OHP update

From: Peter Nuernberg (pnuern_at_daimi.aau.dk)
Date: Tue 07 Apr 1998 - 05:29:31 CDT

Hello all-

   This is just a short note on some of the latest news on the OHSWG front. I've just returned from Southampton where I spent a very interesting and productive few days. For those of you who have not heard about this plan already, we are proposing a demo for HT 98 that will show Chimera, DHM, HOSS, HyperDisco, and Microcosm (and some representative clients from each of these efforts) interoperating using the services defined in the latest version of the navigational hypermedia interface. I think this demo will serve many important purposes, including putting a lid on the basic navigational front, getting our first specification under our belt, demonstrating to some watchers at the IETF and W3C that we are "real", and providing us with the warm fuzzy feeling that only comes through tangible results!

   As I think all of us expected, now that several labs are well into the implementation of the current specifications, we found a few problems and weak points. We tried to balance the need to keep the specification stable (since we are already implementing!) and addressing particularly bothersome parts. We made two non-trivial changes to the specifications during our meeting. I'll briefly describe them here so that you get a feel for them. They can be discussed on their respective subgroup lists in more detail.

  1. Removal of presentation specifiers and scripts. None of the systems taking part in the demo used these features, and both seem ill-thought out in their current form. The current specification even basically says that some of the functionality of the interface as currently defined is basically there as a placeholder until we can think through these issues a bit more carefully. These two areas seem to fall into that category. We are already missing functionality we all agree must be in some later version of the specification (like contexts), so we decided not to implement anything on these fronts for now. Further discussion should be directed to the navigation subgroup list.
  2. Redefinition of the non-component-based OTW encoding. We have ditched the "tagged ASCII" approach of the current specification in favor of an SGML-like markup. This simplifies the parsing considerably and allows us perhaps a more formal vehicle for specifying the OTW issues. Ken Anderson has pointed out that since there are free and reliable XML parsers, it would be to our advantage if we can ensure that our specifications do not use any of the SGML features not in XML. Further discussion should be directed to the OTW subgroup list.

   The Sotonoids (Dave Millard, Hugh Davis, and Sigi Reich) and the Arhusians (Uffe Wiil and I) have agreed to provide free hard and soft copies of the specifications (in IDL and {SG,X}ML) and (pointers to) free implementations of the basic communication infrastructure. Dave and Sigi are taking over the specification generation, and will hopefully post their drafts to the appropriate subgroup mailing lists a few weeks before HT 98.

   I'm looking forward to any discussion and/or helpful comments on this and the subgroup lists, and/or to seeing all of you at HT 98!


Peter J. Nuernberg
Department of Computer Science, Aarhus University, Denmark http://www.daimi.aau.dk/~pnuern/

This archive was generated by hypermail 2.1.5 : Tue 13 Aug 2002 - 07:20:55 CDT