The Collation

Research and Exploration at the Folger

Cataloging questions: How should we display variant titles?

Do you use Hamnet, the Folger’s online catalog? Do you want to help make it better? Of course you do! This is the first in what I hope will be an ongoing series of conversations designed to keep me from playing around with the display settings in whatever way strikes my fancy just because I happen to know the master password. First up, “Variant Title(s).” 

Consider the silkworm picture from last month:

Depiction of the production of silk. In the foreground, seated to right, Emperor Justinian sits under a canopy and converses with monks; in the background, seen through a window, women look after silk worms and weave silk.
Jan van der Straet, (1523-1605). Ser, siue, Sericus vermis. Antwerp: Philippe Galle, ca. 1591. Folger Shakespeare Library ART Vol. f81 no.8.

As the post about V, u/v, and library transcription rules went to great lengths to explain, the title could be converted from uppercase to lowercase  three different ways…

  • Ser, sive sericvs vermis
  • Ser, sive sericus vermis
  • Ser, siue, Sericus vermis

…but only the last one in the list is “correct” according to current rare materials cataloging practice. Luckily, the rules also say that catalogers should provide access to variant forms of the title “if it is thought that some catalog users might reasonably expect” to find the material under that title. 1 (Note that even though this is a record for an engraving, the same rule of reasonable expectations holds for books, manuscripts, and all other formats.)

The list of  reasonably expected titles can get very long, so Hamnet is currently set to display them near the end of the main record, so you don’t have to scroll through them to get to key information like the publication place and date (the block of information that starts with location and call number gets brought in from a secondary record, and always displays last). 

Arrow points to variant titles clustered near the end of the record
Hamnet record for ART Vol. f81 no.8 as it appeared on 20 April 2014.

Notice that in addition to the two variant spellings, the list includes a completely different variant title for the print, “New Hollstein title: Production of silk.” Unlike the library world, which prefers transcribed titles, the art world prefers descriptive titles, in the vernacular. The New Hollstein series is the acknowledged authority on early modern European prints, so it’s reasonable to expect that many people know this one by its New Hollstein title.

When you look at the MARC code behind the online catalog display, you can see a difference between these two types of variant title. 2 All three have the main code “246” (defined as “Varying form of title”) but the spelling differences are flagged with the secondary code “3” while the New Hollstein title is flagged with the secondary code “1”.

Box identifies two different numeric codes
Excerpt from the “MARC view” of the record for ART Vol. f81 no.8

Code “3” indicates that the reason for including that version of the title is easily intuited from the rest of the record. Code “1” indicates that the reason can’t be figured out on its own (which is why this one has “New Hollstein title” in front of it). At most libraries, variant titles coded “3” don’t display in the normal view of the record, only in the “MARC” or “Staff” view. For a long time at the Folger, it was felt that users would prefer to see all variations, even the ones that aren’t normally displayed, so that a person wouldn’t be left wondering why, for instance, typing “Ser, sive Sericvs vermis” into a “Title” search brought up a record that shows only shows the spelling “Ser, siue, Sericus vermis.”

So, dear users, here’s the question: do you like it this way? Would you rather that the obvious variations not display at all? Would you rather that the not-obvious ones display right after the main title, with only the obvious ones pushed to the end?

Here are three options for you to weigh in on:

  1. No change. Keep all “Variant Title(s)” together near the end of the record.
  2. Move any unclear variant titles up so that they’re  immediately below the main “Title” and continue to display the obvious variant titles near the bottom.
  3. Move any unclear variant titles up so that they’re  immediately below the main “Title” and hide the obvious ones so that they can only be seen if you click the “MARC view” button.

Comments are open. Don’t be shy. Anyone who uses our records—readers, librarians, folks browsing from afar—has a stake in this question, so please chime in!

  1. Library of Congress-Program for Cooperative Cataloging Policy Statements (LC-PCC PS) for and Library of Congress Rule Interpretations (LCRI) 21.30J.
  2. If you want to learn how to interpret MARC, there’s a handy online tutorial called Understanding MARC Bibliographic: Machine-Readable Cataloging linked to from the MARC Standards homepage.


  • With electronic cataloging and retrieval, it should be possible to set up a default alternative retrieval for these sort of things. What I mean by this is that the searcher looking up “Ser, siue” automatically gets offered “Ser, sive” entries too. In fact, all searchers get irritated by electronic systems which generate multiple references to a single item, so the system should also filter so that only one reference is produced for each unique catalog reference.
    This would ensure that searchers neither miss an appropriate reference nor get umpteen duplicated ones.
    Cliff Webb

    • Fear not: that’s already the case. Whichever version of the title you search for, you just get the one record. Now we’re wondering whether the spelling variations entered by the cataloger should be visible in the record, or whether the system should “silently” bring up that one record if you use a different u/v combination in your search.

  • My first thought is that I’d prefer either option 1 or option 2. Just knowing that that is how the system works makes both more or less equally helpful in my view. I’ve been trying to rack my brain to think of a scenario in which one would be more detrimental than another, but so far I’m not coming up with anything.

  • Option 3 is clearly superior. What Erin didn’t say about the ‘1’ vs ‘3’ coding is that the technical definition of code ‘3’ is to _prevent_ display. The fact that they do currently display in Hamnet was a past choice to circumvent the meaning of the codes.

    Those who do want to see all the underlying title indexes can display the MARC or Staff view and take a look.

Leave a Reply

  • (will not be published)