Two important issues...
ducasse at iam.unibe.ch
Sat Feb 15 19:47:22 UTC 2003
> I sit on the other side of the world, don't know Stephane, probably
> will. Don't mind him being direct. However, I do mind reading things
> Squeak *must* do this, this, and this or we're leaving.
I'm not saying Squeak should do this. I just notify our current
problems and our current
state of reflection in a transparent way.
Don't you think that we are not frustrated not been able to produce a
changeset for the traits model? Don't you think that I really asked
nathanael to evaluate how difficult it would be to produce them. Don't
think that nathanael spent a lot of time bug fixing the metaclass
kernel (and I can tell that when an excellent guy like nathanael said
that this was difficult, I have problem to imagine it.)
But we have to respect choices and sometimes step back and evaluate how
we can go further.
Also I preferred to let people know what were our problems and how they
because having a decent browser would be beneficial for everybody.
Apparently you would have prefer that I say nothing in this wonderful
asked and I just tell. I think that you should not interpret our
decision as pressure
on Squeak: who is Squeak, nobody!. But we have to make a real
evaluation of where we want to
go and assess our risks because in research we also have business model
> I also mind when people demand the inclusion of things without
> the code that implements it, as if some magic fairy is going to
> some pixie dust and grow a new from scratch approved browser overnight.
But don't worry if we need a browser in Squeak we will do one. Roel did
the starbrowser in VW
because he wanted it, and it is really good. Then I can tell you a true
story one of our old squeakers in the team started to build yet another
browser extension in squeak, we were discussing our the squeaker from
berne could join effort to buidl a new browser.
And yesterday he shows is new version but it was in VW and told me: ok
for now I will experiment in VW because I will go faster but I will
help if we do the browser in Squeak.
So what can I say. Nothing.
I threw away 2 months of night work for my pretty printer just because
I realized that I was stupid not to have build it on RB AST. This was
not the only reason but I learned the lessons: evaluate the choices
before jumping. I learned in the process ;)
> pretty sure those tones were one of the reasons that we lost Henrik,
> contributed quite a bit to the community.
Do you think so? I'm really wondering what you are implying, as if what
I'm saying as any importance in this community. People are doing what
About the fact that henrik disappear, I can tell you that we
***really*** supported the modules
more than a lot of people. Alex helped henrik, we invited him, I hosted
We spend days with andrew and others to understand the module system.
Then I think that the fact that modules where thrown away were the
safest thing to do for squeak. I was sad for all the energy put into
that. I regret because we could have got a packaging mechanism (but
modules were not that one) but this is like that and I hope that this
will not happen to the block-closures.
Now the lessons we learned for traits is that we will not be pissed of
if Squeak missed them.
I will be sad but not pissed off because we are not doing traits for
squeak, we are doing traits because we try to solve problems with
inheritance and mixins. Now if squeak can take
advantage, perfect. In any case I hope that we will be able to play
freely with alternate languages because we are into that dreams.
Jim I have the impression that you are pissed of by me. What can I say?
You can always interpret what I say the way you want.
Prof. Dr. Stéphane DUCASSE (ducasse at iam.unibe.ch)
"if you knew today was your last day on earth, what would you do
different? ... especially if, by doing something different, today
might not be your last day on earth" Calvin&Hobbes
More information about the Squeak-dev