>
>> the update is only working for the core because OB is not part of
>> pharo but an add-on.
>> And the add-on are not managed by the pharo stream of changes but by
>> an update of
>> the projects in the pharo Universe (yes this is complex but we do not
>> want that people can
>> say that we are hijacking squeak tools).
>
> The motivation is fair, but the current solution seems untenable...

Yes
>
> I think we should consider cribbing from the model that Linux distros
> use:
>
> Load (more precisely: use -- we might not be able to load it) a /
> specific/ version of the Squeak tool, and then apply a separate  
> (Pharo-
> maintained) set of patches to make it Pharo compliant.

but only the developer of the tool can do that.

> Define a process to specify how bug fixes are to be pushed upstream,
> and how enhancements are pulled downstream -- reassessing the Pharo
> patches each time a new version is pulled downstream.
>
> This way the tool is never hijacked but Pharo has a clear story on how
> it remains stable/unbreakable.
> If pushing bug fixes upstream is supported directly by tools we might
> even be able to make the argument that Pharo is helping the tool
> builders make better Squeak tools ;-)

:)

>>>
>>
>> Yes
>> Normally with Squeak, severe bugs and selected one where retrofitted
>> in old versions. This way seaside 3.8 users got important fixes when
>> we did 3.9.
>
> It depends on how often Pharo wants to do a 'stable' release, if that
> is once every two years I don't see a problem with this model.

we could see that. It also depends on the pace of changes. because
Pharo 1.0 got a lot of deep changes like closures. I do not expect  
that huge changes in
the future.

> But if that happens more often we will need to maintain a lot of older
> releases to make them viable for commercial use, thus spreading our
> resources thin. Hence my desire to pick only select releases that are
> designated to receive 'long term support' so bugfix back-porting
> efforts are concentrated on only one or two older releases to keep
> those alive for say 3+ years.

Yes I see.

> [...for version management]
>> The vision is bytecode loader
>
> Pleas don't go there. Byte-code loading is a very nice-to-have in
> deployment scenarios (lazy plugin loading, live update delivery to
> deployed apps etc) but if there is one thing that using VW's Store has
> taught me it is that it definitely should not be part of a code
> repository that is used for development code version management.

This is in VW store is bad. I do not see why a package MCZ could not  
ship with
a byte-code synchronized version of the source. Then we could have a  
stripper
or a synchronizer that recompile the source code and produce new  
bytecode.

> In the case of VW+Store standardizing on using only source-loading has
> served us well, it is definitely fast enough.
> Building our 'refactoring' image that holds all our projects (almost
> 100 MB image -- just code) takes less than an hour which is quite
> alright in a scenario where one uses nightly builds and updates the
> resulting image during the day-time development cycle. I suggest to
> get source-loading/diffing et al up to speed first.

we will start by fixing packageInfo

stef

_______________________________________________
Pharo-project mailing list
[email protected]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Reply via email to