Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
tnecniv
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
69 ms
·
781.
▲
by
tnecniv
9y ago
I feel your pain. You can pretty much blame either MATLAB origins or an relentless pursuit of runtime efficiency for most of these problems
782.
▲
by
tnecniv
9y ago
It's a very odd perception that I think comes from their size. They tend to be smaller, more focused departments. The education quality is still high and everyone's favorite companies recruit at them.
783.
▲
by
tnecniv
9y ago
I have both Axler and Strang. I learned from Axler but occasionally use Strang as a reference. I find that Axler emphasizes the algebra part of linear algebra while Strang emphasizes the linear part. One isn't better than the other but
784.
▲
by
tnecniv
9y ago
The other factor being that colleges want to boast about their 99% four year graduation rate.
785.
▲
by
tnecniv
9y ago
> Which is why contracts should include warnings, as Senator Elizabeth Warren proposed. How would this work? Isn't the contract itself supposed to be the warning? Obviously contracts can contain a lot of legalese and potentially une
786.
▲
by
tnecniv
9y ago
I'm from New Jersey and I'd call it either, which I guess summarizes the predicament of my state.
787.
▲
by
tnecniv
9y ago
Why would I use one instead of the other? I use LaTeX often, but normally let my IDE autodetect and deal with all that stuff on its own.
788.
▲
by
tnecniv
9y ago
I feel the overfull hboxes error is symbolic of everything that is wrong with LaTeX.
789.
▲
by
tnecniv
9y ago
> It would be interesting to build a matrix showing how degrees in each STEM field are tied to jobs in each STEM field. Because I suspect CS funnels much less into the other STEM fields than vice versa, providing a view where a CS degree
790.
▲
by
tnecniv
9y ago
Have you considered doing embedded systems work? You can flex that programming muscle without abandoning all the EE stuff you think is cool.
791.
▲
by
tnecniv
9y ago
I forget where I heard it, but someone once pointed out that AI, before the recent DL boom, refers to things that we don't know how to do. As soon as we know how to solve a problem, it gets a much more apt name.
792.
▲
by
tnecniv
9y ago
I don't know if they still do, but I know the Japanese used Fuzzy Logic for a while to schedule their trains. The distinction between fuzzy logic and the Bayesian interpretation of probability has always been unclear to me.
793.
▲
by
tnecniv
9y ago
> Prior to Deep Learning, Machine Learning community don't really associated themselves with Artificial Intelligence very often. The AI at the time, if I recall correctly, related more to 70s symbolism and heuristic search approach.
794.
▲
by
tnecniv
9y ago
For MS and BS students, IIRC, a lot of countries will subsidize the education of their citizens in the US to some degree.
795.
▲
by
tnecniv
9y ago
Some of the things you mentioned, like SLAM, are not tied to cameras and used with LIDAR as well.
796.
▲
by
tnecniv
9y ago
They bother me because sometimes their articles are very good and others they are...subpar.
797.
▲
by
tnecniv
9y ago
Doesn't matter My ABET accredited ethics course in college took the attitude that ethics (for some definition of ethics) can't be taught in a semester at college so we'll just inform you about consequences of different choice
798.
▲
by
tnecniv
9y ago
I worked at a restaurant for a while and had a similar experience. I had no bad experiences with customers like the horror stories you always hear.
799.
▲
by
tnecniv
9y ago
I think season 5 was a little weak but it's still excellent.
800.
▲
by
tnecniv
9y ago
> With that said, I'm not exactly sure what type of career path somebody studying, say, political science could embark on I'm not so sure it matters. Even in engineering, college courses alone won't provide you with a wort
801.
▲
by
tnecniv
9y ago
> Not just for topics that you're knowledgeable on, but also if you pick up on the patterns that writers use or read tvtropes it also ruins the watch-ability. This is true of story telling in general, though. There's really onl
802.
▲
by
tnecniv
9y ago
I've seen probably a half dozen things JJ has worked on, but none of them end up being any good. He normally starts out with a pretty good premise, but it rapidly becomes apparent that he doesn't know what to do with it. Besides S
803.
▲
by
tnecniv
9y ago
That would make for a great interview "Well, we wanted to write this show that was a satire of the culture in Silicon Valley, but in the process we revolutionized the field of information theory."
804.
▲
by
tnecniv
9y ago
Actually it reminds more more of the Darmok episode of TNG. Same cadence as "Temba, his arms wide."
805.
▲
by
tnecniv
9y ago
> My great aunt, before she passed, had a suitcase in her closet with lots of cash in it to survive the coming collapse of society. This in itself is poor preparation because it assumes the currency will still be worth something should s
806.
▲
by
tnecniv
9y ago
The best math class I ever took was probability, where the professor worked history of the field and famous historical examples into his lectures. It helped that probability has a colorful past, but the right motivation really brings things
807.
▲
by
tnecniv
9y ago
In my experience, this stems from a difference in comfort level between students and professors. Professors often begin class on the first day by inviting people to call them by their first name. Students, however, are normally intimidated
808.
▲
by
tnecniv
9y ago
Depends on the context. In the US, I've had a lot of classes where the professor tells everyone to call them by their first name (most people don't end up doing that). Also if you work more closely with them as a research assistan
809.
▲
by
tnecniv
9y ago
That depends if you know calculus, preferably multivariable.
810.
▲
by
tnecniv
9y ago
This is my Facebook experience. I didn't unfollow everyone like the grandparent, but I have no desire to go to the main sight. I, however, use messenger constantly (via messenger.com) because it's a unified service that all my fri
More ›