Slashdot Mirror

Here's my list:

http://user.interface.org.nz/~gringer/iopencd/browser/home.html

Which reminds me... I should update that CD and replace OpenOffice with Libreoffice.

Here's a graphic to go with the protests about this:
http://user.interface.org.nz/~gringer/pics/censor_blacker.png

I have plenty more:
http://user.interface.org.nz/~gringer/iopencd/browser/home.html

I'd rather not go through the hassle of Lenovo technical support again.

http://interface.org.nz/ExchangeSaga

Then again, I don't seem to have a good track record with Acer support either (bulk orders of replacement parts taking >1 month, "courier" pick-ups taking >3 weeks).

In the line of Shakespeare, here's a little fabricated sentence I just coughed up, with the help of a script I wrote:

The Repanal's thiout lived in Reluggia, a very reonancid place with plenty of neute senables that wore large undatinks. The tounthude of Reluggia, luckily, spoke Iingic, telling the thiout that jenigircer left her fers in Resepro. It was an inverish statement, but a bit afty and plumboob, so the thiout believed the tounthude.

Schose, the reshil, was rubbling near the paybacinnion, playing with the harnemy and the canths, who looked very sompery.

So I guess it's time for another stupid/useful tricks post, this time for /usr/share/dict. You could use it to make a list of english-seeming words (that get people really confused when piped through 'festival --tts'), similar to this one:

http://user.interface.org.nz/~gringer/montyword.php

[okay, that's derived from IRC logs... I should probably change that to /usr/share/dict]

A book? Sorry, I'm not well read enough in that area (hopefully, I'll be a bit better read over the next six months...). That six generation stuff is some chromosome crossover/simulation tests I did (in Excel, of all languages!) at the start of my PhD research. I assumed 2-3 points of recombination per chromosome per generation, and picked points randomly across chromosomes. This is a picture I often show to demonstrate how much information can be transferred in a single chromosome — that's 4 generations and there's no "blue" information left in that chromosome, but my rough eyeball of a few crossovers suggests 5-6 is about the best you could do unless you're really lucky. I'm currently in the process of trying to work out how close I can get to that 5-6 generations of information in real life. One of the more tricky things I've been trying to work out is which pieces of DNA are most useful for doing this.

It is somewhat possible to derive ancestry trees from DNA by comparing individuals within a group, based on the number of differences between individuals. This is becoming more common in species determination — researchers wanting to know if the new thing they have found is a new species, or a variant of something discovered previously. I prefer seeing DNA as being a tool for genealogy, rather than a final solution, mostly due to the random nature of mutations and recombination.

Watch this space, I guess. I may have a better answer for you (hopefully in the form of a PhD thesis) in just over a year.

Ha! If you'd come to VUW (the university on a hillside overlooking the wellington CBD) earlier this year, you'd have been able to hear Milton Ngan from Weta talking about it.

Yet another excuse to plug Interface, the VUW Computer Club, which ran that event (or more accurately, took it over from the CS department): if you're a CS student at VUW, you really should become a member.

The talk was co-run by Interface (the VUW Computer Club and ACM Student Chapter) and the School of Mathematical and Computing Sciences.

If you're a VUW student and want us (Interface) to do more stuff like this, you should join us .

The talk was co-run by Interface (the VUW Computer Club and ACM Student Chapter) and the School of Mathematical and Computing Sciences.

If you're a VUW student and want us (Interface) to do more stuff like this, you should join us .