 So good afternoon. Thanks for the invitation to come and speak about what is going on in physics and indeed what has been going on in physics for a long time when it comes to publication. I was asked to talk about something called the archive, the physics archive. And well, I call it the physics archive, but actually it's widely used now by many fields. So this is ARXIV archive. And this is something that was invented for physics and which I and my colleagues use absolutely every day. And it is a totally open access solution to disseminating research in physics. So the history is that especially the high energy physics community in the 1980s already had a very extensive system of mailing preprints for papers around to libraries around the world where researchers would routinely flip through preprints in hard copy. And by the early 90s, the internet was good enough and fast enough that person by the name of Paul Ginsburg at Los Alamos in the United States came up with this way where people could put their preprints on a server and then other people around the world could look at that server and download papers off of it. This was before web browsers were invented. So this thing started in 1991. And it was an immediate hit with the high energy physics community where the amount of snail mail that was traveling around the world immediately plummeted to essentially zero and people started every day looking at the archive to see what was new since the day before in high energy physics. So in those days, it was based at Los Alamos. It was called xxx.lanl.gov. And because it was before web browsing, you had to log into it using some FTP server. And it was tremendously complicated to post papers. And it was moderately complicated to download papers back into your own machine. As was mentioned, it was kind of slow. But the advantage was just so tremendous that people adopted it very with their whole heart. So I remember my first experience loading a paper onto there, it was tremendously complicated. It took me about half a day to go through all the obscure steps with FTP to make that happen. But there was a real feeling of pride. I said, well, now I'm a member of this worldwide community, just my stuff is out there and they can pick it up in Brazil if they want. It's great. So the reason I say that is because it was really just absolutely fantastic for the worldwide community. So it's totally open access, totally free. And the costs were minimal because Ginsparg and just a handful of people at Los Alamos were running the whole thing single-handedly or just a few hands. So the thing then expanded as the years went by so that today the archive is used by pretty much every branch of physics, all of astronomy, computer science, quantitative biology, mathematics, statistics, and quantitative finance. So when you go to the archive now, you can download the latest papers from all of those fields through that interface. In 1999, the small team at Los Alamos got tired of doing all the legwork themselves. And so they made an agreement with Cornell University so that Cornell started, well, they took it over completely. And so then it became archive.org. And Ginsparg moved to Cornell and continued overseeing it. And it's been there ever since, and it's been growing. And so the costs for running it have been carried up until fairly recently entirely by Cornell. And the system, it costs Cornell supposedly around $400,000 a year to give this worldwide access to all these papers. And they have started making inquiries to libraries if they would please, please, please, please donate something to the cause. So I don't know where that stands in detail. But I know Cornell is trying to spread those costs around a little bit. But for researchers, it's still totally free. Not only to use it, but also to post on it. So I've posted, I don't know, a couple dozen papers in the last 20 years on that system. And it never costs me anything. So let's see. So this thing is not peer reviewed. And so there's the issue of how do you keep the crackpots out? How do you keep out the ocean of lunatics out there who would post any rubbish if they would think that some real physicists would read it? And the way that works in practice is first it's still, even today, a little bit complicated to post things. So the average 18-year-old kid can't figure it out too easily. So that's sort of one level of keeping it serious. Each of the subfields represented within the archive has what they call moderators. So moderators are anonymous people. I don't know a single one. But these moderators do have the power to reclassify papers if they're not put into the proper subcategory. And I think they also have the power just to delete papers if they're totally inappropriate. Don't quote me on that though. I'm not totally sure of that. So that's sort of the second level to keep it professional. And then the third thing is a few years ago they started a system of what they call endorsers. So if somebody comes along and tries to post a paper, then if it's a post from someone who's never posted a paper before, then the paper has to be endorsed by some other person who is endorsed by the archive system already. So that's less onerous than you think. So if one of my graduate students wants to post a paper because he's from Carnegie Mellon and he's from a school that's got a certain level of reputation in the world, my graduate student can get in and post without any hassles from the endorsers. But if it's something from left field, they will ask somebody who knows that field, is this a serious paper? It's not peer review, but they just look, is this not totally nuts? And so that's what they, as I say, introduced a few years ago. And I've never actually, I think I've had to endorse one paper in the last six years or so since that's been going. So that's a fairly low hurdle to get posted as well. OK, so the next thing that, at least in physics, my colleagues and I, we all set up a thing where the archive will feed you every morning all the new papers in the field that you're interested in. And so you don't have to even go there. It sort of arrives in your email box, which is terribly nice because then if you see a title, what comes is usually the title and the abstract. And if you see one that you like, you can just click on it and it pops up. And so that's really great in terms of the rapid access. But as I say, it's not peer reviewed. So the way it's treated nowadays is that people will post to the archive and then they will take the same paper and submit it to one of the professional journals. So in my field, that's, in first order, the physical review in this country or the European physics journal in Europe or various Asian journals as well. And so the routine is you post to the archive. You also post to the, so what we call the archival journal. It's not the archive, it's the archival journal like physical review where it then goes through the standard time-tested peer review process. And like I think somebody mentioned, that process still today can take for physical review anywhere from three to six months. So on the issue of getting credit for doing something new, I think Tar was talking about that, about the idea that if you might lose your Nobel Prize because you were submitting to some tardy journal, nowadays submitting to the archive, that's your time stamp for your work. So you get credit for something new even if it isn't peer review yet, reviewed yet if you have an archive entry. So that sort of solves that problem. I mean, if he were still here, I would explain that. So in physics, we don't suffer from priority fights anymore in the sense that if you've got something ready to go, the whole world will know about it the next morning. Okay, so but having said that, it's still viewed to be very important to have peer review. I've been an associate editor for physical review letters and I know that the people who are professional editors over the years, they've worried a lot about, you know, what's the relationship between us, the stodgy old formally print journals with this new quick open system for publishing things. And so their competitive advantage in their opinion is still the thorough peer review that they've always done. And by and large, I think at least the physics community still feels that peer review is worth the effort because that way what ends up finally in physical review are the papers that some reviewers at least agreed are worthy of publication. So in a sense, the traditional publishers like European Physics Journal and Physical Review, they've capitulated to the reality that today the open access archive has essentially taken away from them the newness of anything that they might print. But they have come to accept that to the point where nowadays if I submit a paper to the archive and then I go to the physical review website, they will take the archive files. I don't even have to separately submit them to physical review. They'll just suck the files from the archive. And so that makes it totally painless for me to submit there. So they're trying to work with this. I know the editors are worried about their long-term future. And it is an evolving thing. And like somebody already said, probably 10 years from now we'll have a different discussion. Last comments about the archive is that they do not do any metrics or citation counts and things like that. They don't do that on the archive. They just post the papers. And so a number of alternative sites or systems have popped up that then will scour the archive and count citations and the like. And so the ones I'll mention there is something called Inspire at Slack in California. And Google Scholar, I find, is pretty good too for doing that. So the archive continues to grow like crazy. There are now about 7,000 new submissions per month. And that's grown pretty much monotonically since 1991. And on the last note, they continue to grow. And they are having growing pains. And they're now trying to figure out how to make it feasible at Cornell to keep this going. But certainly in physics, we expect it will continue to be our main source of new information on a daily basis for the foreseeable future.