Mark Taylor wrote:
> However, I think the two points that John raises are important
> ones, especially the one about what you do on a multi-user machine,
> or more generally, if the well-known port earmarked for the service
> is already in use.
This is a crucial point, we can't assume that the user can talk to a "general user" daemon, we aren't looking at a service here. Everything has to take place in user space, so we can't assume that the 'hub' is up and running continuously. We have to assume that the hub is (mostly) started by the user in some fashion, there may be multiple users on the same machine. The "well known" port may be in use, so there has to be some way to communicate this to the user, and to applications looking to connect to that user's hub (not somebody else's hub that's running on the same machine, but the hub that the user started)...
> My feeling is still that the file-based approach, while inelegant,
> presents fewer problems in practice than the socket-based one, but
> I am prepared to have my opinion changed.
I don't see any way around using a file to solve this problem, I'd welcome persuasion, but a file based approach to communicate this one piece of vital information seems to be the only way around it. Everything else, of course, should go down the wire.
Al. Received on 2007-02-16Z13:20:34