[Squeakfoundation]Concrete SqF projects

Paul Fernhout squeakfoundation@lists.squeakfoundation.org
Sat, 26 May 2001 03:34:18 -0400


"Raab, Andreas" wrote:
> [Snip] The most important issues are in my understanding:
> 
> * Presence and Advocacy
> * Credibility
> * [snip] organizational structure
> [snip] I *really* want to hand over the WinCE stuff 
> but where can I go?! How do we make this public
> [snip] how do we give the person
> taking on this task some initial credibility 
> [snip]

I think these are all great points and important issues.

They touch on at least three big themes related to things Cees and I
discussed the other day. One is "marketing" (education), one is
"blessing", and one is "recruiting".

I would agree it makes a lot of sense for a core group of people to
produce coherent marketing and basic education materials for Squeak.
Certainly some people have already tried (and a few books have come from
this). Your mini-faq idea is brilliant, because it is both easy, needed,
and gives the foundation a voice. One might even have several faqs
maintained by different committees with different interests (VM faq,
Morphic faq, Smalltalk faq, Debugger faq, etc.).

I think the blessing part for various packages is for now just handled
informally through someone interested in them in the community.
Obviously, SqueakC is in the best position right now as the biggest
producer of Squeak content to bless things (and keeps such power whether
it wants it or not in part because blessed things would get carried
along in the current monolithic mainline image and kept working through
refactorings, and non-blessed things fall by the wayside). I'm sure if
you announced on the list "I am handing WinCE to XYZ who is now the
maintainer" that would be most of the blessing that person needed (if
they were otherwise credible). 

However, finding that person to, say, pass the WinCE VM to is a
different story. Obviously, good marketing materials mean more people
participate in Squeak which makes it more likely someone might show up
with the right combinations of skills, interests, and spare time. The
only way to force someone to appear is to pay someone to maintain a port
(and it can still be hard to find and hire the right people for any task
when you have the money). Is that one concept here for the foundation
here -- basically raising funds to support specific ports and projects?
We might ask ourselves how this works in Linux. How does say the
StrongARM port get maintained over time? I don't know, but I can say the
bigger the community and the more important the project the more likely
(by statistics) someone will step forward. 

Perhaps what we need is an official way to say "I'm not maintaining this
anymore" without sounding wimpy. Then somehow the original author of
some group could then "bless" someone who comes along and wants to work
on the project again. After all, why should someone be penalized for
doing a good job for a while on something with being stuck with it --
when that thing might pose an interesting challenge for someone new.

Still, I think there is a clash here between a model of paid work and a
model of voluntary work (where credibility and status comes with just
doing something). I'm not denying the social value of formal
organizations, just that the organizations we are talking about need to
reflect the reality of "herding cats" who are all or mostly volunteers.
Perhaps we just have to accept that there will both be failures and
successes. And of course, if you need to move onto other things and no
one wants to maintain the WinCE port, then maybe it is no longer that
important after all. Maintain for who? Why? If enough complaints are
lodged against the WinCE port as an orphan, then maybe someone may feel
the demand to improve it and respond to that demand. Sometimes things
need to get worse before they get better. (The Theodore Sturgeon short
story "The Skills of Xanadu" circa 1952 talks about a model like this 
for managing a voluntary economy based on literally feeling demand.) 

Also, we need to distinguish credibility from being blessed by authority
as opposed to credibility earned by successful effort on behalf of the
community (from active programming, writing, testing, integrating,
educating fundraising, or whatever).

We could think about what is implied by your phrasing of SqF "putting
its weight behind the effort" in relation to coordinating FAQs or other
things. What does this mean in a formal context vs. an informal context?
Does SqF have to be formal to be able to do this? How formal? Would paid
SqF employees help or hurt this? Who are we convincing and what other
options are there? If Squeak isn't convincing enough on its own, could
it be that Squeak itself needs improving as opposed to improving only
the communications about it? [I know there is the "building a better
mousetrap fallacy" but there is also "word of mouth".]

One Smalltalk shop I worked at used to call anything the developers
didn't have time to get to "training issues". Are we saying Squeak has
"training issues"? Does it have too many? How can they be reduced?

-Paul Fernhout
Kurtz-Fernhout Software 
=========================================================
Developers of custom software and educational simulations
Creators of the Garden with Insight(TM) garden simulator
http://www.kurtz-fernhout.com