3 ms·
FTA: ...finding it to be 2.79284734462 plus or minus 0.00000000082 nuclear magnetons... Ok, physicists can measure this quantity to 10 decimal places and Compu
by titzer 9y ago
FTA: ...finding it to be 2.79284734462 plus or minus 0.00000000082 nuclear magnetons...
Ok, physicists can measure this quantity to 10 decimal places and Computer "Scientists" cannot understand the systems _we built_ well enough to reliably measure a 1% speedup.
/sigh
- Kenji 9y agoCome on, that's an unfair comparison. 1% speedup of what? Of a program? With what input? On what architecture? On what hardware? Ask a clear, well-defined question and a Computer _Scientist_ will give you an accurate answer.
- miko1293949 9y agoScience in CS is an abused term.
- moron4hire 9y agoNo, computer science is not the same thing as software development. There is a lot of science in computer science, you probably just haven't been exposed to it.
- phkahler 9y agoI think there is a lot of math in computer science and not as much science.
- perl4ever 9y agoMath and logic may be called "formal sciences" as opposed to "natural sciences".
- moron4hire 9y agoBenchmarking programs is runnimg an empirical experiment. There is a lot of science in computer science.
- ISL 9y agoBut you can measure how many bytes are free on a 10 GB disk.
- Filligree 9y agoSo long you don't care about performance. Sure, "free space" sounds like a simple concept, but the simple version isn't terribly useful.
- saagarjha 9y agoThat’s because “speedup” is a hard thing to measure of the variance of what you’re trying to measure is more than the accuracy you’re trying to achieve. The comparison doesn’t make sense here: physicists are trying to measure a constant while execution time depends on the input data, system load, etc.
- umanwizard 9y agoThe systems we built are a lot more complicated than one proton.
- platz 9y agoA proton is just a location in a vast sea of proton-field
- ihaveahadron 9y agoYou mean to tell me, we can make micro processors, but we can't land a man on Mars. I don't buy it.
- analog31 9y agoInterestingly enough, even the physical constants are all over the map in terms of their experimental uncertainty. The gravitational constant is only good to around 5 digits. https://physics.nist.gov/cgi-bin/cuu/Value?bg https://physics.nist.gov/cgi-bin/cuu/Value?bg