- From: Nigel Megitt <nigel.megitt@bbc.co.uk>
- Date: Thu, 3 Mar 2022 08:57:45 +0000
- To: Philippe Le Hegaret <plh@w3.org>, spec-prod <spec-prod@w3.org>, "James Graham" <jgraham@mozilla.com>, Philip Jägenstedt <foolip@google.com>, Mark Nottingham <mnot@mnot.net>, Marcos Caceres <marcosc@w3.org>, "Michael[tm] Smith" <mike@w3.org>, "Tab Atkins Jr." <jackalmage@gmail.com>, fantasai <fantasai@inkedblade.net>, "Travis.Leithead@microsoft.com" <Travis.Leithead@microsoft.com>, Chris Wilson <cwilso@google.com>, "Tess O'Connor" <eoconnor@apple.com>
Hi Philippe,
Seems to me that this biases against any specs that are not primarily implemented in browsers.
I'd like to see safeguards against inappropriate usage of this information, and ways to define other data sources. By inappropriate, I mean using it as a selection input into those specifications that are or are not suitable for onward development, given that the data sources as they are now are patchy at best.
Also, Spec Editors are important, but specs are owned by WGs, who need to agree to this. Can we assume that this would not be mandatory?
Nigel
On 03/03/2022, 00:18, "Philippe Le Hegaret" <plh@w3.org> wrote:
Hi All,
As part of the W3C's "ClearSpec project" [1], one of the goals is to
help the community understand the level of implementation of a
specification ("Document and inform about the level of adoption of
various documents").
Bikeshed displays information from WPT and MDN, for example:
https://drafts.csswg.org/css-conditional-4/ (WPT test results)
https://drafts.csswg.org/css-conditional-4/#at-supports-ext (MDN data)
Similarly, ReSpec displays information from MDN and Caniuse, for example:
https://w3c.github.io/manifest/ (caniuse)
https://w3c.github.io/manifest/#name-member (MDN data)
Various considerations and trade-offs ave been going into those tools.
I can imagine several questions around this information:
- are we using the right information and at the right level and for
the right audience?
- are we representing the information in the most optimal way?
- what about engine coverage, such as mobile results? (WTP tests
results don't cover mobile)
- Are we ok with this information appearing in documents published
at https://www.w3.org/TR ?
- Are we confident about the test coverage and if it meaningfully
represents feature support?
Note: we're using the Spec Editors Community Group to have those
discussions [2].
Feedback and comments are most welcome. If folks would like to get a
virtual call to discuss, I'm happy to organize one.
Philippe
[1] https://github.com/w3c/tr-pages/blob/main/clearspec2021.md#project-2
[2] https://www.w3.org/groups/cg/speced-cg
Received on Thursday, 3 March 2022 08:58:15 UTC