distribution (was Re: Documentation: swarm-symbols)

[glen e. p. ropella]

I'm going to have to go with Marcus and Alex on this one.
One of the things I tried to hammer into them when they 
first hired on (if I can be so presumptious as to pull 
this [grin]) is that Order, Methodology, and Process are
Good Things, but they are expensive.  Most often I hate the
content of the "Complexity in Management" email list; but,
they have one thing right: that without the influx of 
energy a system tends to disorder.  This includes people.

Just as a nod to the complaints, however, I will say that
it *really* irritated me that I had to change the value
of SWARMHOME in the Makefiles for the apps.  But, I *knew*
that I was engaging in self-flagellation because all I had to 
do was succumb to the pressure to put the assigment of 
SWARMHOME in my .bashrc file.  But, I hate that and I hate
social pressure in general... So, what can I say?  I deserve
to live in such a huge state with so few people and so little
water. [grin]

But, all that aside, how many of you know how much work 
it is to run through all the applications, running a test
procedure on each application that tests each specific 
feature, packaging it up, and distributing it?  Believe
it or not, even that simple task is difficult (on top of 
the sheer boredom of it).  *If* SwarmSFI were a fully funded,
professional software development effort, then the packaging
as it is would be extremely shameful.  But, as the project is,
with it's focus on the technology rather than the packaging, I
think it's being handled quite well (especially now that I'm
out of the way).

The system that's in place isn't good enough.  That's clear.
But, the improvements we implement must be optimized for 
low cost and high impact.  Distributing/delegating the 
responsibility for keeping the user contributed code up to 
date with the Swarm release and isolating the SFI team's
responsibilities to Swarm and a small set of demos seems like
the best way to do it, to me.

And on that note, trashing the Swarm documentation has become 
a bit like a habitual joke.  Kinda like "Workin' hard, Bob?"
.... "Hardly workin', Dave!"   Hahahahahah.   "Got any 
documentation to go with that library?"  "------"  Hahahahaha.

But, I think it's safe to say that I've only seen *one* contributed
app that was even slightly documented.  And that one shouldn't count
because the guy who contributed it was pressured into documenting
it by his ex coworkers.  (I'm talking about Manor.  Are you
listening, Manor?)  I do understand that alot of the contributed
apps had academic papers to go with them.  And Barry's SCL is
a good example of one of these.  In it he explains the model to 
a decent level of detail.  But, I still wonder if I could 
*reproduce* the model from his specification.  I worry that I 
couldn't reproduce Bank from Manor's either.

Now, this shouldn't be taken as a slight to the users who've 
been gracious enough to donate their code.  In fact, they're
the only people keeping my interest in the "large-scope project"
(meaning the whole distributed development/freeware experiment
we're all a part of).  But, they often just don't have enough
time/money/energy (in that order, of course [grin]) to do the
real detail work.

So anyway, the point of this rambling message from too little
sleep was to suggest that the method we have will work (and
it's my belief that it will work better than forced methodology)
if and only if we regard our software as our dependents and 
feed and care for it after we put it in the anarchy directory.

Speaking of which, there was somebody who asked me about the
3dlib I wrote at the SwarmFest, but I haven't heard from him
since.  I suppose I should go see if that library works with
Swarm 1.1, eh?

glen


glen e. p. ropella         =>Hail Eris!<=         <address@hidden>
The Swarm Corporation                             (505) 424-0448

                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.
                  ==================================