Grants talk:Project/Timeless/Post-deployment support

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search

Question[edit]

Can you summarize what are the main differences with the previous submission 6 months ago? Ruslik (talk) 08:21, 7 February 2018 (UTC)

This is largely the same submission, resubmitted per staff recommendation. There were internal WMF issues preventing its acceptance at the time; these issues are in the process of being worked out internally, but in the meantime relevant staff have encouraged me to resubmit, with assurances that such issues would not come up again. If this proposal is rejected again, actual reasons will be provided - to me, to the communities, to the grants staff, and to the committee.
As is, the only major difference is that the amount has changed - partly due to my having more information now from which to calculate the amount in the first place (this actually changed during the previous submission, but I never got an answer at the time whether or not I was allowed to so change a proposal mid-process), but also because due to the extra time that has now passed, extra work will now be required to make up for that lost time. -— Isarra 16:45, 9 February 2018 (UTC)

Columns & general question[edit]

I've tried using this skin today, and it's ... interesting. I'm honestly not sure whether I like it yet or not - it's more colourful, but it's also less obtusive. One concern I have is that it adds another column to the right of the page (which reminds me of some of Brandon's concepts), which doesn't work so well when there's also an infobox to display on the right (you end up with 4 columns: left-menu, content, infobox, right-menu, which is messy). So, I have a question about how this situation could be handled - would it make sense to move the infobox into the right-menu, or merge the right-menu into the left-menu again, or are both OK?

But that's my specific question. My more general one is, what's the long-term aim here? Is it to switch the default from Vector to Timeless at some point in the future? If not, then is there a channel for improvements from this project to feed into Vector improvements? Or if so, then is there a timeline for proposing the template for community adoption, and a long-term plan to transition the maintenance of the template to the WMF? Thanks. Mike Peel (talk) 23:56, 13 February 2018 (UTC)

I don't wanna answer for Isarra, but I want to say there's a discussion in Phab somewhere regarding the third column (my pre-caffeinated searching can't find it though). Regarding Vector: I don't know of any plans to replace Vector, but I would wholeheartedly endorse such a proposal. I don't think Vector is salvageable, and a completely fresh approach (such as Timeless) is warranted. 😂 (talk)
I think it's good to have multiple concurrent skins: the "competition" facilitates innovation and interoperability. So there is no need for such an investment to have a grand plan to replace this or that skin. In fact, if this development support scheme works, it would be nice to see another skin or two emerge. And then it would be great to have a wider and better supported range of skins in usage on third-party wikis, because many (even in WMF) choose to abandon MediaWiki rather than spend some effort thinking how its layout and interface can be tweaked to their needs in the way one changes Wordpress themes. --Nemo 19:30, 14 February 2018 (UTC)
What they said. And when it comes to eventually replacing Vector as the default, having this and other skins already on hand will prove invaluable for getting a better idea what we would actually need and want of such a default. In the meantime, by using a premade skin like this we can experiment with all sorts of things without wasting time on worrying so much about specific designs before we even know what all the designs of any successor skin would need to address.
As for the third column, interesting point. The idea behind the right column as is is that it should only appear when there's lots of spare space already (very wide displays), but we do want the content itself to remain at a fairly consistent width regardless of how many columns there are, too. And having too many columns of things is potentially an issue, but so is having too long of columns when we have plenty of horizontal space, too... but I don't expect we will be likely to move infoboxes into the right menu thing because that would muddy the distinction between page content and page tools/actions/chrome. These things should not be mixing, probably? I'm not sure. -— Isarra 22:08, 14 February 2018 (UTC)
I'm not able to find it right now, but on some point (on Phab) there even was a proposal to not only put the infobox into the sidebar, but the references as well to achieve a more integrated view. I see the problem with mixing content with page tools, but I think it might be worth some thought. --MGChecker (talk) 16:24, 15 February 2018 (UTC)
On the wider point - OK, fair enough. That reduces my interest in this, but there doesn't seem to be any shortage of other people's interest. ;-)
On the columns, I'd suggest moving the 'page tools' list back to the left, and then having the infobox + languages + categories on the right. That way you're keeping content/article-specific navigation together, rather than mixing it with page tools. References are normally way too wide for a sidebar, surely? BTW, [1] is the case that I've been looking at here, with a Wikidata-enabled infobox and lots of language links. Thanks. Mike Peel (talk) 16:41, 15 February 2018 (UTC)
Yeah, actually that might work fairly well. We really should find the relevant task and add this discussion there. -— Isarra 01:15, 16 February 2018 (UTC)
MGChecker, perhaps you mean Magnus Manske's "CSS explosion" proposal at Wikipedia redefined. --Nemo 09:15, 16 February 2018 (UTC)
@Mike Peel: If you move the page tools away, you lose this nice line: Read–Edit–History-Move-Linkshere–… that Timeless has, but other skins lack. All page actions are in one group here. It doesn't have to be like this, but it is something we shouldn't forget when thinking about this. A more readical change to accomplish this would be to switch Read-Edit-History with Article-Talk as well or maybe dump the former completely into the left sidebar. This would provide a complete distinction of two sidebars: The left one for meta-stuff and actions and the right one for content related stuff (infobox, other projects, languagelinks, categories, images?, references?)
@Nemo_bis: Something like this, yeah, but it focussed more on the references. (I believe I saw it as a mockup on Phabricator.) If you look on a typically enwiki responsive references section, you will notice that their width is small enough to comfortably fit in a timeless sidebar on a Full-HD-Screen. An option I just thought about: Don't show all references there, but let them show at hover (or something like that). It would stay less crowded, but would provide additional use of the sidebar (and retain the possibility to use it for other things at the same time as well).
I thought on moving the images into the sidebar before, but I'm not really sure if it's a good idea, as this would be a radical intervention with pagination as it works in other skins. Maybe one, if there's no infobox.
Just a litte brainstorming here :) Do you think there are some things on it worth considering? --MGChecker (talk) 01:35, 17 February 2018 (UTC)

┌─────────────────────────────────┘

References in new sidebar: @MGChecker: That was me in phab:T158012#3797513 (cf. the "I dream" link) which would make the citations more visible and useful to readers. I believe other people have suggested the same idea in previous years. E.g. I think mw:Winter's "Right Rail" idea (I can't find screenshots) also included (or we discussed including) citations at the right.
Unused site-sidebar space: The wasted space from the existing sidebar(s), in both Vector and Timeless, is my major frustration with skins. See File:Wikipedia sidebar complexities highlighted.png. The Green area is what bothers some people (wasted space, makes things harder at small window sizes), and the Orange area is what bothers other people (abundant power-editor-centric links cause many readers to ignore the whole thing, they're wildly inconsistent between all our sister projects, possibly other reasons). I hope we don't make that worse, by adding even more permanent & unused whitespace in a new column. -- Some of the people who like Timeless, and some of the people who like TheDJ's "Mobile Vector" gadget, really appreciate them the most when they reach the thinnest responsive breakpoints, and completely collapse the site-sidebar, and they wish that was permanent at all window sizes.
Fixed width and sidebars: I understand that many people want a sensible maximum fixed-width for the prose content for readability (especially people with widescreen monitors, who use maximized windows), but that has to be done with consideration for basic data tables (let alone the large ones) and panoramic photos, otherwise we're causing problems/downgrades at the same time as fixing others. Quiddity (talk) 02:45, 17 February 2018 (UTC)

Eligibility confirmed, round 1 2018[edit]

IEG review.png
This Project Grants proposal is under review!

We've confirmed your proposal is eligible for round 1 2018 review. Please feel free to ask questions and make changes to this proposal as discussions continue during the community comments period, through March 12, 2018.

The committee's formal review for round 1 2018 will occur March 13-March 26, 2018. New grants will be announced April 27, 2018. See the schedule for more details.

Questions? Contact us.

--Marti (WMF) (talk) 01:53, 17 February 2018 (UTC)

Idgi[edit]

(Since this is not an endorsement, I am not writing on the main page.)

The skin was enabled as an option globally in Wikimedia projects specifically on the premise that it is a volunteer development that is completely optional and it is just for testing reasons so more people could use it and send the bug reports to the developers. OK, that is fine, but now people developing it want 16,000 USD for ‘post-deployment’ support of this skin, although it was not properly discussed anywhere both in communities (outside of handful few that would jump on any new feature) and it wasn’t discussed whether WM sites really needed another volunteer development that would be more than likely used by a small minority and then eventually stop being supported or even taken into account when taking the decisions about the site’s design.

We’ve already disabled a handful of such skins before and I hope that we will eventually disable more. So the simplest question here is: is there any rationale for WMF to support the development of another volunteer skin financially when their current track record shows that almost no people are interested in them, they create additional technical debt for developers and volunteers that created them have to be on constant WMF payroll to keep those skins working for a really small minority of users (less than 5%)? stjn[ru] 20:43, 12 April 2018 (UTC)

I think this comment mischaracterizes the deployment of Timeless. Well certainly the initial deployment was for testing purposes, later deployment requests make no mention of the word test [2] [3] [4] [5] [6]. I also find it puzzling to suggest that english Wikipedia would jump at any new feature. At least in my experiance, they are usually very hostile to new features. Bawolff (talk) 20:44, 13 April 2018 (UTC)
That was directed more to Hebrew Wikipedia, to German Wikipedia etc. I mean, as a matter of fact, they had to make ‘will not change’ bold in English Wikipedia so people could not be worried that another skin is being sneaked through project by project so that eventually it would be enabled everywhere without consent of the communities that they didn’t care to ask. There should’ve been a bigger RfC regarding whether or not Wikimedia community even needs new optional skins, and that, frankly, should’ve been judged on arguments, not on votes like most of those discussions. It already has more than enough, no need to add more and more – there are a lot of community efforts in skinning, but at this rate nothing but Vector and Minerva Neue would ever have continuous support from the WMF (and it is already like that). stjn[ru] 21:34, 13 April 2018 (UTC)
Sneaking some code into the wikis without demand and then demanding resources to maintain it is something that WMF does regularly, so I understand this caution, but it's extremely rare for anything similar to happen with community developers like Isarra. I don't think the worry is warranted here: the request is framed as a way to meet the demand that emerged, rather than something we're forced to. If the work on this skin helps us increase interoperability and hence reduce "skin lock-in", all the skin benefit from it and in the future it will be easier to make hard decisions about skins as well. --Nemo 22:40, 17 April 2018 (UTC)
What Nemo said. Timeless is unlikely to be the new main skin or any such, but it allows us to explore new possibilities for using-facing workflows and back-end practices and perhaps eventually have some idea what an effective new main skin could actually be, while at the same time working to improve all the existing ones as well. Various communities asked for Timeless' deployment to the point where it made sense to globally deploy it, so now it only makes sense to roll with it and put it to use for this purpose. -— Isarra 03:43, 25 April 2018 (UTC)

Feedback from Kunal[edit]

Hi!

I was asked to do a review of this proposal, and overall I'm enthusiastic to see this go forwards. COI disclosure: I'm User:Legoktm, I've known Isarra for years, and am serving as an advisor to the project.

  • I think the main significance of this project would be to provide a real competitor to Minerva/MobileFrontend on adapting MediaWiki for smaller devices. MobileFrontend was a good step forward from the previous mobile gateway, but it was the only serious responsive skin until now. When MF was still being rolled out, there was a decent amount of experimentation upon it because the audience was much smaller than the amount of people who use Vector. But now that MF is pretty widely rolled out, it seems mostly feature frozen. In addition, there are parts of MediaWiki that have moved on and MF hasn't caught up (e.g. Notifications). Timeless will hopefully provide a more suitable audience to trying out responsive design concepts that enhance the interface.
    • Oddly enough Timeless seems to be also shaping up to be a competitor with MonoBook. The double sidebar interface provides a more rich editing experience, and links that editors might want to click on more often (e.g. user contributions) are moved to more prominent places as well. I've heard from editors who couldn't stand Vector immediately switching to Timeless, which is surprising, but great.
    • Overall I think more competition in the skins area is going to be a positive outcome. I hope that some of the positive things that people like about Timeless can slowly be pushed into other skins like MonoBook, Vector and Minerva. It looks like some of it is already going into MonoBook :-)
  • The main concern that's been brought up (and stjn mentions above) is the extra maintenance burden that Timeless would bring, both on the MediaWiki infrastructure side, but also to editors, gadget authors, etc.
    • I think there are enough MediaWiki developers who want to see Timeless succeed that I'm not that worried about the MediaWiki maintenance burden right now. Worst case, the skin is clearly advertised as an experimental feature so it can be undeployed whenever.
    • For editors, the main work would be needing to ensure content displays properly in fixed width skins. Given that this is already a problem for readers on the Minerva skin, this isn't actually a new problem - it already exists and won't go away. For gadget authors, there will be more work to support Timeless. Most gadgets currently only support MonoBook-like skins. Ideally we should work towards making gadgets skin-agnostic as much as reasonable. If MediaWiki isn't providing proper APIs for skin-agnostic gadget development, we'll need to fix that - Timeless should make it obvious where we are missing things.
    • There's an underlying meta question of whether we want more or less skins. We previously removed a bunch of skins because they were outdated, barely used, and problematic to support, pointing the needle in favor of less skins. But at the same time, we've seen very stagnant skin development, suggesting that maybe less skins that no one is brave enough to significantly alter may not be what we want either.
  • The goals seem reasonable for Isarra to accomplish. I think her previous work on Timeless and other MediaWiki skins/extensions demonstrates that Isarra knows what she's doing. In addition, if you follow wikitech-l, you'll know that Isarra is never afraid to ask for help when she isn't sure how to move forward.
  • The main risk that I see is that some giant architectural change (MCR maybe?) forces a large part of Timeless to be reworked, taking up time from fixing/improving other features. I think the likelihood of this happening would be low, since most of the ways I can see this happening would probably be violations of MediaWiki's deprecation policy.
    • Risk #2: TemplateStyles is currently being deployed to Wikimedia wikis to make it much easier for editors to develop responsive content. Since Timeless is not planning to transform content like MobileFrontend does, having some types of content display well will require TemplateStyles. As I understand it, TemplateStyles is a priority for other WMF teams as well, so I would expect any problems with that deployment to be addressed by WMF developers.

Kunal / Legoktm (WMF) (talk) 01:48, 16 April 2018 (UTC)

Aggregated feedback from the committee for Timeless: Post-deployment support[edit]

Scoring rubric Score
(A) Impact potential
  • Does it have the potential to increase gender diversity in Wikimedia projects, either in terms of content, contributors, or both?
  • Does it have the potential for online impact?
  • Can it be sustained, scaled, or adapted elsewhere after the grant ends?
7.1
(B) Community engagement
  • Does it have a specific target community and plan to engage it often?
  • Does it have community support?
7.6
(C) Ability to execute
  • Can the scope be accomplished in the proposed timeframe?
  • Is the budget realistic/efficient ?
  • Do the participants have the necessary skills/experience?
8.1
(D) Measures of success
  • Are there both quantitative and qualitative measures of success?
  • Are they realistic?
  • Can they be measured?
8.0
Additional comments from the Committee:
  • The problem I see here is that it's normal to preview a code review and a bug fixing, but after one year of development it seems to me not reasonable.
  • The project obviously fits with the Wikimedia strategic priorities and have a great potential for online impact. It probably can be sustained and scaled.
  • Timeless successfully received WMF support, thus its impact potential significantly increases. No more block on the WMF side is a very good reason why the project will be sustained after the grant ends.
  • The project is iterative. The risks appear to be low as it is mainly about resolution of existing problems with Timeless skin.
  • Timeless continues being an innovative solutions with a good target and even lower risks than before (it is already used, and it is not blocked by WMF staff anymore)
  • The project (up to 6 months) and its budget appear to be realistic. No problem with skills,
  • Feasible, Isarra and Jack has the skills needed, the third proposal brings even more clarity regarding the budget than the previous two.
  • There is a visible community outreach.
  • Good community support, with a significant number of users at this stage, enough to speak of a real community. It is now already supported by WMF, which makes it even better.
  • To be checked how many times the developer will claim grants to fix the bugs. A bug fixing of a skin after one year indicates that there will be in future similar problems again. It would be interesting to know if the bug fixing of this skin will continue to be on the shoulderso of the WMF grants.
  • I am happy that all problems with the Wikimedia staff have been resolved. So, I support this project again. However I continue to think that the development of new skins should be funded as a core WMF project and not from the regular grant funds.
  • Based on the recommendations of WMF staff
  • Third attempt deserves funding. Reasonable budget, good potential impact for the price, buying a skin for 16kUSD is definitely a good investment.
IEG IdeaLab review.png

This proposal has been recommended for due diligence review.

The Project Grants Committee has conducted a preliminary assessment of your proposal and recommended it for due diligence review. This means that a majority of the committee reviewers favorably assessed this proposal and have requested further investigation by Wikimedia Foundation staff.


Next steps:

  1. Aggregated committee comments from the committee are posted above. Note that these comments may vary, or even contradict each other, since they reflect the conclusions of multiple individual committee members who independently reviewed this proposal. We recommend that you review all the feedback and post any responses, clarifications or questions on this talk page.
  2. Following due diligence review, a final funding decision will be announced on March 1st, 2019.

Questions? Contact us.



Some comments about funding - yes, this is the third attempt at getting funding, but the project has been ongoing regardless. The fact that it's gotten this far without any is indicative of the fact that if it does get funding to actually round things out now, it should be fairly self-sustaining after. On the other hand, the fact that it's gone on so long without funding is also a good chunk of why the amount has gone up so much since the previous proposal - much of the momentum the project has had has been lost, and the longer these delays continue, the harder it will become to rally people into the project to contribute. -— Isarra 03:31, 25 April 2018 (UTC)

Some statistics[edit]

There's more at https://phabricator.wikimedia.org/T180860#4132795

So I was curious how much Timeless is used at present. So I ran some one-off statistics. These definitely aren't exhaustive, but at least satisfied my curiosity, so I wanted to share (I'm also not a stats expert. Its possible I'm using inapropriate ways of measuring these things).

So of public statistics, what currently exists is Wikipedia:Database_reports/User_preferences#Skin. However that's misleading as it treats every user equally, including people who haven't visited the site in 10 years (Note the 1290 using amethyst, which was removed from MediaWiki in 2005). Additionally, you can use quarry to look at only users with a high enough user_touched timestamp (Roughly this means looking at people who have changed a single preference in the last X days) [7]. This is a bit better, but still probably introduces biases (People who are happy with their preferences, don't change them).

So as far as I can see it, there are 2 things we want to measure:

  1. What skins are popular among currently active editors
  2. What skins are popular among readers.

For question 1, I think an obvious way to do this is just look at skin stats for people on Special:ActiveUsers. I did this for several wikis - the data is at https://phabricator.wikimedia.org/T180860#4132795. Key take aways

  • English Wikipedia: 111969 vector, 4200 monobook, 492 modern, 202 cologneblue, 184 timeless, 51 minerva, (30 invalid skins)
  • Most wikis I tested seemed to put the popularity of timeless roughly in fourth place, between Modern and cologneblue. On Wikis where it is behind cologneblue (enwiki, dewiki, itwiki, plwiki, elwiki), its usually pretty close, and its always ahead of or at least tied with Minerva
  • Timeless is third place (behind Vector, and monobook) on no.wikipedia.org, meta, wikidata, wikitech, fr wikipedia, fr wiktionary, he wikipedia [Note, for some of the smaller wikis, the sample size is probably too small to say anything useful]. I consider this to be quite a sign of success for Timeless, as its beating all the skins that didn't have the advantage of previously or currently being a default skin. Additionally, it is second place (behind vector, ahead of monobook) on office.wikimedia.org and mediawiki.org (!)
  • Perhaps 1 edit in last 30 days is too small, and extending to say 60 or 90 days would give better information.

For question 2, what skins are popular among people who don't edit, or only edit rarely, we need a different approach.

I thought for this question, perhaps the best we could do, is look at all users who have registered after timeless was deployed globally (Nov 22, 2017). Here it is for english Wikipedia:

skin editcount # of users
cologneblue 0 151
cologneblue 1 - 10 60
cologneblue 10 - 100 9
cologneblue 100 - 1000 1
minerva 0 228
minerva 1 - 10 95
minerva 10 - 100 11
minerva 100 - 1000 1
modern 0 268
modern 1 - 10 196
modern 10 - 100 10
modern 100 - 1000 2
monobook 0 134
monobook 1 - 10 71
monobook 10 - 100 35
monobook 100 - 1000 10
monobook 1000 - 10000 1
timeless 0 554
timeless 1 - 10 227
timeless 10 - 100 68
timeless 100 - 1000 8
vector 0 848075
vector 1 - 10 231028
vector 10 - 100 28293
vector 100 - 1000 1580
vector 1000 - 10000 66
vector 10000 - 100000 1

Query used:

select IFNULL(up_value, 'vector') 'skin', IFNULL(concat( power(10,floor(log(user_editcount)/log(10))), ' - ', power(10,floor(log(user_editcount)/log(10))+1)),'0')  'editcount', count(user_id) '# of users' from user left join  user_properties  on up_user = user_id and up_property = 'skin'  where   user_id > 32449779 group by 1,2 order by 1,2 limit 60;

The main take away from this (imo):

  • Most people who register an account but don't really edit, never change their skin (Not really surprising)
  • Of users who change their skin from the default, Timeless is clearly the most popular for people who either don't edit or only make a few edits
  • Of new users who become active editors, monobook still seems the skin of choice for people who change their skin (Which is interesting in itself. There's a lot of common wisdom, that monobook fans are mostly people who hated change, but clearly, monobook is still attracting new users in the very active user demographic). Timeless seems to be in a strong second place for highly active new users (albeit the sample size is kind of small to draw conclusions).

All in all, I think this shows that Timeless is already a strong contender considering what stage it is in its development process. Bawolff (talk) 06:13, 21 April 2018 (UTC)

Round 1 2018 decision[edit]

IEG IdeaLab review.png

Congratulations! Your proposal has been selected for a Project Grant.

The committee has recommended this proposal and WMF has approved funding for the full amount of your request, 16,000 USD

Comments regarding this decision:
The committee is glad to support this community-led skin designed to meet the unique needs of editors.

New grantees are invited to participate in a Storytelling Workshop on June 5 and a publicly streamed Project Showcase on June 14. You can learn more and sign up to participate here: Telling your story.


Next steps:

  1. You will be contacted to sign a grant agreement and setup a monthly check-in schedule.
  2. Review the information for grantees.
  3. Use the new buttons on your original proposal to create your project pages.
  4. Start work on your project!

Questions? Contact us.