Harvey,
They are useful—fascinating even. All told, I think the coverage is
good. I like how things are pretty evenly distributed, but SciFi does
a bit better than average and westerns fall off a cliff! That seems
about right; I've never noticed westerns being big on LibraryThing.
They don't look too big in your holdings either, but the percent is
still off.
Thanks for the data. Now all we have to do is figure out how to
present LibraryThing data to libraries. I think we're going to ask for
an ISBN import. Spidering the site and etc. seems too fiddly.
Tim
On 3/6/07, Hahn, Harvey <hhahn_at_ahml.info> wrote:
> I previously wrote:
> |Tim Spalding wrote:
> ||I made an ISBN feed, so libraries could compare their holdings with
> ||LibraryThing
> |
> |I've downloaded your file to compare with our public library's
> |333,000-record bib database, not all of which have ISBNs. (It'll take
> |a while, though--I've got a lot of other things on my plate, too.)
>
> Well (after filtering out microforms, maps, pamphlets, magazines, and
> equipment), here are the results for ADULT materials *only*:
>
> [The first line would be read something like this:
> 145650 (56.6%) of the 257475 ISBNs that our library has in the given
> category (in this case, "grand total") matched ISBNs in LibraryThing's
> ISBN list]
>
> =====================================================
> NUMBER OF ISBNs THAT MATCHED LibraryThing's ISBN LIST
> =====================================================
>
> Matches Items Percent
>
> GRAND TOTAL: 145650 257475 56.6
>
> Book: 139410 216657 64.3
> Nonbook: 6240 40818 15.3
>
> Book subtotals:
>
> FICTION TOTAL: 31948 39755 80.4
>
> Fiction: 19537 23969 81.5
> Mystery: 7928 10296 77.0
> SciFi: 2873 2941 97.7
> Western: 281 864 32.5
> Gen pbk fic: 472 611 77.3
> Romances: 857 1074 79.8
>
> NONFIC TOTAL: 83592 124348 67.2
>
> 000's: 2350 3080 76.3
> 100's: 3504 4753 73.7
> 200's: 4870 5906 82.5
> 300's: 11535 19047 60.6
> 400's: 1142 1460 78.2
> 500's: 3114 4308 72.3
> 600's: 18688 29770 62.8
> 700's: 14331 21615 66.3
> 800's: 6573 10014 65.6
> 900's: 11113 16369 67.9
> Biographies: 6372 8026 79.4
>
> =====================================================
>
>
> As I mentioned, I've got other things going on, too, so it may be a week
> or two before I can post YOUTH results, unless you don't want/need those
> results because of the nature of LT's ISBN list. By the way, these
> breakdowns are pretty easy on our system because we created a
> 5-character positional code (LOCATION) in every item, based primarily on
> formats and DDC/genres. I massaged the output of our III system to
> eventually (after comparing against LT) come up with a file where each
> entry had the form "xxxxxiiiiiiiiii n"--x is the positional code, i is
> the ISBN, and n is 1 or 0, depending on whether that ISBN was found (via
> a binary search) in LT's sorted ISBN list or not. By using VBS's
> regular expression object in an OCLC Connexion OML (VBA-like) macro, I
> can do counts on the file using a pattern match on the 5-character codes
> and summing the 0/1 results of matching LibraryThing's ISBN list.
>
> I hope these results are interesting and useful!
>
> Harvey
>
> --
> ===========================================
> Harvey E. Hahn, Manager, Technical Services Department
> Arlington Heights (Illinois) Memorial Library
> 847/506-2644 - FX: 847/506-2650 - Email: hhahn(at)ahml(dot)info
> OML & Scripts web pages: http://www.ahml.info/oml/
> Personal web pages: http://users.anet.com/~packrat
>
Received on Thu Mar 08 2007 - 15:42:46 EST