The selected-papers network

This post is to report briefly on a new and to my mind very exciting venture in academic publishing. It’s called the Selected Papers Network, and it has been designed and created by Christopher Lee. If you want to know what it is and what you can do to help it become a success, then you may wish to stop reading this post and turn straight away to a post by John Baez, who has been closely involved with the venture and understands it better than I do. But let me just briefly mention the main point that has struck me so far.

A problem with the current situation is that it is easy to come up with ideas for websites where people can review papers, complete with clever protocols for how the reviewing should take place, whether it is open, reward systems, etc. etc. It’s much less easy to persuade people to use the sites that are created as a result: what is going to persuade them to make the effort, when there’s only rather a small chance that the site will become in any sense “official”?

The Selected Papers Network potentially solves this problem in a very interesting way: it is not a website with a system for reviewing, evaluating, rewarding etc.. Rather, it is an environment that makes it easy to build your own systems.

The rough idea is this. If you ever feel moved to write an appreciation or evaluation of any kind you like about any paper, and if you tag what you have written with #spnetwork, then the Selected Papers Network automatically sees what you have written and adds it to the network. So what, you might ask. Well, so quite a lot actually, since if you add other tags, then the Selected Papers Network will make all these reviews searchable in a multitude of ways. For example, it will be searchable by subject matter, or by reviewer, or by some group of reviewers who have decided to club together, etc. etc. I’m slightly hazy about the precise mechanisms for this — and there I would definitely recommend reading John Baez’s post — but the point is that the mechanisms are there.

At the moment, the site works only with Google Plus posts. They were chosen because (i) a number of mathematicians have taken to Google Plus, and (ii) Google Plus posts are easily visible even to people who are not signed up to Google Plus. To give you an idea of what the site does, here is a suitably tagged post by Terence Tao on Google Plus, and here is what the Selected Papers Network did to it. And here is the front page of the Selected Papers Network. You’ll notice that the design is strikingly similar to that of the arXiv. That is of course not a coincidence.

It seems to me that a very good thing to do at this point would be to get a lot of content on the site and not worry too much about how that content is organized. To that end, I have a plan to create a personal list of recommendations. Each recommendation would take the form of a Google Plus post that links to a paper that I feel has influenced my mathematical development and explains why. I can see that as this develops (if it does, but I would like to try to make the posts of a kind that I can write fairly quickly, to maximize the probability that it will), I might start to want to categorize the posts in a finer way. For example, perhaps some of the papers will be ones that have interested me greatly but not actually affected my research all that much, perhaps I’ll want to distinguish between very recently arXived papers and older ones, etc. etc. But I still think that it would be good to get the process started.

Another advantage of getting content up on the site quickly is that one can always reorganize it later. For example, suppose that a number of people with similar interests to mine started recommending papers. Then perhaps it would make sense to combine into a “subnetwork”. It’s easy to imagine that being a very useful resource — a kind of instant annotated reading list in one area of mathematics.

My personal policy for my “reviews”, or whatever one wants to call them, is this. I won’t write about papers that don’t interest me, because I see my job as being positive and encouraging. For a similar reason, I won’t make comparisons, either explicitly or implicitly (by the latter I mean via some kind of ranking, which would allow my attitudes to different papers to be compared). I will explain my reasons for being enthusiastic about the papers I write about, which may allow some kind of reading between the lines if you want to gauge my level of enthusiasm — but that will be an imprecise measurement and therefore I hope that I won’t come across as implicitly negative about any papers I write about, which would be ridiculous because I’ll be writing about them because I find them interesting.

If I don’t write about a paper, it won’t be a sign that I don’t find it interesting. I think I’ll ration myself to at most one a week, so for a long time I won’t have written enough reviews for it to be possible to interpret my not having written about a paper as meaning anything at all. Nor will the order in which I write about papers carry any information: I’ll just sit down and think, “Oh yes, that’s a nice paper. Let me write a few paragraphs about that one.”

One thing I’d like to see is a day when if somebody applies for a job in mathematics, there will typically be several appreciations of their work instantly available online, conveying information that is currently conveyed by reference letters (but not all that information — in particular, not comparisons or negative opinions). Or if you want to know about a particular area and want a feel for what the key papers are and why they are important, you can find all the information you could possibly want by keying in subject tags into spnetwork, or reading the recommendations of your favourite reviewers or groups of reviewers.

While writing this, I’ve thought of another way of explaining what will govern my selections. I have a private concept that I call “the story of mathematics”, though it might be more accurate to say “the stories of mathematics”. A paper belongs to the story of some part of mathematics if it would be very natural to mention that paper when describing that part of mathematics, since for one reason or another that paper changed people’s view of the area — by introducing a key definition, solving an important problem, introducing a technique that is now widely used, etc. etc. This is a matter of degree of course. But it’s papers like that — papers that have contributed to the stories of mathematics that have had an impact on me — that I want to write about.

Let me end with a disclaimer: my saying all this is not a guarantee that I’ll actually get round to doing it. However, it increases the probability that I will, since it will be slightly embarrassing to have written this post and followed it by not writing any reviews.

But the other reason for writing the post is that I hope it will encourage others to do similar things: even if 1000 mathematicians each wrote just one review, that would already create a site worth exploring, and in principle it could happen very quickly. I’m also quite pleased to have a chance to advertise Google Plus. If you’re a Facebook kind of person who thinks Google Plus is a rather lame imitation of Facebook that was introduced too late to be worth taking seriously, then you may, if you’re a mathematician, like to think again. I’m not on Facebook myself, but my impression is that Google Plus, or at least the corner of Google Plus that is inhabited by mathematicians and others with similar interests, is doing something interestingly different. I personally see it as a good place for short posts that aren’t serious enough for this blog, but are nevertheless things I feel like saying, or drawing attention to. And several other people are using it in a similar way. So if you’re a mathematician without a Google Plus account, you may be missing out on something you’d like. And that may become even more true if the Selected Papers Network takes off, though — and this is important to stress — whether or not you sign on to Google Plus or the Selected Papers Network itself, you’ll be able to read all the reviews.

25 Responses to “The selected-papers network”

  1. Bill Says:

    It will be terrible if departments start using this system in their evaluation of job applicants. In my experience, different fields have very different cultures, determined by the attitudes of senior people in the field. I can imagine a young and more talented postdoc being passed over for a job because another one has good comments from people in his field, like Timothy Gowers. It is not the first one’s fault that senior people in his field are not into promoting their field on the internet. Will this put pressure on everybody to use the system to promote their fields? I hope this will not happen.

    Otherwise, this might not be a bad idea. As an author, I would also like to be able to add some informal comments on my own papers in addition to formal abstracts, which can be instantly connected to arxiv. I hope the general culture will not necessarily be to post formal review. For example, I might be interested to comment on some particularly nice idea in a paper, without giving an opinion about the paper as a whole.

    • gowers Says:

      I see this not so much as a “system” and more as a means of making it easier to obtain information of a certain useful kind. It may be that some fields will make use of this facility more than others, but we have to deal with similar kinds of “unfairness” the whole time. For example, some fields are, even now, better at promoting themselves — I think that’s independent of whether they use the internet for the purpose. And some people are known for writing more hyperbolic references than others. And so on. In response to this, one has to develop a skill at interpreting the data. To my mind we should just embrace that, rather than trying to restrict the data. To take another example, Twitter can be a useful way of finding out the very latest news about something. A lot of what one finds on Twitter is unreliable, but as long as you use it with an appropriate level of scepticism, it can be a valuable additional source of information.

    • Bill Says:

      Well, I looked at most of the comments on the SPN so far and here is my interpretations of this small data. People are indeed treating it as a social network, as if they are posting on Google+ as usual, so most of the comments are rather uninformative (to put it mildly). The discussions look like typical comments on a random blog. If the noise level will stay this high, any good quality comments will be lost in it, accept for the people that are participating in a discussion. Certainly, I would not waste my time looking at SPN to make a hiring decision. So, yes, I guess I shouldn’t be worried about this side affect, but as a referee I would still look at the discussion.

      By the way, John Baez states that the purpose of SPN is:

      • selection—assessing papers;
      • endorsement—making the quality of papers known, thus giving scholars the prestige they need to get jobs and promotions.

      So, my concern was right on target, but I predict the SPN will fail in its stated goals but might succeed in other ways.

    • John Baez Says:

      Bill wrote:

      By the way, John Baez states that the purpose of SPN is:

      • selection—assessing papers;
      • endorsement—making the quality of papers known, thus giving scholars the prestige they need to get jobs and promotions.

      Not exactly. I said:

      The big problems are expensive journals and ineffective peer review. But we argued that solving these problems require new methods of

      • selection—assessing papers


      • endorsement—making the quality of papers known, thus giving scholars the prestige they need to get jobs and promotions.

      The Selected Papers Network is an infrastructure for doing both these jobs in an open, distributed way. It’s not yet the solution to the big visible problems—just a framework upon which we can build those solutions.

      It’s an infrastructure, sort of like the phone system. You’re imagining that people will use it in a loose, unstructured way, a bit like randomly dialing phone numbers to ask people questions. I’m imagining a decades-long process of exploring the possibilities of this infrastructure and building up social conventions and institutions for using it effectively. It got started just yesterday; people are just testing it out. In 5 years we’ll know if it’s doing something useful.

    • Bill Says:

      Sorry for being pessimistic, and I do wish you good luck with this infrastructure, I do want this to turn into something useful in 5 years. But maybe it’s a good idea to ask from the start that users follow some basic rules to reduce random chatter…

  2. Bill Says:

    Another possible negative side effect. Quite often, I really like a result in some paper but am not willing to put the time and check that all the details of the proof are correct. Suppose I leave a comment saying that this result is very important and was certainly difficult to obtain, without mentioning that I haven’t checked all the details myself. Is it possible that a jornal referee will take my comment as an endorsement of the paper and skip checking the technical details?

    • Bill Says:

      The example you mention above of Terry Tao’s comment, I bet that paper will be accepted quickly because his comment really sounded like an endorsement. Especially since he is talking about the proof in that paper, it really sound like he found everything to be correct and if I were the referee I would have much less pressure to check all the details. But in reality did Terry Tao check all the details?

    • gowers Says:

      I plan to be quite careful to specify in my reviews whether or not I have checked the details, and more generally to give some idea of the basis on which my comments are made.

  3. launched! | chorasimilarity Says:

    […] Timothy  Gowers  has a post where he explains what he intends to do with/for  the spnetwork. Especially interesting […]

  4. Leonard Says:

    Traditionally, post-publication review in mathematics has been done via the services of MathReviews and/or zbMATH. What implications has this SelectedPapers idea for those services? Is this a way of saying they are outdated? Could such a service like SelectedPapers be somehow integrated within the established MR/ZB services?

    Since MR/ZB are primarily concerned with completeness of the database, whereas SelectedPapers can per definition never aspire to be complete but can instead offer flexibility and immediacy, wouldn’t it be desirable to incorporate both sets of features under a central service? By the way, is Google the right platform, given the latest concerns on how they deal with data and products (privacy concerns, shutting down services like Reader…).

    Does the community need even more venues to comment on mathematics literature, or are these new social-network-based approaches intended to replace the old services?

    Thanks for your opinions and insights.

    • John Baez Says:

      Leonard wrote:

      By the way, is Google the right platform […?]

      If you read my post about the Selected Papers Network, you’ll see that it’s a ‘federated’ system, not relying on any one platform. Chris got it to interact with Google+ first, but other social networks, blogs, etc. are coming.

      Math Reviews and zbMATH don’t freely share their information, which greatly limits their ability to interact with the Selected Papers Network… but it does not entirely rule out their ability to do so.

      If you’re too busy to read my blog article, try just this:

      After reading this, you may be tempted to ask: “Doesn’t website X already do most of this? Why bother starting another?”

      Here’s the answer: our approach is different because it is federated. What does that mean? Here’s the test: if somebody else were to write their own implementation of the protocol and run it on their own website, would data entered by users of that site show up automatically on, and vice versa? The answer is yes, because the protocol transports its data on open, public networks, so the same mechanism that allows to read its users’ messages would work for anyone else. Note that no special communications between the new site and would be required; it is just federated by design!

      One more little website is not going to solve the problems with journals. The last thing anybody wants is another password to remember! There are already various sites trying to solve different pieces of the problem, but none of them are really getting traction. One reason is that the different sites can’t or won’t talk to each other—that is, federate. They are walled gardens, closed ecosystems. As a result, progress has been stalled for years.

      And frankly, even if some walled garden did eventually eventually win out, that wouldn’t solve the problem of expensive journals. If one party became able to control the flow of scholarly information, they’d eventually exploit this just as the journals do now.

      So, we need a federated system, to make scholarly communication openly accessible not just for scholars but for everyone—and to keep it that way.

  5. Leonard Says:

    Thanks for your reply, John. I had meant to read your original blog post and I should have before posting my question.
    The federated concept is definitely brilliant. I appreciate the effort to make the SelectedPapers infrastructure platform-independent and complementary, rather than an alternative, to already established services.

  6. E.L. Wisty Says:

    Reblogged this on Pink Iguana.

  7. Daniel G. Says:

    Dr. Gowers, Dr. Baez, do you have any advice for graduate students looking to contribute to the early content of the SPN? More experienced mathematicians commenting on influential papers is a great idea that I think everyone would benefit from. (Dr. Gowers, your most recent post about a paper of Dr. Ruzsa’s was particularly enlightening, especially since I’d read the paper while totally lacking context behind it.) I’d love to contribute to the cause, so I’m wandering in what capacity we (graduate students) could be the most effective.

    • gowers Says:

      I have a few ideas about this, but they should not be thought of as an exhaustive list of possibilities, since there are many many ways that the SPN could in theory be used.

      The thought I have with reference to your situation is that it is quite common for graduate students to make an effort to read and understand papers in more detail than a more experienced and world-weary mathematician would. Beginning mathematicians need to build up their bag of tricks, while older mathematicians are often more in the business of finding things to do with their existing bag. (This is of course a gross generalization, but I think there is something to it.)

      Currently, if a graduate student struggles to understand a paper, precisely one person benefits from that struggle. But all those lemmas that seemed to lack motivation until you thought about them hard, those typos that confused you for two unnecessary hours, those techniques that looked like magic until you realized that they were modifications of known techniques, those results quoted from other papers that you looked up only to discover that they were easy exercises you could have done on your own, etc. etc. — all that is incredibly useful data. If graduate students were in the habit of writing “What I wish I had realized before reading this paper” type posts, they could form a wonderful resource.

  8. another anonymous Says:

    This is totally off-topic, so that I have no right to expect any answer, but do you plan to blog at some point about David Donoho’s work, for which he has been awarded the Shaw Prize (by a committee of which you were a member)? If you have no such plans, then please interpret this comment as a request for such a post.

  9. The selected-papers network | The De Morgan Forum Says:

    […] Tim Gowers and John Baez discuss in their blogs the new system for distributing and reviewing research papers, […]

  10. FPGA Books and Bets and Beliefs « Pink Iguana Says:

    […] mathematics. It seems like Tao, Green, Baez, and Gowers might be headed in this direction (e.g., selected papers network) particularly if they work to replace/improve on the functionality provided Elsevier […]

  11. sashakolpakov Says:

    Dear Dr. Gowers, this is good news that the SPNetwork has started functioning! I would like to share a thought about it: I see that this platform could have become an OpenSource alternative to AMS Reviews or Zentralblatt MATH in the sense that I may post a review on a paper that I’ve read recently, as well as comment on it. I have written a number of reviews for both AMS and Zentralblatt Reviews Databases, but it was always hard to bear the fact that these are behind the paywall. Well, the AMS gives me “points” though, but I would prefer to post and read reviews and comments freely rather than get an elusive reward for writing things that will generally remain concealed. Please share your opinion on whether it’s worth to write review-like essays in addition to comments and discussions. This will definitely help me to arrange any further contributions to the SPNetwork project.

  12. If one were to invent scientific journals today — Quantum Forest Says:

    […] 2013-12-15 Thomas Lumley pointed me to a couple of papers on the Selected Network, which would be one way of dealing with […]

  13. Sylvain Ribault Says:

    To be able to publicly comment on articles would be nice, but it would be even better if one could also create and share improved versions of articles. This is what I speculate on here:

  14. Notes: Algebraic Methods in Incidence Theory | Mental Wilderness Says:

    […] feed that gathers from all over the Web people’s responses on papers. (See Baez and Gowers on SPN.) Unfortunately it doesn’t support wordpress at the moment, so I will be posting the […]

  15. Personal website, and sharing (esp. learning) resources | Math is hard Says:

    […] perhaps sort of the idea behind the selected papers network (see homepage and Baez’s and Gowers’ blog […]

  16. Links for March 2016 - foreXiv Says:

    […] I’m bummed that I only just now discovered the defunct Selected Papers Network from this (still live) blog post from Tim Gowers. […]

  17. Blogs that are Current | Gödel's Lost Letter and P=NP Says:

    […] Blog Phillipe LeFloch’s blog Process Algebra Reasonable Deviations Science Notes by Greg Egan Selected Papers Network Shtetl-Optimized Since it is not… Sketches of topology Snapshots in Mathematics! Stacks […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: