Journal Series
Richard Pyle
One more question before I get on my last plane ride home:
Many journals have multiple numbered, lettered, or otherwise delineated
series. For example:
Annals and Magazine of Natural History 2
Annals and Magazine of Natural History 3
Annals and Magazine of Natural History 4
Annals and Magazine of Natural History 5
Annals and Magazine of Natural History 6
Annals and Magazine of Natural History 7
Annals and Magazine of Natural History 8
Annals and Magazine of Natural History 9
Annals and Magazine of Natural History 10
Annals and Magazine of Natural History 11
Annals and Magazine of Natural History 12
Annals and Magazine of Natural History 13
Bulletin du Muséum National d’Histoire Naturelle 1
Bulletin du Muséum National d’Histoire Naturelle 2
Bulletin du Muséum National d’Histoire Naturelle Zoologie, 3
Bulletin du Muséum National d’Histoire Naturelle, Section A, Zoologie
Biologie et Ecologie d’Animal 4
Proceedings of the Zoological Society of London A
Proceedings of the Zoological Society of London B
In some cases, it's not so much a series, as a specific locality. For
example:
Bulletin du Muséum National d’Histoire Naturelle Paris
Bulletin du Muséum National d’Histoire Naturelle Santiago
In my implementations, I capture the title separately from the Series
designator, and I put the Series information in the "Edition" field:
Title Edition
----------------------------------------------------
Annals and Magazine of Natural History 2
Annals and Magazine of Natural History 3
Annals and Magazine of Natural History 4
Annals and Magazine of Natural History 5
Annals and Magazine of Natural History 6
Annals and Magazine of Natural History 7
Annals and Magazine of Natural History 8
Annals and Magazine of Natural History 9
Annals and Magazine of Natural History 10
Annals and Magazine of Natural History 11
Annals and Magazine of Natural History 12
Annals and Magazine of Natural History 13
..then I concatenate when formatting output.
So, the question is, should the "best practices" for the exchange standard
be to parse the Series info a separate element (and have an element for this
purpose), or should we just lump it all together in one Title field?
Although it has been handy for me to dump the Series data in the "Edition"
field of my database, I think this is not the best way to represent it in
the Exchange standard. Because we want to support multiple legitimate titles
for each Reference, we should treat the Series information at the same
resolution -- such that each listed title has its own corresponding Series
value (which will be important when representing the same title plus series
in different languages).
Also, what about the Locality qualifier? Should that be treated as part of
the Journal title, or captured separately?
Thanks,
Rich
Richard L. Pyle, PhD
Database Coordinator for Natural Sciences
and Associate Zoologist in Ichthyology
Department of Natural Sciences, Bishop Museum
1525 Bernice St., Honolulu, HI 96817
Ph: (808)848-4115, Fax: (808)847-8252
email: deep...@bishopmuseum.org
http://hbs.bishopmuseum.org/staff/pylerichard.html
Guido Sautter
Storing the series designator separate from the journal title seems like
a good idea to me, as it allows for listing the whole of a journal,
regardless of the series.
The appropriate model for storing the series designator separate from
the journal title would be making the former an attribute of the latter.
This way, we have a clear relationship between the two in the presence
of multiple titles.
- Guido
Richard Pyle
> separate from the journal title would be making the former an
> attribute of the latter.
> This way, we have a clear relationship between the two in the
> presence of multiple titles.
Wouldn't it be better to have the Element contain the full concatenated
title (with Series, etc.), then have attributes for things like
"periodicalTitle" (for the non-series part) and something like "seriesLabel"
for the series bit?
Or, maybe a structure something like:
<ReferenceTitles>
<ReferenceTitle kindOfTitle="StandardTitle" title="Annals and Magazine of
Natural History, Series 9">
<PeriodicalTitle>Annals and Magazine of Natural
History</PeriodicalTitle>
<SeriesLabel>Series 9</SeriesLabel>
</ReferenceTitle>
<ReferenceTitle kindOfTitle="AbbreviatedTitle" title="Ann. Mag. Nat.
Hist., Ser. 9">
<PeriodicalTitle>Ann. Mag. Nat. Hist.</PeriodicalTitle>
<SeriesLabel>Ser. 9</SeriesLabel>
</ReferenceTitle>
</ReferenceTitles>
Rich
sau...@ira.uka.de
>> The appropriate model for storing the series designator
>> separate from the journal title would be making the former an
>> attribute of the latter.
>> This way, we have a clear relationship between the two in the
>> presence of multiple titles.
>
> Wouldn't it be better to have the Element contain the full concatenated
> title (with Series, etc.), then have attributes for things like
> "periodicalTitle" (for the non-series part) and something like "seriesLabel"
> for the series bit?
Good point ... putting the plain data in the textual content and the
parsed data in one or more attributes generally seems to be a good
guideline. This allows for a clear one-to-one association between raw
and normalized data without having to use container elements. Sorry
for not thinking of this one earlier.
>
> Or, maybe a structure something like:
>
> <ReferenceTitles>
> <ReferenceTitle kindOfTitle="StandardTitle" title="Annals and Magazine of
> Natural History, Series 9">
> <PeriodicalTitle>Annals and Magazine of Natural
> History</PeriodicalTitle>
> <SeriesLabel>Series 9</SeriesLabel>
> </ReferenceTitle>
> <ReferenceTitle kindOfTitle="AbbreviatedTitle" title="Ann. Mag. Nat.
> Hist., Ser. 9">
> <PeriodicalTitle>Ann. Mag. Nat. Hist.</PeriodicalTitle>
> <SeriesLabel>Ser. 9</SeriesLabel>
> </ReferenceTitle>
> </ReferenceTitles>
Not so fund of this one, like the other variant better.
- Guido
Rod Page
journals (or at least be treated as such by some databases). For
example, take a look at
http://worldcat.org/xissn/titlehistory?issn=0375-0434
Decisions about whether series belongs in a separate field or as part
of the contents of the title field could based (at least in part) on
whether the series have distinct ISSNs (or other identifiers, such as
OCLC numbers).
Regards
Rod
Richard Pyle
Thanks, Rod. I think most agree that each series represents a separate
journal, worthy of its own unique identifier. I think it's outside the
scope of the exchange standard to map all the different related journals
together. In the example you gave, I see the following:
GUID: 1
Title: Royal Entomological Society of London. Proceedings of the
Entomological Society of London
Series:
ISSN (Print): 1472-0949
ISSN (Online):
GUID: 2
Title: Proceedings of the Royal Entomological Society of London
Series:
ISSN (Print): 1472-0981
ISSN (Online):
GUID: 3
Title: Proceedings of the Royal Entomological Society of London
Series: Series A, General entomology
ISSN (Print): 0375-0418
ISSN (Online):
GUID: 4
Title: Proceedings of the Royal Entomological Society of London
Series: Series B, Taxonomy
ISSN (Print): 0375-0434
ISSN (Online):
GUID: 5
Title: Proceedings of the Royal Entomological Society of London
Series: Series C, Journal of meetings
ISSN (Print): 1946-1496
ISSN (Online): 1946-150X
GUID: 6
Title: Journal of Entomology
Series: Series A, General entomology
ISSN (Print): 0047-2409
ISSN (Online):
GUID: 7
Title: Journal of Entomology
Series: Series B, Taxonomy
ISSN (Print): 0047-2417
ISSN (Online):
GUID: 8
Title: Proceedings of the Royal Entomological Society of London. Journal of
meetings
Series:
ISSN (Print): 0080-4355
ISSN (Online):
GUID: 9
Title: Journal of Entomology
Series: Series A: Physiology & behaviour
ISSN (Print): 0308-5007
ISSN (Online):
GUID: 10
Title: Physiological Entomology
Series:
ISSN (Print): 0307-6962
ISSN (Online):
GUID: 11
Title: Antenna
Series:
ISSN (Print): 0140-1890
ISSN (Online):
As shown above, "Series" is captured as a separate element from "Title".
Are there examples of other journals where it's not appropriate to include
the series label as a separate element? (i.e., where it is appropriate to
treat it as part of the journal title?)
Another question that we need to decide amongst ourselves is whether in the
example of GUID: 5 above wheter this should be treated as one uniquely
identified journal (as seems to be implied by the diagram at the website),
or whether it should really be counted as two (one for print, and one for
online).
Here my gut feeling is the same as what seems to be rendered in the diagram:
it's one journal with two different ISSNs representing the two different
forms in which the same cited reference it is published. But I could be
persuaded to treat them as two separate records (with two separate GUIDs),
which would allow us to, for example, track Code-governed nomenclatural Acts
directly to the print version. On the other hand, if we can reliably assume
in such cases that the two versions are identical in terms of information
content, then it would be a bit cumbersome to have to double-up on all
citation records (not to mention Name-Usage records in GNUB) for every
journal that publishes a separate print & electronic version (with separate
ISSNs). I doubt that anyone would indicate which of the two versions
they're spcifically citing in a bibliography of a published paper; so it
seems that they represent the "same" citation (in the same way that reprints
have been historically treated as the "same" citation as the originally
published form).
Lots of stuff to think about!
Aloha,
Rich
Kevin Richards
"<div xmlns:dwc="http://rs.tdwg.org/dwc/terms/" >Revis. <span dwc:taxonID="urn:lsid:ipni.otg:names:312284-2">Barnadesieae</span> Mutisieae Fl. Colomb.: 68 (1993)</div>"
or even (IPNI HTTP URI if they were implemented):
"<div xmlns:dwc="http://rs.tdwg.org/dwc/terms/" xmlns:ipni="http://www.ipni.org/ipni/id/" >Revis. <span dwc:taxonID="inpi:312284-2">Barnadesieae</span> Mutisieae Fl. Colomb.: 68 (1993)</div>"
useful??
If you ban html formatting you loose this kind of info.
Kevin
Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails.
The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
Kevin Richards
Roderic Page
Re GUID 5, I'd treat it as the same journal. The print and electronic
ISSN distinction never made sense to me, and as far as I can tell
people use the print ISSN by default (for example, CrossRef), unless
only the electronic ISSN exists (e.g., the BMC journals).
There are also other GUIDs to be used, such as OCLC numbers, etc. I
suspect that for pretty much anything we're interested in, somebody
somewhere will have issued an identifier.
Regards
Rod
Roderic Page
Professor of Taxonomy
DEEB, FBLS
Graham Kerr Building
University of Glasgow
Glasgow G12 8QQ, UK
Email: r.p...@bio.gla.ac.uk
Tel: +44 141 330 4778
Fax: +44 141 330 2792
AIM: rodpa...@aim.com
Facebook: http://www.facebook.com/profile.php?id=1112517192
Twitter: http://twitter.com/rdmpage
Blog: http://iphylo.blogspot.com
Home page: http://taxonomy.zoology.gla.ac.uk/rod/rod.html
Richard Pyle
> electronic ISSN distinction never made sense to me, and as
> far as I can tell people use the print ISSN by default (for
> example, CrossRef), unless only the electronic ISSN exists
> (e.g., the BMC journals).
OK, thanks. I'm definitely inclined in this direction as well. Anyone else
object?
> There are also other GUIDs to be used, such as OCLC numbers,
> etc. I suspect that for pretty much anything we're interested
> in, somebody somewhere will have issued an identifier.
Yes -- definitely. We agreed at TDWG that there needed to be an open-ended
mechanism for attaching N-number of GUIDs to any specific citation record.
Some will have many; others (like microcitations/treatments) may not.
Again; more on this later.
Rich
Dean Pentcheff
series number in with the journal title (or at least I'm not opposed
to doing that).
On the one hand, the series number with the text part of a journal
title really does define a run of a journal, e.g. the volume numbers
restart with each new series (usually). In that sense, the series
numbers should be part of the journal title.
The counterargument is that it is then impossible to list all the
numbers of a single journal, since the "journal names" then differ by
the series number.
The counter-counterargument is that to list an entire journal you have
to include multiple titles anyway, since continuous journals often
change their name. Given that client systems will ultimately need to
establish that "Bulletin of the National Museum Singapore" and "The
Raffles Bulletin of Zoology" are the same journal, unifying journals
that differ only in in series number would seem to fall into the same
job.
One issue that arose is the need to minimize title complexity as an
aid to deduplication (identifying which pairs of references actually
point to the same publication). I don't think that should be a
consideration. That's because I think the whole issue of deduplication
is difficult enough that any particular markup or particularization of
journal titles is the least of the problems. Anyone who thinks that
deduplication is a pretty simple endeavor may want to check the
following recent papers:
Kan, M.-Y. and Y.F. Tan (2008) Record matching in digital library
metadata. Communications of the ACM 51(2): 91–94. URL: DOI:
10.1145/1314215.1314231 http://decapoda.nhm.org/references/31053
Lee, D., J. Kang, P. Mitra, C.L. Giles, and B.-W. On (2007) Are your
citations clean? Communications of the ACM 50(12): 33–38.
http://decapoda.nhm.org/references/31052
The bottom line? We're not the first community to want to deduplicate
electronic references; even for "modern" references it's a really hard
problem; and the computer science community is struggling with the
topic (without outstanding success thus far). We'll be struggling with
deduplication for a long time.
-Dean
--
Dean Pentcheff
pent...@gmail.com