• Pages

RDA thoughts

[updated 2017.11.9]RDAlogo_rgb

I’ve been listening to webinars on RDA (Resource Description and Access) and related developments in cataloging.  Most recently, the Amigos webinar “Is RDA on Your RaDAr?”, which has been interesting.  Thank you, Amigos team.

It’s my responsibility, after all, as the Technical Services Librarian.  I do all the cataloging.  I have one person to follow up after all that (a much more time-consuming job, and I’m pathetically glad I have her) with the barcodes, physical processing, etc., much of it by using the tools I worked out for that (and happily delegated). Then we have a student worker to cover/place labels/pockets/etc.

I’ve been doing this long enough to have gone through some changes in AACR to AACR2, MARC formats, etc.  That includes the first records for things that exist online, even.  So I’m no stranger to changes.

However, that tends to make me a little slow to jump onto new things just because they are supposed to be improvements, at least while still in beta stages.  RDA has been out a while, but it was being tweaked for some time.

I don’t want to sound negative about evaluating newer ideas such as RDA, but I’m a big advocate of cost-effectiveness.  This is sometimes considered inconsistent with the obsessive-compulsive, sometimes perfectionist, nature of catalogers (being personally somewhat guilty as charged). When there’s nobody else to take up the slack in the workload, however, one tends to boil down to what’s really going to get used.  The perfect should not be the enemy of the “good enough for actual use by our end users.”  So, that’s how I tend to look at something like this.

I specifically asked a presenter what advantages RDA offers that, say, a Discovery service doesn’t as far as searching and limiting and other patron-relevant functions.  I did not get one specific improvement, just some talk about how it supposedly was going to be better, and Discovery services are using it or compensating for the lack of RDA, or whatever.  

We have a Discovery service now on top of the catalog, so will RDA improve on that?  What can I use to justify the work and expense of converting our existing catalog records?  That’s the sort of question I have to ask.

I’m seeing my own catalog, which would need revisions to essentially every single record to become RDA, at some cost (going through a vendor to reprocess the records, as others have done, I hear).

I have some idea of what RDA advocates are trying to improve.  MARC, admittedly, is rigid, because the standardization of MARC allows computers to search and display the data consistently.  There are coding conventions (including AACR2) that still contain holdovers from the limited amount of data which could be crammed onto paper catalog cards, such as abbreviations which could be upgraded to make records clearer to end users.  That’s a pro, in favor of RDA procedures.

And yet I saw a commenter in a webinar session saying reactions from users about the missing GMD (General Material Designation) in the 245 field (the h subfield with things such as [sound recording] to describe the material RIGHT THERE) which patrons are complaining are missing in RDA records.  Patrons now say they have to open the entire record to see what the item actually is (since apparently they are not using the icons representing format supposedly being displayed using these fields).  Our catalog and Discovery service both offer visual icons as well as GMDs, since some people will look one place and not the other.

Other librarians say the end users haven’t seemed to notice (so why are we doing this RDA stuff, again?).

The reality, I fear, is that the vast majority of patrons/end users may not use — or even care about — the features that RDA and related standards are trying to provide.  I need those supposed enhancements to be able to justify spending the money and time to non-library as well as library administration, during a time of increasingly limited funding, or we’re not getting approved for this.

So, I’m trying to line up the factors that I (not necessarily anyone else) am thinking about at this time, which may change as we progress, and consider what — if any — reason(s) might be valid for putting effort into converting to RDA.

*** There are certainly pros to RDA, such as getting away from the abbreviations used since the days of typing catalog cards, and allowing more flexibility in tracings and description.  However, I’m not seeing why elements such as those cannot be added to our existing records (other than consistency), whether or not RDA is implemented otherwise.  We are definitely overdue for some of these changes.  I could do a lot of that with just the Global Update function in III’s Millennium, however, if it seems useful enough. [No Reason]

*** Don’t we love the 007 and 008 fields?  I wonder how many lone professional-cataloger libraries were involved in creating RDA and/or implementing it…  Of course, that’s not a pro or con; a lot of stuff gets created by the members of larger staffs which benefit — or at least affect — the rest of us.  I certainly didn’t have time to sit down and write AACR2 anyplace I’ve worked.  Several presenters (in this webinar and others) talk about the meetings in committees and elsewhere, over details such as how many tracings to do, while us solo types hold such meetings in our heads.  Meanwhile, I admit I pretty much never bother with 007/008 fields on the rare occasions when I do original cataloging — anyone else who wants them is welcome to add to my record. [Reason Only Useful to Catalogers = Con]

*** I hate implementing anything until it’s completed, so I haven’t rushed to do RDA up to this point (2009 to 2014). I think it’s pretty much out there by now, with some tweaks in process.  So it’s only now that I feel I can seriously look at it, and at how it’s working for places that are using it. [Reason to Consider]

*** BIBFRAME is proposed to replace MARC coding, to allow more links and versatility in handling them.  Someday.  For open source software, perhaps.  It looks promising, but it’s not something I can use at this time. [Reason to Wait]

We have been discussing changing our ILS at some time in the future, and if we decide to do that, we may just have much of this handled automatically by using enhanced records.  [Reason to Wait]

I’ve set our catalog  so it accepts and displays RDA records.  We can use the incoming records downloaded from OCLC.  I keep the RDA fields in records since they don’t display or provide any relevance in the online catalog which users see.  [No Reason]

All in all, I cannot see that RDA helps us, or that not having it hurts us.  At least at this time.  Maybe later.  Wait and see.

Update 2017.11.8

We have allowed the webpub.def in our Sierra OPAC (online public access catalog) to show RDA fields such as 264, since RDA-compatible records show that field instead of 260 for a while now.

I’m seeing another Amigos presentation on RDA and such. With all due respect, I’m not seeing anything that will make a bit of difference to patrons trying to use the OPAC. Still a lot of stuff being developed. Still — quite honestly — arcane.

ARCANE: As in, “known or knowable to only a few people”, as in “mysterious, obscure”. But hey, the idea is still to keep understanding to catalogers, and that function is working quite well. Top marks.

Extra credit for terms such as “the four-fold path” (didn’t Dr. Strange come up with that one?) complete with a graphic and a need for a new ‘toolkit’ to figure out how to apply it. They are now restructuring RDA and the toolkit. Paint your wand a different color, Mr. Potter.  [Everyone following the cultural references okay?]

This is looking more and more like change for the sake of appearing to ‘update’ procedures, which quickly becomes a new way to complicate whatever you’re doing. I don’t doubt the dedication of the people working on this, but any presentation that has to quote the Hitchhiker’s Guide and say “Don’t Panic” at the end, does not fill me with confidence. No panic here – just minimal compliance to remain functional, until they finish digging all the way through the planet and come out the other side.  (“The elves are working hard” as they said.)


how to process over 71,000 books in a couple of weeks… mostly

So, here we are:


Added the Academic collection from ebrary.  Over 71,000 ebooks.

First, I download the MARC records from ebrary for all of these.  I did it in the chunks of about 10,000 per chunk, so as not to overload the server when I put them in.  Also, it lets me experiment on smaller batches.

Ebrary has a list of publishers that do not allow downloading of ebooks, so first I go in and gather up a batch of all of those (total 7,236 records) and process them separately:

  • 500 note “Online reading only by publisher restriction.”
  • 856 link says An electronic book accessible online; click to view
  • no additional 856 link to the LibGuide for ebooks on how to download
  • Location: ebook, status = Internet use only

They also get the following, as do all the ebooks:

  • c  (for call number) status for 050 field is changed to f
  • the 082 field is copied to the 092 field
  • the / and ‘ are removed from the 092
  • the various |3 subfields for the versions of Dewey are removed (versions 19 through 23)
  • if there’s no 082, set the 092 to 000 instead
  • add |bEbook subfield to 092
  • Location: ebook
  • title field 245 has [electronic resource] changed to [electronic book]

BTW (By the Way), pet peeve: it shouldn’t be “electronic book”.  There’s nothing electronic about it other than the means of access.  It should be “digital book” or at least “digitized book”.  Too late now…. Way too late.

  • add a 541 with the acquisition info: ebrary |cFY2012 ebooks |f group purchase
  • add a 583 with a processing date and initials
  • 856: the “” part of the URL becomes “”
  • change the series to 440 (yes, I’m still using the old field  a while longer) and setting the indicators to skip “The ” and “An ” and “A ” (I’ll get to the non-English articles later)
  • change the t 240 to u 240 so the system sort the 240s with the 245s in a t title search (and therefore get listed twice in searching)

For the downloadable ones, the |z subfield in that 856 gets “Accessible online or DOWNLOADABLE as an Adobe Digital Edition; click to view” and an additional 856 field linking to the “Using Ebooks” LibGuide for our library.

Now, it couldn’t go that smoothly.  I had do these in batches of 25,000 or less to even get the Global Update to handle it.  It worked somewhat better when I limited it to 10,000 at a time.

Even so, the server was dragging.  I’m not sure it was the server’s fault, however — it might be more the network speed (since servers do a little of this, some of that, some of the other, check the network, go around again… if the network is slow, every cycle gets slowed down).  This time of year, especially, we have a lot of access going on.

Due to some backup problems, our transaction file wasn’t being cleared in the manner we needed, so we had some of these tasks done to some records in batches, and some not done, and some junk characters dumped in one field in some records, depending on when the transaction file filled up and actions stopped.  So, I’m having to go back in and clean that up, meanwhile watching the count on the transaction file much more closely than before.   I hate that kind of surprise.

In addition, I’ve found that ebrary has different information in different places.  For example, the info on the web site told me 7 days for normal circulation and no early check-in.  I test a download, and it says 14 days circulation and you can use Adobe Digital Editions to check materials in earlier there.  So, now I get to go back in and take the “7 days” out of the 856 link “DOWNLOADABLE for 7 days” and make it simply “DOWNLOADABLE”.  There’s another mass change, but it’s not as urgent.

Meanwhile, since that ties up Millennium Cataloging, I’m going into Millennium Administration and cataloging my print books there using the Create List function, while waiting on the Millennium Cataloging Global Update processing.

Of course, all this processing loads the server and reaction times begin to stretch, but we don’t do this many at once normally.

Ebrary is supposed to have a 710 with their company name in all the records.  Well…. I haven’t found it on all of them.  Must have some older ones that didn’t get it.  But they all have “ebrary” in their 856 link so I can pull by that.

Eventually I need to attach holdings to OCLC WorldCat for all of these, but the WorldCat links are not the same as the ones we have (ours apparently lead to our customized site), so I can’t just use the ebrary ISBNs to download the WorldCat records and work on them instead.  I’ll have to stay with the vendor records and see what I can do to limit the multiple findings per ISBN.  It’s going to be a long, exacting chore, and then… I get to do updates since the collection drops some titles and adds others at intervals.

Also, I have my 690 “PROGRAM” fields for our campus programs.  These are more useful than trying to go just by call numbers (especially on call numbers assigned by somebody else for their campus programs, and for materials that fit under more than one call number: ‘philosophy and television shows’ for example).  These are great for both gathering up program materials for accreditation groups, and for our updates-by-email function in the catalog, if faculty want to use that to be notified of anything new in a specific subject.

First I do use the call numbers to gather up groups, but some groups need to be broken up.  For example, 793 – 795.9999 has some specific items that need individual assignments, and the rest — majority of this batch — are on programming for computer games.  So they go under PROGRAMPROGRAMMING, which seems to have nothing to do with this Dewey section, but is actually relevant to computer programming (normally found back in 005).

Once I get the ones with Dewey done, I’ll gather the ones with 000 call numbers by the LC field (about a third of the total) and assign them section by section.

Job security, that’s what this is.

Cataloging (the) blues

[updated 2010.5.25]

So, I’m cataloging a huge batch (over 1700!) of CDs donated by the family of a man who used to do radio programs on the blues.

It’s not easy, folks.

To do these properly and make them a truly useful reference for people doing research on blues music, I’m trying to make sure that every bibliographic record has (a) a complete list of the musical pieces on it (for keyword searching), and (b) a subject line for the blues artist(s) on it (for grouping by the individual artist or group under a standard entry).

That means that I need to download authority records from OCLC to our catalog, so we can have a standard authority for each person and cross-references to nicknames, aliases, etc.  It avoids variations in spelling, nicknames, etc.

Now, if you know blues names like “Muddy Waters” or “Little Joe Blue”, you might expect these would be easy to find.

Try this one (yes, it’s copied directly from a real authority record) for Robyn,  William:

SUBJ AUTH    Robyn, William, 1894-1996
SUBJ S FRM    Bennett, Bert, 1894-1996
SUBJ S FRM    Brown, Tom, 1894-1996
SUBJ S FRM    Conroy, Frank, 1894-1996
SUBJ S FRM    Conroy, Fred, 1894-1996
SUBJ S FRM    Crane, Thornely, 1894-1996
SUBJ S FRM    Ender, Jack, 1894-1996
SUBJ S FRM    Foster, Al, 1894-1996
SUBJ S FRM    Franklin, Fred, 1894-1996
SUBJ S FRM    Gravelle, Buddy, 1894-1996
SUBJ S FRM    Hamilton, Edward, 1894-1996
SUBJ S FRM    Hamilton, Ray, 1894-1996
SUBJ S FRM    Mack, Bobby, 1894-1996
SUBJ S FRM    Manning, John, 1894-1996
SUBJ S FRM    Norton, Walter, 1894-1996
SUBJ S FRM    Powell, Ray, 1894-1996
SUBJ S FRM    Powell, Roy, 1894-1996
SUBJ S FRM    Ray, Walter, 1894-1996
SUBJ S FRM    Robinow, William, 1894-1996
SUBJ S FRM    Robinson, William, 1894-1996
SUBJ S FRM    Robyn, Willie, 1894-1996
SUBJ S FRM    Ruban, George, 1894-1996
SUBJ S FRM    Rubin, William, 1894-1996
SUBJ S FRM    Rubinoff, Mario, 1894-1996
SUBJ S FRM    Rubinoff, William, 1894-1996
SUBJ S FRM    Scarpioff, William, 1894-1996
SUBJ S FRM    Scarpioff, Wolf, 1894-1996
SUBJ S FRM    Scott, Henry, 1894-1996
SUBJ S FRM    Shaw, Eddie, 1894-1996
SUBJ S FRM    Smith, Harry, 1894-1996
SUBJ S FRM    Spear, John, 1894-1996
SUBJ S FRM    Stanley, William, 1894-1996
SUBJ S FRM    Waters, Frankie, 1894-1996
SUBJ S FRM    Wee Willie, 1894-1996.
SUBJ S FRM    Allison, James, 1894-1996.
SUBJ S FRM    Bolton, Jamie, 1894-1996.
SUBJ S FRM    Clarke, Walter, 1894-1996.
SUBJ S FRM    Forsythe, Reg, 1894-1996.
SUBJ S FRM    Francis, William, 1894-1996.
SUBJ S FRM    Hart, Charles, 1894-1996.
SUBJ S FRM    Henderson, Larry, 1894-1996.
SUBJ S FRM    Henry, Lawrence, 1894-1996.
SUBJ S FRM    Hillman, Bob, 1894-1996.
SUBJ S FRM    Johnson, Al, 1894-1996.
SUBJ S FRM    Lee, Albert, 1894-1996.
SUBJ S FRM    O’Shea, Allen, 1894-1996.
SUBJ S FRM    O’Shea, John, 1894-1996.
SUBJ S FRM    Playman, Edmund, 1894-1996.
SUBJ S FRM    Remick, Walter, 1894-1996.
SUBJ S FRM    Richards, Edgar, 1894-1996.
SUBJ S FRM    Richards, Walter, 1894-1996.
SUBJ S FRM    Rickman, Eddie, 1894-1996.
SUBJ S FRM    Roberts, Billy, 1894-1996.
SUBJ S FRM    Roberts, Lewis, 1894-1996.
SUBJ S FRM    Robin, Wyllie, 1894-1996.
SUBJ S FRM    Thomas, Brian, 1894-1996.
SUBJ S FRM    Turner, Ray, 1894-1996.
SUBJ S FRM    Weston, Les, 1894-1996.
SUBJ S FRM    White, Jerry, 1894-1996.
SUBJ S FRM    Young, Louis, 1894-1996.
NOTE    William Robyn oral history interviews: label (William Robyn) transcript (Wolf Scarpioff, Mario Rubinoff)
NOTE    Enc. of rec. sound, c2005 (Robyn, William; b. Latvia 1894; d. Englewood, N.J., 12 Apr. 1996; tenor; born William Rubin he used more than 50 pseudonyms on recordings published by Victor, Columbia, Cameo, and approx. 50 other labels between 1919 and 1931)
NOTE    Pseudonyms on American records, c2005: p. 401 (Bert Bennett ; Tom Brown ; Frank Conroy ; Fred Conroy ; Thornely Crane ; Jack Ender ; Al Foster ; Fred Franklin ; Buddy Gravelle ; Edward Hamilton ; Ray Hamilton ; Bobby Mack ; John Manning ; Walter Norton ; Ray or Roy Powell ; Walter Ray ; William Robinow ; William Robinson ; “Wee” Willie Robyn ; George Ruban ; Cantor William Rubin ; William Rubinoff ; William Scarpioff ; Henry Scott ; Eddie Shaw ; Harry Smith ; John Spear ; William Stanley ; Frankie Waters ; Wee Willie)
NOTE    ARSC journal, v. 23, no. 2 (1992): p. 223 (James Allison, Jamie Bolton, Walter Clarke, Reg Forsythe, William Francis, Charles Hart, Larry Henderson, Lawrence Henry, Bob Hillman, Al Johnson, Albert Lee, Allen or John O’Shea, Edmund Playman, Walter Remick, Edgar or Walter Richards, Eddie Rickman, Billy or Lewis Roberts, Wyllie Robin, Brian Thomas, Ray Turner, Les Weston, Jerry White, Louis Young, Olympia Quartet (member))

He used more than 50 different names?!?!?!  (I have to wonder, was somebody chasing this guy all the way from Latvia?  Or, whose decision at the record label(s) was it to create yet another pseudonym?)  Maybe this was simply to avoid legal encumbrances made under other names, but come on!

And then we have all the short, fairly common names like “Joe Martin” or “Henry Gray” (and actually, those are relatively easy — less than ten different authority records to check to find the musician among them).

And THREE, count ’em, THREE different Howlin’ Wolf guys.   There’s J.T. “Funny Paper” Smith a.k.a. the Howling Wolf, and Chester Arthur Burnett, and neither of these should be confused with “English rhythm and blues singer Howlin’ Wolf”.  Go figure.  So, I’m paging through each little CD notes booklets to figure out which one is playing on this specific CD….

This is why Hollywood makes each performer have a slightly different name, to avoid this kind of confusion.  Good thinking!

Anyway, it will be a useful and valuable collection when I get it cataloged.  But it is going to be a long, slow, very picky job.

And that’s what catalogers have to be good at doing.


Having griped about William Robyn, I find that his pseudonym Eddie Shaw is not even the correct Eddie Shaw!  Turns out that Eddie Shaw 1937- is the blues musician I needed.

Catalogers got the right to sing the blues, all right….

(Thanks to the person who emailed me with the alert to check this!)