Journal volume and page range not in API

[Tobias Klevenz]

Hey there,

I'm a developer at Scribbr, working on our free citation generator.

I am looking at your API as an alternative for Crossrefs API, for our Journal search. 

I noticed that your response does not include Volume and Page numbers for Journals, however these are quite important for students when citing Journals.

Is there a reason you don't have these? 

Thanks, Tobias

[Richard Orr]

Hi Tobias,

Thanks for your input and sorry OpenAlex isn't meeting your needs here. We do have volume and page numbers for many Works in Works.biblio, but if you've been comparing with Crossref you may have noticed we mostly get them from Crossref, plus some from PubMed. There's significant room for improvement here - we're just now applying the bibliographic data we have to some Works, and in the future we'll be able to extract it from less structured sources like landing pages.

But I may be misunderstanding the question. When you say "for journals", do you mean "in the Venue object"? If so, would that be a listing of the journal's available volumes (assuming the Venue is a journal), and the page ranges of the issues in them?

Thanks,

Richard

--
You received this message because you are subscribed to the Google Groups "Unpaywall discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unpaywall+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/unpaywall/0f655fd1-711b-4d15-8dc8-28657d97868bn%40googlegroups.com.

[Tobias Klevenz]

Hey Richard,

thanks for your reply, you are correct I was referring to the volumes and page numbers that you have in OpenAlex. However I was using and referring to the Unpaywall API, as I was actually not aware of OpenAlex.

What is the difference between the two? They seem to be serving the same purpose.

Thanks, Tobias

[Richard Orr]

Hi Tobias,

Apologies! I was replying from my inbox and didn't notice which list this came from. It sounds like you're pretty familiar with Unpaywall, so hopefully the main OpenAlex documentation page and a few key points will explain the difference.

OpenAlex has a much broader scope than Unpaywall. From the outside, Unpaywall is essentially a function: the input is a Crossref DOI and the output is an Open Access status, hopefully some OA locations, and basic bibliographic data that mostly comes as-is from Crossref. The only first-class, concrete entities in Unpaywall are the DOI and the OA status. Everything else, the journal, the authors, and so on, are referred to by strings (ISSNs for journals, names for authors), but they aren't represented as independent entities - their existence is only implied by these strings.

OpenAlex presents this data in a more cohesive and usable way, as five Entity types: Work, Author, Venue, Institution, and Concept. Each instance of each Entity type has a persistent identifier used to link it to other Entities in a web, or more precisely a graph - initially we distributed the dataset in the same format as Microsoft Academic Graph, which has since been discontinued. The graph enables all kinds of queries that aren't possible with Unpaywall, starting with simple things like listing all the Works published by an Author or affiliated with an Institution. The Entities themselves give concrete representation to things like journals (a type of Venue) that were only alluded to in Unpaywall.

Because we're no longer using Crossref DOIs as a primary key (Works can be referenced by any of several IDs) we're also able to include papers that couldn't be represented in Unpaywall at all. We do pull in all Open Access information from Unpaywall (see Work.open_access, .host_venue, and .alternate_host_venues) so rather than serving the same purpose OpenAlex does everything Unpaywall can do and more. But it's hard to beat Unpaywall for simplicity at its core use case - plug in a DOI and get an OA location - so while development efforts are focused on OpenAlex and many new users may find it more useful, there are no plans to discontinue it.

Best,

Richard

To view this discussion on the web visit https://groups.google.com/d/msgid/unpaywall/85d05b59-4179-4868-b1cd-b507dbe237fen%40googlegroups.com.

--

Richard Orr

Lead Developer - Unpaywall, OpenAlex

OurResearch: We build tools to make scholarly research more open, connected, and reusable—for everyone.

[ric...@ourresearch.org]

Hello again,

I realize I haven't really answered your original question. We didn't include volume, issue and page numbers in Unpaywall because we didn't think they served its core purpose of helping readers find legal copies of paywalled articles. The Unpaywall API response includes the title, basic journal information, and an authorship list for convenience, but as I mentioned this all comes straight from Crossref. If we were to include it you wouldn't see any improvement over what you're getting from them now. In the future we may feed enhanced bibliographic data from OpenAlex back into Unpaywall. Emphasis on may - we haven't talked about this as a team, it's not on our roadmap, and off the top my head I have concerns about breaking consistency with Crossref which existing users may rely on.

Best,

Richard

[Tobias Klevenz]

Hey Richard,

thanks a lot for your detailed feedback, from your description and from exploring OpenAlex a bit yesterday this seems closer to what we'd need for generating citations anyways. 

I can imagine offering students recommendations based on the data from OpenAlex's crossreferencing capabilities, although this is not currently on our roadmap.

One thing I noticed in OpenAlex that is a bit of a problem for us is that the authorship object does only include a display name, instead of a full name, split up in first and last name. 

Since we do support different citation styles, with specific formatting for an authors name, like some require initials instead of first names for example, it's not possible for us to use "only" the display name of an author.

Cheers, Tobias