Quick note on article in recent Tech Services Quarterly (25/3, 200 on NACO training among Caribbean libraries.  “As part of the [planning] process, a project leader was assigned, a budget created, and tasks allocated to the six-member coordinating team under the following areas: budget management, communications and marketing, hospitality and entertainment, facility preparation, and document production.”  Eek!  Of course, this was for coordinating training among libraries from several different institutions & islands…won’t be quite so complicated for us.

TN SL is using ArchiveIt to get DC recs that they then turn into MARC (using MarcEdit?).  This is done by their fed docs librarian?

This wasn’t at BPE 2008, but I read the report on the plane in between knitting. RLG Programs Descriptive Metadata Practices Survey Results, a nice brief report that’s available online from OCLC. It’s connected to  their “Changing Metadata Creation Processes Program” (so not a snappy title) which has the following problem statement:

Creating metadata that suits local needs, readily aggregates across communities, and is easily exposed to Internet search engines remains a costly enterprise lacking adequate tools and guidance. Metadata created by libraries, archives, and museums is generally not available to the user communities that look first to Internet search engines. Although mapping data structures has become a commonplace solution to integrate descriptions, real interoperability across the libraries, archives, and museums communities cannot be achieved without addressing differences of description at the data-content level.

Some of the text of Gather Evidence to Inform Changes Needed in Metadata Practice is a little impenetrable, but I do love that they note this:

There is no evidence available yet of what data elements are truly critical for users searching in a Web networked environment.

Go for it, OCLC!  We need research!  There is so much talk out there about how terrible OPACs are and how useless MARC is and SO LITTLE actual in-depth research on what’s really needed by users.

Being RLG-oriented, there is of course an emphasis on museums and archives in addition to libraries, and only 18 institutions were surveyed for the report.

Interesting that “the single most common[ly used tool to create, edit, and store metadata] … was ‘a customized tool,’ cited by 69% of all respondents. These include homegrown tools for creating, managing, or providing access to archival collections [etc. etc.]“  Which sounds like an awful lot, and makes you wonder what the commercial tools are lacking that makes this necessary.  But, then again, the institutions reported 263 total tools used, so if there’s only one customized tool being used at each institution, each institution could still be using several non-customized tool; it would be much more informative for us to know just what the customized tools were used for.  Were they used for the major portion of the metadata creation process, or for a minor part?  Grumble.

“Structure used to create metadata”: only 19% MODS (poor MODS), 65% MARC, and 59% DC (I’ve combined qualified and unqualified DC there).

Local thesauri: 69% are creating one for genres of materials (like us with that dang Format field!). Smaller numbers for topics, names, places, etc.  Would be interesting to compare all those Genre thesauri to see what they’re providing that other thesauri are not.

Libraries were more likely to expose their metadata to OAI harvesters than museums.  Hmm.

Overall, I’m bothered by the small sample size of this survey.  I realize it’s difficult to carry out surveys, but sometimes I wonder about the quality of library “science” research (this report blithely references another OCLC report, College Students’ Perceptions of Libraries and Information Resources, and its statistic that “89% of college students use search engines to begin an information search”–WELL DUH.  Was “information search” defined or refined at all? I perform oodles of “information searches” every day that are far better handled by the phone book or a company website or an online forum or an agency website instead of a library website, because of the nature of the information I’m looking for. Come back to me with a survey that studies where college students are beginning their information searches for research papers or other library-appropriate stuff and I’ll give more weight to your numbers!)…uh…that was a really long aside…well, anyway, enough grousing.  I guess I’m glad that there’s research being carried out because we definitely need it, but I hope it’s good quality.

Whoa momma I wish I had more time to post on this thing.

From here on out, I think I will quit pretending that it’s in any way for public consumption and just use it for note-taking.

So.  From BPE 2008, (presentations mostly available on the site), the Oklahoma Dept. of Libraries gave a nice update on their CONTENTdm project.  Oklahoma Crossroads (snappy name, I like it!) is their online repository for digital state documents.

[compare our browse display to theirs when our dig rep will FREAKING OPEN]

Note their interesting use of a “custom thumbnail” for digital docs (visible when browsing).  The thumbnail is not for the individual publication, but for the publishing agency.  In some ways I LOVE this–it’s usually a better image than the front page of a legislative committee report (yawn).  Also, I like the “branding” it gives for each agency, and the way it can let you visually pull together different works by an agency.

However, I can’t imagine us making a lot of tiny custom images of agency logos (plus, some of them are hard to read) and maintaining this over time.  And for publications that are issued by more than one agency in cooperation, you have to dick around figuring out which one is the most prominent (though we already do that for our “Agency” field, I guess?).  Also, users probably expect to see a “cover image” more than the author/publisher logo…though I think you could debate that for state publications.  But I think it’s too much trouble for us.

Random notes:

• They do include their call number (is that in a DC Subject field?)
• “Other formats” field is interesting (”Printout; preservation microfiche”)
• I appreciate that they include a detailed list of the contents, but I wonder if users are confused by the fact that the linked words don’t link to that portion of the document (like the “Glossary”) but instead execute searches.  A clickable contents list usually implies that you can access the actual contents by clicking (not that I have any data to back that up…).
  
• 029 in the MARC record is something OCLC added to fix some URL problem (be sure to look into this if we decide to try WorldCat Harvesting)
• They have added a field to their DC records to distinguish monos from serials: Description Bib Type (filled in only for serials).  We can identify our serials by the “ser_” at the beginning of the file name, but maybe this field would be a good idea, too.  Doesn’t seem to display in public view.
• wtf even putting “goat newsletter” in quotes results in a “goat OR newsletter” search–is this typical of CONTENTdm??  I want to see the Goat Newsletter without having to use Advanced Search, not all these other things!  “Goat and newsletter” searches for “goat OR and OR newsletter”–not an improvement!  OCLC needs to work on some ranking algorithms.  (Goat newsletter is disappointingly short on photos of cute goats.)
• Nice images in the presentation of what the error reports from OCLC after a crosswalk look like.  Reports seem a tad cryptic.
• They note that CONTENTdm sometimes chokes a bit on loading compound objects of mixed file types (”especially HTM or other web capture”)
  

I wonder how much of the resulting MARC record (after the WorldCat Harvest) is auto-generated (like the 533).

RLG Annual Meeting MP3s and notes are up, with tons of interesting stuff as always. The keynote speech by Dylan Tweney of Wired is really thought-provoking. All kinds of ideas in there about web history, 2.0 functions, and the (potential) role of libraries and librarians in the 2.0 universe. To wit:

By 2012, the demand for speed will have turned all publishers into de facto bloggers, and all search engines into Google.

There is an opportunity for librarians on the internet, but the window is closing.

What the information universe wants: a hackable library database.

read read read read read! Don’t miss his suggestion at the end:  create a books-digitization site similar to YouTube and let users digitize books themselves. Work out the copyright issues later. Wow!

MP3 of the follow-up Q&A session is also online.

I have to confess that I was totally rolling my eyes about Twitter as being another flash-in-the-pan geekfad, but this nice presentation by David Free and David Lee King from the BIGWIG Social Software Showcase is both brief and informative, and has some great examples of how libraries and librarians can actually use twitter.  Still of limited usefulness for someone like me, who (*gasp*) doesn’t own a cellphone and therefore can’t go around texting and tweeting left and right, but well worth a look-see.

Other presentations from the BSSS covered LOCKSS, folksonomy, and LibraryThing for Libraries.

Thomas Mann (the LC guy, not the Magic Mountain guy) has written previously about the “crisis in cataloging” these days and the apparent attacks on controlled vocabulary and full cataloging, and he seriously rocks.  Library Juice points to his latest effort, “The Peloponnesian War and the Future of Reference, Cataloging, and Scholarship in Research Libraries.” The whole essay is great and while it covers what is familiar territory for Mann–the difference between scholarly research and “quick information seeking” and the tools needed for the former, and value of traditional library strengths such as controlled vocabulary, high-quality cataloging and reference, patron education–one especially nice touch is the inclusion of specific examples of these concepts in action.

There’s so much information in this essay I really can’t do better than recommend that everyone read it, including library administrators, reference librarians, catalogers, digital librarians, advocates of faceted searching and tagging, etc.  Here is my favorite snippet to pull out, however:

If our goal is to promote scholarship, then “least effort” on the researchers’ part means “most effort” on our part, in our acquisition efforts, in creating high quality cataloging, in providing proactive reference service, and in assuring the long-term preservation of our material.

THANK YOU, THOMAS MANN.

I hope that this will be as widely disseminated as Karen Calhoun’s and Deanna Marcum’s recent reports which completely ignore Mann’s points about the very complex searching needs of scholars, which are entirely different from the (also legitimate) needs satisfied by sitting down and typing a couple of keywords into a “single search box.”

Nifty little new feature in worldcat.org : the ability to select records to add to your personal lists–a very cool LibraryThing-like way to make something like a bibliography.  I like a lot of things about this feature.

What I don’t like is not knowing how to get to other people’s lists to look at them–I guess you go to Lists and choose Search for Lists, but what if you don’t know what to search for, but want to browse?  Or what if (gasp) people didn’t use controlled vocabulary when they named their lists?  There needs to be some function that just gives links to the lists so you can click around and find them serendipitously.  But, it’s an exciting little feature, and you can export your list as a comma-delimited file, hmm!

Here is my list of lists. Not much in them yet; will have to play around more.  Say hello to professor P. V. Glob.

Saw an announcement about this service on the handy Library Technology Guides blog. Community Reserve (could they have come up with a clunkier, less exciting name?) is “a digital resource sharing service within the Digital Library Reserve platform that enables participating libraries to manage and share download audio books, eBooks, music, and video in supported formats. Use it to distribute locally produced, non-commercial content and share digital materials across OverDrive’s library network at no added cost.”

A handful of public libraries have signed up and the vendor’s FAQ notes that Community Reserve contains “over 100 titles” (I’m underwhelmed, but it is pretty new). I’m not quite sure I get the purpose of this product given the market strength of YouTube, but perhaps it allows a sort of inventory control (use statistics, etc.) and does it allow you to add these locally-produced digital things into a database that includes your paid digital materials? Or is Community Reserve a separate entity altogether?

The University of Washington Libraries have a neat experiment with WorldCat Local going on, at least for as long as their “Search UW Libraries and beyond” beta test is up. I haven’t had time to play with it much, but became curious about it following a discussion about it on the RADCAT list. Some people in that discussion cited difficulties with being able to perform a traditional “browse” on an author’s name, and pointed out how important that is for humanities research (among other things). So I played around with a little author searching.

Here’s what I don’t like about it. A plain old search for “pam allen” (no quotes) brings up several options in the “Refine your search” sidebar. One of these is “Pam Allen,” so I click on it (that part is pretty cool). I get results for several different Pam Allens, one of whom is the Interweave Knits editor I’m interested in. But there’s no way to differentiate her from the other Pam Allens, I think? And when I go into a record for one of her books, clicking on her name in the record doesn’t lead me to a list of only *this* Pam Allen’s publications; it performs an author search for Pam Allen, bringing up a whole lot of other Pams and Allens. I’m not even getting Pam Allen the Knitter in the first three pages of results (horrors!).

So I’d love to see this search interface tied to the authority file (remember that thing, guys?) so the searching is more sophisticated. It’s fine for the search interface to be made simpler, but why make the functionality so much less?