Minutes – Z39.86 Advisory Committee Conference Call

Date: 1 June 2005
Time: 9:00 – 10:10 EST
Venue: telcon
Scribe: George Kerscher
Last Edited: 8 June 2005
Status: Version 1

Attendees

Advisory Committee:

Participating Experts:

Apologies:

Agenda

The agenda was distributed to the Advisory Committee's list on May 31. there were no amendments in advance of the call.

There were no additions to the proposed agenda.

3. Confirmation of Date and Time for Next Call

Gk proposed that the next call be held on July 6.

Update on API Meetings in Japan

George reported on the very successful API meeting held in Tokyo May 18-20 and the following meetings in Urakawa May 22 & 23.

The meeting was very well attended and had multiple representatives from INRIA (France), CWI (Netherlands), NRCD (Hiroshi's organization in Japan), and DFA and the DAISY Consortium. All of these groups are currently heavily involved in SMIL developments and working on the next generation of SMIL and related applications.

We all agreed that a comprehensive abstract data model was needed, and that a core software engine to manipulate the data was needed. all of these groups have resources available to put on the project and some have deadlines within an 18 month time frame. We all agreed that a comprehensive project description was needed. I will have a draft project proposal by June 15 and a final version by the end of June. This will include the commitment of resources by each organization. The DAISY commitment is expected to come from both the DFA project and from software development sponsored by the DAISY Consortium's budget. This currently looks like 1.25 FTE from DFA and 1.5 FTE from DAISY.

Hiroshi and NRCD were a tremendous host and served to focus our attention on the collaborative project. Markus did a great job in laying out the technical issues and forming an action plan. Markus, Marisa, and I are responsible for getting the project proposal completed by the end of June.

Review of Action Items

ACTION: Lloyd still needs to gather up the DTDs and get them to Markus

After the conference call, we have had contact with Cynthia, the editor from NISO. Neil relayed the following:
The edits mentioned to Cynthia are indeed those regarding namespace issues. I wasn't on the last zed call but it sounds like we've settled on what needs changing. So, I'm waiting to get a final version from Cynthia * she said she was "almost finished editing" * which I'll then turn around check against the version Michael sent her to get a list of differences. From there I'll ask someone on the committee, probably Markus, to make the final namespace and maintenance agency changes. I'll then do another document comparison (against Cynthia's version) and present those along with the final document back to Cynthia for posting.

I suggest that in these minutes we finalize the namespace and file name revision issues. This will give Markus an approved set of changes to make to the standard.

ACTION: Markus to have the site up on June 10.
ACTION: Chris will notify Chuck and Skip about the plans for releasing the 2005 version. This will help in NIMAS moving forward.

Summary of Z39.86 URI and file naming schemes

Thanks to Markus for this portion.

General site URL

No objections raised for the URL being:
http://www.daisy.org/z3986/
This is an entry point listing all assets; specs, guidelines, DTDs, etc. including a page with a listing (and a zip) of the latest DTD releases

Spec URL

Local Spec URL would be in the lines of http://www.daisy.org/z3986/2005/yetunknownfilename.html

DTD file names

A majority was in favor of dropping the three-tier naming approach as in zed2002. The New scheme is two-tier, based on major version (== release year as inherited from NISO) and minor version (errata update). Consequent listing of 2005 first release filenames:

System Identifiers

A majority was in favor of canonical http URLs of the DTDs being:
http://www.daisy.org/z3986/2005/filenameAsAbove.dtd

Issue for tracking: A proposal was brought up re parallel publishing of a "minor version anonymous" DTD. This would drop the trailing minor version integers, and just use "ncx2005.dtd" etc. This would practically be a copy of the latest "minor version explicit" DTD, and made available for those processing circumstances where all you need is the latest DTD, but don't care to find out the exact filename of it. The group decided to postpone final decision on this proposal.

Public Identifiers

PIDs obviously keep NISO as the registry reference. Consequent listing of 2005 first release PIDs:

Namespace URIs

Although all they need to be is unique strings, they still cause geeks to go berserk any time they are brought out into the light. Markus and Dave brought forward the issue of NS URIs preferably being as stable as possible; including the major version in the NS URI might not be an optimal choice. Dave further brought forward the proposal that NS URIs be ended with '/ns#' as practiced in the RDF realm.

Finally, a majority of the group envisioned the NS URIs having form: http://www.daisy.org/z3986/2005/rootElementName.

Issue: In the case of Resource with root resources, and bookmark with root bookmarkset, this scheme does not work as perfectly as discussed on the call. Through email discussion, we decided not to mess with changing things in the spec and plan to address this in a future release.

Consequent listing of 2005 first release NS URIs:

Items on agenda, but not covered

Error in DTBook
   - Dave to explain the use of code (block versus inline)

Communications
   - Who wants to join a communications sub-group?
   -Co-branding with NISO (What do we call the spec?)
   - Announcement of new maintenance agency
   - Announcement of revision of the standard
   - Providing a Road Map to our Members
What else?