RegicideX wrote:
As far as "have no idea who they are".
This is ridiculous.
An argument should stand on its own regardless of who says it. Otherwise experts get away with bullshit and good points made by non-experts are ignored.
And your point would be? I happen to agree with you. Concise, accurately stated arguments are absolutely welcome. George seems to be making some sort of "thinly veiled threat" concerning "we don't know who we are dealing with." That is (to me) irrelevant. And that is what I was responding to.
As far as "identity" goes, I'm hardly a fan of anonymous posters, but that is a different subject. If someone is not willing to stand publicly behind what they say, then there is a question about why.
Furthermore, the discussion so far has not been very technical -- some general knowledge of compilers and general knowledge of probability is all that was required.
Again, I agree completely. Although some here somehow do not follow the discussion with any level of comprehension. This is not a complex process to go from machine to C. And it doesn't require someone with 40 years of experience, as the "Strelka" episode _clearly_ shows. This is not ground-breaking research, it is just hard work.
As far as who I am -- Robert Hyatt knows more than me about computer science and I probably know more than him about probability and statistics. But just as my knowledge of stochastic calculus was not relevant here, deep technical expertise about the internal optimization techniques of particular compilers was not relevant so far. General knowledge of how compilers work and general knowledge of probability was sufficient.
We could argue that point. If _you_ understand the issues, and you understand the direct mapping from C to asm and asm back to C, then you have the background needed. But some do not, and for them, the appropriate technical background _is_ needed. I would expect a graduate from most any Computer Science program to be able to do exactly what others have done in this process. I would not expect non-CS students to have any hope unless they picked up the requisite experience in some other way I know a few engineers that turned into pseudo-computer-scientists by following computers more closely than transits and calculators.
I heard about this debate on the Rybka forum. As a chess enthusiast, I visited that forum from time to time, but I'm not a "regular" there.
But all this is irrelevant, and with all due respect for the original poster, the entire thread is pretty silly.
I would agree. This investigation may turn up nothing more than a coincidental match between two programs. That's why time is needed. It is somewhat akin to actually building a "ship in a bottle". It is a slow and tedious process, but it is not that hard to do in terms of effort/skill.