lliiffee

23 ms·

181.
by lliiffee 17y ago
But they will probably also sell a lot more phones! A lot of people might buy android phones instead if they can get free navigation there, but apple insists on $99. (Of which they actually keep, what, $20-30$?)
182.
by lliiffee 17y ago
It seems like, on the one hand you are criticizing universities for focusing too much on job skills, and on the other hand for producing graduates that can't get jobs. Isn't there a contradiction?
183.
by lliiffee 17y ago
Good thing for me I wasn't arguing that the TSA was actually testing the liquids and thus not wasting everyone's time! Edit: on the other hand, I'm not sure it's so infeasible... If 5% of passengers have liquids, and 5% of those get checke
184.
by lliiffee 17y ago
Let's make the math a little more favorable. If the TSA caught 99% of liquids people try to bring through and then randomly tested 5% of them, the terrorist faces a 1% chance of success in each attempt, and a 4.95% chance of arrest. Assum
185.
by lliiffee 17y ago
His asks if the TSA ever tests if confiscated liquids are actually explosives. This is a brilliant question! If there is no cost to trying to sneak in liquid explosives (unlike, say, a gun), terrorists can easily try again and again until
186.
by lliiffee 17y ago
One of the things EoSL emphasizes is that simple methods can often give very good results. If you have been through the first few chapters of the book, you should be able to do some cool stuff with nearest neighbors or linear classifiers.
187.
by lliiffee 17y ago
Also it has hardware pdf support.
188.
by lliiffee 17y ago
Undone. I do that all the time. Probably I downvote accidentally more than intentionally. Whatever the reasoning for now allowing us to change our votes is definitely off in this case...
189.
by lliiffee 17y ago
"All of Statistics" by Larry Wasserman is very good. It is traditional, fairly mathematical statistics, but intended for a general audience. (It is particularly popular with computer science people.)
190.
by lliiffee 17y ago
This guy tested this, and found that chargers measure 0.0 watts. By plugging a bunch in simultaneously, he was able to get a nonzero reading, but still less than 0.5 watts: http://www.inference.phy.cam.ac.uk/sustainable/charger/ (He is n
191.
by lliiffee 17y ago
About Clinton. That's what I thought at the time, too. But it turns out to be absolutely false. http://dir.salon.com/story/opinion/conason/2003/07/30/terror... > "CIA officials went to the White House and said they had 'specific,
192.
by lliiffee 17y ago
Neural networks have been untrendy since the late 90s. This has decreased research since then. The general opinion of practical people now is that nnets with a single hidden layer are one of the major workhorses in the classification/regr
193.
by lliiffee 17y ago
My point is that there IS no serious "pure" AI research these days. Your image of lots of pure AI researchers wasting their time is a fantasy-- those people don't exist. People work on applications.
194.
by lliiffee 17y ago
> From an outside perspective it seems the "God of the gaps" argument - AI is what AI researchers haven't done yet - is used as a smokescreen to cover up that AI research hasn't really done much in the last 30 years. Here is what you do
195.
by lliiffee 17y ago
Isn't it an accepted journalistic standard of some kind to smooth out the many stutters and small grammatical mistakes everyone makes when speaking verbally? Exact, word to word transcriptions like this are actually a classic dirty trick f
196.
by lliiffee 17y ago
The most interesting thing might be the fraction of workers that are male. There is a large jump in female employment around 1960, which isn't surprising. But apparently, there was also a big jump in 1860!?
197.
by lliiffee 17y ago
Yes, and it is a very sarcastic choice. The author implies that Arrington was just a totally arbitrary choice in the previous article, absolutely irrelevant to his point, and this is more than a little disingenuous.
198.
by lliiffee 17y ago
This would seem a less... sarcastic if the author had tried in good faith to find another real example, rather than woundedly choosing something deliberately boring.
199.
by lliiffee 17y ago
Language evolves. We can help it along! (I probably wouldn't do it in a formal document.)
200.
by lliiffee 17y ago
"I saw someone in the distance. He was riding a horse." Wrong?
201.
by lliiffee 17y ago
As I am sure you are aware, "he" is the correct English way to refer to someone of unknown sex. (Though to be honest, I have been trying to switch to using "they")
202.
by lliiffee 17y ago
He misses the simplest, easiest, and most effective action anyone can take: Don't do reviews for locked-down journals.
203.
by lliiffee 17y ago
I just tried it with psyco. I get about 103 seconds with normal python and about 4 seconds with psyco. (Including the imports.)
204.
by lliiffee 17y ago
The site is down now, but from looking at wikipedia, you are probably right. (Sorry, I just glanced at the way they were using the matrix before, and assumed they did n matrix multiplies to calculate fib(n)) I think you are also right ab
205.
by lliiffee 17y ago
That basically uses dynamic programming, which is "cheating". (It is a linear time algorithm instead of exponential!) Different compilers will almost always just give us constant factor improvements. Smarter algorithms will always beat t
206.
by lliiffee 17y ago
Does fib.py use psyco? If not, that does still seem slow. Standard numerical benchmarks indicate that c or fortran or whatever should have a much larger speed advantage relative to regular python than 40/163. E.g. on the n-body benchmark
207.
by lliiffee 17y ago
I imagine the airline anticipated people doing this, and they are more than happy to have the publicity. (My guess is they are flying a bunch of empty seats lately because of the economy, so the marginal cost is very small.)
208.
by lliiffee 17y ago
I think, maybe it was this article, unfortunately now behind a paywall. http://www.economist.com/displayStory.cfm?Story_ID=E1_TTGNNP... I think your experience is unusual in that you would have chosen easier classes left to your own dev
209.
by lliiffee 17y ago
Fortunately, while all the research I've seen has suggested that it is critical to have different "tracks" for different students, it has also suggested that there is no need for the teachers/administrators to decide what track each student
210.
Skepticism about beautiful people having more daughters.(stat.columbia.edu)
43 points by lliiffee 17y ago|9 comments

More ›