Dave and Todd, (01)
DCH
> Thank you for this [talk by Paul Borrill at
> http://www.youtube.com/watch?v=VKkGqNRlUJM ].
> It is very profound, if troubling. (02)
TS
> I always found the current (and past) conceptions
> of time problematic/troubling (but haven't had a
> more coherent and consistent explanation). (03)
At a conference in Germany many years ago, I met Carl Adam Petri
and participated in an interesting discussion. He not only
invented Petri nets, he had also developed some important ideas
about the way the nets represent processes without requiring
a metric time or even a linear time sequence. (04)
Paul B. mentions Petri nets (and related notations) and discusses
some related points. Both PB (in his talk) and CAP (in various
papers) emphasized similar points: (05)
1. In a distributed process, the logical dependencies are critical. (06)
2. Those dependencies imply constraints on the partial ordering
of the events. But there are no logical constraints on the
exact times or on the linear order of independent events. (07)
3. For distributed systems, the exact times are difficult or
impossible to measure, and they are usually irrelevant
to the results of the computation. (08)
These issues imply that a specification of processes in terms of
time focuses on the accidental, often irrelevant linear sequence.
That can be a very serious source of bugs. (09)
John (010)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (011)
|