No. This forum will only make MediaWiki more abandonware. Invest money and time in making our discussion system better. We have plenty of money, and plenty of workers. -Theklan (talk) 10:01, 29 May 2022 (UTC)[reply]
I'm trying to revie all of the various resources, forums, etc as well. right now, I'm open to all vioewpoints on this. i plan to look these over. --Sm8900 (talk) 20:55, 9 June 2022 (UTC)[reply]
It seems like this new forum will be actively harmful to meaningful collaboration. To collaborate, people need to be able to come together in the same place to discuss things, and by creating a new place, those discussions become divided. Say, for example, the on-wiki community comes to one consensus, the community on this message board comes to another - what then? TomDotGov (talk) 14:14, 1 June 2022 (UTC)[reply]
Discussions are already scattered in our movement, legitimately by language and then kind of accidentally through many other channels that people have been opening in commercial platforms. That is the reality we have. This forum has a chance to increase participation of non-English speakers in central conversations together with only-English speakers and in equal terms. It also has a chance to bring back and consolidate participants and feedback currently "lost" in social media. About having different "consensus" in different places, this is also part of the current reality for many processes. Nothing that would have to be invented because of the existence of this forum. Plus, consensus or decision processes are just part of the picture. Movement Strategy implementation is mostly about... implementation, collaboration between project peers on an ongoing basis, discussing work and getting things done. Qgil-WMF (talk) 09:36, 2 June 2022 (UTC)[reply]
I'm not a fan of creating yet another off-wiki venue, further fragmenting communication across more and more channels. But I have to say that the inline translation is a big thing for this kind of conversation. I enjoyed reading some exchanges simultaneously across English, Japanese, and Arabic. I wonder if there's any plan to make something like this (maybe as a gadget?) available for Mediawiki. It would be useful for many cross-language conversations at Meta. MarioGom (talk) 15:58, 1 June 2022 (UTC)[reply]
@MarioGom that's probably possible, but not that simple, and definitely not doable as a gadget.
Direct requests from the browser to the translation servers (Google Translate, in the Forum's case) would be in breach of the WMF privacy policy.
Translations cost money, we don't want to pay for them a hundred times if a hundred people read the same post, so they need to be stored somewhere.
Translation requests include a user key (so the provider knows whom to bill) which obviously cannot be made public as it would with a gadget.
The entirety of that extension is just a <script> tag that gets inserted on every page, and cross-loads a translation script from Google servers. It fails all three criteria I mentioned above, and then some. Tgr (talk) 14:52, 3 June 2022 (UTC)[reply]
Then let's copy what Discourse is doing. If they could do it, we can do it. We have more budget, more staff and more volunteers. If this is something really important for our movement, we should be developing it in the fast lane. Theklan (talk) 16:58, 3 June 2022 (UTC)[reply]
"Features for collaboration" should be (and I believe is) one of the main things that the WMF focuses on, so that isn't really a great objection.
A problem with hosting movement strategy discussion off-wiki is that it further disconnects the participants, and especially those at the Foundation, from having to use Mediawiki - from having to "eat your own dogfood". That's been a chronic problem with movement strategy so far - the word 'wiki' is mentioned twice in the movement strategy recommendations.
I get that community members are using other platforms. Having the Foundation actively endorse a collaboration platform other than the one it's responsible for seems like a more major problem, because while movement strategy feels like it commands a large share of the foundation's attention, the wikis are fundamentally where collaboration that benefits the mission happens, and it makes sense to devote resources to making that tool more fit for purpose. TomDotGov (talk) 21:55, 3 June 2022 (UTC)[reply]
"Features for collaboration" *is* one of the main things the WMF focusing on - from VisualEditor, to the talk page improvement project, to the new Moderator Tools team, and many more. But the Discourse project is centered on tools for discussion, and in MediaWiki discussion is just one of several mechanisms of collaboration and the others are equally important (or more important, in the case of editing) so naturally less time will be spent on it. Plus, MediaWiki talk pages are stored in wikitext which means everything ten times harder (and certain communities throw a fit when anyone tries to change that, even though it is a historical accident that never made much sense). So I don't expect us being competitive any time soon with a successful open-source project that is specifically discussion-focused.
Wrt the Foundation actively endorsing a collaboration platform other than MediaWiki, we had that from day one in the form of Mailman (and later OTRS). As Douglas Adams once said, "Anything that is in the world when you’re born is normal and ordinary and is just a natural part of the way the world works. Anything that’s invented between when you’re fifteen and thirty-five is new and exciting and revolutionary and you can probably get a career in it. Anything invented after you’re thirty-five is against the natural order of things." That's sadly how many editors react to changes too, just replace "born" with "registered" and make the numbers a bit smaller. Tgr (talk) 15:11, 4 June 2022 (UTC)[reply]
Mediawiki talk pages are not particularly good at their core job, detailed discussions of possible changes on content pages (cf. e. g. this recent example), I agree. Are you aware of any discussion/forum software out there, that performs much better at this specific task? HHill (talk) 10:57, 5 June 2022 (UTC)[reply]
In theory, I like the concept of annotation-based comment systems (although admittedly I haven't heavily used any, and although some of them support revisions to the annotated document, it wouldn't change as quickly or as much as a typical Wikipedia page). In practice, a system for discussing Wikipedia pages needs to be tightly integrated (e.g. support wikitext, share notifications, share search, have a compatible permission and anti-abuse system, handle things like user renames and page moves) so I don't think it would be feasible to use a non-MediaWiki-based system even if a superior one were found. Tgr (talk) 22:32, 5 June 2022 (UTC)[reply]
So, as I said, they have developed everything with just 67 members, a fraction of the WMF staff. I think that a rather smaller sub-team have been working on the one and only feature that seems superior to our discussion: the automatic translation using Google. If they could do it with a small subset of their 67 workers, we can do it with a small team. Because MediaWiki developers are as smart as Discourse developers. Theklan (talk) 07:03, 4 June 2022 (UTC)[reply]
That's it. They developed everything with a fraction of our team and in a fraction of our time. We can do it, a budget is needed (no budget has been shown in this discussion yet). Theklan (talk) 09:06, 4 July 2022 (UTC)[reply]
As far as I can see the Discourse plugin is doing four things:
identifying which are the meaningful targets for translation (ie. where the individual comments are). For the average forum software this is very trivial but see below.
providing its own API so translation requests can be proxied through the site's own servers
providing some sort of caching layer so the same message does not need to be translated multiple times
doing some sort of language detection, I think - the translation icon is only present on messages which aren't already in your own language.
Task #2 is straightforward; maybe #4 too although I'm not sure what exactly the Discourse plugin is doing there. #1 would be pretty hard for MediaWiki where talk page structure largely lives in the human imagination and under the hood it's all just unstructured wikitext. Just identifying where the comments are is already a hard problem, but that one has already been solved by the Editing team via their work on DiscussionTools. I don't know if that works with old revisions, though. #3 is also problematic because of that - stored translations should be attached to comments, but MediaWiki has no concept of comments. What happens when a section is archived? What happens when a page is transcluded? Etc.
So it's a fairly complex problem, compared to the Discourse translator plugin just being there. Nevertheless, it's a valuable problem to solve, as we have several multilingual wikis (Wikidata, Commons, Meta, mediawiki.org) with lots of conversations which are tied to MediaWiki features, processes or content; these conversations are always going to happen on MediaWiki, so we need some kind of multilingual strategy there. I filed T309920 about it. But it's a hard enough problem that I wouldn't expect it to happen quickly or soon. If anything, the strategy forum could help there by demonstrating that this kind of thing is effective, getting a feeling of the costs etc. Tgr (talk) 14:35, 4 June 2022 (UTC)[reply]
Thanks for the analysis, @Tgr. In fact, it is not going to happen quickly or soon (I wouldn't say never, because obviously this is something that will come in the future). And we can't change the past (only the present time), so we can't be always complaining about things not being done in the past. That said, there's no path for the WMF to adopt this even in the near future. And I doubt there has been any analysis of the cost of adopting a system like that, besides taking Discourse "because is done there". If we see a cool stuff, and we can do it, the correct way to proceed is doing it... or at least doing a cost-analysis for the tool. We might discover that it would be easy to do, or it might be plainly impossible. I doubt that this analysis has been done, and there's the huge elephant in the room. Obsolescence by procrastination. Theklan (talk) 16:45, 5 June 2022 (UTC)[reply]
It seems that if a negative comment is done, it gets deleted. Heavy censorship, without option for anyone to see what is being censored. -Theklan (talk) 12:39, 3 June 2022 (UTC)[reply]
@Theklan Moderators are technically able to delete posts but no post has been deleted based on moderation. Are you missing a post you have written? Qgil-WMF (talk) 18:52, 3 June 2022 (UTC)[reply]
IIRC this comment was hidden shortly, before being split to its own topic. Hiding works like the Flow feature with the same name: the comment remains available, but the reader needs to make an extra click to see it, and the author is encouraged to improve it (which IMO in this case was justified). Tgr (talk) 14:39, 4 June 2022 (UTC)[reply]
┌─────────────────────────────────┘
I think that the new Discourse forums are absolutely what the community needs. I absolutely support them as a resource. --Sm8900 (talk) 12:57, 17 July 2022 (UTC)[reply]
Do you think this forum can be useful to welcome and retain new contributors to Movement Strategy?[edit]
No, it will only make newcomers understand less how Meta works. In the long term, this is making newcomers abandon MediaWiki. This is the worst way to make things. -Theklan (talk) 10:02, 29 May 2022 (UTC)[reply]
Probably not. I'd generally think movement strategy isn't and shouldn't be the place to welcome new volunteers, and insofar as Discourse doesn't use the wiki syntax, it doesn't provide skills that transfer to more Mission-central projects. TomDotGov (talk) 14:14, 1 June 2022 (UTC)[reply]
Sorry, the original question was confusing and it has been reworded to specify that we mean newcomers to Movement Strategy. This includes editors with +10 years of editing experience but unaware of the Movement Strategy. See this related discussion in the forum. (I have tried to edit the section title but I'm breaking something when I do.) Qgil-WMF (talk) 09:43, 2 June 2022 (UTC)[reply]
It would have to demonstrate that it makes a valuable contribution to the discussion and solution of an actual pressing problem of some controversy and magnitude. One example of this would be the discussions of the revisions to the universal code of conduct - would use of the Discourse add to the ongoing discussion? Or would it just provided a 'second consensus', which would make coming to a real consensus harder? TomDotGov (talk) 14:14, 1 June 2022 (UTC)[reply]
The Movement Charter has been mentioned as an example (see here and here). Pressing, controversy, magnitude... it has all the potential ingredients. Qgil-WMF (talk) 09:49, 2 June 2022 (UTC)[reply]
"improve community collaboration on a multilingual platform that is welcoming and easy to use"
to
"explore ways to improve community collaboration on a multilingual platform, including making it welcoming and easy to use, that we can adapt to our multilingual wikis"
@Sj The "explore ways to" sounds very good and we could add this right away. Removing "that is welcoming and easy to use" seems to focus the goal on the multilingual aspect only? I'm curious about your reason to remove them. Adding "that we can adapt to our multilingual wikis" can mean many things. Right now, we couldn't commit to this.
Implementing the Movement Strategy 2030 is already a big ambitious goal. This is the reason why we are proposing to use this forum right away, sustaining that Meta and Telegram alone won't be enough. Setting a goal to bring a user experience to MediaWiki on par with the most advanced open-source forum platform is a 2030 strategic discussion in itself. The Improve User Experience recommendation would be a good context for such discussion. Qgil-WMF (talk) 14:19, 12 June 2022 (UTC)[reply]
Ah my confusion: I thought "ways to improve community collaboration" naturally includes "making it welcoming and easy to use". Of course the focus should be both.
"Right now, we couldn't commit to this." <-- then our projects will dissipate and by 2030 we won't really be the ones calling the shots for the future of collaborative knowledge :) That's a possibility, but we shouldn't be planning for it now. We should definitively, unapologetically, adamantly be committed to learning from whatever experiments we can run and adapting them to or integrating them with what we've built and maintain for the world. –SJtalk19:53, 13 June 2022 (UTC)[reply]
It doesn't have to be an all or nothing? The Talk Pages Project run by the Editing group is the current iteration in a long trajectory of development of MediaWiki-based communication features. This team does research and development in this area, they discuss with editors, they look at trends... They have data showing how the features they are developing are bring positive results to their target users (mainly new editors, also the experienced editors onboarding newcomers). We are in touch, they know what our team is doing here. Qgil-WMF (talk) 22:09, 15 June 2022 (UTC)[reply]
"Setting a goal to bring a user experience to MediaWiki on par with the most advanced open-source forum platform" <-- that's definitely not what I meant by 'adapting'. We need to learn from and adapt any external experiments that otherwise have obvious costs in branching and fragmenting attention... learn enough to make this cost worth it! We could even integrate Discourse directly into MediaWiki; many others in both communities might find the combination useful if we do. But not considering that, and not aiming to understand what can be adapted to connect the different target audiences and workflows, is a lost opportunity. –SJtalk19:57, 13 June 2022 (UTC)[reply]
There is a big difference between "we couldn't commit" to something and "not considering and not aiming to understand" something. You suggested that we edit our goal (a goal we can commit to) with an alternative. I'explaining why this alternative is a goal that I think we cannot commit to, as in we don't have the capacity and we cannot promise it. I see your point, but agreeing today only to disappoint you in the future wouldn't be useful, right? :) Qgil-WMF (talk) 22:17, 15 June 2022 (UTC)[reply]
The WMF has about $400 million in assets (including the $120M? $150M? or so in the endowment), and this year has once again broken its revenue record, as it has done every year of its existence. Last year, the WMF took $90 million more than it spent. In 2020, the WMF gave away $4.5 million to Tides Advocacy, but it has no capacity for incorporating machine translation in MediaWiki. The priorities aren't right. AndreasJN46613:29, 16 June 2022 (UTC)[reply]
We have a giant pool full of money, but we can't replicate any single feature Discourse has, because, well... you know... stuff. Theklan (talk) 06:44, 19 June 2022 (UTC)[reply]
Is there an independent team tasked with determining if this proposal is successful or not? It seems like a basic good governance principle would be to not have a proposal evaluated by its direct proponents. TomDotGov (talk) 13:57, 13 June 2022 (UTC)[reply]
It is quite common that whoever runs a project analyzes the results of a phase and plans the next phase accordingly. What team would you propose? Also, the proposal for the MS Forum moderation and administration states that "This forum is organized for and by the community that promotes the implementation of the Wikimedia Movement Strategy". We aim to have a majority of moderators not affiliated to the Foundation actively involved in the governance of the site. Qgil-WMF (talk) 22:27, 15 June 2022 (UTC)[reply]
What do you think about the proposed name and domain?[edit]
The proposed name is fine, but the domain should be transferred to the Wikimedia Foundation. Using a domain with concealed registration data is a big red flag. It suggests it is either not official, or dodgy in terms of security and privacy. MarioGom (talk) 07:15, 1 June 2022 (UTC)[reply]
@MarioGom The domain is being transferred to the Wikimedia Foundation. It is very unlikely that it will become the final domain based ion the feedback received but yes, you have a point. Qgil-WMF (talk) 10:01, 1 June 2022 (UTC)[reply]
What do you think about the proposed process for new moderators and administrators?[edit]
We don't need it. Meta and Wikimedia has moderators and administrators. We don't need a new bureaucracy only because we can't manage this one. -Theklan (talk) 10:03, 29 May 2022 (UTC)[reply]
Just to be clear, we're talking about the moderation of the Forum here? The link to the topic discussion makes me think we're not talking about wiki administrators. 18:26, 9 June 2022 (UTC)
Are there other channels that you would prefer to use in addition to or instead of this forum for Movement Strategy updates and feedback? Why?[edit]
I prefer to use this wiki, the objective of meta-wiki is concentrate content and discussions that are not related to one single wiki. Wiki pages are not only for documentation, they can be used as a forum too. In my opinion what we need is organize better the wiki pages, I was looking for a wiki page to discuss about the implementation of the recommendation "evaluate, iterate and adapt" some time ago and I didn't find it, maybe we are not discussing about the recommendations in meta-wiki just because we are not finding pages to do that. Danilo.mactalk19:22, 24 May 2022 (UTC)[reply]
This wiki #3. I do not plan to use or even experiment the other platform at all, specially after the WM Space debacle. Improve what we already have, instead of dispersing the effort, patience and time of volunteers by multiple new platforms. - DarwinAhoy!10:09, 29 May 2022 (UTC)[reply]
Meta, and possibly local project notifications. Moving Movement Strategy updates to an more obscure platform is a great way to reduce reach. MarioGom (talk) 07:18, 1 June 2022 (UTC)[reply]
Automatic notifications is only part of the story of course, but it's an important one for a scalable solution that takes into account Meta, the big wikis, and also all the rest. Qgil-WMF (talk) 09:53, 2 June 2022 (UTC)[reply]
Meta is the designated place for these discussions, along with notifications to the project wikis to keep them in the loop. At the very least, I'd expect any channel used to integrate with on-wiki alerts, so we don't have to pick between email spam and one more place to check. TomDotGov (talk) 14:14, 1 June 2022 (UTC)[reply]
I'd prefer to use this wiki, with a caveat: during the 2017 strategy discussions, volunteers were merely asked to "endorse" what the WMF had decided, and I remember the phrase "Please note that the strategic direction will not be renegotiated, but will serve as the agreed upon groundwork for phase 2 conversations. In short, the endorsement means: “This is the right way for us all to move forward together. Let’s go!"
In my experience, Meta is often viewed as its own community (or lack thereof :p ) and project. In practical use it's not _the_ central organizing space. There are many spaces being used for various groups, communities, regions, interests. Some entirely outside of Wikimedia Foundation supported spaces (like Discord!). As a movement as large and diverse as ours we can not ever have just one. This forum is project and language agnostic, is topic-focused, and is easier for new folks. I see this like I see my work with Diff. It is part of a dynamic ecosystem with many opportunities for folks to jump in and learn more. We can build "campaigns" around what we're trying to discuss or work on to leverage these systems and spaces. I wouldn't suggest someone just create a page on Meta as much as I wouldn't suggest someone just post a blog on Diff and call it done. This forum is not a replacement for, but an addition too our toolset to work with communities. 19:03, 9 June 2022 (UTC)
@Qgil-WMF:, I fully support the previous reply directly above, but it appears the editor has omitted their user name. did you write this reply? anyway, well said!! --Sm8900 (talk) 18:48, 15 June 2022 (UTC)[reply]
If there is a feature that is required at Discourse and we don't have it in Meta... we MUST develop it in Meta. That's the only feature we need. -Theklan (talk) 10:05, 29 May 2022 (UTC)[reply]
I fully agree in the long run MediaWiki needs to have modularity and flexibility to serve discussions better than it is now to match Discourse, but I fail to see how this will happen in mid-term, let alone short-term for urgent need of facilitating discussions now. -- Zblace (talk) 11:02, 1 June 2022 (UTC)[reply]
Movement strategy often feels like a distraction from what should be the core mission of the WMF, which is providing the infrastructure for free knowledge. Spending resources on a novel platform like Discourse feels like a distraction from the mission - we should focus on improving the infrastructure first, and movement strategy second. TomDotGov (talk) 14:14, 1 June 2022 (UTC)[reply]
Adding a comment box to MediaWiki took about a year for a full team. Setting up Discourse probably took a few weeks for 2-3 people. Reimplementing Discourse features in MediaWiki to a sufficient extent to make it a good discussion platform would take several years. What's a distraction is this weird insistence to build all our tools in-house when there are perfectly good open-source tools available, and save us years of work. Tgr (talk) 13:55, 3 June 2022 (UTC)[reply]
You seem to have very little idea of how software development works. We obviously don't have $100M to spend on a single engineering project (the entire movement's budget is somewhere in the $200M range and that funds a huge multitude of projects, each of which have their advocates who consider it extremely important), but even if we did, time and money aren't inversely correlated (see e.g. The Mythical Man-Month).
Anyway the reality is that getting funding for good discussion features in MediaWiki is not feasible (more than it is funded already, which is quite significant - I'm pretty sure the Talk pages project cost about two magnitudes more money than the Movement Strategy Forum), and getting funding for setting up a discussion site which already has great discussion features is feasible. So feel free to advocate for the former (there are worse ways to spend money, and the team working on the Talk pages project has done a great job IMO), but it doesn't make any sense to oppose the latter and point to the former as a justification. It's a classic case of the Nirvana fallacy. Tgr (talk) 15:10, 3 June 2022 (UTC)[reply]
No, it's not. We have a pretty good tool for discussing. I'm using it here now, and you too are using the same tool. There are some things missing here, for sure. But there's no single feature at Discourse that is not feasible to get here in Meta. And if there's one, it hasn't been pointed out despite I have asked for it many times. Theklan (talk) 16:55, 3 June 2022 (UTC)[reply]
It's not a pretty good tool. Fixing a typo in your comment requires you to work through several pages of dense wikitext. Adding an image to your comment requires you to learn about Commons and the upload process there and conform to copyright policies which are entirely pointless for this purpose. Finding an old comment is basically impossible - people use custom bots to build index pages to make it somewhat easier. Moving or splitting a discussion is ardous. Moderating a discussion is basically impossible so you end up with rampant incivility and often with the real discussions moving to private lists. The software does very little to guide new users in a productive direction. There is no way to identify good or popular comments. Notifications are very limited (well, at least we do have notifications now, although it is still a beta feature). There is no consistent way for users to introduce themselves. There is no way to send non-public messages so that happens on a different platform. And so on.
Maybe it would be feasible to reimplement any one specific Discourse feature in isolation. Bu that's like saying there is no single stone in the Pyramid of Cheops that wouldn't fit in your garden. Adding the full set of features needed for a good discussion platform to MediaWiki would take several years. Will take several years, hopefully, as it should be done eventually. But that doesn't help with needing a good discussion platform right now. Tgr (talk) 15:29, 4 June 2022 (UTC)[reply]
Sorry, but no. We're using Flow in many places and you can archive, split, summarize and edit your own messages. You can easily ping whoever you want and you don't need wikitext to do this. We can even down-up vote other user's comments (in Discourse only upvoting seems possible, weirdly) using mw:Extension:Comments. And we can't promote the use of copyrighted images for a discussion: we should be promoting just the opposite.
Do we need a better discussion system? Indeed. Whose responsability is to do that? Let's check, because I think that there's an institution called WMF that should be accountable for not doing it when it was necessary.
We have one of the most civilized discussion system in all the web. We are far from a harassment free environment, just as the society is. But he have built some of the best projects in all the Internet and we have done them together, discussing and using our own software. We can improve it: we have plenty of money to do that.
There's a discussion about a Chat feature currently here. I copy my answer here, where it belongs:
If the WMF, or a team withing the WMF, thinks that having a chat extension is a good idea, then they should develop it to work inside MediaWiki technology. The technical process for that is responsability of the WMF, and that’s why the WMF has a budget and workers. The money is not for using Discourse, is for making MediaWiki and Wikimedia better. If a chat makes it better, then it’s their job to do it.
Granting consensus at Meta or wherever is also WMF’s responsability. If WMF is not able to sell the virtues of a chat system to Meta then whe WMF communications team have a real problem. But even if they are not able to sell a chat system to Meta, then the WMF can use… the WMF’s own wiki, where they can decide what tools are optimal for communication without dependency on the Meta community.
Furthermore, there was a wish in the wishlist about a chat client but it was rejected WITHOUT DISCUSSION by… yes, the WMF. So the WMF is saying that a chat would be useful, is rejecting the option to develop it even if the community is asking for it, and then is complaining because they can’t develop a chat system using MediaWiki and inside the community discussion places because of the same community.
Use Discourse to test out future wiki integrations + features![edit]
Discourse is nice in a few ways:
Good translation integration. This is reason enough for me to play with it (it's my top MW feature request).
It's free software, so we can learn from parts we like.
It's simple to set up and customize, so we can experiment.
It's totally devoted to a single narrow use case, with a community of practice around discourse that we can bounce ideas off of
It has some definite cons:
Most discourse installations end up being abandoned and going offline. (happened to 5 of the 6 I have set up or used, the other is phasing out)
Not really designed for refactoring or collaborative writing, just for threaded discussion (but: has a quirky model for summarizing threads)
Not designed to be archival / make revisions and diffs easy to see
I'd like to see this explicitly designed as a short-term experiment, with discussions intended to be mapped back into Flow if/when the discourse shuts down.
What does discourse do that Flow doesn't? that Flow can't?
Could we build explicit integration into MW, for those using both? [e.g.: a bot that maintains a "discourse dashboard" on a wiki showing topics and activity, or consolidated RecentChanges]
How does this inform our future roadmap for talk pages + notifications?
What discourse extensions seem useful / used enough that we should make similar things available as MW extensions?
Who found discourse easier to use, and can they be their own interest group for MW design charettes?
Where did the people that abandoned your Discourse installations go? Were they migrations to something else or end of community project? I'm asking to understand better the circumstances of your projects. Meanwhile, there are plenty of successful uses of Discourse, so I don't think the tool alone is to blame.
Approximately zero collaborative writing happens on talk pages, so I don't think this is a particularly interesting use case. For refactoring discussions more generally, the toolset of Discourse is limited but still far better than that of MediaWiki: you can split and merge discussions, which is theoretically possible in wikitext but cumbersome and messes up history/notifications so it is very rarely used. And discoverability is far better (catagories, tags, a search system that understands the concept of topics), it has reminders (somewhat unhelpfully called bookmarks), a functional moderation system, great comment formatting abilities etc. It is several grades above MediaWiki talk pages, with or without Flow (which e.g. to this day doesn't support searching, and has deviated too far from core MediaWiki architecture to be realistically fixable). The one area where they are maybe head to head is change tracking (Discourse notifications are more user-friendly and easier to fine-tune but MediaWiki watchlists scale better).
That said talk pages will of course always be used for discussion related to a single wiki, so it makes sense to see what features are popular and feasible to adopt in MediaWiki. Unfortunately, with unstructured wikitext talk pages, I'm not sure there will be many. Tgr (talk) 15:56, 12 June 2022 (UTC)[reply]
Past Discourses: Wikimedia has two of them: for technical developers and Spaces. The group that works on Massive Wiki used one for less than a year before traffic died down and the shuttered it. Discussion happens in a Mattermost install instead. Another group moved to Github Discussions. (less overhead, more flexible, easier issue integration). &c.
"talk pages will always be used for discussion related to a single wiki" <-- this seems like a tautology. Talk pages are not very useful at our current scale, and as you note haven't implemented most of the well-honed discussion tools and standards that have developed in recent years. We need tools for on-wiki discussion with better threading, page-annotation, translation, permalinking/archiving, &c. There are many ways to do this. Flow made a few decisions in this direction. Discourse showcases a handful of others that we could make. We could even try integrating Discourse directly. Let's not avoid this necessary change. If there's noone leading or thinking about this currently, we should change that. This conversation -- addressing a real need for a 2-month project that many people care about -- is as good a time as any to make sure this part of our roadmap a) exists and b) is healthy. –SJtalk20:11, 13 June 2022 (UTC)[reply]
@Sj it's not a tautology, it's a prediction. There are important benefits to using a discussion system that's perfectly integrated with your collaborative editing system (same user identities, same permission management system, same UI, same notification system, tightly integrated navigation system etc. etc). Those outweigh the benefits of using an external discussion system with better UX if the subject of the discussion is said collaborate editing of content. That's not to say on-wiki discussion tools shouldn't be improved - much to the contrary, since replacing them is not feasible, it is important to improve them. There is not much point in debating that here, though - you are saying that the WMF product department should work on a thing, you should tell that to WMF Product or the Editing team or whoever is writing / accepting the annual plan (which is apparently only published in a back-of-a-napkin format not really useful for such a conversation :/ ). Tgr (talk) 21:52, 13 June 2022 (UTC)[reply]
My apologies for not being clear. I am only talking here about building systems that are integrated with our collaborative editing system. Same identities, same permissions, same notification system, shared UI elements. integrated navigation. We could absolutely integrate a Discourse-style plugin if we chose to, or simply share libraries, templates, or concepts. See, e.g., Github Discussions : which did a beautiful job. In this context, it is absolutely possible to replace traditional Talk: namespace tools. You may not like it, you may think that it 'deviated from core MW architecture', but a number of wikis fully switched to Flow. Being in a mixed-Flow state is, in some sense, naturally inferior to being in a single consistent state across all of our Projects, and we should choose a target, move towards it, and commit to reunifying all talk experiences on all Wikimedia-Mediawiki-instances.
The 'we' here is hopefully larger than "WMF Product", it is "the community advancing our projects together". That includes the active users of multilingual projects like Meta, present and past wiki developers (most of whom are not WMF staff; who may have worked on MediaWiki whether or not they currently do / feel that is currently a constructive place to experiment with new ideas), people spending annual outlays comparable to our tech budget on other open 'tools for collective thought', &c. The only sense in which any of my comments above are related to staff-nature (as opposed to wiki-nature) is that if WMF staff are investing time and energy in Official Non-wiki Collaboration Tools, I'd like to think they are keeping a ten-year horizon in mind and looking to how those tools inform, synergize, feed back into our own tools. Warmly, –SJtalk04:15, 14 June 2022 (UTC)[reply]
About "the community advancing our projects together", since we are talking about software development here, it is useful to remember that introducing novelties in Talk pages has been a very complex social process. I remember when I was spending most of my professional time on MediaWiki.org (with Flow) and Catalan Wikipedia as a volunteer (with Flow) and then reading passionate discussions about Flow in places where this extension was opposed with strong energy like... here on Meta. :) When doing estimates about the time and money it takes to implement this kind of features, based on past experience it is clear that the social process easily outweighs the technical development. The forum, just like Telegram, Discord, etc, people might like it or not, but it's there, works this way, and people adopt it or not. Qgil-WMF (talk) 23:14, 15 June 2022 (UTC)[reply]
Can we see those time and money estimates, please? It would be useful if everyone involved is in the same page of the discussion. Theklan (talk) 07:44, 16 June 2022 (UTC)[reply]
@Theklan Sorry, first I thought your question was rhetorical, and then when you pinged I was away. Back now. What time and money estimates are you referring to? Qgil-WMF (talk) 15:42, 22 July 2022 (UTC)[reply]
Those you mention at "When doing estimates about the time and money it takes to implement this kind of features..." above. Theklan (talk) 20:52, 25 July 2022 (UTC)[reply]
My point was that developing software takes more time and resources than taking a tested product and using it. And developing software related to Talk pages has proven to be even more complex due to community/social aspects. Qgil-WMF (talk) 13:50, 5 August 2022 (UTC)[reply]
So there are no "estimations about the time an monye it takes to implement this kind of features". Then it is not possible to use this as a point in a discussion, because it could be one month, three years or 34.84USD. We don't know. And we should know. On the other side, and in order to continue with the discussion, nothing prevents the WMF to develop a forum using MediaWiki with all the technological novelties they want. There can be any social answer here in Meta, but the WMF can deploy whatever they want at, let's say, strategy.wikimediafoundation.org. There are no social barriers there. If the problem is that our software is not good, then it is, AGAIN, WMF's responsability to improve that software. Theklan (talk) 14:38, 5 August 2022 (UTC)[reply]
About Discourses closing. As said, there are Discourse instances thriving and Discourse instances closing, both in the profit and non-profit contexts. I have closed several Discourse forums indeed, as part of my work (the examples you mention) and also in my hobby/activist time. Sometimes communities won't embrace a tool, sometimes they will leave it for something else. Discourse requires a community of a certain size to shine. Needless to say, any community tool needs admins/moderators to support the community and enough time to go through a first cycle of outreach, adoption, and consolidation. When Wikimedia Space was closed after six months, we got feedback from many volunteers that the time was too short and that we didn't give any notice. This is why for this forum we are approaching support and governance differently. If the forum isn't attractive enough or isn't good enough at retaining participants then, as you say, the force of community gravity will pull it toward a closure. Our goal is Movement Strategy implementation and this forum aims to be a means to that goal. If it's useful. good. If it's not, fine. Qgil-WMF (talk) 22:48, 15 June 2022 (UTC)[reply]
Do you have any other questions or concerns?[edit]
What single feature is present in Discourse that CAN'T be developed in Meta in any way? I would like to hear about one that is completely impossible. Thanks.
I'm concerned about security and privacy. This forum should be hosted at Wikimedia Foundation infrastructure. The domain should be owned and controlled by the Wikimedia Foundation. You don't need to use a wikimedia.org subdomain for this if it's a problem with Site Reliability Engineering. A dodgy domain registration pointing to an IP in a dodgy hosting is... questionable. MarioGom (talk) 07:22, 1 June 2022 (UTC)[reply]
@MarioGom The domain is being transferred. We are hosting the forum in the infrastructure offered by the Discourse team because it was the fastest and most secure way to do this quickly for the community review. Hosting the forum in our own infrastructure requires having someone with the skills to maintain this software on a production server, and this takes more time and money. Let's see how the community review and the adoption of the forum goes. Your point is taken. Qgil-WMF (talk) 10:06, 1 June 2022 (UTC)[reply]
Discourse seems to require users to provide an email to join, while on-wiki editing does not.
Fundamentally, this seems like a repeat of the Wikimedia Space experiment. That seemed not to work the first time around, and so it's hard to understand why we're repeating the same experiment a few years later. TomDotGov (talk) 14:14, 1 June 2022 (UTC)[reply]
There is a discussion about email requirements with a suggestion for an alternative solution (get a free email account somewhere decent) here.
The underlying software used in this forum (Discourse) is the same that was used for Wikimedia Space, but this is not "the same experiment". The lessons learned from Wikimedia Space have been applied here. Qgil-WMF (talk) 10:04, 2 June 2022 (UTC)[reply]
Perennial proposal: build a new silo to unite the old silos[edit]
Alas this never works.
Silo X will not unite the disparate wiki + IRC + email + FB + Discord + TG + WhatsApp + Insta discussions.
Instead let's aim for a small number of lovingly-integrated pieces, with enthusiastic maintainers. –SJtalk14:56, 11 June 2022 (UTC)[reply]
Absolutely - and it should even not work, at least not just for the convenience of maintainers at the cost of diversity of needs and modes of interaction *(this was WM Space fail). Mind you we still miss A/V rich one *(not arguing for TikTok ;-p) IMHO - it is the Coordination and Interoperability that should be the midterm goal. -- Zblace (talk) 03:34, 16 June 2022 (UTC)[reply]
Please eliminate corporate jargon from community discussions[edit]
I don't want to hurt anyone's feelings, but I found certain category names and descriptions on the forum really off-putting. Two examples:
"Coordinate Across Stakeholders – Deepen engagement with spaces for communication and collaboration."
"Increase the Sustainability of Our Movement – Create people-centered approaches to generating and sharing resources."
Please remember that you are talking to a community of volunteers. Try to make simple things sound simple in your communications, rather than using flashy corporate buzzwords that distract more than they communicate. --AndreasJN46613:45, 16 June 2022 (UTC)[reply]
Hi, the titles of the recommendations (that we use as titles for the categories) are set, and we'd rather keep them in the forum for consistency. The second part in the description are texts that we wrote when setting up the forum, and they can be easily changed. We agree that simpler language is better. To keep track of this suggestion, do you want to create a topic in the Forum improvements category? Otherwise we can. Qgil-WMF (talk) 14:51, 16 June 2022 (UTC)[reply]
I've said what I meant to say; you are free to heed it or ignore it. I will not waste any more time on it: you are getting paid for it, I am not. AndreasJN46620:41, 17 June 2022 (UTC)[reply]
I really don't understand this proposal. It seems to be an off-wiki forum for no good reason? Please just host this on-wiki. Thanks. Mike Peel (talk) 22:38, 31 May 2022 (UTC)[reply]
There is not even a warning, that the links on the page are to somewhere outside the Wikiverse. They look like some normal wiki-links, but they point to some outside venue. The minimum requirement would be a proper warning about this unwanted behaviour. Grüße vom Sänger ♫(Reden)04:28, 1 June 2022 (UTC)[reply]
Yes, they do now for me as well, there must have been something wrong on my side yesterday, as the sign for external link after the link was not shown on my screen. Sorry for the intervention. Grüße vom Sänger ♫(Reden)10:14, 1 June 2022 (UTC)[reply]
@Qgil-WMF, thanks so much for your helpful link above. as you know, I am there and participating actively and regularly. i find this forum to be a really valuable resource. I appreciate your help with this info. thanks! Sm8900 (talk) 18:38, 15 June 2022 (UTC)[reply]
I would like to ask why the WMF has been so non-responsive in the new forums which have been established for discussion of the movement strategy. There have been numerous ideas being offered by various people and groups at the new forum website, many of them from regions around the world. The WMF amount of response, at the new forum website, has been very low; on some new threads there has been little or no response at all. I do not understand why. I thought that was the whole point of having the forums in the first place.
Just a mention for the casual reader that @Sm8900 has posted this question in other places public and private, where it has been discussed at different lengths. I have replied to it here. Qgil-WMF (talk) 15:46, 22 July 2022 (UTC)[reply]
@Qgil-WMFis totally correct; this has been fully addressed elsewhere. if anyone in the community wants to comment,. they can, but I have received several helpful replies. I want to thank Qgil for their valuable work. Sm8900 (talk) 17:19, 22 July 2022 (UTC)[reply]
The goal of the community review is to build consensus for a decision by 24 July 2022.
Summaries of the discussions are shared and updated as the review proceeds. Anyone can contribute to these summaries with feedback and edits. A report of the review including the decision will be drafted openly.
[Survey] Please provide your feedback about the Movement Strategy Forum[edit]
In the August 2022 community review report, it was mentioned that every 6 months, a survey will be conducted among users to collect qualitative data about forum use.
Therefore, in order to "evaluate, iterate, and adapt" we would greatly appreciate your feedback about the Forum. Shared below is a link to a brief questionnaire that will ask about your experience in the Forum as well as some limited demographic information.
This survey is for research and evaluation purposes, and will aid the Wikimedia Foundation in assessing the Forum's impact and effectiveness. We will anonymize and aggregate results before sharing with Forum organizers and publishing to Metawiki.
This survey should take around 5 to 10 minutes of your time and will be available until 26 February, 23:59 UTC.
This survey is conducted via a third-party service, which may subject it to additional terms. For more information on privacy and data-handling, see the survey privacy statement.
[Reminder] Please provide your feedback about the Movement Strategy Forum[edit]
Please take 5-10 minutes to let us know what you think! This brief questionnaire will ask about your experience in the Forum as well as some limited demographic information. It will be available until 26 February, 23:59 UTC.
This survey is for research and evaluation purposes, and will aid the Wikimedia Foundation in assessing the Forum's impact and effectiveness. We will anonymize and aggregate results before sharing with Forum organizers and publishing to Metawiki.
This survey is conducted via a third-party service, which may subject it to additional terms. For more information on privacy and data-handling, see the survey privacy statement.
