The FRBR Blog

Last month Glenn Patton said on the FRBR mailing list, in response to a query, that a second draft of Functional Requirements for Authority Data (FRAD) would be available “in the next month or so.” I’ll be keeping an eye peeled for it and will post a link. The first draft was published in 2005 and the IFLA Working Group chaired by Patton has been going over all the comments and feeback.

If the name FRAD seems unknown but strangely familiar, it’s because you’re used to hearing about FRANAR (Functional Requirements and Numbering of Authority Records) and FRAR (Functional Requirements for Authority Records). FRAD is the new name.

I’ve renamed the FRAR category here to FRAD to match the revised nomenclature. Check out previous posts in this category to find links to the earlier draft.

Lorcan Dempsey, of OCLC, posted GBS FRBR on his blog on Monday. GBS stands here for Google Book Search, not George Bernard Shaw. Google’s going to group manifestations into works.

… Dan Clancy, of Google, mentioned that they were pulling together members of work sets in this way at the Working Group on the Future of Bibliographic Control meeting the other day. This is interesting and valuable. Presumably this is happening programmatically.

This will be worth watching.

FRBRizing Shaw would be interesting too, though a smaller project. For example, the play Pygmalion was made into a movie of the same name, then turned into the stage musical My Fair Lady, which was turned into a movie. Shaw drew on Greek mythology for the story, and the Pygmalion myth has been used by other writers and artists. It would be an interesting web of relationships.

Karen Coyle sent out an announcement about Framework for a Bibliographic Future, which she’s created with Diane Hillmann, Paul Weiss, and Jonathan Rochkind. She says they “have been attempting to create what we think MAY be the kind of framework that we need in order to move forward with bibliographic data.” FRBR is mentioned therein.

The WoGroFuBiCo, or Working Group on the Future of Bibliographic Control (created by the Library of Congress), had a meeting last week at Google headquarters. (They wanted a meeting in California and Google offered the space.) FRBR came up during the day. If you’re interested in FRBR, you’ll want to follow this group, both for what they do and what people say they should be doing. Here are some relevant blog posts.

• Karen Coyle diligently took notes on all the talks and posted them on her blog. Start with Preamble: Users and Uses Meeting for some background and then work your way forward.
• Andrew Pace spoke, and posted on his blog in Bibliographic Control Meeting
• Talis’s Panlibus blog comments with The Future of Bibliographic Control… And the Pesky Users and Uses (and says FRBR is “a good enough idea, but insanely more complicated than is really required, and typically over-engineered”!)
• OCLC’s Lorcan Dempsey was there and in LC WGFBC he says a longer post is forthcoming.

XTF from the California Digital Library: “The CDL eXtensible Text Framework (XTF) is a flexible indexing and query tool that supports searching across collections of heterogeneous data and presents results in a highly configurable manner.”

XTF Experimental Features covers FRBR: “To implement FRBR in XTF, we chose to adapt the standard FRBR Work Set Algorithm to our purposes. In particular, we changed the method to dynamically determine work groups, rather than making this determination at index time. This allowed us to play with and tweak the algorithm without having to re-index the entire test collection (over 10 million records).”

Andrew Hankinson sent a message to web4lib last week saying that he’s reviving Morbus Iff’s LibDB project. I’m a bit slow on things and am seven days late passing it on:

As a project for one of my courses, I have decided to update the Drupal module LibDB to attempt to come up with a fully-functioning FRBR digital library implementation. Obviously this is a very large job, but it is one that I feel I can commit to over the long term (and not just the duration of the course.)

As of right now, I’ve gotten it working (mostly) on Drupal 5, and have been working on a database model that would fully implement the FRBR ‘standard’ as outlined by the original IFLA document. LibDB was abandoned (or at least put on the back burner) by its original creator, and much of it is left to do.

I have a fairly good knowledge of the internals of Drupal, but lack a good sense of programming practices. (I could do it, but it would a) be a slow process and b) be inefficient and sloppy code.) If you’re interested in getting involved in this, feel free to reply to me (on or off-list) and I will start putting things together.

This is great news. Good luck to Hankinson! My own OpenFRBR project got off to a lacklustre start but I’ll be getting back to it shortly, fortified by the new edition of Agile Web Development with Rails by Thomas and Hansson.

The LibraryThing helpers log lists what helpful people at LibraryThing have been doing, including which works they’ve been grouping together (so that The Faber Book of Pop and Faber Book of Pop are known to be the same thing) and which authors they’ve been grouping together (so that Robert L. Stevenson and Robert Louis Stevenson are the same person). It would be interesting to take the works groupings, grab the MARC records of the relevant manifestations, and run them through OCLC’s FRBR Work-Set Algorithm (and similar rulesets) to see which kinds of matches are most commonly made and if any matches are made that aren’t covered by the algorithm.

Today, a few pointers to some posts on other weblogs.

• David Weinberger’s Miscellaneous Hamlet (from 31 December 2006) was pointed out in that ngc4lib thread I posted about a few days ago. “There are three established editions of the play, so when you want to point to Hamlet, which one do you point to? Not to mention the various publishers, editions, and versions, from large print to translations to ones with modern spellings to parodies to coloring books.” FRBR can solve that, but Weinberger makes good points about fuzziness and how much this is a kind of that. I listened to a couple of Weinberger’s talks that Tim Spalding pointed to in Prolegomena to a Review of Everything is Miscellaneous and was very interested.
• Speaking of Tim Spalding and LibraryThing, its users are doing an average of 2,000 work-combination actions a day. “That’s not even works combined, which is higher since a combination will have at least two and and high as twenty. It boggles the mind.”
• darth_libris’s FRBR Fun (from 30 January): “My problem with FRBR discussions is that it is actually impossible to create an expression without *also* creating a manifestation and an item. The live performance of Beethoven’s 5th constitutes an item, as well as being an expression/manifestation.” It is possible, actually. The music floating through the air is the expresion: “the intellectual or artistic realization of a work in … musical” form, “the particular notes, phrasing, etc. resulting from the realization of a musical work” (says the Final Report). It’s only if that performance is recorded that manifestations and items come into existence.
• John Fudrow’s FRBR: Creating Complexity or Simplifying Creation? (from 16 February). “The largest gray cloud on the FRBR horizon would seem to be the dollar signs flashing before our eyes.”

This is a follow-up to De Revolutionibus and De Revolutionibus redux. I just noticed that in the external links section of Wikipedia’s entry on Copernicus, there’s a link to a digitized version of a manuscript of the book, written by Copernicus.

The history of this copy (which they call an “autograph” because it’s written by hand) says:

The Autograph De revolutionibus preserved in the Jagiellonian Library is a result of work of the great scholar, intermediate between a rough copy and a fair copy. It had remained in Copernicus’ hands until his death (24 May 1543). His papers and books passed to his closest friend, Tiedemann Giese (1480-1550), a bishop in Chelmno at that time. He bequeathed his library to Warmia Chapter. However, the autograph went to the collection of George Joachim Rheticus (1514-1574), astronomer, Copernicus’ pupil. Rheticus was occupied with publishing the work of his master, but the basis for printing was not the autograph but its copy.

It knocked around a fair bit after that, ended up in Prague, and was given back to Poland in 1956.

What does this mean for we FRBRians? Well, it’s the same work, De Revolutionibus. No doubt there. Is this the same expression as the one we were discussing earlier? I don’t know. The quote above says “the basis for printing was not the autograph but its copy.” Who made the copy? We don’t know. How different is this autograph from its copy, and how different are both from the printed version? We don’t know. We’d have to check what researchers have found out (or get smart students to make a project of it).

For fun, let’s assume that the text of the copy is different enough to seriously count as a new expression. We won’t be picky about a different letter or word here or there. Let’s say there are noticeable differences that merit distinction. Maybe textual or numeric changes were introduced prior to printing, or Rheticus edited it, or some such. We’re pretending.

Now we have two expressions: the text of the autograph and the text that made up the editions (manifestations) printed in 1543 and 1566. Let’s summarize how things stand, viewed with FRBR, before any digitization was done:

Work: De Revolutionibus by Nicolaus Copernicus

• Expression 1: text (in an abstract sense) of autograph version
  • Manifestation 1: Copernicus’s handwritten copy (with that text on paper)
    • Item 1: Copernicus’s handwritten copy
• Expression 2: text (in an abstract sense) that went to press (which we’re pretending is sufficiently different)
  • Manifestation 0: Rheticus’s copy of the handwritten version (with that text on paper)
    • Item 1: Rheticus’s copy, wherever that is now
  • Manifestation 1: first edition, Nuremberg, 1543 (where the printer put the text into physical form by printing it on paper)
    • Item 1
    • Item 2
    • Item 3 and so on …
    • … up to item 400 or 500 or so, Gingerich estimates
  • Manifestation 2: second edition, Basel, 1566 (where another printer put the text into physical form)
    • Item 1
    • Item 2
    • Item 3 and so on …
    • … up to item 500 or 600 or so, Gingerich estimates

How does the Polish digitization of the autograph copy affect the first expression? How do we express what’s on their web site — and what’s on your hard drive when you’ve looked at a page and it’s stored in your browser’s cache — in FRBR terms? Aha! Now you’re scratching your head!