perfect! go ahead.

> Please let us know what you think about this proposal. What we're 
> trying to achieve is a release that is in sync with Squeakland as much 
> as possible without introducing some more "experimental" changes at 
> this point (e.g., we really don't want to impose large changes to the 
> Squeakland users at this point).

What is experimental in 3.8 unstable? Diego's look? Because else I'm 
curious to know what are the changes that impacts squeakland end-users.

Stef

On 6 oct. 04, at 06:56, Andreas Raab wrote:

> Hi All,
>
> Apologies for the belated reply on the issue but things have been 
> moving quickly (and somewhat unexpectedly) on a variety of fronts so 
> that I didn't have the time to write an earlier message on the 
> subject.
>
> Marcus recently posed the question of whether those people who have an 
> interest in a stable release with m17n would be willing to step up for 
> making it happen. Unfortunately he didn't wait very long for an answer 
> before posting changes into the unstable stream which really don't go 
> too well together with a stable m17n release that should be in sync 
> with Squeakland as far as possible.
>
> After some talking with Michael and Ned, we would nevertheless like to 
> propose that we (e.g., the Squeakland people) manage the release of 
> the 3.8 release under precisely the initial assumptions of it 
> basically being "the m17n release" and not much more.
>
> To this effect here is what we'd like to propose: "We" (that is 
> Michael, Ned, Yoshiki, me and whoever else wants to join) filter the 
> appropriate enhancements and fixes that have already been posted in 
> the unstable stream and repost them into the stable 3.8 stream. This 
> will address most of the issues that needed to be addressed for 3.8 
> and should take no longer than two weeks (the goal would be to have 
> the release out in six weeks at most).
>
> This in turn would mean we go beta as of YYY after which only fixes 
> (and no further changes from unstable) will go in. For the fixes, we'd 
> propose the following mechanism: We will "address" every reported bug 
> in Mantis, which means *you* have to put the bugs there if you want 
> them fixed. "Addressing" the bugs means that we will look at the 
> issues and possibly ask for help in fixing them but we will NOT 
> promise to fix any of bugs ourselves (again it is *your* 
> responsibility to help with fixing the issues). However if there is a 
> fix for a bug we *will* look at it and either accept, modify, or 
> reject it (or ask for improvements). In any case, by the end of the 
> process every bug that has been reported will have been looked at and 
> some resolution will have been found. And of course, help is more than 
> welcome.
>
> Given the number of open bugs at Mantis we expect to complete this 
> phase around mid november. After which we can immediately open 3.9 
> alpha and start to put in the things that would remain in the unstable 
> queue.
>
> Please let us know what you think about this proposal. What we're 
> trying to achieve is a release that is in sync with Squeakland as much 
> as possible without introducing some more "experimental" changes at 
> this point (e.g., we really don't want to impose large changes to the 
> Squeakland users at this point).
>
> Cheers,
>  - Andreas
>
>