Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
Estragon
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
52 ms
·
511.
▲
by
Estragon
14y ago
> I still wish that the course would try to cater for the stronger > students as well, so that completing it with a high grade would give > a real sense of accomplishment. Maybe the problem is that you were taking th
512.
▲
by
Estragon
14y ago
I wonder whether it would be possible to make a lint-like verifier which rejected code whose local complexity exceeded some objective thresholds. I imagine it would be pretty easy to come up with an objective measure of pure control flow c
513.
▲
by
Estragon
14y ago
Pretty unlikely, at least in the medium term. What's the business strategy which would motivate that?
514.
▲
by
Estragon
14y ago
My taste for design is pretty weak, so I'm always interested in comments like this. The design looks OK to me. What are the problems with it?
515.
▲
by
Estragon
14y ago
This is like asking whether god exists. Until there is some evidence one way or the other, it is pointless to consider the question seriously. Though it's worth noting that the total ignorance implied by the lack of evidence about the pur
516.
▲
by
Estragon
14y ago
> he makes a very compelling case, and even gets into the details He always made a compelling case, I would go so far as to say that he was almost always right. The problem with this style of communication is that it hardens pe
517.
▲
by
Estragon
14y ago
> Yes, it's a rant. Why? Actually, the rant's author just liked/felt compelled to communicate that way, pretty much all the time. It really hurt him socially.
518.
▲
by
Estragon
14y ago
Maybe you can improve the app's design so that those support questions don't come up? It sounds like your customers are giving you a gift.
519.
▲
by
Estragon
14y ago
The embarrassing demo failures in the "Advanced Security Topics" at the front and center of the blog post were painful to watch and unenlightening, and discredit the effort which has presumably gone into curating this video collection. Mig
520.
▲
by
Estragon
14y ago
These techniques are quite sensitive to the distribution of the data under study. For instance, if the numbers being processed by the LogLog counter aren't uniformly distributed, the cardinality estimates you get from it could be arbitrari
521.
▲
by
Estragon
14y ago
That's how I interpreted it. They do attribute it all to "development costs," so it must mean something like that unless they brought in outside help.
522.
▲
by
Estragon
14y ago
Thanks, I understand.
523.
▲
by
Estragon
14y ago
This confuses me. I believe Zynga has earnings, because in the comments of the OP Fred Wilson estimates their P/E at 13x ("Zynga trades at something like 13x pre-tax cash flow when you back out the $1.5bn of cash they have.") And I know P
524.
▲
by
Estragon
14y ago
Is this essentially a VPS with python batteries included? If so, if my app on PythonAnywhere gets slammed by HN how quickly can you respond with extra provisioning? Also, do you have scipy installed, too?
525.
▲
by
Estragon
14y ago
What are the laws and/or precedents which establish the risks you're claiming here? Why does giving a candidate general feedback on their interview presentation expose a company to accusations of discrimination? I have received straight-up
526.
▲
by
Estragon
14y ago
The Normal distribution is part of the parametric derivation of the Black-Scholes pricing model [1]. It really does seem to be part of the model's key assumptions. Can you expand on how the Volatility Smile demonstrates that traders model
527.
▲
by
Estragon
14y ago
I've played around with this a bit. It is a real bondage-and-discipline distro. Seems designed to prevent access any local storage of any sort. Probably a good thing for the DOD, but a bit extreme for my purposes. I don't mind being mad
528.
▲
by
Estragon
14y ago
I'm convinced that businesses could save trillions of dollars by applying existing weak AI to their problems. Why are you convinced of this? Do you have any case studies with obvious broad applications? I have had a website a
529.
▲
by
Estragon
15y ago
That's interesting. What are you expecting to change?
530.
▲
by
Estragon
15y ago
I just do most of my bash work in emacs shell mode.
531.
▲
by
Estragon
15y ago
As far as I can tell, you have provided no contact information at this stage, so the information that you're hiring may not do anyone any good. :-)
532.
▲
by
Estragon
15y ago
Got a cite to back that up? Because Ritchie claims that no serious methodological issues were raised during the review process, other than the ridiculous "You can't get positive results from an ESP test unless you believe in it."
533.
▲
by
Estragon
15y ago
I would add "The Heart of Research is Sick" to that list. The problems it outlines closely match my own experience. http://www.labtimes.org/labtimes/issues/lt2011/lt02/lt_2011_...
534.
▲
by
Estragon
15y ago
There's a movie based on this guy's struggle, but it's pretty terrible.
535.
▲
by
Estragon
15y ago
How much does google know about you because your friends use gmail? It knows quite a lot about me.
536.
▲
by
Estragon
15y ago
Such people are useful if you can stand them. Avoid them while you work out your idea, but let them tear it apart after you've developed it but before you commit serious resources to implementing it. It can only make your plan stronger.
537.
▲
by
Estragon
15y ago
> this is really the kind of content that should be coming > out of Google itself. What does this blog post do which the official docs don't? E.g., https://developers.google.com/chrome-developer-tools/docs/sc...
538.
▲
by
Estragon
15y ago
I don't agree with "mutually beneficial" in either case, and "agreed upon" is murky in the case of HFT and at least somewhat misleading in the case of domain squatting. And "I take money from people to increase available funds for others'
539.
▲
by
Estragon
15y ago
I think it's terrific that you're making choices based on what really interests you, rather than optimizing your income. You're obviously the kind of person pclark was talking about, and that's admirable. I do think, though, that if the in
540.
▲
by
Estragon
15y ago
Or if it inferred the exercises you've been doing from some initial training and your phone's accelerometer data/camera.
More ›