[DCRM-L] DCRM(B) public hearing verbatim notes

Thu Aug 11 13:38:06 MDT 2005

Many thanks to Stephen Skuce, who took these verbatim notes at the
DCRM(B) public hearing in Chicago on 25 June 2005. I hope this document
will serve as a springboard for further discussion. --DJL

____________________________________________

Initials = Names 

ja = John Attig

vb = Virginia Bartow

ac = Ann Copeland

cc = Charles Croissant

lc = Larry Creider

jd = Jackie Dooley

jg = Jane Gillis

djl = Deborah Leslie

jl = James Larrabee

jm = Juliet McLaren

km = Kate Moriarty

rm = Robert Maxwell

rn=Richard Noble

es = Elaine Shiner

ecs = E.C. Schroeder

mt = Manon Theroux

pw = Penny Welbourne

At 5:10 p.m., Deborah (djl) addressed the room, gave history and
background of the development of DCRM(B). [See
http://www.folger.edu/bsc/dcrb/publichearingintro.html] Approximately 10
minutes later, she asked for questions and comments.

Richard Noble (rn): A commendation, particularly for the provisions that
allow for the change in relationships within the trade in printed books
in the 19th century.   The difference was not necessarily limited to
hand press vs. machine press; the evolution reflected differences in
responsibilities and financial channels of the trade, often disconnected
from machine vs. hand press.

Elaine Shiner (es): Question about the option to record "double
punctuation." How does DCRM(B) handle a date on the t.p. that is
followed by a period?  How does the 260 look? Is there a double period?

djl: Transcribe it.

rn: Might ISBD be disappearing with RDA in any case?

John Attig (ja): Developers of RDA are still deciding whether it
will/won't continue to prescribe ISBD punctuation.  Those of us
developing DCRM(B) have been very aware, all along, of the 2
not-quite-parallel processes occurring here, and the Bibliographic
Standards Committee (BSC) made a decision early on that the AACR2
revision process was proceeding too slowly to inform its DCRM(B)
development process.  There likely will, however, be follow-up for
DCRM(B) once the AACR2/RDA process is complete.

djl: Bear in mind that retention of original punctuation is a separate
issue from the application (or not) of ISBD.

djl: New topic: DCRM(B) marks a departure from DCRB regarding the
separation (or not) of ligatures into separate letters.  AACR2 says to
separate all.  The LCRI makes an exception for French, for the oe/ae in
Scandinavian languages, and Anglo-Saxon. BSC asked why. How does
transcribing ligatures fit with the whole transcription process? Why the
exception for certain languages?

The editorial team consulted modern French dictionaries, Allen's Manual.
Result: Only in Danish and Norwegian did it appear that digraphs are
actually different characters than, with different meanings from, for
example, the letters "a e."

rn: There are two considerations here. The main thing is the question of
how languages in the Latin alphabet are affected by filing rules.
Composite characters file differently. The other consideration: though
we already depart from AACR2 transcription (e.g., with i/j and u/v), and
then add "filing titles" in the 246, we may want to minimize departures
from AACR2 during this already-mentioned period of co-evolution of the
two documents (RDA and DCRM(B)). Should we work to reduce instances in
which interoperability between old and new records is a question? rn's
recommendation: follow the LCRI, simplify the transcription.

Virginia Bartow (vb): There are retrieval issues. The ascii value of a
ligature is not the same as 2 separated letters. In title browse and
author/title browse, there can be a large separation between, say, an
early edition and a newer edition depending on how the ligature has been
treated. Collocation in certain indexes may be lost.  So we should
transcribe in a manner that approaches quasi-facsimile, but add tracings
without the ligature, to match up later editions with earlier.

ja: Individual systems handle this differently; most have a table of
equivalents. They're less careful for sorting purposes, perhaps. We hate
to design rules based on what a typical system currently does. And then,
Unicode will raise the same issues, exponentially.

vb: Might BSC prepare a planning statement for systems?

ja: There are some such things currently on the table at MARBI. 

Charles Croissant (cc): How are we to transcribe the Ess-Szet? It wasn't
available to us before, so we used 2 s's [the letter "s" repeated]. But
with unicode?

djl: Sounds like an argument FOR separating ligatures!

James Larrabee (jl): The LCRI exceptions really had more to do with
print styles. The oe/ae digraphs were used universally till 1800, but
their usage had fallen away by the late 19th century. In French, on the
other hand, it's still the current style. Scandinavian languages, I
don't know.  If style is the thing, that should be the basis for any
decision.

djl: [Noted the Oedipus dilemma (separated in English, not in French)]

rn: Is the LCRI based on a long history, perhaps of following the
Bibliotheque nationale? "Let's not offend the French"?

Juliet McLaren (jl): The British used, sometimes still use, Latin
medical terms with ae/oe digraphs [there followed a roomwide attempt to
name such terms]

rn: Are we looking at pure typographical treatment, or orthography? 

vb: Are we looking at attempts to retain or represent a given font in
our transcription? No.

rn: In English, ae/oe and the split versions are interchangeable for
filing purposes.

ja: That's a favorable argument for faithful transcription.

Larry Creider (lc): What about the same edition in Latin, different
states?

vb (almost interrupting) Exactly, piracies for example. How do we bring
this out in cataloging?

rn: Given the relative crudity of our transcription, why worry about
this at all.

Jane Gillis (jg): The LCRI comes from the ms. tradition. [Cites AMREMM,
noting it is a "Bib Standards publication"]. So are we part of this or
something else? There seem to be lots of things pointing us in that
direction, there's weight behind it.

Robert Maxwell (rm): Will we always be using the AACR2 rules? Is there a
REASON to do it, if we were starting over? "We've always done it" is not
a good reason.

jg: This was not even brought up during AMREMM discussions.

jl: It's a transcription issue and a style issue. We have different
styles of dealing with capitalization, digraph usage in English vs.
French, for example.

lc: I'm not in  favor of breaking them all out in general. [something
unclear in notes about "current system" -- I'm not sure if Larry meant
DCRB or an online system--SS]. And, speaking of AMREMM and the ms.
discussion, we normalize spelling with ms. because the spelling is all
over the place. Digraph? Subscript? It can be very hard to tell.

djl: That of course is not the case with printed materials. It's time
for a straw poll.  Vote for:

1. Separate all digraphs and ligatures

2. Transcribe all digraphs as digraphs

3. Only transcribe as digraphs Danish, Norwegian, (Anglo-Saxon if
appropriate, TBD)

4. Keep things as is: follow DCRB and the LCRI (separate all
Scandinavian and French and Anglo-Saxon if appropriate)

[Results:

1 = 5

2 = 2

3 = 12

4 = 15]

djl: DCRM(B) instructs the cataloger to transcribe Roman numeral dates
as Roman numerals, as a coherent extension of faithful transcription.
This was not controversial in Bib Standards, but what do people here
think?  Note that we transcribe just digits, not spaces or periods.
Anyone opposed?

[Response: combination of silence and audible, enthusiastic "No" from
audience.

djl: There was a great deal of fussing at the DCRM conference re: the
publication, distribution, etc. area in order to deal with how different
imprint information is in machine press publications.  The editors then
worked hard to devise a single set of rules for ALL relationships
between publishers, printers, manufacturers.  One result is a fairly
long discussion at the beginning of 4A6.  What do people feel about it?

rn: I haven't looked at the justification.  The main thing is that
previously, 19th century information had been confined to a note.  Note
information is not treated the same by systems as information in the
260. Now we can include in that field information that may, for example,
be indexed as publication information. This is a tremendous advantage.

djl: Did anyone read 4a6 all the way through?

Annie Copland (ac): Yes. You have to hold on tight, but it really works.

Kate [Moriarty?]: It's brilliant.  It's clear and helpful. Don't change
it.

vb: I hope we're not confusing primacy with the number of names?  The
printer had to reduce the type size for, say, 3 printers and one
publisher. Are we reading into, e.g., legal requirements mandating the
publisher's name on the t.p.?  What about vanity works? The publisher is
vague. In job work, the printer's name is tiny. Depending so much on
typographic prominence gives me the heebie jeebies.  19th century
typographic stuff comes into it, attempts to fit things on the page.

rn: We face a choice here. "Subordinate" publishers (for example, those
not in London, or those acting as agents) may in fact be subordinate
[but may not be? unclear to note-taker]. We only have 260 $a and $b to
deal with hem.  It really has more to do with the bibliographical habits
of historians ... we put information on the same level [i.e. $a $b) when
printer and seller are operating at same level.  We've put them together
because in the hand press period there was too much fuzz, and it's
appropriate.  There will continue to be judgment involved.  You can tell
when a publisher is saying, "this is my book".  It's different with
stereotypers.  But it's still the same issue: production vs.
dissemination.

jg: I see a problem with page 60. What if it's the Office of the
Governor? 

djl: The new provision takes care of that.

[NOTE: it became very difficult to keep up with, and more important,
even to follow some of the following exchange. sorry--SS]

rn: Government publications are published by, and are a product of, a
corporate body.

jg: But the imprint: it's printed by someone, say the state or city
publishes something from the Waterworks of San Francisco: to say that
the Waterworks is the actual publisher, put it in brackets, when the
t.p. has a definite imprint, that should be transcribed as an imprint.
That's sufficient. [?]

ja: [We could? we couldn't?] do without the paragraph beginning,
"However ..." [?]  One of the strengths of this is: evidence: what's on
the piece? There's a tension between recording information for
identifying the item, and recording factual information which may have
nothing to do with what you see. 

jg: If an imprint appears in the regular imprint area, don't you record
it as publisher? Because for government publications, I'd have to
bracket the Waterworks in San Francisco.

jm: You don't plunk what you know to be true in the middle of a
transcription area.

jg: Yes you do.

rn: It's a mixed field. It contains mostly information extracted from
the piece, but it's still a mixed process. You must exercise judgment:
to assert the facts of publication, is it sufficient to put the truth in
a note?

Jackie Dooley (jd): 4A6 is indeed a slog, but once you've read it, it's
really, really clear. It's very logical.

es: This codifies the way many of us now actually work. It's good.

rm: What about replacing the "However ..." paragraph with "Optionally
..."?

djl: [Expresses some jocular anger at jg: this was done at her bequest.
Reference to railroad schedules]

jg: I think it works. I want to use it when I have a pamphlet with no
real t.p.  But with a real t.p. that says "Printed by" in an imprint
area: it's put out that way for a certain reason.

jd: Don't make this optional.  You already have the option to surmise.

rm: The rules are the rules. If you know something to be true, you must
record it. That's why it wouldn't make sense to make it optional.

djl: I see 3 options, as follows:

1. Drop the paragraph entirely, say nothing about "reasonably surmising"
the publishing agent

2. Keep the paragraph as written

3. Keep something like it, but distinguish, per Jane's concerns, the
freedom to insert/supply the publisher

djl: (in response to a vague comment): No, we will not add the
"optional" phrasing to this vote.

[Results:

1 = 0

2 = 12

3 = 17]

E.C. Schroeder (ecs): The Bib Standards members all voted to keep it as
it is!

djl: We will consider whether a different, or an additional, example
will do it [in response to a statement noting failure to understand the
example, because of its unfortunate line break]

Manon Theroux (mt): [Asked of jg]: Is it strictly government
publications that are problems? When the information is on the t.p. or
somewhere else?

rn: It's a problem generally with self-publication. The poet issued a
book of his own work, and the printer simply jobbed it.

ja: The manufacturer is working for the author or body that's
promulgating this thing. The author as publisher is a reach in some
cases, not in others.

es: If you are reaching or guessing, you don't have to do it.

jg: Something printed in St. Louis, by the newspaper. But put that into
a manufacturer area, and bracket in the St. Louis Waterworks, and
[unintelligible]. The publication presents itself in an uncertain way,
and we have to take that into consideration.

djl: [Ending the discussion and beginning a new topic]: The issue of
transcribing copyright statements vs. just summarizing them when there
is no publication date. DCRB now follows AACR2; use the copyright as the
date with "c."  But we want to hold firm with transcribing the date
area, not silently changing these data. Thoughts? Do you agree that the
copyright date should be [more fully] transcribed?

cc[?]: The copyright symbol as well as the word copyright spelled out?
"Copyright c1982" looks funny.

djl: What about using "sic" as we do elsewhere?

lc: It's no more an error than a printer name plus a printer's device.

[several voices]: But we don't transcribe a printer's device.

[brief yapping about the meaning of "sic"]

rn: This opens a can of worms.  Copyright information often provides
information of a different kind about the genesis of the text.  It is
often entirely separate from the actual production and dissemination of
the piece. "Copyright 1881" in a 1905 edition. It's probably better to
bail, and retain the practice of defaulting to a second standard with
the option of a note quoting the copyright statement.

es: What does rn mean?

rn: I'm saying we should continue the practice of "c1966" as in ordinary
AACR2 practice because the information does not always present itself
well.  This is extracted information, very distant from the notion of
imprint.

ecs: [agrees with richard].  If it's important, put it in a note. We're
not gaining enough by transcribing it.

djl: How do we justify not transcribing in a transcription area?

jg: I would put c1966 all in brackets; this says you've normalized it.

rn: Look at normal bibliographic practice. It's just not done. [Unclear:
"it adds information regarding the possible date of issue" [?]]

djl: [expresses firm opposition to silent normalization in a
transcription field]: When we've used DCRB for such materials, we've
bracketed the copyright date and added a note with the actual
transcription.

rn: For purposes of $c in the 260, it is not imprint information at all.
It's different altogether. You can infer information from it, but all
information from it IS inferential.

djl: It is a transcription field.  We don't normalize there.

rn: But copyright is different.

djl: But silent normalization is wrong.

rn: Follow AACR2. 

lc: We have a tradition of not regarding the imprint as necessarily a
pure transcription field, especially the date, which explains the Roman
numerals in DCRB.  There have been problems all along. If we try it
we'll likely go back to redo it when we revise these rules yet again.
It can help for identification of states and issues, whatever. But I
remember an LC statement: "don't worry about consistency in this area."

ja: Do you express copyright by the symbol or not? is the question. But
remember, this area is for date of publication, and a copyright date is
NOT a publication date. You can record it as evidence, sure. 

vb: I'm working on Longfellow now. The copyright date is clearly and
demonstrably not the publication date of volume 1 of the set. I infer
the [actual publication] date, and leave it at that [i.e., she does not
record the misleading copyright date]

ja: But we are trying to preserve transcription. The fact that it isn't
a date of publication gives you permission to put it in a note.

es: We do treat the date as transcription.

rn: Copyright statements are about registration, not publication. It may
be good, or it may be misleading, data for determining the publication
date. 

djl: Another poll. The options are:

1. Leave text as it is, and work out some difficult issues

2. Silently normalize copyright data when there's no publication date
(the AACR2 rule)

3. Drop out any instruction regarding copyright dates in this element.
Only record the publication date, whether actual or inferred.

[Results:

1 = 3

2 = 8

3 = 18]

mt: That isn't compatible with DCRB, another consideration informing the
construction of these rules.

rm: Is there a "rare books" reason for treating this differently [than
AACR does]?

es: In modern publications, the copyright and publication dates are
usually pretty close.

[Muttering in the room: Not really!]

rn: Again, it's just not a publication date: it's evidence of the
history of the text. 

jg: I voted for #3, but I still think copyright has to be addressed.
What about the notes area?

ja: Add something regarding "how to conjecture"?

lc: There's a fundamental problem with what AACR2 is doing. We can lead
the way. The evidence for current practice is weak. Every year, there's
an Autocat discussion along the lines of, "It's July 2005, what do I do
with this c2006 book?" 

rm: That's because they faithfully transcribe.

vb: Recording all these copyright dates doesn't help with Longfellow,
for example. I now have 5 "title issues" and all are identical except
for the t.p.; they're from the same stereoplates.

Penny Welbourne (pw): What about bracketing a copyright date, but adding
a 500 note? 

jd: This concerns being required to record copyright dates that are just
printings. Go back to the DCRM(B) introductory matter, under
precataloging decisions. Number 4: exercise judgment, and be consistent.

jl: Throwing out the copyright is OK, but if your only basis for a
bracketed date IS the copyright date, and if you have a reasonably
strong copyright date, why not put it in the 260?

djl: (ending discussion): Now for another departure from DCRB: we are no
longer treating engraved title pages as leaves of text. It had been
treated as text, not a plate. See rule 5B9. The physical extent should
reflect the physical structure of the book.

[scattered applause]

rn: It will be difficult. It is possible to provide gaps in numeration
of letterpress, for example, leaving space for plates to be inserted.
But this is a more faithful transcription.

djl: Are there strong feelings for privileging engraved title pages as
leaves of text?

[general noise signifying "no."]

At 7:05, djl thanked the audience.

The editors were applauded loudly and long.

Deborah J. Leslie, M.A., M.L.S.
Head of Cataloging
Folger Shakespeare Library
201 East Capitol St., SE
Washington, DC 20003
202.675-0369
djleslie at folger.edu <mailto:djleslie at folger.edu> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listserver.lib.byu.edu/pipermail/dcrm-l/attachments/20050811/fb98b97b/attachment.htm 