Dec
16th
Fri
16th
Nov
26th
Sat
26th
A graph that is so bad it should make you cry. Thanks to the US government. http://www.netl.doe.gov/coal/refshelf/ncp.pdf . Useless 3D. Awkward scale reading. Shithouse data-to-ink ratio. Absolute failure at illuminating what the actual problem is (read the blue text under the x-axis to read what they wanted the graph to show).
Nov
23rd
Wed
23rd
…something to take the edge off…
— A router.
Nov
22nd
Tue
22nd
scikits gmm
scikits.learn.mixture.GMM needs the training values to NOT be Int64 values. Otherwise you get ”invalid value encountered in log”
Nov
20th
Sun
20th
From The Economist
“This is ironic, because black men are statistically very open to marrying outside their race—more than one in five does. But fewer than one in ten black women intermarries.”
”More than one in five does” or “More than one in five do” ?
Nov
17th
Thu
17th
Same data. Different view.
A plot for work.
Nov
14th
Mon
14th
remove +1 from google reader to speed it back up
Add the following as a custom rule to adblock:
plusone.google.com/u/0/_/+1/fastbutton
Nov
10th
Thu
10th
Back from a trip to Tokyo. Lusted after all the tokyobikes we saw.
Found a great place to rent one (at bikecity-tripper and cycled the hell out of Tokyo.
