## Comment: Stonekeep (Score 0) 382

Stonekeep - Excellent dungeon crawler. The graphics is still the best game graphics I have ever seen, the music and feeling is superb, and the story is great.

This question has been asked, and answerred positively, see
http://stackoverflow.com/quest...

Actually, no. Almost all numbers that are defined, are defined in a manner that describes an algorithm on how to construct them. Hence, they are computable.
You need to brush up your theory on computable real numbers. It is all about the definitions...

Yes, it is hard to solve TSP on a classical computer, but NP-complete problems can be solved efficiently on a quatum computer...

Might it be that all this pseudoscience are things not taught in school?
I mean, there is no decent sex education, so, most people need to discover all the nice imortant bits all by themselves.
Maybe this translates to a broader acceptance of pseudoscience; if schools cannot be trusted to teach a wonderful thing like sex, how can it be trusted to teach anything? Hence, the suspicion about science, and acceptance of pseudoscience?

I use this language (Mathematica) every day. I am a mathematician, but also consider myself a programmer.
The language is quite efficient if you know a little bit of the inner workings. It is super-easy to implement say hash-maps,
just write a["key"]=value, and as pointed out, all functions fit togehter.
I really like it since it is consistent, and the functions you should use for efficiency have nice syntactic-sugar shortcuts. There are also ways to compile segments of code that are critical, and there are (I think) CUDA-bindings (although I have not used it).
Making code run in parallel is also very easy. The language is very good for prototyping algorithms and ideas.
Now, the downside (feature?) with Mathematica is that it mixes the input and output: the format is not really plaintext, so if you accidentally try to output a lot of graphics, you may easily make everything unresponsive, and crash, taking all unsaved data with you. (There is also possible to code in plain-text mode, but this has other issues).
Anyway, the syntax of the language is really nice, and I think language designers really can learn from this.

I like this language since it is consistent,
the code that needs to be run quickly can be coded in C and compiled,
and for really computation-intensive stuff, there is support for GPU programming,
and also availability to run on a cluster.

I went on a combinatorics conference (mathematics) a few years ago about juggling sequences, I think by J. Graham.
It is related to sorting algorithms and various other mathematical objects.
This was also the first time the speaker illustrated his work by actual juggling.

Part of what got me into computers and programming was computer games.
I started with just messing around with config files for games, and proceeded to creating character editor programs for Diablo II.
Gaming is a male-dominated area (see how many female pro-gamers there are in Starcraft, for example, and compareto the number of men that are pro-gamers). Somehow, the gaming scene is very much oriented towards men.
The question is, is there a biological reason why females are not as much into gaming, or is it a society problem?
Are females raised in an environment, where men are considered more childish, and therefore it is considered ok to be a gamer, while females are supposed to be mature?
In my opinion, learning to do programming is about letting your inner child loose, and explore unknown terrirory. Boys are encouraged to play when they are young, while girls are supposed to be well-behaved and well-developed. Certainly, this has some advantages for girls, as it seems like interaction with other people are encouraged more, and women are more common in such occupations, while boys seems to be allowed a wider range of games, where more "childish" aspects are allowed.

A few of my math colleges and I are a bit worried that arXiv, (a huge database where mathematicians put their results before sending them to journals), will be shut down. It is most probable that some material in that database coincide with material published in journals, and most journals have the requirement that you sign over the copyright to them, thus making the arxiv version an infringement.
However, arxiv is the main source for mathematicians to quickly discover results that might be needed,
or to avoid working on a problem which has already been solved.
On a side note, there are a few extreme religious groups that oppose almost all form of science,
so some might get tempted to shut down theoretical physics or other alternatives to "god did it all".

I'd suggest Wolfram Summer school, http://www.wolframscience.com/summerschool/2012/
It is math-oriented programming, in Mathematica. I have not gone there myself, but Mathematica is a quite nice language.
However, Stephen Wolfram is sort of strange, being obsessed by cellular automatas and all that,
but otherwise, my guess is that it is a nice school.

Encryption is essentially a function, f(x), that sends plaintext, x, to encrypted text, f(x).
The public key is the recipy on how to perform f, and the private key is essentially the inverse of f.
Note, the only assumption so far on f, is that it has an inverse, which is hard to find given only f.

Ok, now, a homeomorphic crypto algorithm f has the following additional properties:

f(x+y) = f(x) (+) f(y),

f(x*y) = f(x) (*) f(y),

where (+) and (*) are some mathematical binary operators, that are known. Thus, if I know f(x), f(y), I just perform (+) on these two, and arrive at f(x+y). Notice, I did not need to know x or y, the plaintext data!

The trouble is, finding a crypto with good strength with both (+) and (*) is hard. (Finding only (*) is easy for RSA-like cryptos). Now, for this to be useful, (+) and (*) should not be too slow to perform in practice.

Now, why both + and * ? Well, these operators alone can, when composed together, give any other function:

Wikipedia:

Ok, now, a homeomorphic crypto algorithm f has the following additional properties:

f(x+y) = f(x) (+) f(y),

f(x*y) = f(x) (*) f(y),

where (+) and (*) are some mathematical binary operators, that are known. Thus, if I know f(x), f(y), I just perform (+) on these two, and arrive at f(x+y). Notice, I did not need to know x or y, the plaintext data!

The trouble is, finding a crypto with good strength with both (+) and (*) is hard. (Finding only (*) is easy for RSA-like cryptos). Now, for this to be useful, (+) and (*) should not be too slow to perform in practice.

Now, why both + and * ? Well, these operators alone can, when composed together, give any other function:

Wikipedia:

Using such a scheme, any circuit can be homomorphically evaluated, effectively allowing the construction of programs which may be run on encryptions of their inputs to produce an encryption of their output.

You can create circuits of redstone, and "and", "or" and "not"-gates.
http://www.minecraftwiki.net/wiki/Redstone_circuits
This is what I refer to.

The game was well worth the 10 euro I paid for it.
Found a nice multiplayer server with a great community, lots of people are regulars where I play.
The amazing thing is to see all the new creations since last visit:
Minecraft sort of have the same charm as Lego, I mean there is no ultimate goal in Lego, you just build stuff.
The world of minecraft is complex enough to be Turing complete.

Mathematica supports greek letters as variables, as well as variables with subscript and much more.
However, everything CAN be typed from a keyboard, \[Alpha] yields the greek letter alpha.

OS/2 must die!