Monthly Archives: October 2006


Light and time

There’s lots of stuff in the popular press like “Hubble has let scientists make direct observation of the universe as it was 12 billion years ago.” (CNN). But that sort of construction doesn’t really make sense, does it, in light of relativity theory?

And when sources like CNN are making such scientifically suspect remarks in passing, what has the world come to?

[This has been #33 in the series “Things you didn’t know you don’t care about”.]



Fun site to put in your feed list. It’s “a daily filtration of ideas+aesthetics+amusement” inĀ  “picturebook” form.



withAttach is good

I hope they move a feature like this into Thunderbird’s default install. It would help prevent the prevalent failure-to-attach syndrome.



Bad math again?

I wonder who’s really, really bad at math: me or George Gilder and the Wired editors:

“To handle the current load of 100 million queries a day, its collective input-output bandwidth must be in the neighborhood of 3 petabits per second.”

(From page 2 of a Wired story).

I’m willing to concede that there are things I don’t know about Google and/or Gilder’s calculation (not to mention his penchant for big numbers), but according to my calculation, that’s off by a factor of about 20 million. What a big number _that_ is! I’m impressed! I can’t think of any junk I could add to my calculation to get anywhere near that.

For your reference, since you care enough to read to the end of this article, my calculation was thus: I did a Google query and added the 5kB page to the two 5kB images (which are usually cached for me, but we’ll assume they never are) plus 1kB up for the request, to get 16kB per query.

16kB * 8b/B * 100e6/day / 24hr/day / 3600s/hr = 148Mb/s

3e15b/s / 148Mb/s = 20e6