May. 28th, 2007

Failure

May. 28th, 2007 02:45 pm
notmatt: (eeyore)
Well, it looks like I lose.

We just got back from Fanime con. It was fun and worth an entry of its own which will probably never be written. But I came home to some terrible news.

Apparently my PhD thesis advisor used my code to run a similar case to a very imporant one in my thesis and got different results. Meaning that some undocumented nontrivial change was made to the code after my original results and before I archived the final version he's using now. And no, we have no fancy versioning system, and I did a rather spotty job of archiving earlier versions on my own. I've given him all the information and files I have on this problem in a futile attempt to resolve, but I'm pretty sure it's hopeless. My thesis is irreproducible and therefore wrong.

The only thing I can do now is start drinking heavily.

Profile

notmatt: (Default)
notmatt

September 2010

S M T W T F S
   1234
567891011
12131415161718
19202122232425
2627282930  

Most Popular Tags

Page Summary

Style Credit

Expand Cut Tags

No cut tags
Page generated Sep. 7th, 2025 11:07 pm
Powered by Dreamwidth Studios