CONF - conference paper vs conference proceeding

John_muccigrosso

Zotero interprets the RIS field "CONF" as an item type of "Conference paper", but the RIS spec says it's "conference proceeding" and has the field "CPAPER" for "conference paper".

Is there a further nuance that I'm missing or is Zotero doing it wrong?

adamsmith

IIRC we've talked about this before - it's a bit tricky, both because a fair number of sites uses CONF for conference paper and because Zotero doesn't have a separate conference proceedings item type. Maybe Aurimas has a better recollection?

John_muccigrosso

My own preference would be that Zotero follow the spec and the other sites be encouraged to as well. WorldCat's RIS files are just a mess (as is noted elsewhere in these fora), and it's nice that Zotero's scraper does a better job.

aurimas

I'm not sure what you are requesting. Do you want CONF be imported as something other than Conference Paper (if so, what?), or do you want Conference Paper to be exported as CPAPER instead of CONF? For the latter, my opinion is that the RIS spec provides more relevant fields in the CONF type than in CPAPER (from my quick overview)

John_muccigrosso

I'm suggesting that Zotero interpret the RIS fields according to the spec (which I think I'm reading correctly).

There CONF = Conference Proceedings and CPAPER = Conference Paper. So having Zotero do CONF = Conference Paper is wrong (again, if I read the spec correctly).

adamsmith

What Aurimas is saying is that it's not quite as simple as that, though.
The conference paper item type in Zotero is principally intended for papers published in proceedings. CPAPER in RIS doesn't even have a field for a proceedings title, so just mapping that would lead to data loss.
CONF on the other hand doesn't just have a mapping for proceedings title (C3), but also one for title (the standard TI). If it's just intended for the entire proceedings, what would TI be for? So I think it's not unreasonable to understand CONF as "paper published in conference proceedings" and CPAPER as "unpublished paper given at a conference". And if that's the case the current mappings are the best we can do. If anything we could consider mapping CPAPER to presentation.

John_muccigrosso

I understand your logic, but I think this is more a gap in the RIS spec, than that CONF is intended for single papers.

For example, CONF has fields like "Number of volumes", "Edition", and "ISBN" that would seem to belong to a book-like thing, more than just a single paper.

I would agree that CPAPER seems to be meant for papers that are not in proceedings, but they are perhaps published, as it does include PB, SP, VL and other fields that suggest a publication in some form.

It's also the case that I got on this by downloading an RIS from WorldCat for a book of conference proceedings that had the type CONF. WorldCat isn't great at implementing RIS, of course.

I suppose in the end Zotero should have a good conference paper type, but not interpret CONF to be that. What is the conference-proceeding type in Zotero then?

aurimas

What is the conference-proceeding type in Zotero then?

That's precisely what adamsmith and I were trying to get at. There is no dedicated conference-proceeding type. Conference paper is currently meant to serve both purposes.

adamsmith

well - or book, right? If we're talking about an entire volume for a conference (with editors and multiple contributions by different authors) we'd expect that to be a book.

aurimas

Right, but I can't think of a lot of cases where a book of conference papers would be cited directly.

adamsmith

I don't know. In my field conference proceedings practically don't exists, but the way I understand it, that's what John says CONF is meant for - he's talking about something like this: http://www.worldcat.org/title/proceedings-of-the-16th-international-conference-on-defects-in-semiconductors-icds-16-lehigh-university-bethlehem-pennsylvania-22-26-july-1991/oclc/25330567
which, as he says, Worldcat e.g. has as CONF

aurimas

Ah, yes. If we're talking specifically about WorldCat, then I'm aware of this. I thought I had submitted a patch for this that changes CONF to BOOK, but perhaps I forgot. Well take a look.

adamsmith

well, the question (and John's point) is whether Worldcat's usage may be correct. But looking at the item type and fields as specified, I don't think so.
It seems pretty clear that CONF is principally meant for papers in proceedings. (the fact that it includes ISBN, # of Vols. and editions doesn't speak against that. So does CHAP)

DWL-SDCA

In the safety science world, most conference proceedings are treated as journals issued annually. The series have ISSNs. Proc Human Factors (Sage), Eye & Auto (Elsevier), Proc IRCOBI, etc.

adamsmith

those would be covered by the periodical item type once we get it:
https://github.com/ajlyon/zotero-bits/issues/23

aurimas

But looking at the item type and fields as specified, I don't think so.

Precisely. Sorry for not being more explicit.

In the safety science world, most conference proceedings are treated as journals issued annually. The series have ISSNs. Proc Human Factors (Sage), Eye & Auto (Elsevier), Proc IRCOBI, etc.

This seems to be similar to the issues that we recently discussed about citing letters. Depending on the medium of publication they may need different fields, but I think in most cases, more general titles for these fields, like publication title instead of book title or journal title, could let us avoid this.

On the other hand, is there something specific about citing a conference paper published in a periodical vs. citing a journal article? Same for book section.

those would be covered by the periodical item type once we get it:
https://github.com/ajlyon/zotero-bits/issues/23

I'd have the same comment as I did above. I can't imagine someone citing the periodical as a whole. I looked thorough some of the referenced threads in the proposal, but I didn't really see a very compelling reason besides categorization within Zotero library.

adamsmith

(entire periodicals are cited routinely in history, so there's an established need for that, but that's a different topic).

aurimas

I updated the WorldCat translator to import conference proceedings as books.

You can update your translators via Preferences -> General -> Update Now. The update will not affect importing RIS files directly, so you will have to use the URL bar icon to import these correctly.

John_muccigrosso

In my field (Classics) it is very common to find conference proceedings published as books.

I don't mind turning WorldCat's CONF into BOOK (that's what I've done in Zotero), but the reality is that proceedings have fields that books don't, like "date of conference" and treating CONF as BOOK throws out that information (or puts it into a note of fields not found in Zotero).

I also remain unconvinced that CONF is intended for papers. That CHAP contains book-related info is likely to allow citation of the chapter in the context of the book, no? That is, I might have a chapter specifically cited and also provide enough book info for the book to be found, and in fact the type CHAP has T2 for book title. Chapter citations will normally include info on the book. Likewise CONF has the same info because proceedings may be published as books.

In the end this isn't a great set-up in RIS. CONF and CPAPER clearly don't have the same relationship as BOOK and CHAP, even though one might expect that.

adamsmith

In my field (Classics) it is very common to find conference proceedings published as books.

But a) are those ever cited as a whole? and b) if so are those citations different from regular edited book citations? According to the Chicago Manual, at least, they're not:

A paper included in the published proceedings of a meeting may be treated like a chapter in a book (see 14.125). If published in a journal, it is treated as an article (see 14.175–98).

(CMoS 14.226)

I don't mind turning WorldCat's CONF into BOOK (that's what I've done in Zotero)

Aurimas fixed the translator and it now does that automatically when you use the URL bar icon.

I also remain unconvinced that CONF is intended for papers. That CHAP contains book-related info is likely to allow citation of the chapter in the context of the book, no?

exactly. And what we're saying is that the fields RIS specifies for CONF mirror pretty exactly those for CHAP, which strongly suggest they're for a similar type of item, i.e. a paper/chapter/article published in a edited collection.

John_muccigrosso

Nolo contendere. :-)

I think we're beginning to spend time over a inadequately explained standard no one here is responsible for. I appreciate the attention and the change to the translator.

adamsmith

I was actually most interested in:
"But a) are those ever cited as a whole? and b) if so are those citations different from regular edited book citations?"
which is relevant for how we deal with this in Zotero

John_muccigrosso

Yes, I think they are cited as a whole, but most often pieces are cited. Isn't that typical for collections of short works that appear as books? And no, I don't think they're different from regular book citations, as long as the book in question has multiple authors responsible for those parts.

The difference to my mind lies in the book vs. the proceedings, not the smaller pieces. The conference has place and date info that a book does not. CONF allows this info, but BOOK doesn't. Overall I'd say CONF *is* very similar to book. Here's a list of the fields CONF has, but BOOK doesn't:

A4 - Sponsor
C1 - Place Published
C2 - Year Published
C3 - Proceedings Title
C5 - Packaging Method
CY - Conference Location
M1 - Issue
PY - Year of Conference
T2 - Conference Name

Obviously C1 and C2 do appear in BOOK in different fields (which I don't understand the reason for), but the rest are specific to conferences.

Looking the other way, CONF lacks the following BOOK fields:

A4 - Translator
C3 - Title Prefix
C4 - Reviewer
CY - City
J2 - Abbreviation
M1 - Series Volume
M3 - Type of Work
OP - Original Publication
PY - Year
RP - Reprint Edition
SP - Number of Pages

Year does appear in CONF, but in a different field, but arguably very few others would be necessary.

In short - despite my nolo above - I'd still hold that CONF looks like it would be the place for a book of conference proceedings.

adamsmith

The difference to my mind lies in the book vs. the proceedings, not the smaller pieces.

right, but when a book of proceedings is cited - how does it look? Is the conference date and location mentioned?

John_muccigrosso

I'd say yes, but that's also often the kind of thing that ends up in a long title, like the one you cited above. See, for example, this guide:

http://libguides.reynolds.edu/content.php?pid=143373&sid=1220383

aurimas

And as far as WorldCat is concerned, the fields it exports match much better the fields given for CHAP than it does for CONF. It also does not export anything specific to conference proceedings. So us fixing the RIS to say CHAP instead of CONF actually results in the best import you could get with the exported data.

So the discussion is now focused on how to best import CONF in general. If citing a section of conference proceedings in some citation style requires including conference title or conference date or location, then we would probably want to stick to "Conference Paper" and expand some of the available fields. If nothing special is required, then we can import it as a "Book Section" or "Journal Article" depending on whether an ISBN or ISSN (or some other criteria) is specified. If we need to cite the proceedings as a whole (and again, I do not think that CONF refers to the whole collection of proceedings), it would probably have to be imported as a Book or (the possibly upcoming) Periodical. I would argue, however, that it is always safer to import something as a section (of book or journal), because this retains the most information.

And my argument against CONF referring to the entire collection of proceedings stems from the fact that CONF has fields for Author (AU) in addition to Editor (A2), Pages (SP) instead of Number of Pages that we would see for a book, and Title (TI) in addition to Proceedings Title (C3)

adamsmith

f citing a section of conference proceedings in some citation style requires including conference title or conference date or location, then we would probably want to stick to "Conference Paper" and expand some of the available fields.

oh, we'll definitely want to do this. Conference Paper need two locations and two dates (see also https://github.com/ajlyon/zotero-bits/issues/6 ) CSL already has event-place and event-date. APA style as in the above example by John is a good example for that need.

For me the only remaining question is we need to worry about citing entire proceedings in a way that Book doesn't capture.

John_muccigrosso

WorldCat clearly exports books that are proceedings as CONF. For example, the link adamsmith cited above. WorldCat, as I said at the start, is not a great RIS exporter, so I wouldn't want to put much faith in that.

Again, I think the RIS standard is poor here and we're arguing over imprecision in someone else's work. For example, is it "Conference Proceedings" in the plural, as in the spreadsheet, or "Proceeding" in the singular, as in their accompanying pdf?

And what in the end is supposed to be the difference between CPAPER and CONF, if both are meant to be conference papers? CPAPER has "SP pages", so the idea that it's meant for unpublished content seems odd.

I see a need to have something like a type of "book", which includes conference date and place info. Despite some oddities (like SP = "pages" and not "number of pages") CONF looks to me like the right RIS type. WorldCat seems to agree, but they're not a stickler for detail in RIS , it seems, so I'm not sure that agreement means much.

aurimas

APA 6th edition says:

Proceedings of meetings and symposia can be published in book or periodical form. To cite published proceedings from a book, use the same format as for a book or book chapter (see Example 39). To cite proceedings that are published regularly, use the same format as for a periodical (see Example 38). For contributions to symposia or for paper or poster presentations that have not been formally published, use the following templates.

So I don't think the example on the page linked above is right. It also does not match the example given in the manual:

Proceedings published in book form
Katz, I., Gabayan, K., & Aghajan, H. (2007). A multi-touch surface using multiple cameras. In J. Blanc-Talon, W. Philips, D. Popescu, & P. Scheunders (Eds.), Lecture Notes in Computer Science: Vol. 4678. Advanced Concepts for Intelligent Vision Systems (pp. 97-108). Berlin, Germany: Springer-Verlag. doi:10.1007/978-3-540-74607-2_9

MLA 7th edition does say:

Treat the published proceedings of a conference like a book, but add pertinent information about the conference (unless the book title includes such information)

The example given is as follows (I underlined the conference-pertinent information):

Brady, Brigid, and Patricia Verrone, eds. Proceedings of the Northeast Region Annual Meeting, Conference on Christianity and Literature: Christ Plays in Ten-Thousand Places: The Christ-Figure in Text and Interpretation. 22 Oct. 2005, Caldwell Coll. N.p.:Northeast Regional Conf. on Christianity and Lit., n.d. Print.

But even though they give a format for such citation, I'm still not sure that the whole book would be cited. However, I would imagine similar requirement would be for citing a section of said proceedings, so storing this as a chapter would not provide sufficient fields.

Edit:
FWIW, NLM style guide also gives a large number of examples where the entire book is cited ( http://www.ncbi.nlm.nih.gov/books/NBK7272/ ). We may be unable to avoid adding a conference proceedings item type. It would essentially be a book/periodical with 3 additional fields, so perhaps as far as maintenance is concerned, this wouldn't be a huge deal for Zotero. A bigger problem is the corresponding CSL mapping. I'd still like to find real life examples of where such citations actually occur though.

aurimas

For example, is it "Conference Proceedings" in the plural, as in the spreadsheet, or "Proceeding" in the singular, as in their accompanying pdf?

RIS "standard" is essentially created and implemented by EndNote (a successor to Reference Manager). Both the spreadsheet and EndNote labels indicate "Proceedings Title" for that field.

I see a need to have something like a type of "book", which includes conference date and place info.

I'll go back to my question a couple posts up: do you need to cite this "book" as a whole or do you just want to have this for cataloging purposes?

WorldCat seems to agree, but they're not a stickler for detail in RIS , it seems, so I'm not sure that agreement means much

Yes, WorldCat is not a good source of reference for this. We have to do quite a bit of fixing for their RIS exports before they can be properly imported.

adamsmith

>I see a need to have something like a type of "book",
> which includes conference date and place info.


I'll go back to my question a couple posts up: do you need to cite this "book" as a whole or do you just want to have this for cataloging purposes?

We should really take this away from RIS - where John won't be able to convince aurimas and me ;) - and focus on this question, which is something we can actually determine. I think citations of this in MLA and NLM are sufficient evidence that it is needed.
On the CSL side we don't need an item type. We can just treat this as a book and add event place and date when it exists in the data.

Which leaves us with the representation of this in Zotero. Adding conference data to the book item type seems odd to me, but so does adding a very rarely cited type as a new item type - we'll have too many of those anyway. If we had hierarchical item types this would be a no brainer, but those are a long way out. So I'm not sure how best to solve this.