Report of the meeting of SC-03-07 Working Group on Audio Metadata of the SC-03 Subcommittee on the Preservation and Restoration of Audio Recording, held in San Francisco, 2008-10-04
The meeting was convened by chair C. Chambers
The agenda and the report of the previous meeting were accepted as written.
The chair thanked Chris Lacinak for agreeing to be vice-chair on this group.
Development projects
AES-X071: Liaison with SMPTE Registration Authority
A discussion was held on the merits of having a formal liaison with the SMPTE RA. The role of this particular project was explained by the standards secretary as the place where information on the registration of metadata was discussed (Class 13 in particular).
AES-X098A: Descriptive metadata for audio objects
Project scope: To collect information on all metadata issues pertaining to digital audio objects and all aspects of the digital documentation of digital audio objects. This scope includes field structures to describe and provide access to the audio content contained in digital files. It includes transfer, preservation and restoration information.
AES-X134: Core audio metadata set and XML tag definitions
Project scope: Develop a core audio metadata set and to study and produce a standardised set of XML tags, along with their definitions, that describes the core metadata for audio exchange to accompany professional audio when being transferred by file or streamed formats and define these within the SMPTE metadata dictionary.
Projects X098A and X134 were discussed together. The situation was summarized thus:
Good work had been done several years ago in the AES to capture the requirement for descriptive metadata for the audio industry generally. However, it has become clear that each segment of the professional audio world (Broadcast, Music recording etc.) have their own requirements at this level so the value of the AES proceeding independently with descriptive metadata sets seems low. The proposal is that the AES works with individual industry segments to use their proposals in providing standardisation appropriate for that segment, leaving the AES to publish core administrative and technical areas centrally.
One of the group members had been asked at the last meeting to go and compare the AES X098A draft with the draft EBU Tech Doc 3293 to see what differences they contained with respect to the needs of the Broadcast industry. This group member reported minor differences that could easily be addressed because bothdrafts were originally based on Dublin Core, and had evolved to address similar issues. There remained issues with a user-role list and a common method of describe the way that sample-accurate time is described. If these differences could be accommodated, a single document could be produced.
The standards secretary suggested that, in order to ensure an ISO approved standards document was published, the AES could publish a version of EBU T3293 itself with the EBU as the recognised source. A question arose concerning the video references of EBU T3293 but it was felt that it would not be sensible to strip all the video references from an AES publication, particularly where these helped to put audio elements into context.
It was agreed that the AES standards secretary would have discussions with the EBU with dual publication in mind. Meetings of EBU P/MAG will be taking place in November and this would be an opportunity to discuss and AES proposals.
As referred to earlier, one area that needs work is to extend the sample field to enable a sample accurate reference to be generated. The chair of SC-03-06, D. Ackerman, suggested that the string could be supplemented with a TCF schema to give sample accurate reference as already developed. Ackerman was asked if he could produce a suitable schema for the sample accurate timing area.
The other area of difference between the AES and EBU documents was in the area of roles. In the AES document we have three publishing fields while the EBU has a single publisher field followed by a role type in a table. A discussion then occurred on the benefits or each approach. It was felt that the AES could use the method that the EBU have employed as it seemed more flexible.
AES-X114: Metadata review
It was reported that IEC 62379 part 2 has now been published which has produced a standard audio SNMP MIB for the control of network attached audio equipment.
It was also reported that the following work is going on in ISO TC46-SC9 which may have relevance to the AES.
The international standard recording code is to be revised. This will include a metadata schema.
The digital object identifier is now under vote.
The international standard name identifier is now being produced and is out for ballot. This is the identifier that provides for the registration of public representation of parties.
The NISO web site should have some additional information.
More generally it has been noticed that products are starting to have more metadata and tools to manage this data. This will start to put pressure on workflow standards and the ability to pass metadata between processes as this builds.
The group agreed and a general discussion developed around the issues that will be raised in the coming year. The warning of several years ago was repeated that in the standardisation of metadata structure, getting too engrossed in workflows is a danger as these are often specific to organisations while standards should to concerning themselves with the understanding of interfaces and identifying the workflow requirements across these.
Liaisons
There is a more general informal liaison with SMPTE and this was thought to be very important as they are producing standards with significantly more metadata than before. How this is handled needs a wider discussion within AES Standards. How, for example, should AES groups request and handle class-13 metadata? It was felt that a web page should be produced to provide a focal point to the collection and distribution of class-13 metadata. The question remains on how the governance of such a structure should be carried out but it was felt that a small working group could be set up to advise.
There is informal liaison with the EBU P/MAG group as at least one member of this group is also a member of that EBU group.
New projects
No new projects were proposed.
New business
There was no new business.
The next meeting will be scheduled in conjunction with the AES 126th Convention in Munich, Germany, 2009-05.
