kmicklas

38 ms·

301.
by kmicklas 11y ago
> Google doesn't need to do anything about it They don't, but I'd argue if they want to have any hope of competing with iOS in the high end market they absolutely need to. So many people have super negative impressions of
302.
by kmicklas 11y ago
> I think there is a BIG reason why Haskell has not grown in popularity in language usage and is beat by Scheme and almost beat out by ML. Is it? I would think there is easily WAY more Haskell currently in use in both industry and academ
303.
by kmicklas 11y ago
Google needs to do something about the general load of crap that comes on most Android phones. I just switched from a Nexus 4 to a Galaxy S5, and even considering the massive jump in hardware specs I am consistently less satisfied with my p
304.
by kmicklas 11y ago
If you don't see the point of generics... just go back to using dynamic languages...
305.
by kmicklas 11y ago
The camera issues on the Nexus 4 are ridiculous. Google has done absolutely nothing to address the issue. On the other hand, I'm pretty sure Snapchat is doing really stupid things with the camera API, since the stability issues are not
306.
by kmicklas 11y ago
Linguistics is the only field I can think of where the general public does not take seriously the knowledge of the experts (linguists).
307.
by kmicklas 11y ago
I think it depends on the office. The food in NYC is amazing and many Googlers tell me the MTV food is not nearly as good.
308.
by kmicklas 11y ago
I don't think algebra is super useful in itself for standard PL stuff or for, say, writing verified code in Coq. However (perhaps unfortunately?) it is a common source of examples for texts on type theory, and you'll definitely ne
309.
by kmicklas 11y ago
Why on earth would you ever want to _return_ to big-endian?
310.
by kmicklas 11y ago
> I don't think your examples are good though, Max polling reduces noise. RuLU learn faster than Sigmoid or tanh. That's not theory, that's just observation of the results. Why should we expect it to work that way?
311.
by kmicklas 11y ago
That would be equivalent to 2017 == -President, which is unlikely.
312.
by kmicklas 11y ago
I for one don't use a huge number of tabs because of the poor browser support for it... I suspect a lot more users would use a lot more tabs if it were actually convenient to do so.
313.
by kmicklas 11y ago
Do these guys seriously not realize that "the terrorists" will use end-to-end encryption whether it's legal or not? This literally makes no sense to me unless "the terrorists" is code for "the local weed dealer
314.
by kmicklas 11y ago
Why would those things be genetic? If there were a "serial killer gene" or a "cheater gene", those traits should manifest themselves in all kinds of obvious differences in behavior - which we do not see. It's hard t
315.
by kmicklas 11y ago
It's worth noting that Univalent Foundations (and type theory in general) really has nothing to do with computers. You can do it on paper like traditional set theory based math, and you can formalize set theory on a computer. The reaso
316.
by kmicklas 12y ago
I take issue with the single-number-type thing. This guy is conflating two different reasons/ways to divide up number types - for transparency w.r.t. machine (optimization) and by conceptual types (integer vs natural number vs fraction
317.
by kmicklas 12y ago
I casually started reading this, and then had this feeling that the style and humor was familiar. Then I realized this was by my abstract algebra professor! Rich Schwartz is a brilliant guy, and pretty much the reason I became a math major.
318.
by kmicklas 12y ago
It's very possible that life arose independently multiple times, but one group grew fast enough to drive all the others to extinction. The life forms we observe now almost certainly came from one ancestor due to the numerous molecular
319.
by kmicklas 12y ago
I thought this was more or less solved with the recent developments on the cubical sets models? There is even a toy interpreter with univalence and higher inductives: https://github.com/simhu/cubical
320.
by kmicklas 12y ago
Is there any reason not to just use artificial sweeteners besides unfounded fears in the public? And would all this research money not be better spent on advertising and attempts to influence public perception?
321.
by kmicklas 12y ago
Moore's Law is about transistor density, not computing power, so no it won't. We could probably already build a computer more powerful than the human brain, it would just be huge (millions of cores or something). But that wouldn&#
322.
by kmicklas 12y ago
Regardless of the relative merits of tabs vs spaces, tabs make the most sense _specifically in a Python-like language_.
323.
by kmicklas 12y ago
20 year old college student in the US here. He got a lot of things very right, but also from my experience some things very wrong: Instagram: No one uses it. Maybe 10% of students have one and those that do spend little time on it. Facebook
324.
by kmicklas 12y ago
I think it is anything but clear that Haskell is harder. What may be clear is that functional languages are harder to learn for people who have already learned imperative programming. But then you're forgetting the massive time investm
325.
by kmicklas 12y ago
Yes and some religions object to the death penalty?
326.
by kmicklas 13y ago
> If you're transmitting any data that could be illegal, your safest bet is to use some kind of encryption that would take longer than the statute of limitations for whichever crime you're committing to break (you'd probab
327.
by kmicklas 13y ago
Well at least it's not Cygwin...
328.
by kmicklas 13y ago
Tin foil much? Just wait until someone puts Linux on it (i.e. not Android).
329.
by kmicklas 13y ago
Is that really any different than carrying around a cell phone all day?
330.
by kmicklas 13y ago
The really sad thing about Go is that it actually is mostly better than its (non-obscure) competitors. I am disappointed in Google because they could have done so much better. All the problems in Go are basically "solved" problems

More ›