[whatwg] WHATWG on Google+

On Mon, 21 Nov 2011, Boris Zbarsky wrote:
> On 11/21/11 3:39 PM, Ian Hickson wrote:
> > If you can tell me which pieces those are, I can see what I can do 
> > about updating the annotations mechanism to make those checkins easier 
> > to filter out.
> That's the problem.  The set of changes that matter to a particular 
> person is not static...
> Up until about 6 months ago, for example, I would probably have been 
> interested in any change involving browsing contexts; at this point I'm 
> probably not.

If the number of people who would benefit from explicit annotations is 
small, I would be happy to add explicit annotations for those people. 
Since I have to do those by hand anyway, I'm happy to change the list as 
time goes on.

Another option is for someone (possibly me) to create a system whereby 
people can subscribe to specific portions of the specification, and for a 
tool to detect when a diff affects that portion and e-mail them. I'm not 
exactly sure how to make that work, but if it's something people are 
interested in, we could figure something out.

> > Have you been able to make use of the existing annotations at all?
> I'm not sure what you mean by "existing annotations" here.  Something 
> separate from the changeset commit comments for the spec?

Each checkin starts with an annotation saying which conformance classes 
the checkin affects, e.g.:

   [html5] r5891 - [c] (0) Continue the conformance chain for inline	
   <script> elements. Somehow I'd [...]

"c" means "conformance checkers"

   [html5] r6813 - [e] (0) Forgot to remove this now false note

"e" means "editorial"

   [html5] r6829 - [acgiowt] (1) extend timezone format to also allow	
   omitting the colon

"a" means "authors", "g" means "gecko", "i" means "internet explorer", "o" 
means "opera", "w" means "webkit", and "t" means "tools".

The Tracker page converts these to icons:


By looking only at those changes with the Mozilla icon, you can easily 
ignore things that affect only conformance checkers or authors. The 
tracker explicitly has a way to completely hide purely editorial changes.

(The number in brackets in the checkins is supposed to indicate how stable 
the relevant section of the spec is, and controls how red the lines are in 
the tracker; the theory being that changes to stable sections are more 
critical than changes to new sections.)

Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Monday, 21 November 2011 12:54:10 UTC