Don't I wish.

A cynic might argue that the key difference in this case was that, for a change, the ISP's, and not merely defendants, were challenging the subpoenas; but of course we all know that justice is 'blind'.

An ingrate might bemoan the Court's failure to address the key underlying fallacy in the "John Doe" cases, that because someone pays the bill for an internet account that automatically makes them a copyright infringer; but who's complaining over that slight omission?

A malcontent like myself might be a little unhappy that it took the courts ten (10) years to finally come to grips with the personal jurisdiction issue, which would have been obvious to 9 out of 10 second year law students from the get go, and I personally have been pointing it out and writing about it since 2005; but at least they finally did get there.

And a philosopher might wonder how much suffering might have been spared had the courts followed the law back in 2004 when the John Doe madness started; but of course I'm a lawyer, not a philosopher. :)

Bottom line, though: this is a good thing, a very good thing. Ten (10) years late in coming, but good nonetheless. - R.B. )

Ctrl-S stops text from scrolling on my terminal. I don't see how auto save helps with that.

Yes, these elliptic curves address defined over a finite field, but there's no known connection between the discrete log problem for the field and for the elliptic curve.

The fields of rational, real, and complex numbers are certainly not finite.

Squaring key lengths would be entirely impractical. That said, the improvements only apply to a case of discrete log which isn't actually in use. Cryptographic algorithms generally depend on hardness of discrete log mod p (p a large prime), not in the field with p^k (p fixed, k large).

Yes; the preprint was posted to arXiv when the research was completed. Obviously Science Magazine (the source for the slashdot posting) prefers to write about results when the journal article comes out later, because otherwise the magazine would to check the preprints for correctness on its own, which it can't be expected to do.

Reading the paper, the most notable feature is that their algorithm is efficiency for constant characteristic, including the common case of fields of characteristic 2. It's also okay for the characteristic to grow somewhat with the size of the field, but not very fast.

This is not at all relevant to most implementations of DH, which use prime fields of large characteristic. For example, DSA depends on discrete log modulu a large prime p. In particular, I wouldn't worry about forward secrecy of current internet traffic.

TFA links to the published version on SpringerLink, which is paywalled. A free preprint is available on the arXiv.

If DNA sequencing means taxonomy is now straightforward, then it's good students are switching to other fields. The goal of science is to solve problems, not to ossify. In this case, while taxonomy may cease to be a significant research field, morphology (understanding the structure and evolution of plants and animals) is surely going to continue. The people doing it will simply not be called "taxonomists" anymore.

During the 80s and 90s there were different projects trying to determine the cosmological parameters (mass density, curvature, cosmological constant, Hubble constant, etc). Then WMAP was launched in 2001, and by 2006 (release of 3-year data) the previous techniques were obsolete. Do you think many students in 2001 started working on the old techniques? Should they have? But we haven't lost interest in the cosmological parameters.

Many publishers will try to bribe professors to use their book for a course. Either you're very honest & kind or your class is small.

Do you have first-hand experience with this? It has never happened to me. Publishers routinely send me free books with the hope I'll use them for a course. Almost all have a policy of giving you a free copy of any book you make mandatory for a large enough class (say 100 students) -- which is the closest it gets to a "bribe" -- but in fact it's basically irrelevant to the decision. First of all, the only reason I need the book is for teaching purposes, I'm not particularily motivated to own it except for use during that particular class. Second, since the book is for teaching, if I don't have a copy the academic department (my employer) will buy one for me to use during the course. So the only thing this "desk copy" policy do is save some money for my boss; it has no effect on how I choose a textbook.

I'm a professor, actually. In two words, you're wrong . If the book is only used in the class of the professor who taught it, the book will go out of print in a jiffy, and in any case the total harm to a single class of students is negligible. For a book to actually stay in print, many professors in many universities must use it. In this case very few will, and the problem will solve itself.

But they don't have to sell the books. They can just lend them for a fee, and that would be perfectly legal.

As a university faculty member I consider the cost of textbooks whenever I choose one for a course. I try to never require students to buy the book (I'm not always in charge of the course I teach, so I can't always do this), and I prefer books that are available on SpringerLink (whole-book DRM-free PDFs are available to all our students since our university subscribes). I doubt many faculty members will actually assign this textbook.