Thu, 10 May 2001 23:37:48 +0200
I have just finished reading up on what has happened so far.
Things seem to be stalled. Is everybody waiting for Dan Ingalls's
announced list of projects and processes?
My stab at two most important principles:
* have clear deliverables for each activity (code in the distribution
image, documents, whatever)
* accurately document the current status
But I admit I have difficulties with this chaordic thing (I nearly
fainted when I read that many organizations take a year or more to go
through the process...).
Maybe we should just all blurt out what we want to see in terms of
activities, and draw abstract concepts from that later, if we want.
I'll just start:
* First, let's change the purpose from "To assist in the evolution" to
"Drive the evolution"
Ok, granted, that was still on meta level, but here goes:
I want the Squeak Foundation ...
* to appoint a person as "keeper of the VM" for each platform
this person is responsible to provide an up-to-date VM to a central
location and a list of supported/unsupported features of this
particular platform (Sound, Tablet, whatever)
(typpical example: when ian piumartas releases new sources,
take patches from lex spoon's patches page, apply them, ship
new vm; when an update changes the interpreter, rebuild vm and ship)
* in a similar vein, to strive for a full implementation of all the
existing features that are possible on a respective platform
* to modularize the image (however that may look) and then maintain a
small baseline image and add-on modules
* to maintain a web-based defect tracking system for the bits
distributed by it
* appoint someone(s) to harvest the list for contributions/fixes
* keep some sort of regression test suite to help prevent bit rot from
release to release
* have an easy way to accept project proposals/requests and let people
vote on what the want most
* to be different from the squeak mailing list in terms of ... well, I
don't know exactly how to explain it - I have a feeling that often,
on the list someone suggests something, zillions of people agree,
and then nothing happens; SqF should say "we want xyz (New Compiled
Methods, Block Closures, Jitter, a slicker Look, get rid of Apple's
fonts, ...), and then go for it, (let someone) create it, and then
* connected to that are rules for decision-making - a broad discussion
is already possible on the mailing list - decisions should be made
in a smaller circle
* now then, how do we select that small circle? voting? who can vote?
all list-members? I like the Apache foundation's approach of
members. of course, then, how do we select the initial member(s)? I
suggest let's start with "President Tim", or maybe SqC could choose?
or maybe (probably!) somebody has a different idea
On the other hand, I don't want the Squeak Foundation ...
* to only be a philosophizing (?) debating club
* to be like STIC - can't really describe that, too - it's just that
the scent of irrelevance surrounds it
What else should be thought of:
* how to work with Stable Squeak World Tour, Camp Smalltalk, whoever
I realize much of this is easier said than done. And some of it may
not be desirable to most people. Please say so, I think it is as
important to know what you do NOT want as to know what you want. If
enough people say what they would like to see, eventually something
will emerge that most people want.
If not, we can scrap the Foundation anyway.