[rc5] meta-communication about distributed process evolution

Richard Ebling rebling at xmission.com
Mon Aug 4 03:21:13 EDT 1997


  I think Steve has a good point here.  Like the talking dog [1] the
bovine effort is amazing for having involved so many people so quickly,
each with a miniscule chance of receiving any cash for their trouble.  I
think there are at least 3 major components to this (or any) distributed
computing effort: 

  1. Development (maintenance, improvement) of software for the wide
variety of platforms being used;
  2. Communication of necessary information among all participants; and
  3. Development of an organizational structure/process which maintains/
enhances the desire of participants to remain involved, and facilitates  
items 1 & 2 above.

	IMO, part 1 is (organizationally speaking) a trivial task when a
small band of interested programmers provide software for a large number
of non-programming users.  Part 2 has become technically trivial due to
availability of email and WWW.  

	Part 3 is the hard part.  As others have said (references
desired:) "As an administrative approach, anarchy doesn't scale well." 
It seems to me that a review of the history [2] of distributed computing
will demonstrate a stepwise evolution of useful organizational
structures/ processes, with some methods showing superior effectiveness,
and others heading for evolutionary dead ends.  

	One of the "evolutionary pressures" is that in any but the most
idiotic dist.computing projects [3], the nature of the task and the
internet is likely to result in attracting many participants who are
quite accomplished in programming _and_ who want to contribute, using
their skills and not just CPU cycles.

	I can't imagine that the core Bovine team (those who make and
implement decisions of any kind about direction of the project) was
selected based on their understanding of human interactions, or
interests in applying organizational principles to large number of
humans.  Few projects are.  I suspect that there may be some individuals
around with that sort of expertise; my guess would be that (for many
reasons) <understatement> they're not reading this list.
</understatement>

	Has there been discussion of the "social development" of
distributed computing?  Is anyone else interested in that?  And if so--
is that discussion (sufficiently) on-topic for this list?

 - Richard "Opinions shared freely; purchase price refundable." Ebling

[1] ...what's remarkable is not that it speaks fluently, but that it can
speak at all; source unknown.
[2] I think today, WRT to development of #3, we're at the equivalent of
that scene in 2001:ASO where the monkeys started using tools.
[3] MMF schemes being the canonical example

On Sun, 3 Aug 1997, Steve wrote:
:   [other stuff snipped; this stuff reformatted] 
: It's actually pretty damn amazing what's been accomplished so far. 
: Organizing literally thousands of people that've never met to a single
: task, all of whome have wide varieties of processors, operating
: systems, and knowledge in only the space of a few months.  This is
: literally the biggest distributed computing effort in history, and
: easily the most computing power ever thrown at a single problem.
  [snip]
: -Steve 

----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list