persistence of service URLs

[Ivan R. Judson]

> -----Original Message-----
> From: owner-ag-dev at mcs.anl.gov
> [mailto:owner-ag-dev at mcs.anl.gov] On Behalf Of Robert Olson
> Sent: Tuesday, December 10, 2002 3:59 PM
> To: ag-dev at mcs.anl.gov
> Subject: persistence of service URLs
>
>
> An issue of mild importance is the longevity of a particular
> service URL.
> This becomes more important if they are used as long-term
> identifiers, say
> for connections between venues on different servers.
>
> This is a situation similar to the GSH/GSR distinction in OGSA; their
> solution is to define persistent names that are resolved with
> a mapping
> service. I'm not sure we care enough; I'm inclined to solve this
> differently for the time being.
>
>     o	Assign a persistent port for each server we care to make
> 	persistently accessible.

By server here you mean a server that is serving multiple services, right? I
think that's what you mean, and if so, by port I think you mean listening
port; which I think is reasonable. If those aren't what you mean, I don't
understand the above sentence (it has words that seem obvious but are
heavily overloaded).

>     o	Assign a persistent service identifier for each service
> 	we care to make persistently accessible (instead of the
> 	sequentially or randomly generated service id we use
> 	now).

Is there any reason to not expose the UniqueID for the service as a
guaranteed to be unique thing that can make URLs unambiguous? We can declare
these URLs not intended for human consumption right? If we do that, we can
make all services persistent or otherwise referenceable, which feels right I
think.

> With these rules and by taking some care about when we apply
> them we should
> be able to have enough persistence to accomplish our task.

--Ivan