My blog has moved! Redirecting...

You should be automatically redirected. If not, visit http://ripper234.com and update your bookmarks.

Showing posts with label Computer Science. Show all posts
Showing posts with label Computer Science. Show all posts

11 August 2008

Thesis Complete!

Finally, after years of research, months of writing and correcting, and weeks of bureaucracy, I finished my M.Sc thesis!

19 March 2008

Robot Dog

A bit scary, but very cool.

17 March 2008

The On-Line Encyclopedia of Integer Sequences

I was trying to answer this question - how many labeled rooted trees with n nodes are there?

A quick search didn't find the answer (I'm sure a more detailed search would have). Then I had this idea - find the answer for small values of n, and look in the OEIS. I typed 1,2,9,64 in the search and quickly found the answer (which is n^(n-1) for those interested). I thought about it for a couple of minutes but still hadn't come up with an answer as to why this is true.

13 March 2008

Google Publication Fumble

Update - it appears the link I gave is simply a table of contents of an ACM symposium. So where is the actual paper?

---

One of the feeds I recently subscribed to is Papers Written by Googlers, (the web version is here). Apparently every link on the page is to some Google search instead of a definite link to a paper. I wanted to check out a paper titled Towards Temporal Web Search, by Marius Pasca and got a strange result page containing totally unrelated papers. Only the focused search I ran myself for the exact title gave me the actual article.

05 February 2008

2008 Turing Awards

Just thought to tell everyone that the 2008 Turing Award will be given to two guys that invented Model Checking, that has a very loose connection to my own master thesis. How loose? Both relate to Temporal Logic, that's it.

21 November 2007

Using Google to reverse MD5 and how I almost revealed my password to the world

In this article Steven explains how he used Google to find the password for a given MD5 hash for a user that hacked into his site.

In one of the comments a reader points to this website that offers a direct database of md5 hashes. You enter a string and get its MD5, you enter an MD5 and (if it's known) you get the original string.

The database only works on known (text, MD5) pairs. If I ask for the text of an MD5 the db hasn't seen before, it won't give an answer.

I use a single password to all my internet activities, because I'm lazy. So I almost went ahead and entered that password into the md5 database in order to check if the md5 is known. Then I realized how stupid this would be - it would actually add the information to the db, and actually reveal to the world my password.

Instead I privately checked what my MD5 is (using this C# code), then entered the MD5 into the DB to check if it knows the original password.

The result? No it doesn't :)

13 November 2007

Some xkcd

http://xkcd.com/341/
http://xkcd.com/342/

What's amusing, beyond the Kill Bill 2 analogy, is that "hacking" and "algorithm complexity" are almost orthogonal concepts. An 1337 hacker doesn't necessarily knows his algorithms, and of course algorithm designer (אלגוריתמיקאי?) is usually a lousy hacker if anything.

IMHO.

02 September 2007

Good Programmers

See a discussion on the selection of programming languages on Eli's Blog, and my note about general qualities of good programmers.

26 August 2007

Contet-Aware Image Resizing

A new way to resize images based on the content, while maintaining the way "important areas" appear.

Watch the YouTube demonstration.

10 August 2007

Cool Computer Science Stuff!

First, the simpler one. A really cool algorithm that finds holes of any shape in a given image, and "patches" them from an exiting bank of other images. This might be similar technology to Photosynth.

Second, a historical moment for Computer Science. It appears the first NP-Complete problem has been solved in polynomial time. They use some sort of "optical solution" and not a Turing machine, and the number of photons used is proportional to N^N. I don't know if this will have deep theoretical implecations (haven't read the article yet), but it's interesting (to C.S geeks).

20 July 2007

Checkers Solved

I always said the game is not interesting, compared to chess and Go.

This Slashdot post confirms it - someone solved Checkers (שש בש), by creating a big ass database containing all possible positions...