Re: Drones

[glen e. p. ropella]

Manor Askenazi writes:
 > which is, as far as I can tell, a 100% legitimate request (especially
 > coming from someone who is already a Swarm user), delivered thoughtfully
 > and quite tactfully.
[...]
 > While the former does not deserve to be reproduced in the presence of
 > the latter, they are both symptomatic of a common cause. Namely, that
 > the Swarm team has stated _publicly_ that its current form of funding
 > ends in September, without addressing _publicly_, the obvious
 > implications.
 > 
 > In the interest of the tremendous good spirit and good will generated
 > during SwarmFest'97, I believe an official statement about this matter
 > ought to be produced.

pragma POSTURE (Type => Defensive, Status => On);

Manor, part of the point of discussions is for all who are interested
and have an opinion to get involved.  If there is something that you
feel should be "officially" stated that has not been stated, please
let us know.

pragma POSTURE (Type => Defensive, Status => Off);

Otherwise, Chris is right in stating:
 >   So, as far as an "official" statement - we can "officially" state
 > that we are now looking into all of these options and pursuing a 
 > number of opportunities that have arisen since SwarmFest..
 > 
 >   I am much more confidant at the moment that funding for Swarm will
 > *not* stop at September, we've had several offers from a number of 
 > different groups to continue funding here at SFI...however, the
 > details haven't been worked out yet. 
[...]
 >  In case y'all missed the discussion, it was under the heading
 > "back to the future of swarm"....

Just for reference, it's at:
http://www.santafe.edu/projects/swarm/archive/list-archive.9702/0306.html

pragma OPINION (Status => On);

Now, with respect to the light-hearted comment that David posted and
my totally inane response about membership in hives....  Well, I
understand that that type of chatty traffic is not critical to the
livelihood of Swarm.  In fact, it could be considered annoying by
some.  But, I like it!  It gives a softer, more human dimension to the
list.  However, if alot of people would prefer a more "professional,"
less personable list, let's hear about it!

Now, back to the future of Swarm....[grin].... We do have a general
idea that there should be two parts to the Swarm development entity,
an edu/research effort and a sibling company.  I think that this is
natural in that there are two main thrusts of Swarm, a general purpose
simulation toolkit aimed at the cross-disciplinary use of simulation,
and a robust simulation toolkit to be used to solve real-world
problems.  (That is horribly worded; but, as usual, I will continue on
in the hopes that what I'm trying to say will fall out of the bunch of
words I write. ;-)

In the former case, Swarm is intended to address the "act of
modelling" in the many disciplines where modelling is a fairly new
concept.  It's supposed to provide a common way of specifying problems
across as many disciplines as possible.  In this direction lay alot of
the ideas we have about fully specifying a Swarm language, building
general purpose library objects (like a parameter manager) and
methodologies (like experiment protocols), and making Swarm a
modelling aid.

In the latter, there lay the aspects of Swarm development that make it
more of a workhorse w.r.t. the *execution* of the models after they've
been formulated and specified.  The common issues here are || Swarm,
software robustness, repeatibility, V&V, etc.

We might classify the former set of goals as "functionality" and the
latter as "effectivity."  I tend to think this is a natural cut.

Now, having said that, the other aspect of Swarm development that
cross-cuts both regions (non-orthogonally) is "Swarm as research."
And even though the goal of things like parallelism is to provide more
effective solutions to large scale problems, it is a general area of
ongoing research, in itself.  So, some of these "effectivity" elements
of Swarm development will not fall entirely under the domain of the
commercial development effort.

So, I see the two-part entity of Swarm development as necessary to the
continued progress toward the varied goals for Swarm.  My reasoning
could be flawed, however. [grin] And the important issue, here, is
that our users (including Manor) should *speak up* if they feel that
there is something that's not being addressed.  I don't think the SFI
hive should make decisions like this without regard to the community
and then "officially" dump the word of God onto that community, and I
think Chris and Roger agree with me on that.  But, in absence of
dynamic debate on what form the development entity should take, we
will do the best to provide a cover set for all the development issues
that need to be addressed.  And, it's, in part, due to thoughtful 
comments like those we got from people at SwarmFest and (to a lesser
extent) on swarm-support that have convinced me that we *need* a 
continued educational/research entity to address some aspects of 
Swarm that might atrophy if only a company were formed.

Whew!  I wish I had as much impetus in attacking Swarm bugs as I
do in filosofizing about possible organizational structures. [grin]

pragma OPINION (Status => Off);

glen
-- 
+-------------------------------------------------------------------------+
|glen e. p. ropella (address@hidden) |                                  |
|Hive Drone, SFI Swarm Project         |            Hail Eris!            |
|http://www.trail.com/~gepr/home.html  |                                  |
+-------------------------------------------------------------------------+