grayclhn

95 ms·

421.
by grayclhn 12y ago
I'll do it right now. These aren't issues I've personally run across, though, just stuff on julia-users. edit: Nevermind! On rereading the mailing list thread (especially this message [1]), I might have mischaracterized the i
422.
by grayclhn 12y ago
current HN discussion: https://news.ycombinator.com/item?id=8576411
423.
by grayclhn 12y ago
Octave occasionally runs faster than Julia for transliterated vectorized vector and matrix operations. Julia then runs much faster when those operations are devectorized or replaced with builtin functions. But that's a pretty common so
424.
by grayclhn 12y ago
"Idiomatic" might have been the wrong way for me to put it, but it seems like one of the biggest advantages and fundamental design decisions of Julia is that it makes numerical computing through side effects _very easy_. Most perf
425.
by grayclhn 12y ago
Did you read the performance tips?[1] The mailing list is pretty good at helping with problems like that. Using global variables is a pretty common performance killer. [1] http://julia.readthedocs.org/en/latest/man
426.
by grayclhn 12y ago
The text is pretty hard to take seriously, e.g.: "Market Qawba: Premium themes to make your things elegant" Surely there's a more specific word than "things"
427.
by grayclhn 12y ago
That seems like a pretty decent description of python, right? :)
428.
by grayclhn 12y ago
I've seen julia described as "functional" a lot, but is that accurate? There are first-class functions for sure, but the "idiomatic julia" approach to a lot of problems is to allocate a large block of memory, pass a
429.
by grayclhn 12y ago
R is actually homoiconic. It's somewhat awkward and not a language feature that many packages exploit, but it's there. I wouldn't describe R as having Lisp-like macros, though... maybe "macro-like functions." edit:
430.
by grayclhn 12y ago
You can "print" the text file to a pdf for your CV and get the best of both worlds. My department only allows pdf uploads for CVs, so that's my approach.
431.
by grayclhn 12y ago
Trying to use R to do statistics -- graphs, linear regression, t-tests, etc. -- is the best way to get your feet wet. It's ironically not great for data management and relations. (Although there are very good packages for that sort of
432.
by grayclhn 12y ago
Emacs orgmode is an exceptional LaTeX editor with a markdown feel. Citations, numbered/aligned equations, tables, footnotes, etc.
433.
Harvard Engineering Recommends Open-Access Deposit for Faculty Review Process(officeforscholarlycommunicationharvardlibrary.createsend.com)
1 points by grayclhn 12y ago|0 comments
434.
by grayclhn 12y ago
The value can be small (or nonexistent), but I like to know that everything in my inbox needs to be acted on/processed, rather than having to determine whether I've already done it or not. You can do this while leaving everything
435.
by grayclhn 12y ago
Yes, but bread-machine bread is f-ing horrible compared to the real stuff. I assume that picobrew at least produces good beer.
436.
by grayclhn 12y ago
For some (usually specialized) languages, emacs is easily the best IDE/editor, so if you want to use one of them, give it a try. Its key combinations are less ergonomic than vim's, so you should definitely spend some time configur
437.
by grayclhn 12y ago
There are namespaces, but you have to put your code in a package to use them.
438.
by grayclhn 12y ago
Thanks!
439.
by grayclhn 12y ago
Can someone please provide some context to the submission? Is this something new? edit: politeness.
440.
by grayclhn 12y ago
Just a tip: you don't want to be one of the people building a pyramid. You may want to think of a non-enslavement-and-death example when you explain why somewhere might be a good place to work. I can just imagine the recruiter's p
441.
by grayclhn 12y ago
Update: Ah, apologies for assuming it's "his" post. The author is clearly a woman and I'm a moron. Sorry. Original: I think it's a pretty reasonable assumption that someone who's trying to run a web services co
442.
by grayclhn 12y ago
Not necessarily recommended, but wrapping smoke alarms in plastic bags usually prevents them from going off. Of course...
443.
by grayclhn 12y ago
I would say sometimes you can control for this statistically. When the selection problem is exactly along the dimension you want to measure, (i.e., measure the impulse-control of children of parents with poor impulse-control, but only the
444.
by grayclhn 12y ago
Or scalded and illiterate. It's a tossup. :)
445.
by grayclhn 12y ago
No, even that wouldn't work. "Angry parenting" is going to be massively correlated with many other factors that will affect how parents raise children and with the children's impulse control in general. That's likel
446.
by grayclhn 12y ago
Um... better than backpacks, I'd say.
447.
by grayclhn 12y ago
It's treating "student" as the individual's current occupation, which seems reasonable. But if you are conducting your own unaffiliated independent study that eats up most of your workday, you can probably get in touch w
448.
by grayclhn 12y ago
> Anecdotally speaking, such people know more than a beginner, but are very, very far from being an expert themselves. I've met some assholes who are experts. It's just that real experts are rare, asshole or not.
449.
by grayclhn 12y ago
No, you would need a larger monetary base to support the same size economy. But printing more money is easy.
450.
by grayclhn 12y ago
Ah, you must have never had to deal with administration. Or so I've heard. All of the administration at my university is of course irreproachable. (Hi boss!) But the stories my friends tell me... ;)

More ›