Re: [HTMLWG] CR Exit Criteria redux

On Thursday 2012-09-20 09:23 -0700, Maciej Stachowiak wrote:
> 
> On Sep 20, 2012, at 9:11 AM, Adrian Bateman <adrianba@microsoft.com> wrote:
> 
> > On Friday, September 14, 2012 6:11 PM, Maciej Stachowiak wrote:
> >> == implementation
> >> 
> >> A user agent which: (1) implements the "Web browsers and other interactive
> >> user agents" conformance class of the specification. (2) is available to the
> >> general public. The implementation may be a shipping product or other
> >> publicly available version (i.e., beta version, preview release, or "nightly
> >> build"). Non-shipping product releases must have implemented the feature(s)
> >> for a period of at least one month in order to demonstrate stability. (3) is
> >> not experimental (i.e., a version specifically designed to pass the test
> >> suite and is not intended for normal usage going forward).
> > 
> > Since the goal of testing for CR is to confirm that it is possible to independently
> > implement the spec and achieve interoperability, why is one month of "stability"
> > necessary? If someone ships an implementation that passes tests and demonstrates
> > that interoperability has been achieved, why is it necessary to wait another month?
> 
> That clause is copied from CSS WG's stock conformance criteria, which were used as a model. I don't know the original motivation. I suspect it is there to limit evasion of the "not experimental" clause. It's also my experience that code that someone wrote yesterday is in general less credible evidence that something is implementable than code that has been lived on for a while. Just-written code is likely to have showstoppers. Note that the requirement is that the feature has been *implemented* for a month rather than *publicly available* for a month. So it's likely not necessary to wait a month after shipping in most cases.

My interpretation of this is that it has to be shipping in a
nightly, preview, or beta release for a month in order to count.

My memory of the motivation for this, in the CSS WG discussions, was
that we didn't want to require waiting until something shipped in a
final release, but if it hadn't shipped in a final release, we
wanted to have a decent indication that it wasn't going to need to
be reverted in order to be Web-compatible.

(I actually tend to think it might be best to just require waiting
for a final release; there have certainly been cases where it's
taken more than a month of something being in nightly builds -- and,
in particular, getting it into betas with a wider audience -- to
discover that it wasn't Web-compatible.)

-David

-- 
𝄞   L. David Baron                         http://dbaron.org/   𝄂
𝄢   Mozilla                           http://www.mozilla.org/   𝄂

Received on Friday, 21 September 2012 04:16:16 UTC