4 ms·
There should be something like reddit for academic papers. With upvotes and what not. But I guess it takes people longer to read a paper than to read reddit con
by adamnemecek 10y ago
There should be something like reddit for academic papers. With upvotes and what not. But I guess it takes people longer to read a paper than to read reddit content.
- Analemma_ 10y agoIt's a neat idea, but I would want identity verification - only upvotes from people well-versed in the field should "count", precisely so it doesn't become Reddit. Which means you would have a chicken-and-egg problem when the service got started and few experts were on it yet.
- ampvchen 10y agoThat's actually something we are working on. We are working with verified researchers in the field (industry and academic) to help surface good papers and foster an open discussion.
- Al-Khwarizmi 10y agoThere are websites that try to do that (e.g. scirate) but the problem is that people doesn't participate much. The problem is not the website, but convincing academics to comment and vote (academics are typically starved of time, and reading a paper and writing a good coment is not easy...) It would be nice if someone solved the problem and managed to create a working one, though.
- adamnemecek 10y agoCan I have a beta when it's ready? I've been thinking about this for a while and I like to think that I could provide useful input.
- ampvchen 10y agoAwesome, yea. I'll send you an email.
- paulsutter 10y agoWhat about collecting tweets from verified researchers? It could help getting to critical mass. You could even consider papers tweeted by researchers that are followed by your known researchers, and so on, with the right weighting (something resembling pagerank). With the right weighting this could really boost the size and quality of your dataset.
- ampvchen 10y agoThat would be interesting to leverage Twitter. Right now we are working on helping the community surface information and working with verified researchers to build their "curated" lists for different topics.
- austinjp 10y agoAlso very much something is be interested in, if invites are available, thanks.
- sampo 10y agoPerhaps you could get an ordering based only on the upvotes from your friends, and maybe friends of friends with a lesser weight. Maybe also include upvotes from strangers whose past voting pattern is similar to yours. Maybe one can construct some PageRank-esque structure, whose votes should be weighted heavily and whose not, in the light of your own voting history.
- chrisseaton 10y ago> based only on the upvotes from your friends That would create an echo chamber. You need to know about research that challenges your assumptions.
- sitkack 10y agopretty sure this is reddit, just make a subreddit for the topic and then start feeding it.
- dredmorbius 10y agoIt's not. Reddit doesn't allow subreddits to limit who can moderate posts or comments except by taking the subreddit private and limiting the membership. It's actually a bit of a major pain, particularly for smaller public subreddits. Reddit's moderation system in general is just hugely problematic. It kind of works, but it really doesn't, and has received very little love. The first question for any such system should be "what is your goal?". Reddit serves popularity relatively well. Accuracy, relevance, information: rather less so. Some non-brief thoughts on that from a few years back: https://www.reddit.com/r/dredmorbius/comments/28jfk4/content_rating_moderation_and_ranking_systems/ https://www.reddit.com/r/dredmorbius/comments/28jfk4/content...
- sitkack 10y agoGreat writeup. Too much of these web "platforms" and use the word loosely with air-quotes, don't support the level of compositional delegation that could/would enable what you are looking for w/o having to make your own platform. The only thing I can suggestion w/o understanding your needs on more than a superficial level would be to create bots that have admin access, that attach "flair" that denotes rank that the bot uses to move stories around, etc. Network effects and availability still make sites such as reddit very attractive. I have always wanted a multidimensional discussion, so that joke posts and memes automatically diverge from the current hyperplane of the discussion.
- dredmorbius 10y agoThanks, and largely agreed. What Reddit does offer, and woeprks fairly well, is moderation tools and teams sufficient to scale out pretty well. A bigger problem is that conversation sinply doesn't scale well, something old-timers have been realising for a while. I've got a Dave Winer quote somewhere to tthat effect, and was rereading Shiky's "A Group is its Own Worst Enemy" which suggest what I'm increasingly concluding: with the right people, from 2-3 through maybe 50-100 people can actually discuss something. More than that and it's broadcast or a large number of comingled side conversations. I'm coming to appreciate Wordpress and blogging platforms' capabilities, and sheer size. There's a ton of blogged content out there, it's mostly that finding and commenting on it is challenging. Another element that's lacking is filtering tools, for which I think randomness and/or community ought play a larger role -- filtering content up through smaller groups. Also both implicit measures and known trusted quality "roots" (vetters / editors). I'm coming to appreciate Wordp
- trurl42 10y agohttps://scirate.com/ https://scirate.com/ Is something like that for papers on the arXiv.
- wodenokoto 10y agoI think this is what academia.edu is trying to be. It seems to be a mix of reddit and linkedin for academics.