Citing ebook locations

sdspieg

I keep reading that Chicago, MLA, APA, etc. all state authoritatively that one should NOT cite ebook locations (instead of pages) because they would all be device-specific. It seems to me that we can - and should - do better than that.
It just baffles me that e-pub producers on the one hand and these bibliographical 'standard'-setters haven't been able to find a solution to this. Because what would be better for all parties involved than a hyperlinked citation (in a pdf document, for instance), that one could click on to jump to the exact location via Google Books or Scholar or any other (legal of course) source? They might even be able to charge for some of this!
But even in the absence of such a more 'structural' solution - what would prevent us from using more intelligent 'locations' like in (the amazing) Kovid Goyal's public domain Ebook viewer (part of Calibre)? Those locations, as far as I understand are NOT device-specific.
I realize this is not a Zotero-specific question. But I know that there are many e-savvy academics on this forum who must have ideas on this. Could we add an e-location to the Zotero fields? That we we may be able to show that this CAN be done, and put pressure on all of these parties to get with the time.

bwiernik

Most well-designed ebook systems at this point do have universal locations for referencing (e.g., more recent Amazon Kindle ebooks include page indicators that correspond to the printed book pages), so I think this is a problem that will by and large resolve itself in the short to medium term. For Google Books, the locations also correspond to physical book pages, and this seems to be the standard that is being widely adopted.

sdspieg

I still fail to see why 'we' (the scholarly community) would want to accept the imprecise 'page' as the unit of reference, as opposed to the precise passage. Especially since we do have that ability now...

adamsmith

I think this is a bit harder as a technical problem than you make it out to be. You need to develop a standard that allows for stable, permanent links that work across different file formats and device types and ideally make sense for paper copies of books, too (or for journals: for printed out PDFs).

That's both a significant technical challengs _and_ a sociological one, because you need to get everyone on board with it.

I'm generally quite interested in this, but I think a system that focuses on ebooks, which play a tiny role in academic writing, is a bad idea -- this needs to work for all item types and ideally not just text but also data, video, audio, images, etc. At a minimum, though, this needs to work for journal articles, which make up the bulk of citations.

The group that, to my knowledge, does the most interesting work on this is the open annotation community, which, for obvious reasons is quite interested in anchoring information to specific instances in a text. Hypothes.is would be one way to look. There are other, often related initiatives.

sdspieg

Thanks Sebastian! I'll look into those.

Does anybody know of any 'hard' data on:
* for actual books: on acquisition trends on academic library spending - especially ebook vs printed books?
* for periodicals: do we know what % of academics now read journal articles in e-format?
* for citations: what % are references to data, video, audio, images, etc.

And btw - Kovid Goyal confirmed (on mobileread) that both his old and his new ebook viewers follow the epub CFI standard which allows linking to arbitrary locations inside documents. See http://www.idpf.org/epub/linking/cfi/epub-cfi.html

sdspieg

It seems like progress is being made on this - IDPF/W3C. Has any of this been discussed in the Zotero dev community? I find it hard to believe that I am the only one who now reads virtually all of my books on my kindle. And so I COULD (easily) cite not only the page number but the precise passage I am referring to, if Zotero would let me use it. BTW Sebastian - do you 'cite' anything you annotate on hypothes.is, for instance?

joanfigueroa

APA is the one who should simply accept the cited material and the reference. To look for the cited material, you just go to the cited source and using the search feature, you just "copy-paste" the exact phrase and voila!!! The exact page was only necessary with paper books, in which, of course, you wouldn't be searching page by page.

DWL-SDCA

The current e-book technology doesn't (yet) allow that sort of search to provide reliable phrase-to-single-place referencing. Unless a searched phrase contains a sufficiently large number of words there are likely to be a need for the searcher to evaluate several options -- mostly false drops.

The purposes of citing references are to give credit to the thoughts and findings of those who came before; point to exact quotes; and to help the reader to easily find the reference. It seems to me that, if a phrase is of sufficient length to allow a searcher to find it within a source; the phrase is of a length to require it to be cited as a quotation of the author's exact words.

sdspieg

Any updates on this?

emilianoeheyns

1. Not all e-books allow copy-paste (I have some academic Kindle and Google books that come locked this way)

2. You don't always want to include whole phrases when referencing something. Consider what "some claim (Hanowski, Hickman, Wierwille, & Keisler, 2007, Chapter 3; Huang & Sadek, 2009, p. 6; McCarthy et al., 2017, p. 118; Meiring & Myburgh, 2015, p. 231; Zheng, Suzuki, & Fujita, 2012, p. 12)" would look like when the page numbers are replaced by whole (multiple) phrases.

3. The phrases-location might in some instances work if you are sure that everyone has the article in e-book format. Even when I do have it in PDF, I will usually have a hard-copy. I prefer hard-copy for concentrated reading.

sdspieg

1. Not all e-books allow copy-paste (I have some academic Kindle and Google books that come locked this way)

But all of mine are

2. You don't always want to include whole phrases when referencing something. Consider what "some claim (Hanowski, Hickman, Wierwille, & Keisler, 2007, Chapter 3; Huang & Sadek, 2009, p. 6; McCarthy et al., 2017, p. 118; Meiring & Myburgh, 2015, p. 231; Zheng, Suzuki, & Fujita, 2012, p. 12)" would look like when the page numbers are replaced by whole (multiple) phrases.

They wouldn't - they would just get the exact location of (at least) where the reference starts and ends. So to give the Calibre ebook example - sthg like 392.0 / 2387: 17.91-17.95 (whereby the last two bits are the actual paragraphs). See here for more info. And then imagine that hyperlinks were possible.

3. The phrases-location might in some instances work if you are sure that everyone has the article in e-book format.

This is more for e-books. Which should be identifiable by the ISBN. Although I wonder whether PDF-coordinates shouldn't work either.

Even when I do have it in PDF, I will usually have a hard-copy. I prefer hard-copy for concentrated reading.

For reading, I personally really don't care - a kindle is just a lot easier. But for annotating (and that's what I was talking about - being able to provide references to the passages one may have highlighted in an ebook), it's of course no contest. 

bwiernik

@sdspieg It doesn't really seem like Zotero is the appropriate place to be raising these questions, as there is no universal standard for Zotero to implement at this time. Kindle links might be useful for you (and you already can add kindle:// links to Zotero items or notes to go to specific places in items for your own use), but for citation purposes, they aren't very useful because they are specific to that device.

sdspieg

I guess I agree it's not specific to Zotero. Bu they are relevant to people interested in referencing, which, I guess, is all of us here. But just to correct you on that last point - those locations are not specific to any device, but they ARE specific to Calibre, which provides standardized locations across all devices.

bwiernik

That’s still not generally useful as a bibliographic tool to point other readers to a specific location, as most readers don’t use Calibre or any other single software.

DWL-SDCA

This discussion seems to focus upon personal use (as suggested by @bwiernic). Even though you can include pointers to devices such as Kindle locations in the extras field; a citation including that location is, at best, only useful to someone who has access to a Kindle version of the document. Annotations in documents are only useful to the owner of the document.

I may be on the extreme of the OCD spectrum in that I always verify the sources provided in manuscripts I review or papers submitted by students. I am not unique in that way. (Incidentally, I too frequently discover not only incorrect metadata but cites to things that don't support the authors' statement.) If I can't find the source I insist on seeing a copy of the source -- typically I receive a pdf file. I don't know how to verify a Kindle source citation without viewing the document on a Kindle. I guess that there are ways to accomplish this but all I can think of is making an appointment with a student to see the document on their device (extraordinary effort for everyone involved). I'm certainly not willing to accept responsibility for a student's device on temporary loan. Even those extreme ideas aren't possible to a reviewer of a journal manuscript submission.

All this said, my own preference whether reading for pleasure or knowledge is to view the document on my tablet and I have a Kindle app on my iPad. I look forward to a time when both technology and citation review protocols allow e-books and similar devices to be viable tools for document review.

sdspieg

Wondering whether there already was a better solution for this, I stumbled upon this unsent saved draft. But so is there anything new in this discussion?

This discussion seems to focus upon personal use (as suggested by @bwiernic).

Why do you say that? To me, personally it is not. I now almost exclusively books on my Kindle (or on my pc). I do want to able to share the exact locations that I'm citing in my works. So for me, the ideal solution would be that there would be an extra field in the Item type 'Book', which would allow us to add the Calibre location. Or any other standardized form of ebook location that the people working on style manuals etc. come up with. This would then also allow us - at least for open access journals or other online materials - to add drilldowns to the exact citation.

Even though you can include pointers to devices such as Kindle locations in the extras field; a citation including that location is, at best, only useful to someone who has access to a Kindle version of the document. Annotations in documents are only useful to the owner of the document.

Not with the solution I propose. You could open the pdf file in Calibre (which is open source, free AND pretty cool), and immediately check the precise quote. With much more precision that a crude 'page'.

adamsmith

Nothing new I'm aware of.

I still really don't see this as something Zotero needs to worry about in the near future. For in-text citations, you can already add arbitrary suffixes as locators, including Hypothesis anchor links, Kindle, or Calibre locations.

Since these are to specific locations in a given work, I don't see how they'd be included in Zotero metadata for an item (just like book items don't have page ranges: they're for the whole book), so no need for an update in the Zotero data model.

So from Zotero/CSL's perspective, I fail to see major functionality that isn't already available. The rest is a socio-technical question of adoption, and this, being mostly a tech support forum, is probably not the place for that.