Re: DOM reference in WebVTT Draft CG Report

On Wed, Oct 19, 2016 at 2:58 PM, Tab Atkins Jr. <jackalmage@gmail.com>
wrote:

> On Wed, Oct 19, 2016 at 2:52 AM, Simon Pieters <simonp@opera.com> wrote:
> > On Wed, 19 Oct 2016 08:26:56 +0200, Thierry MICHEL <tmichel@w3.org>
> wrote:
> >>> FWIW: the HTML reference is to the WHATWG version also, so we should
> >>> be consistent.
> >>
> >> Right. we should be consistent.
> >> The HTML reference to WHATWG version is the same issue.
> >> It should be also update.
> >>
> >> as I said earlier, if we start having diffs from the CG draft and the WG
> >> draft, it may be a nightmare to synchronize those documents by updating
> refs
> >> and other stuff for each TR publication.
> >
> > I believe bikeshed (the tool that generates the spec) is able to switch
> to
> > stable references for TR publications (+cc Tab Atkins). No manual labor
> > should be necessary. In fact TR publications for WDs should be automated
> > also, but is not yet done for WebVTT. See
> > https://github.com/w3c/webvtt/issues/205
>
> Yes, Bikeshed can switch between "current" and "snapshot" versions of
> a single spec. WHATWG HTML vs the W3C fork is a different situation,
> however, and Bikeshed takes the explicit position that you should be
> linking to the actual HTML spec, as written by WHATWG.  You can link
> to the W3C fork as a biblio reference with `[HTML5]` or `[HTML51]`,
> but all autolinks go to the actual spec.
>

I believe there is a difference of opinion as to what is the "actual spec",
certainly in the case of DOM4 and elsewhere.


>
> ~TJ
>
>