[IMPORTANT] Concrete proposals!

goran.krampe at bluefish.se goran.krampe at bluefish.se
Mon May 12 14:36:24 UTC 2003


Hi Andreas!

"Andreas Raab" <andreas.raab at gmx.de> wrote:
> Hi Göran,
> 
> [Comments from the back seat ;-]

;-)
 
> > > Your proposal will help us remember that, yes, so-and-so proposed
> > > such-and-such, but won't help code into the image unless there is
> > > a change in attitude about what gets in and what doesn't.
> > 
> > What do you mean with "change in attitude"?
> 
> I guess Josh means the same thing that I mean - a concentration on pure
> minimalism. For example, your quote says "as long as it doesn't introduce
> inter-package dependencies". Now there are packages where this is perfectly
> reasonable; for example if one would take the facial animation package and
> hook it up with speech synthesis you'd get plenty of dependencies and
> rightfully so.

Well, don't make everything so darn black and white! Of course we are
talking about judgement calls here. But you must surely agree with the
long term goal of partitioning the image into packages with well
understood interdependencies? I surely hope.

> > Reread the plan I quoted above and explain what you
> > think is wrong.
> 
> Well, there is nothing wrong with the *words* in that quote ;-) But I think
> they can be easily interpreted along the lines of the above. To give an
> example, one might interpret the thing about "improvements that could be
> merged into their respective packages in the image" like "really, we don't
> want this but unfortunately the packages haven't been removed yet".

Well, in essence we *don't* want this. :-) What I mean by that is that
it is much better if the packages are broken out and then maintained
outside the image (or at least halfway like Avi and I are planning to be
able to do with Monticello, more on that below) instead of getting even
more intertwined *in* the image.

But again - we shouldn't let code rot either. I haven't looked at the
TrueType-package yet but perhaps it would be better if someone took the
time to carve out a Text package or whatever and put this stuff in there
instead of simply trying to push even more stuff into the update stream.

But if that is very hard work (as I suspect it is) then we should
probably get it into the stream. But never forget that eventually it
should end up in a package. How on earth are we otherwise going to reach
the 200kb kernel?!

We will never get to a partitioned image if people insist on paddling in
the other darn direction.

> > > I understand that different people have different interests and
> > > reasons for participating in the Squeak community.  However, I
> > > strongly believe that it is critical to acknowledge the vision that
> > > lead to Squeak's creation, and incorporate a commmitment to 
> > > continuing that vision into the mission statement.  The quote from
> > > Dan had something like "number-one priority"; we don't have to be
> > > that emphatic about it.
> > 
> > But Joshua - please remember that this is now a *real* Open Source
> > project.
> > In open source people pursue the itches and needs they have.
> > We can't tell people what to do.
> 
> But, Göran, even if Squeak has been an "Unreal OSS Project" (somehow I like
> this term - I have some interesting visual associations with it ;-))) noone
> ever told people what they should do. People have always been following
> their own interests. And that's not what Josh is asking for.

Well, I hope not. I just want people to realize that we are in a
different situation now.
There is no SqC anymore. Simple as that. As Guides we will of course try
to revise the mission statement - I have already promised to work for
that. But we can not "promise" to work in a certain direction just
because SqC did that, everybody must understand that. We can only work
in those directions that the community is showing real interest in
going. And by that I mean not just posting on the list "Please go in
this direction.". You need to do the walk and not just the talk!

So all you Dynabookers or eToyers or SaveTheWorldWithSoftware-ers -
stand up and form a group or whatever! :-) Seriously.

We have you Croqueteers - though I am not exactly sure where you are
going. We have the MCP group obviously caring a lot about Morphic. We
have the KCP people caring about cleaning up the heart of Squeak. We
have Squeakland - which I am not following and I do not know their
status, because they don't talk much on this list AFAICT. We have
Squat/Craig and similar people interested in smaller things. And we
probably have a bunch of other groups I don't remember right now.

> > > Repeat: Squeak is the result of a vision that the community seems
> > > to be in danger of throwing away.  Alan has said that Smalltalk is
> > 
> > I can't see anyone throwing anything away. Squeak will be what we make
> > it! And if noone steps forward pursuing the original vision (which
> > one is that btw? eToys? Dynabook?) then it simply will not be pursued!
> 
> The flaw in that argumentation is that you assume our vision would be about
> computers per se (such as a piece of software or hardware). It is not. It is
> about the shape of the world, the state of eduction, many things which have
> no direct relation to computers at all. Computers can *help* realizing the
> vision, but as Alan likes to say "the music isn't in the piano" and
> similarly the "ideas aren't in the computer".
> 
> By claiming that all it takes is someone from the computing community to
> "pick up the vision" you are ignoring that there are many people out there
> who need software engineers in order to follow the vision. Check out the
> Squeakland site, its mailing list and subscription. Hardly any of the
> subscribers there will be able to build systems which enable them to do the
> important thing - go into the classroom and change an entire generation!

The whole Squeakland project is based on eToys, right? And who is
leading that project?
Michael? Fine. Are the people involved in Squeakland active on
squeak-dev? I sure hope so. But frankly I don't hear much from that
direction. Hardly anything actually. To me the Squeak development
community more or less equals this list. If you want to affect where
Squeak is going this is the place to be. If you are not active here
then, sorry - what can I say?

> If we are true to the vision, then it's our duty to help those people
> helping us, helping the kids to grow up in a better world. If you proclaim
> that "it will simply not pursued" by the computing community you are taking
> away an important level of support that the vision requires.

Eh, what? These three paragraphs are all just super fine words. And I am
not kidding.
But what do you want from me/us? Explain to me what I am "taking away"
and from whom.

To me it sounds like you want us Guides and thus the community to follow
in the footsteps of Alan and SqC - but what if it turns out we are just
ordinary mortals with down-to-earth needs? I use Squeak for work. So
does Ned. We want Squeak to be better, better, better for pretty much
everything. Simply put - we have our own agendas just like you have with
Croquet. What gives you the right to tell us what to pursue?

And sure, I like the visions painted by Alan etc. I am not saying I
don't. I am just saying that I will never fool people into thinking that
I will spend a lot of my time on trying to save the humanity!

I am having a hard time just saving Squeak. Seriously.

> I'm sorry to say it again, but you are thinking *way* too small here.

I am starting to suspect you will not be pleased with us until we
promise to simply just continue on the same path that SqC/Alan staked
out. And I will not do that. Sorry.

What if *I* have a grand vision that simply has nothing to do with your
vision?! What makes your/Alan's vision more important to pursue than
mine? In short - what makes you or Alan better than me?

Sorry if this sounds provocative, that is not my intention. I am just
having trouble getting through it seems. Or I am simply all wrong - that
happens too, but then you will just have to convince me.

> Cheers,
>   - Andreas

regards, Göran



More information about the Squeak-dev mailing list