Indeed. When in grad school (in Math) I googled my best friend's name (a fellow grad student) and the academia.edu page turned out to be the second hit. Turned out that they had scraped his professional website, as well as pretty much everyone else's, and created a mediocre copy.
There was a tempting link at the bottom: "Are you [X]"? In a frivolous mood, I claimed that indeed I was. I was asked for no verification whatsoever -- but I was asked to set a password. I changed my friend's profile to "I'm pretty much a putz who sits around and plays bridge all day", and it stayed that way near the top of Google's search results for some time.
I might have felt guilty, but honestly I got the overwhelming position that nobody could possibly take them seriously, they were just an SEO farm with no content.
That link is gone now, and the site looks much more professional. I searched for people I know... and people I know are using it! (I confess to being surprised.)
The front page is a bit ambiguous. "Share your papers" solves a problem no academic has. But getting notifications when anyone from a list of people I know shares a paper would be genuinely useful.
I'm not creating an account just yet, and my cynical instinct tells me that in the long run this will prove to be more trouble for academics to use than they find it worth. In any case they face the chicken-and-egg problem. But, this site looks more useful than I had (ten minutes ago) imagined any such site could be.
When we first launched in September 2008, we did try creating a few thousand unclaimed profiles as an experiment, seeing if academics would claim them. Other sites like Spock.com and Spoke.com were also trying the 'claim your profile' approach. It turns out that it's a really bad strategy, and, in particular, you can't build a community like that. We turned the experiment off after about 3 months. It was an interesting lesson in how not to build a social platform. It's relatively difficult to know this a priori. You wouldn't believe how many people since we launched in September 2008 have said 'Hey, you should create profiles of academics at departments, and encourage them to claim them'. We learned by experiment that that's just not a good strategy for growing a community.
Thanks for your thoughts about the our current product. Many academics write in to tell us that they are printing out their Stats Dashboards on Academia.edu and are including them with tenure track portfolios, as they are keen to demonstrate the global impact of their research, and we help them with that. Many academics feel that a tiny number of people read their papers, and they are incredibly happily surprised to see that, when they upload their papers to Academia.edu, their Stats Dashboard reveals that quite a few people read their papers, from several different countries.
You say you were in Math, so you probably used the incredible www.arxiv.org for disseminating your Math papers. Some areas of Math and Physics (e.g. high energy physics) are lucky enough for a lot of their community to be using the Arxiv. For various reasons, the success of the Arxiv hasn't been mirrored in the pre-print repositories of other disciplines. None of the other pre-print repositories (SSRN, Repec) has got much traction - nothing close to the traction of the Arxiv. One way I think about building Academia.edu is to bring some of the magic of the instant distribution you find on the Arxiv to the other 95% of research - biology, medicine, chemistry etc. We think that bringing instant distribution to the 95% of research that Arxiv doesn't have strong penetration in could yield huge societal dividends.
Faster sharing in, say, biology and medicine, could mean that cancer is cured 12 months before it otherwise would have. That would equate to millions of lives being saved. It's widely thought that science right now is too slow, and too closed. We are trying to change that, and accelerate research sharing.
A couple of thoughts on things I would love to see.
First of all, if research papers could be associated with discussions that would be great. The arXiv is great for math, but you can't ask questions other than e-mailing the author. Imagine that someone posted a paper, and you asked some question "Doesn't XYZ also relate to ABC"?, etc., etc. There is some of this on people's blogs I know but it would be great if there was a lot of this in one place.
You might also read http://meta.mathoverflow.net/ -- MathOverflow (mathoverflow.net) is great, but I feel there are a lot of pain points you'll notice there, and there is room to do things which are outside of MO's scope.
Also, check out Tim Gowers's blog -- http://gowers.wordpress.com/ -- even if you don't know abstract math, it's well worth reading. His math posts lead to the kinds of discussion I would love a more systematic venue for that is not run by a single person.
I can more or less speak for my discipline (AMA), but I understand different disciplines have very different cultures! Even CS is quite different from math from what I've heard.
Thanks for these additional thoughts. I totally agree with you that it would be wonderful to have discussions around papers. It is something that we want to do.
I love MathOverflow. It's really a terrific site. Also I'm a huge fan of Tim Gowers. His polymath project was awesome; I hope that's indication of what the future of scientific collaboration looks like. We are definitely going to do what we can to help push things in that direction.
'"Share your papers" solves a problem no academic has. But getting notifications when anyone from a list of people I know shares a paper would be genuinely useful.'
Indeed when I worked for http://philpapers.org this was quoted by users as the main reason to use it.
Google Scholar last week (soft?)-launched personal pages that seem have much more comprehensive coverage of the literature and accurate automatic attribution of authorship.
I'm not sure what to think yet, but this general space (including the citation/paper archiving tools) is getting incredibly crowded. There's not really a problem finding anyone's homepage (which I have much greater faith will be kept free and available). And despite the evil publishers, if you're affiliated with a university, there's no problem accessing papers. The only hard part is deciding what to spend time reading.
Distribution time is still very long in many fields - pretty much all fields except a few sub fields of math and physics that use the arxiv. Most fields see publication cycles of 6-24 months. I think in the future it will be the norm for the distribution time for a paper within a given community to be measured in hours and days, rather than months and years. We are trying to make that happen with Academia.edu.
I definitely agree with you that it's hard to decide what to read. Each month 60,000 papers are published in the biomedical field alone. We think that seeing what your friends are reading, as well as seeing which papers in your field are trending right now, could really help with this.
The Google Scholar profile product is very different from what we are trying to do with Academia.edu. There is no sharing on the Google Scholar profile product, or News Feed where you can see papers from academics you follow.
I see the Google Scholar profile product as a way of Google Scholar enhancing the accuracy of its search index, rather than as a way of building a research sharing platform, with first class research sharing features. Google is getting serious with social, as Google+ shows, but I don't think think the Google Scholar team sees itself as building a social product, a la Google+, for the research community. I could be wrong, but it just doesn't look like that right now.
According to the eligibility requirements [1], yes. However, this was not always the case and domains registered before 2001 were grandfathered in. According to the whois record for academia.edu [2], it was first registered in 1999.
.edu domain names registered prior to 2001 were grandfathered in and not made subject to the requirements of being a higher educational institution. This is the case with www.academia.edu.
I just tried it out. Clean design, looks smart and runs fast enough. It found a bunch of my papers.
One important thing: I can't find a way to give/get formal references to the papers. This means I can't use any of the papers I find here without googling to find the text (or if I'm lucky, BibTex) of its conventional reference. This is such a glaring omission that I wonder if it was intentional. Does anyone know?
Also, scribd rendered PDFs look like hell, and the scribd frame captures scrolling events. If the frame doesn't fit in a browser window (which it doesn't by default) you have to move the pointer in and out of the frame and scroll in both places to read a page of the paper. yuk.
We're shortly going to switch to using Scribd's HTML5 embeds, which will eliminate the double scroll bar that you see when viewing a paper (the iframe scroll bar and the browser scroll bar). We should be switching to that in the next few weeks after we've ironed out a few kinks. Our goal is to make a really beautiful reading experience.
We have a few fields for each paper (title, abstract, other info etc), but I agree that we could have more. As we build out the platform, this is definitely an area that we'll be developing.
Looks pretty smart, I really hope this project succeeds because as an ex-academic I find it almost impossible to get hold of information about research, and particularly papers.
One thing which could be improved though is the lack of canonical departmental names. For example, if you look at the University of Manchester there are entries for 'Classics & Ancient History' and 'Classics and Ancient History', so you have to look at both to see all the members.
Localisation would also be good - in the UK we don't tend to refer to people as 'Faculty' or 'Graduate Student' (lecturers see themselves as being tied to a department or school rather than the mostly administrative faculty).
It didn't take me long to find papers which are under the copyright of fee-charging journals, which should not be shared outside the institution they are licensed to - how do you plan to tackle this?
Indeed. When in grad school (in Math) I googled my best friend's name (a fellow grad student) and the academia.edu page turned out to be the second hit. Turned out that they had scraped his professional website, as well as pretty much everyone else's, and created a mediocre copy.
There was a tempting link at the bottom: "Are you [X]"? In a frivolous mood, I claimed that indeed I was. I was asked for no verification whatsoever -- but I was asked to set a password. I changed my friend's profile to "I'm pretty much a putz who sits around and plays bridge all day", and it stayed that way near the top of Google's search results for some time.
I might have felt guilty, but honestly I got the overwhelming position that nobody could possibly take them seriously, they were just an SEO farm with no content.
That link is gone now, and the site looks much more professional. I searched for people I know... and people I know are using it! (I confess to being surprised.)
The front page is a bit ambiguous. "Share your papers" solves a problem no academic has. But getting notifications when anyone from a list of people I know shares a paper would be genuinely useful.
I'm not creating an account just yet, and my cynical instinct tells me that in the long run this will prove to be more trouble for academics to use than they find it worth. In any case they face the chicken-and-egg problem. But, this site looks more useful than I had (ten minutes ago) imagined any such site could be.