Posting to NGC4LIB
The way I look at it, the original need for centralized databases to find copy for cooperative cataloging no longer really holds. There are other technologies that are far more powerful and better than z39.50 transfer of ISO2709 MARC21 records. OCLC also has ILL capabilities, but I think these are similar to some of the web2.0 tools and could be developed without too much trouble.
Technological change is not a bad thing, and it should be a catalyst to move into other areas. IBM started as the “Tabulating Machine Company” but they moved on as the technology changed to become one of the giants of the world and society. Today, there is a genuine need for library metadata to leave the confines of our libraries and go out into the world where users are doing their research and information searching. We can’t expect users today to come to our library catalogs when they have online databases of all kinds, government information, excellent materials on the WWW, and of course, Google Books. There is and will be plenty to keep our users occupied in these projects, and they won’t have time to come to our stuff. If we are to have a chance, our records must go to our users and not wait for them to come to us.
This is why I think we need OAI-PMH so that we and the rest of the information universe can help create innovative tools. OCLC will still be vital in this world as a library think tank, and I think libraries will be happy to contribute to OCLC’s researches. But we must remove beyond proprietary z39.50 databases.