For those who have access the subject is [reading-wmf] Future work idea: skin system and in the e-mail @bd808 describes a skin system and points to us doing T483 (which is now done) with replies from myself and @Tgr .
This placeholder task is probably not helpful, so being bold and resolving, but I would welcome @bd808 and @Tgr taking a fresh look at the skin system as it is now (maybe trying the tool at https://skins.wmflabs.org/#/add for some inspiration) and opening any new tasks on reflection of that email conversation compared with where we are now. :)
Nothing secret in what I wrote in the email that @Jdlrobson mentioned, so I'll happily share it here for others:
From: Bryan Davis <firstname.lastname@example.org> Date: Tue, Jun 9, 2015 at 9:58 AM Subject: Future work idea: skin system To: Internal communication for WMF Reading team <email@example.com> I added these bits to the Q1 brainstorming etherpad today: * Begin researching skin system revamp with primary goal of describing a "better" skin system that allows easier skin creation and responsive CSS ** Better support for extensions adding widgets to skin in a manner that gives the skin control of presentation (eg Echo, watchlist enahancements, etc) ** Better support for template based skin layouts ** Easier to make and distribute a skin in general ** Better support for layouts that aren't just Monobook/Vector with a small amount of CSS/JS added on ** (perfect world) Ability to replace MobileFE with a non-traditional devices skin ** Will require collaboration with other teams and the community ** At least one RfC would be needed I haven't tried making a new skin in ages (like 7 years?) and I'm not familiar with the guts of the Mobile Frontend extension. I have heard lots of smart people talk over the last couple of years about how hard making a skin can be and how useless it it to make a skin that doesn't start off as a fork of Monobook/Vector as many/most/all? extensions that introduce new UI components will not know how to add themselves to your skin. On Wikimiedia properties and industry wide the page view trend is skewing more and more to non-traditional devices (eg "mobile"). Lila made several speeches over the last year that declared that we should be thinking mobile first across the movement. Having an ability to create responsive skins that expose the full functionality of the wiki seems to me to be a necessary component of such a push. The current process of implementing UI features in 4 places (Vector, MobileFE, Android, iOS) doesn't scale. A related but separate project would be to start serious work on https://phabricator.wikimedia.org/T483 (RfC: Allow styling in templates) which could unblock community transition from tables to semantic markup for many widely used templates. Moving away from tables to semantic markup that can be styled by the skin for infoboxes and other Template:Sidebar content would be another huge leap forward for responsive presentation. It would be technically possible to do this without per-template CSS, but I think adoption would go easier and greater flexibility would be achieved if CSS styling was available when work on this was begun by the community. Bryan -- Bryan Davis Wikimedia Foundation <firstname.lastname@example.org> [[m:User:BDavis_(WMF)]] Sr Software Engineer Boise, ID USA irc: bd808 v:415.839.6885 x6855