> But Google would be able to produce two PDF files that would, as git sees them, appear to be same just as easy as these that were produced.
Right, but they would have to re-do their enormous calculation. ("This attack required over 9,223,372,036,854,775,808 SHA1 computations.")
Google started with a common prefix p (the PDF header), then computed blocks M11, M12, M21 and M22, such that (p || M11 || M21 || S) and (p || M12 || M22 || S) collide for any suffix S. Given p, M11, M12, M21 and M22, anyone can make colliding PDFs that show different contents quickly. But to generate a collision with a different prefix q, e.g. one including the file size, one would have to do the expensive computation all over again, I think.
Note: I'm not trying to argue that SHA-1 can be made secure with padding. I was just trying to say that the statement "The PDFs have the same size" misses the point.
The PDFs have the same size, but they do not have a header in the file that states their overall size. If PDF had a header at the beginning of the file that states the file size, then it could be harder to find a collision. From what I understand, the attack works by inserting garbage data after a fixed file prefix and before a fixed file suffix (anyone please correct me if I'm wrong).
> I thought Hacker News was above Nazi-comparisons
> [makes Nazi-comparison]
Edit: Parent changed
> Downvoted without response? I guess I shouldn't be surprised, groups that resort to tactics like those at Berkeley are hardly above trying to silence discussion here with downvotes.
into
> (But I'd rather we avoid Nazi comparisons altogether. They can be an effective way for an author to get attention but they can be used by both sides and they rarely lead to useful discussion.)
As a keen cyclist, I often find myself thinking "I should lose 2kg." This amount of weight does make a difference for getting up mountains fast. I'm quite thin already, but it would clearly be possible to lose that much fat. But I've also found that losing these last kilos is quite hard.
The choice of desktop environment is quite a personal matter. I would recommend Gnome 3, which is especially polished on Fedora.
The early transition years from Gnome 2 to Gnome 3 were rough, and I switched to KDE for this time. KDE was fine, but Gnome 3 has been polished a lot in the last view years. So I switched back to Gnome 3 and never looked back :).
> And that was out of the question. I too would have left the GDR immediately if that would have happened, and pretty much everybody I know too.
So what? Depopulation happened anyway due to unemployment. I'm not convinced that substantially more people would have left if reunification had been planned for 1995, for example.
I also was at the demonstrations and I also think that there was no realistic alternative to a reunification.
But at the time people were discussing different ways of performing the reunification. The SPD under Lafontaine proposed a slower approach rather than an immediate reunification. He made the point that because of the crumbling infrastructure and desolate state of the economy, a unification wouldn't be easy, resulting in unemployment etc.
These warnings turned out to be true, and people should have known. But people didn't listen and voted for CDU in great numbers, because they promised it all. And I think one reason for this is that the warnings also came from the old media, so many people thought "They've been lying for so long, this must be false."
This all reminds me of the situation in Eastern Germany just before the German unification. In the elections campaigns in Eastern Germany just after the fall of the Iron Curtain, politicians were lobbying for reunification with rather overblown promises for the future of Eastern Germany after the reunification. It should have been clear that these promises weren't realistic, and this was in fact pointed out at the time. But people dismissed these warnings, as they came from the (now free) Eastern German media. It seems that many people now thought that everything the old media said was a lie and dismissed as false everything they were saying. This feels quite similar to the attitudes about the "lying press" that are currently circulating.
Yes, I agree. Also, this text is quite old and isn't all that representative of his blog either. The blog takes controversial positions, but is entertaining at the same time. It's sort of like a yellow-press newspaper for CCC hackers, if that makes sense.
Looks like it's a concurrency bug that is only triggered non-deterministically under certain circumstances. It is probably hard to reproduce reliably in a test, if this is possible at all.
"There exist only two kinds of modern mathematics books: ones which you cannot read beyond the first page and ones which you cannot read beyond the first sentence." -- Chen Ning Yang
It's Meta Language. The ML language was developed in the early 1970s as a meta-language for the LCF theorem prover. ML was developed as a language for programming proof tactics. The strong type system and type soundness guarantees of ML were important to guarantee that such tactics could only prove correct theorems.
A good understanding of randomness has much further implications than just generating random numbers for cryptography. This work is not indended as a practical method for generating random numbers, I think.
Randomness is very important for algorithms. It is currently unknown if being able to make random choices adds expressive power to efficient algorithms (the P=BPP question). Being able to generate good random numbers from a weak source can lead to more efficient determinsitic implementations of randomized algorithms, for example.
At first, I thought the title was a mistake, as the letter speaks of 'neutrons' throughout. But it turns out that the term 'neutrino' came only later:
'Pauli earlier (in 1930) had used the term "neutron" for both the neutral particle that conserved energy in beta decay, and a presumed neutral particle in the nucleus, and initially did not consider these two neutral particles as distinct from each other'
Right, but they would have to re-do their enormous calculation. ("This attack required over 9,223,372,036,854,775,808 SHA1 computations.")
Google started with a common prefix p (the PDF header), then computed blocks M11, M12, M21 and M22, such that (p || M11 || M21 || S) and (p || M12 || M22 || S) collide for any suffix S. Given p, M11, M12, M21 and M22, anyone can make colliding PDFs that show different contents quickly. But to generate a collision with a different prefix q, e.g. one including the file size, one would have to do the expensive computation all over again, I think.
Note: I'm not trying to argue that SHA-1 can be made secure with padding. I was just trying to say that the statement "The PDFs have the same size" misses the point.