On 08/11/2011 17:23, J. McRee Elrod wrote:
<snip>
Jim said:
Getting rid of a *single* main entry would be the equivalent of DC's
<creator> and<contributor> where<creator> is repeatable, thereby creating
multiple main entries.
How would you produce single entry bibliographies? How would scholars
cite in footnotes? How would cataloguers construct subject and added
entries for works?
Libraries are part of a larger bibliographic universe, and should
adhere to its standards and practices, which would include returning
to compiler main entry.
</snip>
Could you point me in the direction of a bibliographic citation format
that demands someone choosing a *single* main entry? I have worked a lot
with them and have never found anything resembling a single main entry.
While the practices vary, the main rule is, copy the authors in the
order they appear on the title page. Some stop at a maximum of four,
none more than seven. Some want the forms of names as spelled out on the
item, others say to abbreviate first and middle names. These formats
mostly want people to differentiate between authors and others, e.g.
editors, compilers, and translators, by putting in (ed.) or mentioning
translations. Here is the Chicago format
http://writing.wisc.edu/Handbook/DocChi_WC_book.html. Another nice page
is from Ursinus
http://myrin.ursinus.edu/help/resrch_guides/cit_style_mla.htm. Here is a
guide for the Harvard rules
http://libweb.anglia.ac.uk/referencing/harvard.htm "For books with two,
three or four authors of equal status the names should all be included
in the order they appear in the document. Use an *and* to link the last
two multiple authors." These rules, and others, actually use "et al."!
I admit that these considerations would provide a reason to go back to
the practice of adding relator codes (which I do *not* think is the
right thing to do, by the way).
Now, as far as cataloging items for subject or added entries for works
with two or more main entries, it can be done in XML quite easily, but
more difficult with ISO2709. With XML, for a subject entry for Masters
and Johnson (two main entries), you could have (an abbreviated MARCXML
record. I think catalogers can follow):
<record>
<100>
<a>Smith, John</a><d>1960-</d>
</100>
<245>
<a>The book by Masters and Johnson</a>
<b>some thought</b>
<c>by John Smith</c>
</245>
<260>
<a>New York</a>
<b>Random</b>
<c>2011</c>
</260>
<subjectUniformTitle>
<100>
<a>Masters, William Howell</a>
<d>1915-2001</d>
</100>
<100>
<a>Johnson, Virginia</a>
<d>1925-</d>
</100>
<240>
<a>Human Sexual Response//</a>
</240>
</subjectUniformTitle>
</record>
The same could be done with an analytic or series, just replacing
<subjectUniformTitle> with <analyticUniformTitle> or
<seriesUniformTitle>. How this could be done in ISO2709, I do not know,
but I won't say that it cannot be done because somebody may figure out a
way, but I can't imagine why anyone should want to. XML can do it right
now and it could be utilized by browsers the world over--right now. Once
we get away from ISO2709, there will be all kinds of novel bibliographic
structures that can be implemented. ISO2709 leads catalogers to think in
certain ways about how information in structures. There is no need for
that any longer.
--
James Weinheimer weinheimer.ji...@gmail.com
First Thus: http://catalogingmatters.blogspot.com/
Cooperative Cataloging Rules: http://sites.google.com/site/opencatalogingrules/