http://fm.ea-tel.eu/flashmedia/flashmeeting/fm7863_bb7cab-7863 http://fm.ea-tel.eu/fm/memo.php?pwd=bb7cab-7863 http://fm.ea-tel.eu/public lomi 0:5196:0 61:5198:0 222:723:0 894:2434:0 2535:2979:0 2992:4165:0 4255:5223:0 324:5196:0 1909:4015:0 2857:4702:0 Hi Jon, can you hear us? ;-) yes sorry for being a bit late... Hi Erik that all sounds like a good start, Mikael http://www.bl.uk/services/bibliographic/meeting.html re involving others -- yes, I think that's a good idea, Erik, but we may need to hold meetings with a single topic focus not really Only in a vague sense! ;-) AACR3 I'm back -- dropped out for a few minutes welcome back ;-) what would aacr3 be an application profile of? sorry ;-) Yes ;-) plus fully optionalness of LOM No, IMHO FRBR specifies the set of "things in the world" to be described OK - then what does RDA try to add to that? DCAM specifies the nature/structure of DCMI's descriptions of the things in the world It still presents a challenge for how to keep all this stuff as part of a coherent picture! In DC, I have found a need to use the notion of a "base Application Profile" i'll give it a look Don't know too much about FRBR myself No ok - understand the different manifestations of a "work" clearly, RDA & FRBR are very important in the library world. FRBR doesn't get bogged down in the ambiguity of just talking about a "resource" -- the 4-level conceptualisation is very rich FRBR: http://www.ifla.org/VII/s13/frbr/frbr.htm The FRBR doc itself is quite long, but it is quite readable Work - Expression - Manifestion - Item When I said "no" to Erik, I meant that FRBR does not make any assumption that an instance of any of those entities is described by only a single metadata description/record It is perfectly possible to have multiple descriptions of a single Work e.g. distributed on the Web THANKS for the clarification! For eprints DCAP see e.g. http://www.ariadne.ac.uk/issue50/allinson-et-al/ yes Current DC Libraries DCAP: http://dublincore.org/documents/2004/09/10/library-application-profile/ Yes, agreed: steps in themselves have considerable value http://ariadne.cs.kuleuven.be/lomi/index.php/Log20070419 The Australian Learning Federation took a "mix n match" approach some years ago but I know that they regretted doing so because doing so has created a range of implementation/interoperability issues. But I don't think there's any one answer on this issue. Context determines most things! I think strictly speaking the "element" labelled "Coverage" used in a LOM instance and the "element" labelled "Coverage" used in DC metadata are two different things ;-) well put An interesting metaphor for this issue is that we might all speak English but that doesn't guarantee in any way whatsoever that we'll communicate effectively and parse semantics in the same way. So diversity will continue to reign -- yes, I like the snowflake effect too! But at least if we all speak English we share a common "grammar" it might depend upon where you come from! :-) But grammar is syntax! agreed - relates to "flat" and "hierarchical" structure... I'm aboutto leave the meeting as my next phone call is approaching. sorry but these days are too busy, This metadata interoperability discusson is quite interesting although, as Erik says, difderences exists and agreements are very hard. One issue could be to understand what is the driver to agreement. Reuse? Interoperate? Exchange ontologies? haha "Noone will fire you for using LOM" - it might go into my sig agree that vocabularies are really important. At the end of the day I think application profiles are created for pragmatic reasons - to serve a particular requirement for a community of practice. They will always have limits to their versatility & interoperability because they just represent a constrained convention Jon: yes, but why reuse DC or LOM at all? so as not to reinvent wheels when it is not necessary to do so at a very abstract level the whole world could be modelled in terms of containers and networks (or relationships). Even with 0's & 1's, bits & packets represent containment. This is why organisational "silo-isation" is probably the core "knowledge management" problem. If DC & LOM hadn't emerged in parallel one would have certainly followed. In a sense, there is still containment & hierarchical relationship within DC. It just depends how you look at it. (thinking out loud) the meeting is ending in 8 minis yes, documenting & understanding the nature of the problem is important - agree 14h00 CET Fine for me. Sure... OK - I'll send an email ... Thanks! Annotation profiles as tool to counter balkanization