Re: [MUD-Dev] World Persistence, flat files v/s DB v/s ??

[J C Lawrence <claw#under,engr.sgi.com>]

On Thu, 26 Mar 1998 01:55:50 PST8PDT 
Chris Gray<cg#ami-cg,GraySage.Edmonton.AB.CA> wrote:

> In the area we started out discussing, that of attempting to run
> world-changing threads on the clients, that isn't good enough. You
> need to make the entire transaction that the code does atomic. There
> was an example about this a few months ago, involving two agents
> trying to go through a door. I don't remember the details, but one
> agent could block the second, but that second agent had two steps to
> its progress.  The key is that the first step must not be allowed to
> update the DB unless the second step also completes, but the second
> step cannot complete if the first agent blocks the door. Simple
> transactions from the client to the server cannot handle this - the
> entire sequence of steps in running the action must be tentatively
> done, then committed or not, depending on things that may have
> happened in the mean time. I'm not explaining this well...

This is the "Dragon's Dinner" scenario posited by Demos which I just
posted.

-- 
J C Lawrence                               Internet: claw#null,net
(Contractor)                               Internet: coder#ibm,net
---------(*)                     Internet: claw#under,engr.sgi.com
...Honourary Member of Clan McFud -- Teamer's Avenging Monolith...