-
Notifications
You must be signed in to change notification settings - Fork 718
[css-2025] The Current Work page is, well, not current #9790
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
My suggested resolution is to point instead to https://drafts.csswg.org/ for the list of all modules. |
The Current Work page also suggests https://meiert.com/en/indices/css-properties/ as a list of all properties. That page says
and so my suggested resolution for that is to instead point to https://drafts.csswg.org/indexes/ for the list of all properties, at-rules etc. |
The best would probably be to update the status and specifications pages automatically, as far as possible. And the descriptions on that page generally need to be updated. E.g. it still refers to "the development of CSS3", it still advises users to participate on the www-style mailing list and misses the FXTF specs. It would also be good to know where the code of those pages lies, so they can be maintained by more than just one person. Regarding the list of CSS properties, we also have https://www.w3.org/Style/CSS/all-properties.html, which is dynamically generated and so always up-to-date. That aside, I agree it makes sense to refer to https://drafts.csswg.org/ in CSS 2024. Though to get a quick overview of the different spec. statuses, and especially for authors who don't want to take a deep dive into the specs, I also think it is ok to keep referring to the Current Work page ‒ assuming the outlined issues get addressed. Sebastian |
I'm not sure, it says:
I don't understand how an auotomatic index can "miss" properties. |
Thanks, Chris!
I indeed maintain the current work page. Some parts of it are
automatically updated, but not all.
For the list of publications there is an automated process that checks
daily for new publications that are not yet on the page and sends me
email if one is missing. I then add that publication by hand, because,
if it is a new draft, I also write a summary for it. I could automate it
and copy the abstract, but the abstract of first drafts is usually not
very good. The downside is that, if later drafts add significant new
features, I may not notice. (Similarly if editors are added or removed
in a later draft, I usually don't notice.)
The classification such as ‘testing’ and ‘refining’ is somewhat
subjective and hard to establish, so I only reclassify a specification
when it's very clear. (Fantasai, who originally created the
classification, has on occasion helped.)
The status of Color level 4 was an editing error. I found a few more and
I just fixed them.
B.t.w., Color level 5 *is* listed, as a WD under ‘Exploring’.
I removed the link to Jens Meiert's property index.
As for the code used by these pages that @SebastianZ mentioned, it is
well documented. But, for security reasons, only the W3C team can edit
the W3C web site (with very few exceptions).
So, Chris, if you have time to improve the page, please do!
The reason https://www.w3.org/Style/CSS/all-properties.html says that it
may miss properties is precisely *because* is it automatically
generated. It relies on the markup in the specifications and there is no
manual verification. I think incorrect markup is rare nowadays with
specs using Bikeshed, but nine years ago there were still editors'
drafts with properties without markup.
Bert
--
Bert Bos ( W 3 C ) http://www.w3.org/
http://www.w3.org/people/bos W3C/ERCIM
***@***.*** 2004 Rt des Lucioles / BP 93
+33 (0)4 92 38 76 92 06902 Sophia Antipolis Cedex, France
|
Thanks, Bert. Yes, it would be good if you could show me how the current system works. I see a bunch of *.tmpl and *.html files in a variety of languages, but it isn't immediately apparent what is generated and what to edit for updates. |
Bert, please send me email (ie, not here) explaining how the current system works, so that I can correct the various errors people have pointed out (particularly the summaries, but also to update the editors for each spec). I would like to get this sorted before TPAC, so that CSS 2024 (probably published towards the end of 2024) can continue to point to the current work page. Thanks! |
I can also help with that, @svgeesus, if I am allowed to as invited expert. Sebastian |
As requested on the other thread: I researched and couldn't find much info on the history of Specs CSS Modules, and I wanted to know what are the most important changes in terms of modules since CSS3, an idea would be to add a table that can sort by date on official Recommendation and maybe to add fields such as first draft date.
|
Uh oh!
There was an error while loading. Please reload this page.
The snapshot has for some years contained the advice:
This isn't true. I'm not sure who if anyone maintains it (perhaps @bert-github ?) or if it is just some software that happens to keep running. But it certainly doesn't reflect all CSS modules and doesn't reflect current status.
Parts of that page appear to be maintained, perhaps automatically. For example the "What's New" part seems to have an automatic feed, and lists the 2023-12-19 CRD of Backgrounds and Borders 3 as the latest CSS /TR publication. The official CSS /TR page agrees that this is the latest (and also lists some TTML specs, presumably because they reference CSS properties).
Other parts seem to be manually generated and are quite out of date. I certainly didn't check it all; I just happened to notice on a spot check that things were clearly not up to date.
To take an example, CSS Color Level 4 is listed in the "Testing" category, current status WD,Upcoming status WD. While the actual category (from the bikeshed source) is CRD, and the spec went to CR Snapshot on 5 July 2022. The Current Work table also links to a summary:
That summary isn't extracted from the current or earlier abstract, omits a bunch of things (Oklab, Oklch, color interpolation, gamut mapping, and so on) and the list of editors is out of date by 3 years.
CSS Color 5, on the other hand, is not listed at all despite being first published in March 2020.
The text was updated successfully, but these errors were encountered: