Read on Twitter
1/ @cwilso ran a TPAC breakout today about the challenges to web standards caused by the reduced number of browser engines over the past few years. This thread has my own observations, for minutes of the actual discussion see https://www.w3.org/2020/10/29-engine-scion-minutes.htm.l">https://www.w3.org/2020/10/2...
/2 Once Apple, Google, Microsoft, Mozilla, and Opera all implemented a https://en.wikipedia.org/wiki/Browser_engine">https://en.wikipedia.org/wiki/Brow... . Now there are 3, and most shipping browsers are powered by the engine from the Chromium open source project https://en.wikipedia.org/wiki/Chromium_(web_browser)">https://en.wikipedia.org/wiki/Chro...
/3 Since the web platform is so complex, even large companies and well-funded OSS projects have trouble even reviewing, much less implementing, all the innovative features being driven in Chromium. Those that ship in Chromium browsers can be seen as "de facto standards"
4/ The breakout explored the challenges this creates. @jensimmons pointed out the challenge for end users: Website developers find it hard to figure out which specs they find with a search engine can be considered "standards", that is, are widely supported and interoperable
/5 @t noted the parallel problem that a spec being a W3C "Recommendation" doesn& #39;t mean it is supported in shipping browsers. This causes problems both for both non-browser companies hoping to promote their innovations, and for governments proposing reality-based regulations.
/6 @othermaciej (Webkit) noted a complication: it& #39;s not a binary "support/don& #39;t support" decision to discuss or implement a feature innovated in Chromium, it& #39;s a complex decision based on customer interest, available bandwidth, and philosophy about core values such as privacy.
7/ Various solutions to these challenges came up, including:
Maybe web standards orgs should get into the business of certifying which products conform with their official standards? That would give end users clearer guidance.
Maybe web standards orgs should get into the business of certifying which products conform with their official standards? That would give end users clearer guidance.
8/ Maybe W3C should work harder to "visually" distinguish early-incubation specs from those that are widely implemented, and those which are mature standards? (I like the way WHATWG puts inline feature-level "support tables" in specs to inform readers about support and interop)
9/ @briankardell noted @igalia & #39;s approach to working in all of the OSS repositories to get features users need implemented across browsers. https://www.igalia.com/open-prioritization/">https://www.igalia.com/open-prio... allows anyone to vote on priorities (with their wallet 
https://abs.twimg.com/emoji/v2/... draggable="false" alt="đ" title="Grinsendes Gesicht" aria-label="Emoji: Grinsendes Gesicht">)
/10 Maybe the W3C Process or team practice needs some refinement to encourage single-implementer specs to get documented but NOT confused with "standards".
/11 The TAG could use hints and help to prioritize their architecture reviews to focus their limited bandwidth on the specs with the most real-world momentum
/12 I& #39;m sure this summary missed some key points that came up, it was an information-dense hour of conversation among very well-informed people! My biggest question: How can we continue this discussion in some online forum to try to get consensus on problems and solutions?
