Hacker News Reader

Comment by grsites | Hacker News Reader

This kind of gouging soon won't be a problem anymore. I'm in the process of creating an online service (launch date sometime this summer) that will provide completely free full-text access to all academic journals in one fell swoop. Think Napster for academic research.

I won't give implementation details for now, and I'm still evaluating different domain names, but it will be called Acropolis.

Yes, I know it will create a legal shitstorm, but I feel it's a small price to pay for what is at stake. To hell with it. Academic research is supposed to be free.

View on HN

The problem that needs to be solved isn't distribution, it's how to transition the traditional model of peer-review into one with much less overhead. Building a web app to allow researchers to submit their articles and have them peer-reviewed by other researchers would be an interesting project that could actually disrupt the industry. I don't think things like arXiv do that yet.

Your project seems disruptive and may cause people to talk about the issue but it doesn't solve anything, as it just undermines the system that is producing the very journals you are exposing. The same way Napster didn't solve the dependency between musicians and outdated industry business models, it just exposed how the models were outdated.

> Building a web app to allow researchers to submit their articles and have them peer-reviewed by other researchers would be an interesting project that could actually disrupt the industry.

True, but the chicken and egg problem prevents that from taking off: researchers submit to the highest impact journal they can get the article in, not the open access journals. Similarly, they won't submit their articles to such a service until it is already well established.

I'm hoping that bringing the problem to light will provide an impetus to evolve to a better system.

In a few fields the bootstrapping has taken place by having the entire editorial board of a top journal leave en masse to start a new one, which in a few cases then quickly takes over. The most successful case I know of is the Machine Learning board leaving to set up Journal of Machine Learning Research, which is now considered top-tier: http://en.wikipedia.org/wiki/Journal_of_Machine_Learning_Res...

We've seen Zotero and Mendeley trying to offer a platform for that (unconvincingly if you ask me); from mass adoption of those, points for good comments can be awarded to incite proper behaviour (reddit, HNews and Quora are here to attest it works, and abuses and their proper correction are well typed); I'm expecting the Livfe extention of Mekentojs' Paper 2 to come closer to something usable.

Hey, I work for Mendeley, so I can comment on a few things. Commenters here have noticed that it's the lamentable reliance on journal prestige that is keeping this whole dysfunctional system held together. The good news is that there are a number of smart people working on alternative metrics for papers (and for datasets and code as well) which focus on reuse, basically an extension of citation. Mendeley helps because we can collect and display some of these stats back to the academics who use our service. Not only does this allow research to proceed much faster by radically shortening the feedback cycle, but it provides a possibility that as the alt metrics get better uptake and respect, the lack of a suitable IF won't be a barrier anymore.

As far as I can tell both Zotero and Mendeley tackle the personal paper management space. Did they also move into being peer-review platforms?

How are you doing this? Citeseer caches the free versions people upload to their personal sites, so they're already making a clearing house for papers that skirts the legality.

I am guessing you're crawling from behind some library account, but I'm not entirely sure how you'd be avoiding detection from the local library (assuming, perhaps wrongly, reasonable comp sec competency).

Maybe it will work like RECAP does for PACER. https://www.recapthelaw.org/ Basically, a bunch of libraries and schools who already have access will grab the articles and send them to the free archive. Of course the legal implications were less intimidating, since the documents (court records) in PACER are in the public domain and they only charge a processing fee.

Yes, that's essentially it...

I'd very much consider uploading a few of mine to that service, it's hard to justify what the IEEE deems as reasonable. Heck I don't even get access to my own papers.

Having published in an IEEE journal myself, this is false.

Unless the IEEE journals differ, you should have full rights to give people a preprint version of your paper, as well as to put a preprint version of your paper on your personal website. Not to mention that you need to ok the final version of the paper, it's hard to see how you don't have access.

I look forward. Is there a twitterfeed / alert mailing list?

No yet, but once it launches, you'll know. :-)

Well, there are private torrent sites with most journals already. Or so I've heard ;)

I haven't found any, but I guess that's the problem: they're private.

Reply on news.ycombinator.com