Hi,
i am only a lurker (and i will be until i can open linux-texts in filelist :)
But this is more about processes than squeak and i am curious:

1) Why is Lex not in this group? His "universes"-approach sounds
pretty reasonable to me. He is actually *doing* leadership here IMHO.
Locking him out sounds like a very bad move.

2) What is bad about forking? I read a lot about packaging and debian
here. Well, there are the BSD's too. Lots of forks, but sharing a lot
of code.

2a) Is smalltalk ready for a single base-image + packages? IMHO this
leads to a java-approach. Then you should stop biting java and addapt
the rest of it. Not only packages and namespaces. Add strongly typed
interfaces. Forbid adding methods to system-classes. You get a lot
more "just add this packages"-features then! But you lose fun and
coding-speed. Lots of hassle to prepare hooks where you now can add
just a method in object. IMHO the strength of smalltalk is, you have
all the code in the image all the time. And when conflicts occur, they
bite immediate. Thats why you can go without static typing: tests
touch the whole image often and trap conflicts quickly.

2b) How much effort is it to keep the own base-image current? I
mean if a guru-squeaker/group wants to make his/her own image and
apply the current patches to it. Not automatic, but by hand and
reviewing it. It seems there are some succesfull forks (etoys, various
stripped images, croquet, or Lex sorting out his universe).

2c) if 2b) is practical, how about a "race" to the next image? Some
groups of people make their images, and after some time one is choosen
as the official next one. The others take it, port their favorite
stuff, improve, and hope to make be official the next time.


On Tue, 15 Feb 2005 18:15:45 -0400, Lex Spoon <lex at cc.gatech.edu> wrote:
> This coup is illegitimate.  I hope that the Squeak community does not
> fall for this power grab, and that they remember this incident well.
> This power is neither legitimate nor a good idea.
> 
> If there is a Squeak reorganization, we should hear about it from Squeak
> Central itself.  That has not happened.
> 
> This comment about Squeak Foundation supporting them is flatly wrong.
> The new Squeak Foundation does not even exist yet--there is only a
> mailing list discussing the formation of such a thing.  The only way
> such a fledgeling proto-organization could support such a group, is
> through unanimous agreement of the members.  Not only is such agreement
> lacking, but the idea was never circulated at all.  It wasn't even
> discussed in general, much less was a specific list of names sent
> around.
> 
> The proposed "in" list is not representative.  Where are the educators
> and education researchers?  They are Squeak's raison d'etre.  And where
> are the people representing *my* views on how Squeak should move
> forward?  Am I supposed to send supplications to a private mailing list?
>  Get real.  I have been hoping for a new Squeak organization to support
> *more* Squeakers than merely squeak-dev.  This effort moves in the wrong
> direction, and represents only a subset of squeak-dev, which is itself a
> small subset of the Squeak community.
> 
> Overall, of the three tiers of support that a true leadership structure
> of Squeak should have--Squeak Central, Squeak Foundation, and the
> general community--this group has none.  It has no legitimacy except
> outside its own claims.  They can say they are our commanders all they
> like, but it will never make it so.
> 
> And what is the point of this power grab, anyway?  There is no crisis!
> The community has done well over the years by two approaches: put up
> services that people can voluntarily use, and--when we have to agree on
> something--discuss it on the mailing list until a rough concensus is
> reached.  For examples of the first, witness the swiki, SqueakMap, BFAV,
> SqueakPeople, and SqueakSource.  For examples of the second, consider
> the harvesting process.  What excuse is there for someone in the group
> to start bossing the rest of us around?  We don't need a tyrant,
> temporary or otherwise.
> 
> The only plausible reason I can think of for this power grab -- and it's
> regrettably ugly -- is that this is an attempt to sidestep annoying
> people on squeak-dev who disagree with people on the designated list.
> Is this truly a good idea?  I actually think it is a good idea that, at
> the few times we do need to agree on something as a community, that we
> take our time and work on wide agreement.
> 
> Community members, if any of you go along with this, the result will be
> massive fragmentation of what community we have left.  Already, external
> groups have left the squeak-dev-led Squeak because they do not feel
> represented.  Now we are seeing an effort to divide squeak-dev into
> factions: those who follow the designated list, and those who (like me)
> reject it.  If a large portion of the community does follow them, then
> an even larger portion will fork off and make NetSqueak... then
> FreeSqueak... then OpenSqueak....
> 
> We would end up like the Scheme community, with 100 different fiefdoms
> all mildly incompatible, and all pathetically re-inventing the same
> basic infrastructure.  That's a major waste of effort.  Let's work
> together, and let's do what we think is right -- not what a self-styled
> "leader" says we should do.  It is not too late for us to be a loosely
> coupled comunity using a shared codebase.  We have already lost too many
> people who have decided to fork at old versions of Squeak instead of
> dealing with the current crowd.  Let's try and get some of those people
> to come back, instead of forking even further.
> 
> Gentlemen on the "in" list:  Instead of spending those 6-7 days
> discussing how best to make a power grab, I wish you had spent that time
> working on a concrete proposal for a proper democratic organization.  I
> have seen zero effort from you on that front.  Not one post to the
> formation mailing list has proposed concrete details of an organization.
> 100% of your visible effort has been to attain personal power over the
> Squeak community.  Please stop.  Put your energy into working out a
> truly good community, instead of in this destructive direction.
> 
> Consider carefully that you are not going to take over "Squeak" even if
> you try.  Even if it is called Squeak, you will have but a subset of the
> population that you see now.  You will be kings of a small fiefdom, and
> you will be reviled by all those who leave.  You could instead be a
> mid-level leader of a thriving community, loved by those you are
> helping.
> 
> Is there some issue that is frustrating you guys?  What is it that you
> would like to see happening, exactly?  What do you think will not
> happen, unless you someone gain the power to command us?
> 
> Everyone:  Let us keep moving forward as I proposed before.  Let us
> thoroughly packagize Squeak, so that there are fewer central decisions
> that have to be made at all. The more we can packagize, the more becomes
> optional, and the more people can share efforts with the main group
> instead of needing to fork off separately.  This seems to be going along
> well, as far as I can see.
> 
> And let's proceed to *carefully* build a proper democratic organization,
> for those few times that we must make group agreements.  I expect these
> times to be few, because most action in Squeak happens in individual
> sub-projects which will have their own structure.  For those few times
> that we do need to make a decision, we should do it gently and
> carefully.  Let's take our time and build a proper organization that
> includes everyone.
> 
> Quite sincerely,
> 
> Lex Spoon
> 
> 


-- 
-Volker

"Any problem in computer science can be solved with another layer of
indirection. But that usually will create another problem." David
Wheeler