Re: [css-experimental] catch-all place for new spec ideas

On May 21, 2014 12:44 AM, "Tab Atkins Jr." <jackalmage@gmail.com> wrote:
>
> On Wed, May 21, 2014 at 10:21 AM, John Daggett <jdaggett@mozilla.com>
wrote:
> > Looking over CSS drafts these days, something that strikes me is that
it's often hard to distinguish experimental feature proposals from ones
that are intended for actual implementation and usage.  I think it creates
confusion to have new properties listed in specs labeled "CSS3 XXX" or
"CSS4 YYY" when those are really pie-in-the-sky items unlikely to ever see
the light of day.
> >
> > I think it might be interesting to have an explicit "spec" labeled "CSS
Experimental" for containing various ideas that haven't been quite flushed
out or need more work.  Having them all together would help in curation
too. Related features belonging in different modules could be grouped
together.
> >
> > Keeping features in a catch-all bucket like this makes it clear to
authors that these features are experimental.  For implementors, it
provides a simpler way of discovering existing ideas for solving given
problems.
>
> I don't think it makes much sense to have a ton of unrelated features
> thrown into a single spec just to indicate they're "experimental".
> Why not just figure out a way to better indicate that some feature is
> experimental?
>
> For example, we could add a small "Under Construction" icon next to
> the headings of experimental sections.  Would be easy to build into
> Bikeshed.
>
> ~TJ
>

I agree with the concept about idea vs official.  I'd like to suggest
perhaps we think bigger.. This isn't unique to CSS, and its something
several of us in public-nextweb@w3.org (the Extensible Web CG) have
discussed a bit... We really need to figure out a way as a community to
progress from idea to rec where we expect ideas (and even prollyfills and
polyfills) to flourish and rise or fail.. providing a vision for how
developers can find submissions in the w3c queue (like these) as well  as
quickly understand their status seems like a necessary ingredient
(official, but experimental here - we have unofficial on github examples
before that already).. It would be a really awesome thing if CSSWG would
help lead the charge in helping to make a proposal to figure that out-set
an example and begin the discussion.  <3