[dojo-contributors] dojo.Stateful - long live r23032!

Rawld Gill rgill at altoviso.com
Thu Mar 10 17:47:50 EST 2011


On Thursday 10 March 2011 13:53:40 Rahul Akolkar wrote:
> Hi Rawld,
> 
> Inline.
> 
> On Thu, Mar 10, 2011 at 4:16 PM, Rawld Gill <rgill at altoviso.com> wrote:
> > On Thursday 10 March 2011 06:31:47 Rahul Akolkar wrote:
> >> On Wed, Mar 9, 2011 at 11:48 PM, Bill Keese <bill at dojotoolkit.org> wrote:
> >> > We didn't agree to this at the meeting (or at any other time).
> >> 
> >> Correct, its MO -- tautologically, I disagree with anyone who
> >> disagrees with me on something as fundamental as this :-)
> > 
> > Why is it fundamental?
> 
> <snip/>
> 
> Its an important interface in Dojo in that its increasingly pervasive.
> 
> I believe I have replied to most of your questions to me already,
> including why connect() is not the answer, please see comments #4 and
> #7 on #12399, or last meeting's logs. If you do want me to further
> clarify on anything after reading those comments, let me know and I'd
> be happy to provide more input.

Yes, I've read all that and was at the meeting yesterday. I'll give my 
feedback on the ticket.

> 
> I will jump to your summary opinion and make one comment:
> > I save my *opinion* for last: dojo.stateful.watch docs seem fundamentally
> > correct, which make the implementation incorrect.
> 
> <snap/>
> 
> Perhaps (I cannot speak to the original intent), but note that this is
> how dojo.Stateful has behaved in 1.5, 1.6 and continues to behave in
> trunk today. Anything else is a regression (and its worse, because it
> allows no way to get old behavior without breaking the Stateful
> paradigm).

I agree that the argument to not break back compat is strong, although not 
based on strictly on engineering rational.

Best,
Rawld


More information about the dojo-contributors mailing list