[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Hobbsian processes

• To: "'Bullard, Claude L \(Len\)'" <clbullar@i...>, "'Paul Prescod'" <paul@p...>, "'Mike Champion'" <mc@x...>
• Subject: RE: Hobbsian processes
• From: "Didier PH Martin" <martind@n...>
• Date: Fri, 23 Aug 2002 11:21:43 -0400
• Cc: "'XML DEV'" <xml-dev@l...>
• Importance: Normal
• In-reply-to: <2C61CCE8A870D211A523080009B94E430752B87A@HQ5>

Hi Len,

For our fellows in that list here is a good link about Nash's
equilibrium: http://william-king.www.drexel.edu/top/eco/game/nash.html

Len said:
So can you explain to me how adopting or not adopting 
the data standard affects that equilibrium?  Are you 
saying that in such an equilibrium, nothing changes 
unless the whole industry changes?  Or that nothing
changes until some threshold of the number of installed 
software systems of a given type use that standard?

Didier replies:
Yes you are right on that, this is caused by the network effect and the
well known technology introduction curve. You always need a critical
mass before the mainstream barge in. I think that we see an important
modification of the ecosystem when these numbers take off, for instance,
the HTML standard (or de facto standard), HTTP, SMTP and several other
technologies reduced the overall cost by reducing transactions'
friction, by reducing the transaction costs or by reducing the costs of
adaptation. Pure Adam Smith behavior leads to increased cost in the case
of communication artifacts (i.e. networks). As soon as we speak of
network, the groups behavior is becoming important. I a group of neurons
do not play the same communication game as the others we face important
thinking problems. When a network is too fragmented it does not work
properly. 

Maybe it is also a question of mindset and demographic characteristics,
the millennium cohort will probably know that by intuition because they
were immerged in the connected world since childhood. This is quite
different for boomers and GenXers raised in a top down and one to many
world. 

The critical mass effect, at least in networked systems, can be
accelerated when a certain consensus has been reached on common ways to
exchange information. This will push the overall equilibrium toward the
ideal solution set. Otherwise, the resultant solution set is still at
the edge. in equilibrium yes, but a lot more instable than the former.
We now have to think not only of equilibrium but also in term of
stability of the equilibrium. Off course, constructive destruction(1)
will occur in times but the scaled reached when the network has reached
some standardization is way beyond what would happen without any.

(1)
http://transcriptions.english.ucsb.edu/archive/courses/liu/english25/mat
erials/schumpeter.html

Cheers
Didier PH Martin

• References:
  • RE: Hobbsian processes
    • From: "Bullard, Claude L (Len)" <clbullar@i...>

• Prev by Date: Re: Hobbsian processes
• Next by Date: Re: How to get realistic expectations for Namespaces (was Best Practice = Vanilla XML? )
• Previous by thread: RE: Hobbsian processes
• Next by thread: RE: Hobbsian processes
• Index(es):
  • Date
  • Thread