About squeak image compatibility (3.6/7/8)

stéphane ducasse ducasse at iam.unibe.ch
Sun Jan 8 21:13:30 UTC 2006

On 8 janv. 06, at 21:50, Samir Saidani wrote:

> stéphane ducasse <ducasse at iam.unibe.ch> writes:
>> So what can I do?
>> If you take the latest 3.9 version 6707, and publish the mcz files in
>> the inbox and scripts I will include them
>> after review. You can also send them to me by email.
> Stéphane, really this question of reviewing sounds sometimes for me as
> a friction into the energy we are deploying for squeak : why don't we
> start from reponsability instead of mistrust ? I don't say that
> reviewing is bad, feedback are important as an educational process to
> understand what you miss, but Edgar has done some work, why don't let
> him to include himself ?

Because right now I put my name on 3.9 and I want it to work.
I will not say to edgar send me your files and I push them without  
daring to look at
what you did.

> Another example is that I proposed to include
> on the squeak core a little enhancement 5 years ago and it was never
> done because X wants to review Y who wants to control Z.

Which one?
5 years ago was another time.

> Giving responsability is a different kind of organization : ok Z,  
> let's
> include it, but be really careful, don't hesitate to ask us, and
> nevertheless we have designed things to step back if there is a
> problem. Notice that this is also an educational process, because we
> give the possibility to make errors and to step back.

Sure normally I ask and give feedback.
Now for example, I ask the morphic team to look at the  
PluggableListMorph refactoring
and they said that this was slowing down the UI stuff for none  
essential functionality
so I did not push the PluggableListMorpRefactoring.

> But in squeak, stepping back seems to be a very difficult task, and I
> think that a lot of problems come from the fact that squeak is not
> "reversible", implying that each step must be carefully done (and so
> leads to endless control and painful reviewing). The question is : Is
> it possible to get a "reversible" squeak in order to put
> responsability in the center of the organization, and not the
> reviewing ?

It should but I do not know it.
I think that having external packages and a build process is the way  
to go and
this is what we are trying to put in place.

> "Lic. Edgar J. De Cleene" <edgardec2001 at yahoo.com.ar> writes:
>> Cees De Groot puso en su mail :
>>> However, the unloaded
>>> packages *must* be adopted by someone who will keep them in sync  
>>> with
>>> 3.9a development so we can build up a Full image when wanted, and as
>>> long as that isn't in place it would be unwise to just drop these
>>> packages...
>> So , who is in charge of Alice ?
>> And the all Games ?
>> And MorphicWrappers ?
>> And many , many orphans ? (today I answer to how load StarBrowser)
>> I wish a Cheetah, not a Dinosaur. And if you wish a "tortilla",  
>> should break
>> some eggs....=:)
> If we unload the package, we would like that a maintainer exists and
> that he dedicates some work to make the same code working with
> different images. There is another way that I'm trying to stress here
> in this thread : the squeak core workers dedicating some work to make
> the same code working with different images. I really think that it is
> the responsability of people changing something to take care that
> their changes do not hurt external code, and if does, to to minimize
> the damage as much as possible. Same applied to coders working on
> library packages. Presently, squeak core workers says : "Ok, we are
> changing the core, it's up to you package maintainers to work to
> maintain your code, we don't care". Sorry, but I fundamentally
> disagree.
> Regards,
> Samir

More information about the Squeak-dev mailing list