a definition of frustration

[pantryslut]

I am looking for a book.

I know I own it.

I know exactly which shelf it used to live on...

...two apartments ago.

Comments

[tyrannio.livejournal.com]

This is why you need all your books cataloged and shelved by LC (or Dewey, if you must) call number.

[pantryslut.livejournal.com]

Oh, no, Library of Congress for sure.

As soon as I recruit that intern, it's done.

Meanwhile, I did eventually locate the book.

[manomano.livejournal.com]

ISBN is much more useful

[pantryslut.livejournal.com]

Let's you and him fight!

[tyrannio.livejournal.com]

If you mean it's useful to be able to search by ISBN, sure. My cataloging software does that, but I think that both my recall being better for author or title than for ISBN, and the possibility of multiple ISBNs for a given work, make searching by ISBN less useful than being able to search by author, title, subject, etc. (The availability of OCLC's xISBN project mitigates the latter problem a little, true.)

If you mean that shelving by ISBN is much more useful, I don't see why. Such a scheme would mean that, for example, my Philip K. Dick collection, being from many publishers, would be scattered across my shelves instead of all shelving together.

[manomano.livejournal.com]

Yeah, I'm partial to ISBN for cataloging too. And you're right, shelving by ISBN is not useful unless you're recall is by publisher, not author.

Shelving by number is only useful if you assume a fairly uniform or modular shelving surface, or uniform unit size. I think having a cataloging software, with a categorization scheme that is personalized (one's own faceted metadata system for instance, or adopting Dublin Core), with the ability to include format as well as book is what I'd aim for. Something actually fairly simple is probably best. Something like catagories such as "Paperback SciFi" "Hardback Fantasy" and "Oversized Cookbooks". Adopting someone else's catagorization scheme may limit browsing and make it difficult to fit everything on the shelf.

Although, having that information is also useful for any software that you're using.

[tyrannio.livejournal.com]

I find that I can get away with two sizes, regular and oversize, with all but 2% or so of my books being regular. Shelving different-sized books in different areas certainly saves space and is clearly appropriate for closed-stack libraries, but also inhibits browsing the shelves. Also, if there's a possibility of moving books among shelves with different sizing constraints, it would seem more useful for location and size to be maintained separately from subject.

Developing my own internally consistent controlled vocabulary for classification for all fields I am currently interested in or might in the future be interested in would be a large undertaking in its own terms, which is one of the major reasons I use LCSH (Library of Congress Subject Headings). I wasn't aware that DC came with its own subject heading: I thought it just provided a syntax for carrying a vocabulary decided on out-of-band. I'm also used to LC call numbers and LCSH, so I know where to browse. (One additional advantage of LCSH is that I can automatically link from one of my books to holdings in academic libraries which have the same subject headings: here I find that network effects outweigh any benefit I might derive from my own cataloging system.)

I also think that any one-dimensional classification system will inhibit shelf-based browsing. For example, it'd be impossible to group all my cookbooks (South Asian and non-South Asian), my South Asia books, and my religion (SA and non-SA) books together.

Obviously all this is my experiences and preferences: I'm mostly arguing that my experiences with LC call numbers and LCSH in my ~2500-volume library and homegrown software have been positive.

[pantryslut.livejournal.com]

My experiences with your library and homegrown software have been pretty positive, too. Though the ease of scanning in and retrieving bibliographic data via ISBN is also very attractive, owing to my chronic (and tragic) lack of interns.

[tyrannio.livejournal.com]

Oh, you can enter books into my system via ISBN lookup and copy cataloging from LC (or other libraries), too. (I find it easier to type the ISBN in manually than to locate the scanner, but that's just me. Well, maybe not just me.)

[manomano.livejournal.com]

Oy, I have to go back and review that lecture on DC, you're right it doesn't include subject headings of its own. It's a nice format for data exchange.

And you're right, a faceted metadata approach is much more useful than a single catagorization scheme.

[tyrannio.livejournal.com]

At least some people find it underspecified. I was talking to a friend about a project he did integrating DC metadata from a number of sources: he complained that there were so many wacky date formats that he had to give up on doing anything useful with date data.