Posting to public-schemabibex
On 01/07/2013 00:36, Young,Jeff (OR) wrote:

I think we need to have an open mind on this. Clearly, the disjoint WEMI classes with properties that have tightly bound domain/range are a major bottleneck. But if we stop trying to pigeonhole all bibliographic descriptions into one of these classes and focus on a more open sense of broader/narrower (call it what you will), I think we could actually accommodate/develop some of the stronger notions of WEMI on a per case basis, even if that takes some time. Some resources can clearly benefit from having a few identifiable levels of abstraction, be it 2 or 20.

We can have open minds–I actually try my best to do so, but that doesn’t mean that the new models should not be tested–and tested strictly–and these models must be allowed to fail. I’ve discussed a lot of this on other lists and in my podcasts. But I will discuss a specific point here: I have tried very hard to imagine what the purpose of “manifestation” is for a digital resource. The purpose of the manifestation record has always been to bring together the various items (or copies) of a work/expression in some sort of coherent way for the searcher; otherwise he or she will be overwhelmed by the multitudes of copies. Aside from the fact that the very ideas of “manifestation” and “virtual” seem to clash, and that a file can display on my computer quite differently from how it looks on your computer because of different software and stylesheets, there are other problems, and these problems have always existed.

What constitutes a copy can be quite different from one community or organization to another. ALA has guidelines different from LC, and individual libraries normally have more variation. Rare book/antiquarian dealers have even different ideas of what is and is not a copy. I have worked in non-library environments where copies are defined in completely different ways from any of the others. Some of these I agree with and others I do not. While I understand the difficulties nevertheless, I agree that the manifestation record (what has traditionally been called the “edition”) is vitally important. That is, in a physical environment.

But on the internet, what is a “copy” of a website, or even of a webpage? That is, other than duplicate books being scanned in Google Books, the Internet Archive, HathiTrust, etc.? The only places where I have found duplicate websites are mirror sites (relatively rare), and sites such as my blog where I send copies of my emails, and where I may edit them slightly for stylistic purposes (also very rare). Even in the Wayback Machine of the Internet Archive, what we see there are actually different “states” of a website and not actual “copies”–unless the webpage has not changed in the meantime. But even if we do want to consider all of those as copies or manifestations, do we really want to treat each one in the traditional ways? For example, the Guardian currently has 6,241 captures there

Aside from these possibilities, there are no copies of websites, but each is unique. I don’t want to get into the issue of mashups which adds another few layers of complexity.

If copies (items) don’t really exist on the web today, what is the purpose of a manifestation? That is, if we determine that except for very rare circumstances it turns out that “item = manifestation”, why keep the manifestation? It can no longer be to bring together the multitudes of copies for the ease of the searcher. What purpose can it serve?

This is one reason why I have said that cataloging web resources is similar to cataloging manuscripts, because we are cataloging unique items. This is what happened before printing. A medieval librarian would probably recognize a situation he already knew. But the idea that a website can suddenly morph into something different would probably blow the old librarian away.

Therefore, it seems to me that the entire WEMI scheme is premised on physical items that do not change. Similar questions can be raised about each of the other points of FRBR and how the public responds to them.

I have an open mind about these things, but the scheme must make sense in today’s environment–and tomorrow’s too. Too much is riding on the outcome for libraries to just hope and cross your fingers. So, I do mostly like what I understand of BIBFRAME so far, but realize that the project could have much broader implications for everyone.