For about a year now I’ve been following the math.AG and math.NT postings to the arXiv through Google Reader. The good side of this is that you find out very quickly about papers in areas of special interest to you. The downside, I guess, is that it can be sort of distracting; you sit down to work, you flip through the latest listings, maybe there’s one paper that’s interesting enough to read through the introduction and think about a bit, and before you know it the morning’s gone.
On balance, I like following the arXiv and intend to keep doing it. But I have no sense of whether this is a standard practice, the way it used to be a standard practice (and maybe still is!) to go down to the library and flip through the latest issues of your favorite journals.
So: do you follow the arXiv? If so, what do you get out of it?
Quomodocumque 
I get daily email digests from arXiv with my selected subjects. The mail comes around 10 or 11 each evening. I read it first thing in the morning, the next day.
I typically look the old-fashioned way at the arXiv front web-page every morning, and go through the list of titles of all math papers. I usually look at the abstracts of a few of them, then open the PDFs of the most interesting-looking (sometimes none). But I may leave the window with the PDF open for a few days (or a few weeks sometimes) before having a look beyond the abstract…
I do the same as Ken, so I get a digest of all the math.AC and math.RA every morning. (Mine doesn’t come till 2:30 in the morning, though — wonder how that works?) I subscribed to math.AG too for a while, but the volume was just too much.
And then I do the same as Emmanuel — PDFs stay open in tabs till I print them out, or skim them onscreen, or just accept that I don’t really care enough to do either.
Like Ken and Graham, I read an arXiv digest every morning (GT, DG, and QA in my case). I just skim the abstracts mostly, but sometimes I’ll read the intro of an article or two. I must confess I’m relieved when the mailing is short or there’s nothing of interest in it — one less thing to feel like I’m behind on.
I haven’t gone to look at the new journal shelf in years, and indeed my visits to the library have become a lot less frequent over time, a combination of things like Google Books and the fact that I own most of the key texts in my area.
I get the email digests too. I probably subscribe to a half-dozen groups. It usually feels like a chore to read through them, but I weed out most of the papers by looking at the titles and authors, so really it just takes a minute or two. There are usually a few abstracts I read each day. Maybe I download a few PDFs each week, but I usually just end up looking over them without really reading much. Every once in a while, there’s something I’ll want to have a closer look at. Then I’ll print it out and save it till I need something to read. A few times a year, I’m interested enough to read them closely right then.
I used to feel the pressure to read a lot more. Then I became pretty focused on my own thing and didn’t really care what most people in my field were doing. The funny thing is that then I also became more interested in some low-tech fields which I would have thought were too boring before.
It’s a bit of a walk to my library, so I don’t browse the journals anymore. Keeping up with the Archive is just something I do to have some idea of what’s going on. I wish everyone put their papers on it, though I can understand why some people don’t do it. What I think would be way better is if the MathSciNet sent out emails with abstracts of newly published papers. With some very basic filtering, based on which authors, subjects, key words, etc you like, they could probably keep the emails reasonably small. That would be much more useful than the archive digests. Then you could truly keep an eye on whatever fields you want without much effort. The more I think about it, with the number of papers growing and universities contracting their libraries, this seems like a really good idea.
I get the daily e-mails as well — CO, AG and RT in my case. I used to also read NT, but I realized I didn’t have time for it. I usually take about 30 minutes to read the abstracts each morning, and mark 0-3 papers as things I might want to read. Those abstracts I copy to a file until I either read them or decide I don’t care. This is my main way of learning about new results; I almost never look at journals unless I am looking for a specific article.
I get the daily arXiv mailing — AG, AT, DG, DS, GN, GR, GT, KT, MG, NT, RT, and SG — and I read through it every morning while I drink my coffee. It takes me about 20-30 minutes. I usually download a few papers and print them out. They go in the “to read” pile on my desk, which is disturbingly large.
I should probably try to cut down on the arXiv categories that I have mailed to me, but I’m too lazy to do so.
Unlike most people, I usually try to spend an afternoon in the library once a month or so flipping through recent issues of a number of journals.
Now I feel like my eminent commenters are going to intimidate into silence anyone who doesn’t follow the arXiv…
I used to follow AG and NT (with daily mailings), but I decided that anything in AG that was interesting to me was always cross-listed in NT. I often download and then skim a paper to see what the “key idea” or “key result” is. Unfortunately, many people in arithmetic Langlands don’t post to the ArXiv, and a certain subset don’t even have a webpage. (Indeed, some of them don’t even seem to _write_ papers on results they announced over 5 years ago, but that’s another issue…)
I’m not one of you’re regular readers, so I’m looking at rather different subject areas with presumably different local cultures than most of the other commenters, but I also get abstracts from several areas emailed to me, which I skim first thing each morning. I keep a file on my computer of abstracts of papers which I think I might want to read eventually (generally after glancing at the PDFs); I go through this file every few months at which point I delete many of the old abstracts and then look more carefully at the PDFs of the rest. (The value of all this is partly just to give me some better idea of what people are doing beyond what I think is directly relevant to my current work.) Maybe once every couple of weeks a paper is posted that is important enough to my current work that I read it immediately.
It would probably fit the way I read the arXiv better to stop getting abstracts emailed and start using the RSS feed, but I’ve stuck with email through inertia.
Regarding your remark about intimidation, do you make your graduate students follow the arXiv? Having read this post, I think I might suggest it to my student…
I check the NT listings every night.
I check the AG listings promptly at 8:01pm EST :)
I look over the new postings for NT most evenings. If something seems interesting, I will download the PDF and skim it. The chance that I will print something is very small, but the PDF file may stay open in the background of my machine for a while.
I have various motivations. General curiosity is a big one; restricting just to NT than acts as an enforcement mechanism to prevent me from spending all my time on the arxiv. Also, my impression is that most young people post on the arxiv, so looking at it regularly helps keep me abreast of what is going on; in particular, I learn about papers that are interesting, but not earth-shattering enough that I would learn about them directly from conversations with friends and colleagues. I will sometimes also look over papers in somewhat distant areas from mine (analytic NT, explicit results about Diophantine equations) just to see what kinds of methods people are using.
Also, once a month or so I type in Jacob Lurie’s name into the arxiv search engine, so as to not miss the latest installments of his ongoing saga. (This is one of the few times I venture outside NT; the wider world out there is fascinating, but way too distracting.)
OK, let me come out and say that I don’t look at the arXiv on a regular basis. I just haven’t got into the habit of it. I’m not sure whether that’s a good thing or a bad thing, and can easily imagine that at some point in the future things might change. Having said that, I do have other ways of accidentally wasting mornings.
Sometimes you have to reload a few times if you try to look it over too close to 5:00 PM (Pacific)! Valuable seconds wasted.
The only problem with the arxiv is that one should basically assume the papers are wrong. An amusing exercise is to search for listings containing the word “withdrawn”.
I rarely look at the recent papers, and definitely have not subscribed. I search the arxiv and google scholar (lately more often than math reviews) about topics that I find interesting, usually after a talk, or a conversation with a friend, or a comment on a blog, or a train of fancy. Sometimes I check during the talk.
Looking at abstracts on the arXiv every day does remind me of Atiyah’s famous saying, “Don’t read books, it’ll only make you depressed. If you want to know something, ask me.”
I check the arxiv almost every weeknight, often as soon as I possibly can. For instance, I know that there is sometimes a window of a couple of hours when the “recent” papers are more recently updated than the “new” papers. I have never subscribed. Psychologically, it seems that would change what is often a treat (I feel like websurfing somewhere; ooh, I wonder what just went up on the arxiv?) to a chore (Three more papers I’m supposed to look at? But I haven’t had time to look at yesterday’s stuff!)
On average I download at least one paper a week, but many of these I squirrel away until something or someone else reminds me about them. Occasionally I get downloader’s remorse, realize I’ll never find time to read a certain paper, and then delete it. But overall, being dimly aware of the existence of a large amount of recent work is very helpful for me in my work, in line with Feynman’s famous advice on how to look like a genius (keep hard problems in mind, not just one or two but a dozen or more. Whenever you hear about anything new, try it out against _all_ of your hard problems.) He describes it almost as a joke, but it’s a seriously good idea. (Though of course it’s no get rich scheme: it’s hard work that takes lots of practice and skill to do well.)
I think it’s a bit of a stretch to see any “downside” to following the arxiv: so there’s
a lot of interesting, or possibly interesting, stuff out there, and you might get a little lost in it for a while — that’s the bad news?!? Can we not just agree that the arxiv is a great thing?
get rich QUICK scheme, I meant to say.
Can we not just agree that the arxiv is a great thing?
Certainly we can! But that’s not the same as “reading the new posts on the arXiv every day or every week is a great thing.”