Their definition of "free" is interesting. Only paid subscribers can search for articles, and apparently others can view articles only if subscribers share links with them. For many purposes, articles might be "free" but still hidden from the public.
Nature gets seen side by side with free access journals there, and so now we get to see them for free (regardless of delivery mechanism). It likely doesn't matter what Hacker News thinks about this. Our impression of free, say github style open source free is otherworldly different than Nature's free: paywall, peer reviewed, exclusive, ultra prestige. Tomorrow there will be celebrations in my old lab about how amazing this is.
This has already been done many times with regular pdfs [1-3]. There are mostly hosted illegally in places like Russia. I'm sure you can find more on TPB or similar.
Unfortunately, these don't host many of the articles I would love to have access to. If I didn't know anyone still in uni, I would probably have no way of doing good research. Looks like one of them is proxying through a university that has access though.
The data needed for searching (titles, abstracts, and other metadata), at least for biology-related articles, is already available in an XML dump from NIH. The public interface is called PubMed, so you can already search for Nature articles, just not within full text.
I have actually tried to do what you are suggesting before, so I could bulk-download PDFs (I have an institutional subscription). The problem is that the URLs are different for every journal, and some have Javascript that makes automated downloading hard. Since there are thousands of journals, it's very tedious to try to invent a generalized solution.
Hah, come on. :) That's not a very high barrier. A 'pirate' simply has to get a throwaway account (or steal one, if he's truly bad) and he doesn't even have to worry about that.
I expect all Nature articles made "free" today to be available on TPB in one month. I'll put .1 BTC on it. :)
Not if the URL encodes the originator of the link, and all links tied to that user stop working when Nature flips a switch.
But regardless of what DRM scheme they choose, it's doomed to fail, all it takes is one person to crack the scheme and liberate the underlying documents. Alternatively, since 100 anointed blogs and media outlets will apparently have the right to download raw PDFs, all it takes is for one of their accounts to get hacked.
