[Openmcl-devel] documentation format and structure

Hamilton Link helink at sandia.gov
Thu Mar 10 16:09:13 PST 2005


On Mar 10, 2005, at 12:49 PM, Gary Byers wrote:
> On Thu, 10 Mar 2005, Hamilton Link wrote:
>> Speaking of which, what's the state of the rest of the updates to the
>> development environment? Is it in CVS at all? Can I test it out?
>>
>> h
>
> I missed the "first week of March" deadline, and am now hoping for
> "real soon now".  I officially feel guilty.
>
> Barring the discovery of some (new) catastrophic bug, what's in
> bleeding edge CVS now (with the possible exception of the Cocoa
> Hemlock stuff) should be very, very close to what goes in the
> next release.  If people have time and inclination to pound on it
> a bit, that'd be very helpful (it'd be better to find a catstrophic
> bug a little before 0.14.3 than a little after ...)
>
> If anyone else (Alex's trace stuff) has stuff to check in, it'd
> be good to get that done soon, if that's at all possible.
>
> I'm not sure if the Cocoa Hemlock stuff is stable enough for
> public consumption yet.  There seem to be about as many arguments
> for holding off as there are for labeling it "still very experimental"
> and putting it in the main release; if the primary purpose of 0.14.3
> is to consolidate about 8 months worth of bug fixes, that might tip
> things in favor of the "holding off" point of view.

Actually my question of the state of things was directed at Alex as an 
FMyI and not intended to be a pester or a suggestion that his stuff be 
included in 0.14.3.

Technically (although I don't know if anyone has been fooled into 
believing this) isn't 0.13 the stable thing? 0.14 has several 
developmental features, like the ObjC FFI, native threads, and the IDE.

I would like to see the Hemlock etc. stuff included in some functional 
form and not entirely broken, but I've been sorta eagerly waiting for 
Alex to finish doing whatever he's doing to make the IDE better.

Possibly 0.14.3 should have the current (reasonably functional) IDE in 
it so the release package is in a known kinda-works state, but I would 
like to see even broken stuff from Alex get committed back into CVS 
right after that so I can try it out when I build the bleeding-edge 
head.

h





More information about the Openmcl-devel mailing list