troels

30 ms·

271.
by troels 11y ago
> Here’s the thought experiment: (...) imagine I showed you an Apple II running World of Warcraft. If you knew anything about computers you’d say, “Impossible. Impossible!” (...) He should have look at some of the stuff that demo scene p
272.
by troels 11y ago
What kind of log data do you mean exactly? E.g. what's the granularity? We have web server logs going 30 days back, on disk, managed by logrotate. Then we have error logging in Sentry. For user level events, we track in Analytics, but
273.
by troels 11y ago
In my experience, CSV is very fuzzy. Do you include headers? What character is used for separator? Comma? Semi colon? Tab? Are fields enclosed in quotes? How are quotes within quotes escaped? What encoding do you use for non-ascii symbols?
274.
by troels 11y ago
With a small enough volume, you could do manual inspection. I'm guessing most fake accounts are easy to spot by an actual human, putting an effort into it.
275.
by troels 11y ago
There's a huge mobile phone recycling industry with its center of operations in asia (Hong Kong). Old phones are traded in very large bulks and a lot of these will have their sim cards still in. I believe the buyers take the phones apa
276.
by troels 11y ago
Considering how vaguely defined CSV is as a format, the challenge might have specified that a bit better. Fun nonetheless.
277.
by troels 11y ago
I used Ubuntu on the desktop/laptop for a couple of years, before switching to mac. I still use Ubuntu extensively on server side environments.
278.
by troels 11y ago
Quite the opposite. A punch on the chin will knock you out easily.
279.
by troels 11y ago
Sure, but there is overhead of communication with a large group. One smart person can be extremely efficient, but 10 smart people are probably not 10 times as efficient. And it gets worse with scale.
280.
by troels 11y ago
One thing I don't see discussed much in these superintelligence/singularity scenarios, is the possibility that there is an upper limit to intelligence. Maybe it simply isn't possible to be much smarter than what we already ar
281.
by troels 12y ago
Zero is an arbitrary limit - It may have a psychological signal, but there is no reason it should be more shocking that the rates drop under zero than when they dropped under 0.1% or any other number.
282.
by troels 12y ago
Maybe I just don't get the terminology right, but isn't this exactly a statistical method?
283.
by troels 12y ago
Depends on the jurisdiction of course, but generally salaries would be first in line.
284.
by troels 12y ago
Also, removing the "l" from cluck would leave you with a word that still sounds obscene, making the whole process pointless.
285.
by troels 12y ago
I wonder what a "super clean commit" looks like anyway?
286.
by troels 12y ago
Care to share your thoughts on why?
287.
by troels 12y ago
It looks more and more useful, but I wonder why fnord was ported from ruby to c++?
288.
by troels 12y ago
In the Australia comparison, the CDN version spends almost 1.5 second on DNS lookup for the main domain. In the non-CDN version this figure is much lower. It seems disingenuous to attribute this overhead to the CDN, no? And as an aside - Ma
289.
by troels 12y ago
Especially on tablet/mobile browsers, a very small number of pages cache before the browser has to reload and regenerate (and that's REALLY slow). Haven't heard this before, but it sounds reasonable. Do you have any
290.
by troels 12y ago
A pity they can't use dates before epoch.
291.
by troels 12y ago
The avatar is brilliant
292.
by troels 12y ago
I think it's generally safe to assume that a large chunk of people who happen to hold opinions you disagree with don't hold them because they are retarded. Even Fox News viewers.
293.
by troels 12y ago
Except of course if there is no oss alternative and you can't conceivably build it on your own. But yeah - Pretty much agree. The fact that most oss products are best-of-class makes the choice even easier.
294.
by troels 12y ago
podio.com sort of tried that.
295.
by troels 12y ago
From what I understand, it is also commonly used to predict optimal poker plays.
296.
by troels 12y ago
If it's still relevant, I've found the Ruby whois lib ( https://github.com/weppos/whois/tree/master/lib/whois/record... ) to be quite comprehensive.
297.
by troels 12y ago
Proxies to the rescue. And take your time - whois data don't change that often and you can know in advance when it might (E.g. expiry date).
298.
by troels 12y ago
I got a bit baffled by that too, but came to the same conclusion as you did after a minute or two. In an actual interview, I suppose I would just start by clarifying what is meant by those terms before going on with a discussion about it?
299.
by troels 12y ago
It seems a bit disingenuous to present numbers like that. For the lay person, it may sound scary that there as 44 cookies on a given page, but that's a completely arbitrary measure. I would think that the important thing isn't the
300.
by troels 12y ago
Very nice. What did you base the data on? Some 3rd party source or did you compile stats from a corpus of random texts?

More ›