebrand> Let me have a look...
[22:39] <matanya> But before this, i would like to appricate your work as a team, and tell you I love the uls idea
[22:39] <YuviPanda> (that seems to have been from Nemo_bis)
[22:39] <siebrand> thank you.
[22:39] <siebrand> matanya: Im pretty certain it's an incorrect revert artefact.
[22:40] <matanya> so, is 2BG ram needed?
[22:40] <MatmaRex_> (probably not, but a fairly powerful PC seems to be required right now. :( )
[22:40] <siebrand> matanya: That is not my understanding.
[22:41] <MatmaRex_> outstanding issues are performance-related, so that should talk about processor specs, if anything
[22:41] <siebrand> matanya: powerful CPUs generally make all things quicker.
[22:41] <matanya> and my second question would be regarding the drop of support for IE7 (though i hate to stand for IE)
[22:41] <siebrand> matanya: the reason that is not supported is in a JavaScript feature.
[22:42] <matanya> I have seen the commit, but what is the fail back? users using IE7 will see what?
[22:42] <siebrand> In IE7, the positioning of the panels doesn't work properly, for example.
[22:43] <siebrand> matanya: users with IE7 will simply not have any of the ULS features.
[22:45] <matanya> as for the place in the pages, why was the interwiki section the chosen place?
[22:45] <siebrand> matanya: Pau will answer that.
[22:46] <pginer> Users primary use case is to change content: they access an article in a language and want to change to a different one.
[22:46] <pginer> the interlanguage links are the current mechanism to do so.
[22:47] <pginer> From our tests with users, adding it to the personal area as we did with wikis lacking interlanguage links was confusing
[22:47] <pginer> because the top-right area was more prominent, and they expected it to be able to change content (main use case)
[22:47] <matanya> Well, that makes sense, I was expecting it to be near the prefs. but this is a good reasoning
[22:48] <pginer> So for wikis with an existing language-related area it was better to not create a new one especially if it becomes more prominent.
[22:48] <matanya> such as wikis with webfonts?
[22:48] <pginer> I described it at this bug in more detail: https://bugzilla.wikimedia.org/show_bug.cgi?id=50789
[22:51] <matanya> users without the relevant fonts don't see some of the optinal langs, is this by design?
[22:51] <siebrand> matanya: not sure if I understand. Can you elaborate?
[22:52] <aharoni> hi matanya .
[22:52] <aharoni> what do you mean by "optional langs"?
[22:52] <matanya> steps to reproduce:
[22:52] <matanya> click the uls circle
[22:52] <matanya> click the ... button
[22:53] <matanya> scroll down. if you don't have east asian fonts, you see brackets instead of some lang name
[22:53] <siebrand> pginer will answer.
[22:53] <Nemo_bis> YuviPanda: I added that at a time when the freezes were thought to be about unsupported systems
[22:54] <matanya> hi aharoni :)
[22:54] * MatmaRex_ has a couple questions
[22:54] <Nemo_bis> and memory consumption by FF 5.0
[22:54] * matanya asked enough
[22:54] <pginer> Depending on the language availability and quality of fonts some language do have a default web font and others do not.
[22:54] <pginer> for the ones that do not, the system font is used, and squares are seen if it is lacking in your system.
[22:55] <siebrand> thanks for your questions, matanya.
[22:55] <matanya> Thank you very much for all your answer and hard work.
[22:55] <matanya> *s
[22:55] <siebrand> MatmaRex_: what do you want to ask? One question at a time, please.
[22:55] <andre__> Re Performance issues: Personally I wonder if any of https://bugzilla.wikimedia.org/buglist.cgi?bug_id=51070,51073,49935,50836 are duplicates, but I don't feel knowledgeable to judge myself, sorry for that :-/
[22:55] <MatmaRex_> yup.
[22:55] <MatmaRex_> why are ULS components being developed on GitHub instead of on gerrit?
[22:55] <MatmaRex_> and why is the extension split into so many components anyway? this makes development and volunteer involvement much harder.
[22:56] <MatmaRex_> testing each change becomes an annoying multi-step process.
[22:56] <siebrand> aren't those 2 questions?
[22:56] <MatmaRex_> that's one bigger question
[22:56] <MatmaRex_> am i expected to rephrase it now? :/
[22:56] <siebrand> MatmaRex_: https://www.mediawiki.org/wiki/Project_Milkshake should explain more than I'm about to say, but in a nutshell:
[22:57] <siebrand> Project Milkshake is a project that aims to make generic jQuery components for commonly needed internationalisation features.
[22:57] <siebrand> It currently has 4 modules, query.ime, query.i18n, query.uls and query.webfonts.
[22:57] <siebrand> The MediaWiki extension ULS uses all 4 of them.
[22:57] <siebrand> Milkshake exists so that these features can also be used outside of MediaWiki.
[22:58] <siebrand> Ideally, the milkshake components should have QUNit tests.
[22:58] <siebrand> http://www.koha.org/ for example uses Milkshake components at the moment.
[22:59] <siebrand> At Mozilla, some people are considering using it.
[22:59] <siebrand> jQuery.IME will be used to create browser plugins, so that the input methods can work on any web site.
[22:59] <siebrand> All that is not possible if we only developed a MediaWiki extension.
[22:59] <siebrand> The cost is a little higher, as are the benefits.
[22:59] <MatmaRex_> oh, of course. but they could be kept in one repository.
[23:00] <MatmaRex_> that's what i asked about.
[23:00] <siebrand> The thought is that GitHub may make participation of non-Wikimedia developers easier.
[23:00] <MatmaRex_> well, it makes participation of Wikimedia developers harder for sure
[23:01] <siebrand> We for example had a lot of contributions by RedHat developers to jQuery.IME, that would have been a lot harder for them if they'd had to use Gerrit.
[23:01] <MatmaRex_> andasisaid, the testing process becomes much harder than it should be.
[23:01] <siebrand> MatmaRex_: yes.
[23:01] <siebrand> MatmaRex_: That's not necessarily correct.
[23:01] <sumanah> Can we possibly use the bridge YuviPanda developed?
[23:01] <siebrand> It's very modular now.
[23:01] <MatmaRex_> sumanah: no, becauseof how the repositories are structured
[23:01] <siebrand> Something we'd like MedaiWiki to be much more often.
[23:01] <sumanah> (To help people contribute via GitHub pull requests, and then pull those into Gerrit)
[23:01] <MatmaRex_> that's exactly my gripe
[23:02] <MatmaRex_> the repositories are not even in gerrit at all
[23:02] <sumanah> Or perhaps go *the other way* and construct Gerrit proxies?
[23:02] <sumanah> so a Gerrit changeset could be turned into a GH pull request?
[23:02] <YuviPanda> sumanah: can't be done easily because of how github is structured
[23:02] <greg-g> sad
[23:02] <YuviPanda> sumanah: hmm, actually there might be a way. I'll think about it
[23:02] <matanya> didn't we reject the use of GitHub for it being closed source?
[23:03] <sumanah> matanya: Could you be clearer? What specifically? ah
[23:03] <sumanah> matanya: yes, we rejected the idea of using GitHub as our primary host for our git repo.
[23:04] <matanya> but for uls it is ok?
[23:04] <sumanah> So MediaWiki won't be in GitHub primarily.
[23:04] <matanya> i'm missing something? or ULS isn't part of the mediawiki?
[23:04] * matanya is confused
[23:04] <sumanah> MediaWiki core and MediaWiki extensions have their canonical repo in Gerrit.
[23:04] <MatmaRex_> YuviPanda: siebrand: the primary issue is that files from the component repos have to be manually copied to the main repo to be tested, and depending on one's workflow they might have to be committed and pushed to testing environment as well
[23:05] <sumanah> However individual teams (such as mobile and language engineering) have sometimes used GitHub for secondary components, the mobile app, etc.
[23:05] <MatmaRex_> i've suggested using git submodules, but received no reply
[23:05] <MatmaRex_> but well, this is getting technical now, andi have other questions as well.
[23:06] <siebrand> MatmaRex_: The performance issue with web fonts could have been analysed and tested without using MediAWiki.
[23:06] <YuviPanda> sumanah: (OT: but mobile team is moving off GitHub, slooowly. We've only one active component there now. We use the GitHub bot for the others)
[23:07] <siebrand> MatmaRex_: Let's go to the next question, please.
[23:07] <greg-g> YuviPanda: (that's great to hear!)
[23:07] <MatmaRex_> siebrand: well, if you generated suitably large test pages, sure
[23:07] <sumanah> (YuviPanda: that's great, and worth a post to wikitech-l &/or a blog post.)
[23:07] <siebrand> MatmaRex_: "View source"?
[23:07] <MatmaRex_> which, again, would be a very annoying and unnecessary thing to do.
[23:07] <MatmaRex_> anyway.
[23:07] <MatmaRex_> being MediaWiki-independent results in worse performance, for example by being unable to use ResourceLoader to load… resources. is this considered acceptable tradeoff?
[23:07] <MatmaRex_> (this is what the async: false issues stemmed from.)
[23:08] <siebrand> the async call from query.i18n is not being used in the MediaWiki extension ULS.
[23:08] <siebrand> So that's a non-issue at this point in time.
[23:08] <MatmaRex_> now it isn't. it was until i pointed itout.
[23:08] <siebrand> Thank you for that, MatmaRex_.
[23:08] <MatmaRex_> and yes, it';s still an issue, that's an unnecessary HTTP request
[23:08] <siebrand> We acknowledged before, we need you and anyone else to make things better.
[23:08] <MatmaRex_> while the data could be delivered via RL together with the code
[23:08] <siebrand> (no, I never worked at Philips)
[23:09] <MatmaRex_> …i didn't say that to be praised. :/
[23:09] <siebrand> I meant to praise you.
[23:09] <MatmaRex_> so, whyisn't this done?
[23:10] <siebrand> I cannot answer that at this time.
[23:10] <matanya> actully RL might help here, no?
[23:10] <siebrand> Niklas and Amir are in a debugging session at the moment, and Santhosh has a day off.
[23:10] <MatmaRex_> this should have been considered at the design stage. :/
[23:10] <MatmaRex_> okay, last one from me
[23:10] <siebrand> MatmaRex_: Thank you for the suggestion.
[23:10] <MatmaRex_> non-technical this time
[23:11] <MatmaRex_> why is ULS still disabled on ml wikis? it's clear from the bug that's not what the community asked for
[23:11] <siebrand> MatmaRex_: it satisfied the bug reports that the community had open.
[23:11] <MatmaRex_> (talking about https://bugzilla.wikimedia.org/show_bug.cgi?id=51019 )
[23:11] <sumanah> I saw a suggestion:
[23:11] <sumanah> set ULS to system font by default for Malayalam
[23:11] <MatmaRex_> if you disabled ULS, you should reenable Narayam there.
[23:11] <siebrand> 20 minutes before we disabled, we disabled the default font for the "ml" language.
[23:11] <sumanah> I don't know whether that makes sense
[23:12] <siebrand> Immediately an issue was opened to disable ULS.
[23:12] <siebrand> We decided to revert the previous change, and honour the new request.
[23:12] <MatmaRex_> as it seems rather crucial for wiki creation that one can type in its language
[23:12] <siebrand> What do you mean by "wiki creation"?
[23:12] <MatmaRex_> also, the bug says "by default", which to me implies a possibility of changing that default
[23:12] <sumanah> BTW who is here *from* the Malayalam community?
[23:13] <MatmaRex_> article creation*
[23:13] <greg-g> creating content, I believe MatmaRex_ means
[23:13] <MatmaRex_> yes, thanks greg-g
[23:13] <sumanah> I think it would be nice if we could hear from them :-)
[23:13] <MatmaRex_> ViswaPrabha is, i think?
[23:14] <balasankarc> sumanah, Vedmaka , Jyothis , manojkmohan and myself from mlwiki
[23:14] <balasankarc> SOrry, ViswaPrabha not Vedmaka
[23:14] <sumanah> balasankarc: Sometimes I make tab-completion errors too. :-)
[23:14] <siebrand> At this point, it is not clear to us if "ml communities" want https://gerrit.wikimedia.org/r/#/c/72702/ or have it reverted.
[23:15] <MatmaRex_> siebrand: https://bugzilla.wikimedia.org/show_bug.cgi?id=51019#c7 seems rather clear to me. :/
[23:15] <sumanah> I know we don't have too much time left in this hour so it would be great to understand how we can best understand ml consensus (my opinion is that's the best way to use a few minutes here)
[23:15] <sumanah> About how many community members are active on mlwiki?
[23:16] <sumanah> What would constitute consensus?
[23:16] <MatmaRex_> sumanah: have you looked at the bug i linked? :/
[23:16] <MatmaRex_> sumanah: http://hexm.de/5dw
[23:17] <MatmaRex_> that seems rather clear consensus to me
[23:17] <sumanah> siebrand: I think you are implying that there are a few different views in the Malayalam Wikipedia community ? Maybe I misunderstand you
[23:17] <MatmaRex_> they even provided an english translation
[23:18] <MatmaRex_> "Therefore, until and unless all technical issues regarding ULS (as far as Malayalam language is concerned) are solved to the general content and satisfaction of Malayalam Wikimedia users and/or until the Malayalam user community reaches a common consensus on its acceptability, it is proposed that the ULS extension should be initially presented as disabled in all wikimedia projects."
[23:18] <siebrand> sumanah: Communication has been very difficult. I've addressed some issues in my opening statement.
[23:18] <MatmaRex_> "initially presented as disabled"
[23:18] <MatmaRex_> it says this right there in the community discussion linked from bug descripotion
[23:18] <MatmaRex_> i have noidea howcould anyone make thisclearer
[23:18] <Nemo_bis> MatmaRex_: it was just a big misunderstanding in the rush of deployment
[23:18] <sumanah> MatmaRex_: Right; I'm checking whether there are additional views that might have been presented to the lang team in some other medium
[23:18] <MatmaRex_> (sorry about the typos)
[23:18] <ViswaPrabha> For ml, we almost always have a clear consensus, especially on this issue, it is almost unanimous.
[23:19] <MatmaRex_> Nemo_bis: yes, iunderstood that
[23:19] <MatmaRex_> now i'masking why it wasn;t fixed
[23:19] <siebrand> sumanah: andre__ and I have repeatedly asked for actionable problem descriptions. None we received.
[23:19] <MatmaRex_> wasn't*
[23:19] <Nemo_bis> MatmaRex_: https://bugzilla.wikimedia.org/show_bug.cgi?id=51019#c23 is the most up to date comment
[23:19] <MatmaRex_> siebrand: have you read the bug description?
[23:19] <Nemo_bis> now we're waiting on that, from what I understand
[23:19] <MatmaRex_> and the statement linked there?
[23:19] <MatmaRex_> really
[23:19] <Nemo_bis> MatmaRex_: the bug was perceived as a response to the deployment
[23:20] <MatmaRex_> you seem to be getting out of your way to misunderstand that request :/
[23:20] <sumanah> MatmaRex_: hey, let's assume good faith here.
[23:20] <sumanah> misunderstandings happen.
[23:20] <Nemo_bis> MatmaRex_: "your"? I didn't misunderstand anything, personally
[23:21] <siebrand> This is what the request states:We are facing lots of issues with ULS such as
[23:21] <siebrand> functional as well as User friendly problems. Also please don't install this
[23:21] <siebrand> extension until community agree with the changes.
[23:21] <MatmaRex_> Nemo_bis: that was directed at the langeng team.
[23:21] <MatmaRex_> siebrand: you replaced Narayam with ULS. that's right?
[23:21] <ViswaPrabha> Actually, the voting at ml.wikipedia right now, (which is a quick-harvest of some kind of collective opinion) is about the comment mentioned up here.
[23:21] <MatmaRex_> siebrand: now you removed ULS, but Narayam is still gone.
[23:22] <siebrand> Yes. The request was not to install the extension until the community agrees with unstated changes.
[23:22] <ViswaPrabha> And wwe have also translated that proposal to English to make it as clear as possible and to avoid any misunderstandings.
[23:23] <andre__> I agree that it was hard to get actionable feedback (good bug reports) for ULS issues, and I'm happy that we have an initial list at https://bugzilla.wikimedia.org/show_bug.cgi?id=51019#c21 now
[23:24] <andre__> however in hindsight, I'd also appreciate if L10N Eng team sometimes found time to elaborate the reasons for specific decisions a bit more. Which is hard when time is limited, but might pay off in the longrun.
[23:25] <Nemo_bis> ViswaPrabha: had amir1 told you about this discussion he had: "[09:45:55] <Nikerabbit> Amir1: there is deployment tomorrow" http://bots.wmflabs.org/~wm-bot/logs/%23mediawiki/20130708.txt ? (my guess was not)
[23:25] <Nemo_bis> * balasankarc
[23:25] <Nemo_bis> or anyone
[23:26] <ViswaPrabha> Not to anyone, that I know of.
[23:26] <balasankarc> Nemo_bis, no I believe
[23:26] * Nemo_bis nods
[23:26] <Nemo_bis> most of this incident is just a matter of remembering names )
[23:26] <ViswaPrabha> Perhaps, We are all in good touch with each other in ml.wikipedia on our community issues.
[23:28] <Nemo_bis> I mean on the devs' end
[23:30] <siebrand> andre__: Let me give you a little background - we've gone through some pretty tough stuff recently in working with the ml wiki community.
[23:30] <siebrand> andre__: specifically, one of the ml community actually went so far as to launch a personal attack on the *spouse* of one of our team members - on Facebook....
[23:31] <siebrand> andre__: so we had to work through that alongside all the information and opinions we were getting, and that has slowed us down -- and demotivated us, unfortunately.
[23:32] <ViswaPrabha> Please cite! Please cite!
[23:32] <sumanah> ViswaPrabha: hey, there might be privacy issues involved there
[23:32] <siebrand> I'm not going to cite harassment of the wife of a team member.
[23:32] <Nemo_bis> might be inappropriate to spread personal information and attacks
[23:32] <ViswaPrabha> These are serious and false accusations!
[23:32] <Nemo_bis> ViswaPrabha: do you feel accused? how so?
[23:33] <Nemo_bis> is there something you want to tell us
[23:33] <Isarra> Accusing people of harassment without citing something is pretty serious.
[23:33] <ViswaPrabha> Not me personally. But the ml community as a whole!
[23:33] <siebrand> I'm sorry this diluted attention for ml.wiki issues.
[23:33] <ViswaPrabha> I invite all of you to that facebook group which is very open and how a particular discussion proceded.
[23:34] <sumanah> ViswaPrabha: my understanding is that it was on someone's personal page.
[23:35] <ViswaPrabha> Oh then I am not aware about it at all. Neither most of our members.
[23:35] <sumanah> certainly!
[23:35] <ViswaPrabha> And ml community has no responsibility on such private posts/discussions
[23:35] <matanya> thanks all. good night
[23:35] <sumanah> this is not an accusation against "all mlwiki" people. It's an acknowledgment that there's been trouble here that got in the way
[23:35] <ViswaPrabha> In fact, we would have appeared there and tried to ease or defend any such issues.
[23:36] <ViswaPrabha> We at ml are very concious of true universal gender/race/privacy issues
[23:36] <ViswaPrabha> And that is actually part of most of our members' social networking activity
[23:36] <sumanah> and the attack was so demotivating - and personal, and had demands re how to address mlwiki tech issues - that it affected their ability re fixing the ULS issue and serving mlwiki right
[23:37] <Pavanaja> How much time will it take ULS to work in VE?
[23:37] <ViswaPrabha> Please remove this point from our discussions. This should not be a factor for our common ULS problem, unless we were introduced to this formally.
[23:37] <sumanah> it is super great that the mlwiki is so passionate - ViswaPrabha you and I have met, I know how great it can be if you're an example :)
[23:38] <sumanah> ViswaPrabha: but it is, because it was so hard for the team to face, so demoralizing
[23:38] <sumanah> it slowed things down and confused things
[23:38] <ViswaPrabha> BTW, the team member is a very important and likened favourite person in our community too. (Majority) has no issues with them. And we are even much proud of that person
[23:39] <alolita> viswaprabha: it has been really a nasty experience for the language engineering team to deal with the ml issues
[23:39] <sumanah> Pavanaja: sorry, they don't have those details yet but will provide them as soon as possible
[23:39] <sumanah> Pavanaja: (regarding Visual editor -- see http://bots.wmflabs.org/~wm-bot/logs/%23wikimedia-office/20130710.txt I think)
[23:39] <ViswaPrabha> I am sorry for whatever may have happened there, Sumana. But had we ever come to know such personal attacks, I would have been the first person to react and defend. I am also sure most of our members would have marched with me.
[23:40] <Isarra> Perhaps it was nasty, but when there is nothing we can do with it here, and when we cannot even appreciate what happened for lack of details, how is it relevant? How does it help to lecture those trying to smooth the situation now?
[23:40] <sumanah> Isarra: I beg you to think of it not as a lecture but as an explanation.
[23:40] <sumanah> Of why this didn't go super great.
[23:40] <sumanah> (Part of why.)
[23:41] <Nemo_bis> Impatience and frustration are a vicious circle
[23:41] <sumanah> And the thing that we can all do to move forward is to, well, be nice to each other. It helps.
[23:41] <Isarra> It came across like a lecture. Perhaps just something to consider.
[23:41] <Isarra> Being nice does help.
[23:42] <sumanah> Isarra: got it.
[23:42] <sumanah> So, I know we have run over, and I think there's probably some wrapup to be done?
[23:42] <ViswaPrabha> Sumanah and others, please. Please understand that ULS was also one of our own, personally my own, dream for many years. We are only happy if we have a SUPER input/output interface for all our wikis universally.
[23:42] <sumanah> ViswaPrabha: agreed!
[23:42] <Jyothis> sorry, on a meeting, so didnt read much of the above - alolita - it would help if there were better communication. People are not nasty by design. and this is not the firt time community is running into such trouble on decisions that affected it.
[23:43] <ViswaPrabha> But at the same time, we have to take care of the neo-literates, hardly digitally enabled, sometimes rural, sometimes physically compromised, etc.
[23:43] <alolita> jyothis: we have been communicating for the past couple of months on uls deployment schedules, testing, call for input
[23:43] <Isarra> Yeah, what andre__ said before about actually explaining decisions would probably help a lot. Communicating why something is the case instead of just closing stuff as invalid doesn't help anyone.
[23:43] <Isarra> Er, would help people.
[23:43] <sumanah> Evidently there's a deployment meeting happening right now so the lang team has to go, so they can make things better faster :)
[23:44] <alolita> sumanah: yup we should wrap up
[23:44] <ViswaPrabha> We have cases where tribal school childrens working with the most primitive bandwidths, most primitive hardware, most primitive OD etc. etc. Wiki is actually going to grow with them with symbiosis
[23:44] <sumanah> btw anyone who wants better faster heads-ups about upcoming changes - https://meta.wikimedia.org/wiki/Tech/News can get delivered to your talk page every week
[23:44] <arrbee> ViswaPrabha: At any time please feel free to ping me if there are matters you think you would like to bring to the notice of the languag engineering team
[23:45] <sumanah> we thoroughly encourage anyone interested in this topic to get into the https://meta.wikimedia.org/wiki/Tech/Ambassadors community for even more in-depth discussion and advance notice
[23:45] <Jyothis> alolita: good to hear, but it doesnt seems to be working well, does it? :)
[23:45] <sumanah> of changes
[23:45] <sumanah> Jyothis: you'd be a great candidate for a tech ambassador I bet :)
[23:45] <Nikerabbit> hi
[23:46] <alolita> jyothis: happy to work together on getting more testing + feedback from the community actually using the tools :-)
[23:46] <sumanah> so, we're 15 min over so I hope people don't mind if we wrap up and move discussion of this back into the relevant mailing lists, onwiki, & (when necessary) BZ
[23:46] <ViswaPrabha> +1 for Jyothis's ambasadorship :)
[23:46] <sumanah> ViswaPrabha: and you as well!
[23:46] <sumanah> mlwiki could use several tech ambassadors!
[23:46] <ViswaPrabha> Sure sumana, Thanks for the patience to listen. And to you all., and alolitha too!
[23:47] <ViswaPrabha> Thanks Siebrand, amtrex, Isharra, Jyothis et all
[23:47] <Nemo_bis> sumanah: Jyothis is a steward..
[23:47] <alolita> sumanah: thanks! let's wrap this up and Viswa and Jyothis - please help and join the tech ambassadors list
[23:47] <ViswaPrabha> Sorry for the typos. I am overtalking!
[23:47] <siebrand> So do the "ml communities" want ULS enabled again as it was a few days ago, or are you better without it?
[23:48] <siebrand> We can re-enable like "right now".
[23:48] <sumanah> siebrand: "as it was a few days ago" - enabled, but off by default? or on by default?
[23:48] <ViswaPrabha> Please wait. We are still gathering some more comments/opinions/votes till tomorrow
[23:48] <siebrand> "as it was a few days ago", as there are no changes.
[23:48] <alolita> viswa: can you please collate all the comments/opinions/votes and send us a comprehensive summary
[23:48] <ViswaPrabha> We may be almost there but some of us still want to be sure things will workas we NEED to get it worked in several scenarios
[23:49] <siebrand> ULS has input input methods enabled, no input method active.
[23:49] <siebrand> There are no other features to be on or off.
[23:49] <Nemo_bis> sumanah: the ULS revision with system font as default lasted about 20 min on the cluster
[23:49] <ViswaPrabha> It's already here; http://ml.wikipedia.org/wiki/വിക്കിപീഡിയ:പഞ്ചായത്ത്_(സാങ്കേതികം)#Explanatory English translation
[23:49] <sumanah> siebrand: where can volunteers test ULS? mlwiki on the beta cluster?
[23:49] <sumanah> I think after that question we need to go :)
[23:49] <siebrand> The next deployment opportunity would be Tuesday in the regular LangEng deployment window.
[23:50] <Nikerabbit> there are very few wikis on the beta cluster, I don't think mlwiki is one of them
[23:50] <ViswaPrabha> http://ml.wikipedia.org/wiki/വിക്കിപീഡിയ:പഞ്ചായത്ത്_(സാങ്കേതികം)#Explanatory_English_translation
[23:50] <Nemo_bis> http://deployment.wikimedia.beta.wmflabs.org/wiki/Special:SiteMatrix
[23:50] <sumanah> so like for instance https://meta.wikimedia.org/wiki/IRC_office_hours?setlang=ml ?
[23:50] <siebrand> sumanah: logged in, https://meta.wikimedia.org/wiki/Main_page?setlang=ml
[23:51] <sumanah> so people who want to test ULS can test on, for instance, meta.wikimedia.org -- it would be great to have bug reports about specific problems!
[23:51] <sumanah> thank you all
[23:51] <sumanah> have a good day/night!
[23:51] <Nemo_bis> Meta has trigger in personal tools; only option is en.wiki on beta
[23:52] <siebrand> Thanks for the discussion.
[23:52] <ViswaPrabha> Thank you all on behalf of ml community as well.
[23:52] <alolita> thanks for the wrapup sumanah
[23:52] <alolita> thanks viswa
[23:52] <Nikerabbit> thanks all
[23:52] <alolita> thanks everyone
[23:52] <Pavanaja> Bye. Good night/day all