Minutes NCIP-IG Conference Call 
Dec. 6, 2002

Attendees:

• Steve Gregory
• Robert Gray
• Janaki Shankar
• Mark Wilson
• Tony O'Brien
• Mark Needleman
• Kevin Stewart
• Barbara Vance
• Jens Dill
• Ed Davidson
• Gail Wanner
• John Bodfish
• Frances McNamara

This Conference Call was hosted by epixtech, inc.

1. Announcements, heads up. 
Mark Wilson told the group that it may be that he will be replaced by a colleague as the TLC representative. He also highly recommended an article in the January C C++ Users Journal on template parsing. He posted the citation to the list during the discussion.
Frances warned that her email had been unavailable for the past 2 days in case anyone had tried to reach her that way before the meeting. John asked to add discussion of a proposal for a guideline to the agenda.

2. Action Items from the last meeting:

• Gail will come up with web seminar (for consultants). Gail reported she has not done this yet and will not try to schedule the session before ALA Midwinter as originally planned. Barbara had volunteered to work on this with Gail. It was decided that Gail will write up her ideas for this session and post that summary to the list for discussion after that, work will begin on putting together the session with the goal of offering the seminar in the spring before ALA.
• Frances: App profiles will be labeled as ‘drafts’. Frances reported this has been done on the NISO site.
• John will maintain ‘messages/profiles supported’ grid. John had not yet compiled this. After discussion it was decided that John will post an example grid to the list and solicit information. In a week, by Dec. 16 he will send a first draft to Frances who must get the password protected pages set up on the NCIP-IG website. John will continue to collect and expand the grid as other implementers contribute their information. Frances will make sure the grid gets posted and passwords distributed to the IG member implementers.
• Frances: Grid will be password protected on NCIP-IG website. Frances reported this had not been done yet but she will follow up and get it set up when she receives the first draft of the grid from John.
• Tony will produce/own a set of dummy XML test data for agencies, users, items (2 of each?), for interop testing, starting with some data from Sue. Tony reported he had not received the data from Sue discussed at the last meeting but had proceeded to compile a set of "simple" test data to start with. He hopes to post this to the list before Christmas but if it is not completed by then he plans to post soon after the New Year. Tony explained he planned to keep it simple by not testing Unicode character set support in the first set of test messages. The others encouraged him to include non-ASCII in authentication prompt and also in Patron Name elements..
• Organizations will report on Z39.50 holdings record structures & data elements which map to NCIP data elements. ExLibris, TLC and Sirsi have contributed their information, the other members have not yet done so. The group asked Frances to post these to the pass-worded section of the website even though there are warnings in most that they can change with time. There was discussion about how difficult it is in the current Z39.50 implementations to deal with all of the variations. There was general agreement that "the whole thing is a mess" (meaning how you get holdings data from library systems via Z39.50) and that the expectation is that the Bath and National Z39.50 profiles should improve the situation. Unfortunately Mark Needleman, who is on the standards committees for those profiles, reported that agreement has not yet been reached. Most expected that NCIP will be needed before those profiles stabilize and as a result implementers will need to continue to struggle with the individual variations. As a result collecting this information system by system will be helpful to early implementers and some of the group have a lot of experience in dealing with the issues involved. It is hoped they can help NCIP implementers who have to implement Z39.50 as well as NCIP for the first time. Barbara said she would provide the Innovative information in the next few weeks, John will provide Epixtech information in January and Ed can provide Fretwell Downing information. There is a hope that there can be some generalization of how best to deal with the Z39.50 information when implementing NCIP and that those who have already done this can share what they know.
• Mark W. to look into NCIP data elements that can potentially be derived from Z39.50 data. Mark Wilson reported he had talked to Barbara Shuh of National Library of Canada and she had suggested that he not try to use Bib1 to do this analysis. This was discussed and disputed. Mark Needleman reported that Bib2 is not actually being used by anyone so it was unclear whether using that would be helpful. Mark Wilson will talk to Barbara again and will probably go ahead and do this analysis based on Bib1 because it is what most people have to deal with right now.
• Mark N. will find and review the academic LDAP schema. Mark Needleman had not received information he needed from David Millman at Columbia to proceed with doing this however he reported that he would see Dave at the CNI meeting both were attending and would get the information and provide the review.
• Mark N. will map open URL to NCIP. Mark Needleman reported there is nothing to map right now as the new OpenURL standard is not yet available, however it was agreed that he should go ahead and map from the de facto Open URL standard that is being called version ".1". Mark reported the new standard will add information. It will be more extensible by providing a registry of metadata formats. Mark will work on this after the New Year.
• John will post proposal on how to do patron name queries in an NCIP application environment. John reported he had not done this but promised to post it to the list for discussion by Monday.
• NEW TOPIC. John Bodfish explained that in doing case modeling for their NCIP application they had run into a situation where they wanted to propose a guideline that all the implementers could agree on. The situation is when an initiator sends a request, responder fails the request because there is a condition (e.g. building use only), the initiator sends a new request specifying agreement with the condition, requestor fails the request with a second condition (e.g. there is a fee) The question was how best to handle this situation. After discussion it was generally agreed that the best thing would be if you said the responder always had to answer the initial request with all applicable conditions, i.e. building use only and the fee. So all conditions need to come back in the response to the Request Item. This requires all the NCIP responder applications to be able to do this, so John will write this up and propose it to the list. Frances will start a page of guidelines for implementers and will put this proposed guideline up in a section marked as Pending. John will ask for discussion and agreement within a 2 month period. At the end of the 2 months, if agreement is reached from the listserv Frances will remove the PENDING and it will be a guideline. It was suggested that the same thing be done with the explanation for how to use invariant character set in NCIP implementations. This had been recently discussed on the list and it would be useful to keep the results in the guidelines section of the website.

3. Other
Mark Wilson recommended ICU IBM Unicode classes for C, C++ and Java as highly useful. He will post a message to the list about this.
The group thanked epixtech for hosting this call.

4. Next meeting. 
The next meeting will be a conference call hosted by Sirsi Friday January 17 at 10AM EST. Mark Needleman will schedule it and provide call in information closer to the time. Tony asked about scheduling the next face-to-face meeting. Frances asked that this be postponed until the January meeting because she hopes NISO will be able to announce the NCIP Maintenance Agency and that the meeting could be coordinated with that agency to be sure they are represented. She promised to contact NISO to try to get a Maintenance Agency representative to participate in the Jan. call.