NCIP Conference Call Minutes
Jul. 30, 2007
1 Call Participants
2 Agenda
2.1 Fall Meeting
Scheduled for Atlanta - Nov. 13-14 or 15 (T, W, and possibly Th)
2.2 Topics from previous list discussion
Koppel suggested that we discuss the various topics that ciruclated on the list following the June meeting.
2.2.1 Message Compression
Wilson raised the XSLT compression algorithm dicussion.
Koppel noted that someone who had not been previously involved on the list and who does not attend the meetings had some valuable insight on this topic. Koppel wanted to ensure we continue to include the viewpoints of people like this.
Wilson recounted that the crux of the argument as dicussed on the list is whether there is a true need to compress NCIP messages. He acknowledged that many in the group feel it isn’t really necessary, but the end-users and administrators in the field do see it as an issue. Others on the call agreed that we as the Implementer’s Group had decided that this deserved attention as a result.
Wilson added the Dill had expressed a desire that the compression algorithm use meaning codes so that messages could be read and/or parsed prior to being fully decoded. Walsh suggested that anyone with an alternative view on the process used to compress messages submit a competing proposal. Zemon indicated that the issue would probably come up again when a draft is made available for open discussion.
Wilson asked whether we should post a rationale justifying the need for com- pression. Several on the call agreed that this would be a good idea, and Zemon agreed to draft something to post on the list. Feedback from the list will be incorporated into the statement, and then something can be posted to the web- site.
2.2.2 Proliferation of Application Profiles
Walsh opened a discussion on the concern over the profileration of Applica- tion Profiles. Koppel and Wilson both indicated that, while undesirable, using Application Profiles to define the “more interesting” details associated with im- plementation may be something we are stuck with for the time being. Koppel continued that he hopes we will reach a point in the future where we can move away from the dependency on Application Profiles. Wanner added that this may indicate a need to better publicize the true role of Application Profiles so implementers do not assume that implementing NCIP solely according to the protocol is sufficient for interoperability. Zemon suggested that the group could redefine the scope of the version 2.0 efforts to rethink Application Profiles if it decides that they no longer represent a desirable approach to defining behav- ior. That may be better than perpetuating something we now feel is less than optimal. Wilson indicated that part of the problem is that “the tasks we want NCIP to do are not very well defined.” He suggested that we might address this through bundles of services more focused around a given task. Wanner mentioned that this was discussed at the March meeting, but we decided not to pursue it. Zemon recalled that John Bodfish felt the “founding group” for
NCIP believed Application Profiles were the best way to approach this issue. Walsh added that if we truly want to move away from Application Profiles in the future, then we should stop recommending them today (unless moving away from Application Profiles is a project of such magnitude that it requires signifi- cant reworking of the existing infrastrucure). Jackon volunteered to experiment with creating a bundle of messages and/or services focused on a single task (like checkout), and Wanner volunteered to help.
Zemon asked if anyone had additional topics for discussion. Hearing none, she adjourned the call.
2.3 Next Call
The next call is scheduled for August 27, 2007 at 3pm Eastern / 2pm Central / 1pm Mountain / 12pm Pacific.
- AutoGraphics - Mary Jackson
- EnvisionWare - Rob Walsh
- Ex Libris - Ted Koppel, Mike Dicus, Dalia Mendelsson
- Innovative Interfaces - Lynne Branche Brown
- Overdrive - Russ Fuller
- Polaris - Candy Zemon
- SirsiDynix - Gail Wanner
- TLC - Mark Wilson
2 Agenda
2.1 Fall Meeting
Scheduled for Atlanta - Nov. 13-14 or 15 (T, W, and possibly Th)
2.2 Topics from previous list discussion
Koppel suggested that we discuss the various topics that ciruclated on the list following the June meeting.
2.2.1 Message Compression
Wilson raised the XSLT compression algorithm dicussion.
Koppel noted that someone who had not been previously involved on the list and who does not attend the meetings had some valuable insight on this topic. Koppel wanted to ensure we continue to include the viewpoints of people like this.
Wilson recounted that the crux of the argument as dicussed on the list is whether there is a true need to compress NCIP messages. He acknowledged that many in the group feel it isn’t really necessary, but the end-users and administrators in the field do see it as an issue. Others on the call agreed that we as the Implementer’s Group had decided that this deserved attention as a result.
Wilson added the Dill had expressed a desire that the compression algorithm use meaning codes so that messages could be read and/or parsed prior to being fully decoded. Walsh suggested that anyone with an alternative view on the process used to compress messages submit a competing proposal. Zemon indicated that the issue would probably come up again when a draft is made available for open discussion.
Wilson asked whether we should post a rationale justifying the need for com- pression. Several on the call agreed that this would be a good idea, and Zemon agreed to draft something to post on the list. Feedback from the list will be incorporated into the statement, and then something can be posted to the web- site.
2.2.2 Proliferation of Application Profiles
Walsh opened a discussion on the concern over the profileration of Applica- tion Profiles. Koppel and Wilson both indicated that, while undesirable, using Application Profiles to define the “more interesting” details associated with im- plementation may be something we are stuck with for the time being. Koppel continued that he hopes we will reach a point in the future where we can move away from the dependency on Application Profiles. Wanner added that this may indicate a need to better publicize the true role of Application Profiles so implementers do not assume that implementing NCIP solely according to the protocol is sufficient for interoperability. Zemon suggested that the group could redefine the scope of the version 2.0 efforts to rethink Application Profiles if it decides that they no longer represent a desirable approach to defining behav- ior. That may be better than perpetuating something we now feel is less than optimal. Wilson indicated that part of the problem is that “the tasks we want NCIP to do are not very well defined.” He suggested that we might address this through bundles of services more focused around a given task. Wanner mentioned that this was discussed at the March meeting, but we decided not to pursue it. Zemon recalled that John Bodfish felt the “founding group” for
NCIP believed Application Profiles were the best way to approach this issue. Walsh added that if we truly want to move away from Application Profiles in the future, then we should stop recommending them today (unless moving away from Application Profiles is a project of such magnitude that it requires signifi- cant reworking of the existing infrastrucure). Jackon volunteered to experiment with creating a bundle of messages and/or services focused on a single task (like checkout), and Wanner volunteered to help.
Zemon asked if anyone had additional topics for discussion. Hearing none, she adjourned the call.
2.3 Next Call
The next call is scheduled for August 27, 2007 at 3pm Eastern / 2pm Central / 1pm Mountain / 12pm Pacific.
