> The documents that I sent around before discuss using a vector time
> stamp where there is a counter for each "site" aka each instance of
> abiword. There is a famous paper here from a MS research guy detailing
> it, unfortunately I don't understand it well enough to explain it fully.
>
> http://research.microsoft.com/users/lamport/pubs/time-clocks.pdf
>
> I think we can avoid using the system clock in order to get GOCollab to
> work.
Whatever the solution will be, when it involves the system clock to be
accurate, _it will never work_. Really. [Disclaimer: I've done quite a
bit of research on system timing/synchronizing].
As a simple example take the clock in my laptop. It is, when I don't run
ntp (something normal users don't do), 15 minutes off each day.
Marc
Received on Mon Oct 24 22:53:45 2005
This archive was generated by hypermail 2.1.8 : Mon Oct 24 2005 - 22:53:46 CEST