neilc

102 ms·

931.
by neilc 18y ago
Stories that aren't 100% accurate are on YC fairly regularly; deleting just the inaccurate stories that affect YC companies would be quite unfair.
932.
by neilc 18y ago
It's a great read, although it hasn't been updated since 2006, so it should be read with care.
933.
by neilc 18y ago
This is only true if you think your application can do a better job of scheduling than the underlying OS kernel. In many (if not most) cases, this is false. Why is it false? As the other reply notes, you have more domain knowledge than th
934.
by neilc 18y ago
~1 process per core, that is (unless the daemon is completely I/O bound).
935.
by neilc 18y ago
If you're hiring candidates that wouldn't consider learning a new language to join an interesting startup -- well, good luck!
936.
by neilc 18y ago
In what sense do are those sites "service businesses"? The whole point of the service vs. product distinction that PG was making is that a service business is less scalable: its revenue is basically linear in the number of employees that
937.
by neilc 18y ago
Reddit and Octopart are product businesses, not service businesses.
938.
by neilc 18y ago
Yeah, the "Y" favicon is really good.
939.
by neilc 18y ago
If favicon is important part of an online company's identity there must be something seriously wrong Why? You're stating an opinion, not an argument. Surely you would agree that Google's logo is an important part of their visual identity
940.
by neilc 18y ago
How does that follow from the above definitions? I don't agree that if you build open source software, you aren't a startup -- but I do think that a big part of being a (VC) backed startup is the potential to achieve "hockey stick" growth,
941.
by neilc 18y ago
A favicon is a pretty important part of an online company's brand/identity. I'm not surprised they thought carefully before deciding on what to change it to. Plus Google have a surplus of very intelligent people; it's not like they couldn't
942.
Why Pure Computer Science Theory Is Wasteful(daniel-lemire.com)
2 points by neilc 18y ago|2 comments
943.
Nassim Taleb's Top 10 Life Tips(ben.casnocha.com)
78 points by neilc 18y ago|23 comments
944.
by neilc 18y ago
Feynman worked for Thinking Machines on the hardware design for the Connection Machine, as well as how to use the machine to solve some problems in scientific computing.
945.
by neilc 18y ago
... except that Microsoft doesn't ask questions like that for software engineering positions, at least not when I interviewed there. I found the interview questions to be reasonably similar to Google's.
946.
by neilc 18y ago
Yes -- what else did you think I was paraphrasing?
947.
by neilc 18y ago
It seems quite obvious that he's quoted out of context: FF3 is obviously not "bug free", but there aren't any further browser fixes slated for 3.0 final (as opposed to fixes that are "server-side and Web-site related"). The headline is just
948.
A short review of Collective Intelligence in Action(daniel-lemire.com)
1 points by neilc 18y ago|0 comments
949.
by neilc 18y ago
It's not "my" startup, but I work for a startup that makes a database system for stream processing -- i.e. it lets you run SQL over live streams of data, to understand data as it changes rather than waiting for it to hit disk.
950.
by neilc 18y ago
You don't necessarily need to "beat" Venter -- startups are not a zero-sum game, especially in a field as early as synthetic biology.
951.
by neilc 18y ago
I'm not sure what lessons are to be learned here -- "be transparent"? The author mostly complains that they are still taking new customers during downtime -- which is probably not good, I agree, but seems like nitpicking.
952.
by neilc 18y ago
Bring on the cloud. A massive service outage is a reason to encourage the move to cloud computing? Such an architecture would only be more vulnerable to service failures outside the user's control, not less.
953.
by neilc 18y ago
Presumably you know the people beforehand whom you're visiting unannounced. That is rather different than imposing yourself on a complete stranger, for the explicit purpose of pitching your business. Calling it a "break in" is unreasonabl
954.
by neilc 18y ago
I thought it was an interesting algorithm that I hadn't seen anywhere else. I think that is perfectly sufficient justification for submitting it; as for making it to the frontpage, ask the people who modded it up.
955.
by neilc 18y ago
BTW, more info on cuckoo hashing: http://mybiasedcoin.blogspot.com/2007/06/cuckoo-hashing-theo... http://mybiasedcoin.blogspot.com/2008/05/more-robust-cuckoo-...
956.
Cuckoo hashing(en.wikipedia.org)
25 points by neilc 18y ago|9 comments
957.
by neilc 18y ago
That's been my general impression with the Unix half of OSX: it is Unix, it's just ... mediocre. Everything is just more of a pain in the ass than it would be with FreeBSD or Linux.
958.
by neilc 18y ago
Perhaps -- but even if that is true, interacting with more machines is presumably a trivially parallelizable operation, so it shouldn't double the overall response latency.
959.
by neilc 18y ago
Yeah, I think this approach makes a lot of sense (and it is another instance of "more data beats better algorithms", in a sense). It's also similar to some of the themes Paul Buchheit talked about his StartupSchool talk -- you want to liste
960.
by neilc 18y ago
Google found that when the results increased to 30 per page ... it took about twice as long to display the longer results list for the user. I'm surprised that increasing from 10 to 30 results increased total latency by ~100%. I would hav

More ›