Re: Requirements document for the September OHS meeting

From: Hugh Davis (H.C.Davis_at_ecs.soton.ac.uk)
Date: Tue 10 Jun 1997 - 03:39:47 CDT



At 16:14 09/06/97 CDT, Jim Whitehead wrote:

>How broad should the OHP be?
>
>I think the OHP would be a raging success if it only accomplished the
following:
>
>Browse functionality:
>- develop an interoperable locSpec
>- allow an OHP client to retrieve a document from an OHP server
>- allow an OHP client to initiate a link traversal

This is a *very* important issue, especially following the Soutrhamptonm Workshop.

The original spec for OHP as proposed by Antoine was that we produce a simple protocol for communication between a simple application (e.g. Word4Win) and a link service. OHP Version I aimed at being this. However, subsequent meetings suggested that actually OHP was a rather more complex beast, and was primarily a protocol for communication between 1 (or more) linkservice(s), and a Session Manager (or whatever we call it today) on the client side. The protocl would thus need to define the behaviour of the Session Manager.

The protocol as it looks now is probably too complex to be implemented in Word Basic and omits many important things on the grounds that they are part of some other protocol (mostly DMP). However, I still believe it should be possible to use most of it in the simple way intended.

Sigi and I are shortly hoping to come up with a new version of the protocol paper for inspection by the community, and I believe that one of the most important issues is ..

"Do we procede with something that we know is not perfect, but which will actually perform a rather limited functionality which will be adequate in 95% of cases, or do we try to come up with the perfect protocol/architecture which will meet all eventualities".

My fear is that if we push the thing much further now, before actually getting some implementations done, then we will end up with a leviathan like HyTime that few will be able to understand and noone can implement!

So - when we do put out the paper draft (aiming for start of next week) I would prefer it if any critisisms aimed at reducing complexity rather than increasing it!


Hugh Davis                        ( hcd_at_ecs.soton.ac.uk )
The Multimedia Research Group,    ( http://www.ecs.soton.ac.uk/~hcd )
Electronics and Computer Science, ( Room 1051, Building 16)
University of Southampton,        ( Tel: +44 (0)1703 593669 )
Southampton,  SO17 1BJ, U.K.      ( Fax: +44 (0)1703 592865 )
            


This archive was generated by hypermail 2.1.5 : Tue 13 Aug 2002 - 07:20:52 CDT