I’d promised some librarians that I would write up a comparison between PhilPapers (PP) and the Philosopher’s Index (PI), because choosing between the two of them might be a budgetary necessity for librarians who wanted to subscribe to PhilPapers under the new terms. This has been delayed somewhat because I knew PhilPapers was planning to announce some important changes, and until then a comparison would be premature. The changes are on the website now, so I feel comfortable writing. The big news is that PhilPapers will be merging with the Philosophy Research Index (PRI). This will still be a comparison, but the incorporation of the the PRI into PP is something of a game changer. But first, some comparisons.
If we’re going by sheer number of entries, PP is ahead. As of July 21, there were 1,104,558 entries from 1,032 journals. According to the Philosopher’s Index website, PI “has a total of over 540,000 journal article and book citations from over 1600 journals collected from 139 countries in 37 languages.” This is qualified somewhat in that only about half of the PP entries are classified according to the categories of its philosophy bibliography. That makes the number of controlled indexed entries about the same. However, PP is, according to David Bourget, “categorizing hundreds, sometimes thousands a day,” and will soon be improving the categorization process. Thus, in the not too distant future, most if not all of PP’s entries will be categorized, making them even more accessible than they are now. In addition, about 700,000 of the entries have categories or associated keywords, and I’ve been told by PP that the most prominent method for accessing entries is search, not browsing via the bibliography. So most of the entries are available to search. In addition, PRI is larger than both PP and PI, with more than 1.3 million bibliographic records. It also covers 800 journals in 30 languages. Once PP incorporates PRI, PP will definitely be by far the largest philosophy literature index. The coverage will also go back to the beginnings of many library journals, instead of just back to 1940 as with PI, and the addition of more foreign language coverage will broaden the scope considerably.
The PP/PRI merger also means that PP will incorporate the Philosopher’s Index Thesaurus. For those unfamiliar with the history of PI and the Philosophy Documentation Center (PDC), a little background might be worthwhile. PDC and PI were both founded in the 1960s at Bowling Green State University, and until 1995 PDC published PI. In 1995 the editor of PI left BGSU and took PI with him. PI is now published by the Philosopher’s Information Center. However, the PDC still owns the Philosopher’s Index Thesaurus, which is the thesaurus PI still uses, and which PRI has been using to build up its own index. The thesaurus is available in print from the PDC, which explains why it cannot be accessed from within PI, comparable to thesauri from other indexes. Thus, when PP incorporates PRI, PP will have both its robust and developing bibliography of philosophy and the thesaurus that PI also uses, plus more extensive coverage of the philosophical literature.
There’s also a difference in how the entries in PP and PI are classified. PI uses the Philosopher’s Index Thesaurus. I couldn’t find any information on the website by whom the indexing is done, but presumably it’s a by a team of indexers with some knowledge of philosophy (if anyone has more complete information, please let me know). PP entries are classified according to the entries of the philosophy bibliography either by the authors themselves or appointed editors, all of whom are professional academic philosophers. I haven’t noticed any problems with either classification process, so I’m not sure the comparison would help anyone make a choice. If others disagree or have found issues, please leave a comment.
One problem I had early on was using SFX from PP. I was getting incomplete results. The problem could be solved only by creating an account with PP and going through a relatively simple process of choosing a link resolver (very simple if anyone from your institution had ever done it before). The accounts can be completely private if you choose, but I disliked the extra steps someone might have to take to get to articles that PP doesn’t have OA but which a library might subscribe to. However, PP is improving OpenURL and SFX linking, and subscribing institutions shouldn’t have a problem. It should work as seamlessly as PI when everything is done.
The final comparison is platform and price. PI is a proprietary index available through Ebsco, Ovid, and ProQuest. Princeton uses the Ebsco interface, which I happen to find very user-friendly. The PP website is also very user-friendly in my opinion. On whatever platform, the cost will vary among institutions because of differences in FTEs or consortial agreements or whatever. Princeton pays a few thousand, and the PP expectation from Princeton is $1200 because Princeton is a philosophy PhD granting university. That makes PP cheaper than PI for my library. I don’t plan to cease subscribing to PI yet, because I’m awaiting further PP developments and I want to have a conversation about it with the Philosophy Department, but I imagine that will matter for a lot of libraries. However, with the incorporation of PRI into PP, I will be canceling the subscription to PRI if it continues to exist as a standalone database, and that money will go to PP instead.
And then there’s the open access of PP. PI is available only to subscribing institutions, while PP is available to everyone in the world. As those of you reading in the spring might remember, my major objections to the PP subscription drive were the unmanageable budgetary timing (asking to subscribe by June 1 or face penalties) and the list of institutions expected to pay (basically every institution in the world from which anyone had ever accessed PP). I thought the first pointlessly hasty and the second unjust. Both those objections were met soon after. The announcement of the PP/PRI merger says, “The service will continue to be available on the model where non-institutional use is free and only institutions located in high-GDP countries and that offer degrees in philosophy are asked to subscribe.” Although I still would have preferred the Stanford Encyclopedia of Philosophy (SEP) endowment method, those are the same kinds of institutions that SEP targeted and the kind that should be subsidizing this sort of open access project if possible.
Of course, PP is more than just a competitor to PI. In addition to the growing index and the structured bibliography, it has the huge OA archive of philosophy articles. It also has announcements for philosophy events and job, and generally serves as a community portal for professional philosophers and philosophy grad students around the world to share work and stay informed. I’m not aware of anything quite like this for other academic disciplines. If PP can gather enough subscriptions to continue to develop, it will remain an important resource for anyone interested in philosophy. And when PP is used in combination with the SEP, philosophy has perhaps the most robust OA reference support of any academic discipline.
Thanks for your thoughtful analysis. You should the link for Brian Leiter to post to his blog. I’m not crazy about the fact that PP mixes up open access journals with subscriptions in the article citations. That may be a source of confusion to some of our students in using SFX, and getting full text. …. They need to research and make sure all OA indexed journals are reputable. Students are often told to only used peer-reviewed or refereed sources. I don’t see a way to filter for those on the PP advanced search enginge. Just filtering for “professional authors” doesn’t quite do it. Anyone can call themselves that. I’m not crazy about the fact that in PP the “export” function for article citations does not include RefWorks as an option. Refworks is more ubiquitous than Endnote, which tends to favor science disciplines. –SAA
[Edited by request to remove one factual error. WBT]
Susan, PP would probably be interested in knowing about the alleged predatory journal issue. And I should have considered Refworks. PP works just fine with Zotero, which I greatly prefer to Refworks, but we also have a Refworks subscription.
Also, Susan, what evidence do you have besides Beall’s list that the “Cosmos and History” indexed by PhilPapers is a “predatory journal”? I just checked out the site. It has lots of content, clear mention of sponsorship by two Australian universities, and no mention of fees to publish. One problem with Beall’s list is that he’s so biased against OA publishing in general that just about everything is considered “predatory.” If this journal is on his list, that’s good evidence his list is arbitrary and flawed, not that the journal is predatory.
I just looked up Cosmos and History in the Directory of Open Access Journals, which indicates it does not charge fees. If this journal is on Beall’s list, it’s an example of him adding reputable green-OA journals to his list. To him, all OA publishing is “predatory.”
Just quickly, Cosmos and History is not listed on Beall’s list. And further, Cosmos and History is listed on Raja’s short list of reliable open-access journals, which you can read what criteria has to be met in order to be listed: http://openaccess.commons.mla.org/rajas-list-of-reilable-open-access-journals/
I also want to thank the author of the original post. I believe this is a very useful service.
Susan: As Wayne said, it’s not obvious what the problem with this journal is.
RefWorks export is on its way. It’s been in the works for some time. Your email request prompted me to accelerate its implementation and I believe that I will go online next week.
Incidentally, the “Endnote” export option is not just for Endnote. A lot of bibliographic software supports this format (too many for us to list them). In fact, I think our Zotero export is just another label for the Endnote option. Every bibliographic software I know of supports at least one of the formats we offer, except maybe RefWorks.
Re mixing OA and subscriptions, one of the main reasons people use PhilPapers is because it provides a single point of access to all online research in philosophy, whether OA or not. It would be a huge loss if we didn’t do that.
I’m not sure how confusion could arise with respect to SFX. SFX is for looking up published works at your library. That can be used with published OA material. For unpublished works, we don’t offer the SFX lookup option, so people just access the paper directly by clicking one of the other links we provide.
I want to clarify that people don’t self-ascribe professional status on PP. This status is ascribed by the system based on such things as publication records. Generally speaking, people who don’t have the status aren’t inclined to game the system. The system actually works. Of course, it’s not a complete protection against poor quality philosophy — we don’t think it’s our role to do that. We’re not going to replace teachers and textbooks as far as guiding students goes.
That said, we are aware of the need for more guidance and we are trying to provide some with the introductory blurbs that we’ve added to about 500 categories (more on the way). This points learners to important works and good introductions on the topics in the bibliography.
Filtering by peer-reviewed status would leave out a lot of excellent and important material published in collections that isn’t really peer-reviewed. As far as I know, we offer the most restrictive “quality filters” of all search indexes with our optional “published only” and “professional author only” filters.
Thanks for the feedback!
Pingback: Philosopher’s Index vs. PhilPapers | Daily Nous
Pingback: PhilPapers vs. Philosopher’s Index « PHILOSOPHY IN A TIME OF ERROR
“PI uses the Philosopher’s Index Thesaurus. I couldn’t find any information on the website by whom the indexing is done, but presumably it’s a by a team of indexers with some knowledge of philosophy (if anyone has more complete information, please let me know).”
PI indexer here. The indexing is indeed done by a team of Assistant Editors, most of whom have PhDs in philosophy:
http://philindex.org/about/editors/
What I do to index each article is: read at least the title and (if it’s provided) the abstract, introduction, section headings, and conclusion. Then I skim the paper and, depending on whether I need more information, I might fully read some or even all of the sections. My job is to tag the article with several keywords from the thesaurus, although I have the freedom to add new keywords when I judge that they’re needed. I’m paid on a per-article basis, and am expected to spend about 10 minutes indexing each article. The journals and books I’m sent are generally keyed to my areas of expertise.
It’s my understanding that the thesaurus currently in use by the PI is—since it’s frequently updated by assistant editors in the course of our work—not the same thesaurus used by PRI, and is up to date with the philosophical literature.
Adam, thank you very much for adding to, and correcting, my limited description of this.
I’d be interested to know how your Philosophy Department enjoys the new PhilPapers, and whether they’d support cancelation of PI. We subscribe to both at my university, but I’d like to cancel PI if PP makes it redundant (and it sounds as if it might). I hope you’ll return to this subject with an update.
There are advantages and disadvantages to PI and PP. Basically, PI is better for accessing all and only the published articles from high credibility journals, and PP is better for getting a pulse on the current state of the field. If I were an undergrad, I would probably be better off sticking to PI’s more limited resources, but as a grad student or professional I would prefer PP. And this seems to be how it’s actually playing out. PP has more web traffic (7/10 Google PageRank vs. 6/10 for PI). Presumably, this is because more of the professionals are engaging with it. PP enjoys 30 times as many Facebook mentions (shares/likes/comments = PP: 624/283/217 vs. PI: 24/5/1). Twitter: 434 vs. 6. Google+: 90 vs. 1. Sites linking in: 1,646 vs. 109.
Undergrads don’t rush to the chance to tweet about the online archives they accessed, so I think what this social media data tells us is that PP is the preferred resource of professional philosophers. But, still, I think PI is better for undergrads, or anyone who doesn’t know the field they’re researching.
Andrew: I have to disagree with the suggestion that PI is better for accessing all and only the published articles from “high credibility journals.”
About the “all” side: My impression is that PI is much slower to index anything, even top journals. We’re also better for getting historical material (PI only goes back to the 1940s). We have more content that is updated more frequently and is better organized for every slice of the field, even high credibility publications.
About the “only” side: PI has a wide diversity of content, including a diversity of languages and approaches to philosophy.I suspect that when you say “high credibility journals” you have in mind a certain set of journals in English-language analytic philosophy (phil review, mind, jphil, etc). I believe PI praises itself in not being focused on this type of philosophy at the detriment of other traditions. Their index is much smaller than PP’s but it’s not the case that it’s “purer” by anyone’s standards because of that. If anything people have complained that PP was too focused on English-language analytic philosophy compared to PI (which is no longer true — that was in the early days when we only allowed English publications).
Regarding the suitability of PI and PP for undergraduates, we go much farther than PI in orienting beginners thanks to our bibliographies and our introductory texts pointing to key readings and introductions. Some of this is a work in progress, but already over 500 categories have introductory texts. Even if PI’s index had a higher ratio of high credibility journal publications (which I doubt), it would still be 99% material that is not a good starting point for beginners, so these features constitute a more important difference between the two services. Our integration with the Stanford Encyclopedia of Philosophy is also something very useful for learners. Arguably, what undergraduates need is guidance from their teachers and encyclopedia-like resources, not an incomplete index. Through integration with suitable resources, we aim to provide them the guidance they need without compromising the completeness of our index.