lukesandberg

15 ms·

91.
by lukesandberg 15y ago
You should read the article. The article discusses how low latency can be achieved in a data center . The specifically discuss why they are interested in intra-datacenter latency vs. attacking latency across the internet as a whole.
92.
by lukesandberg 15y ago
In a language that requires TCO then you are right, it's not about optimization it's about correctness. In a language (like c) that has no such semantics it is 'just' an optimization. In scheme however it is a correctness issue.
93.
by lukesandberg 15y ago
Iterative is definitely the way to go, and it's the way you are (generally) expected to answer a question. If you are asked a question a good first approach is almost always to describe the simplest/dumbest solution you can and then iterat
94.
by lukesandberg 15y ago
Practice! Coding in an interview is unlike any kind of programming you have probably done so it's only natural to freak out a little. Personally the whole exercise of writing code not on a computer is bizarre enough on its own! Find some p
95.
by lukesandberg 15y ago
Nice article. I wonder if it would make sense to add a similar kind of support to redis. Maybe if a sorted set's read/write ration goes above some fixed threshold then it could substitute an implementation based on a sorted array instead o
96.
by lukesandberg 15y ago
Stimulants are the classic answer here (caffein b vitamins). But i don't think that there is anything that acts a a proper substitute for sleep. your best bet is to restructure your schedule so you can get more sleep rather than looking f
97.
by lukesandberg 15y ago
I love coffee way too much, but i am increasingly finding that this is a bad idea. The more coffee i drink throughout the day the worse i feel in the morning, even when i get a reasonable amount of sleep. http://youarenotsosmart.com/2010/
98.
by lukesandberg 15y ago
I think that would create a sort of turtles all the way down situation. Google has so much internal proprietary technology infrastructure that it would be pretty difficult to provide a compelling explanation that doesn't have you wondering
99.
by lukesandberg 15y ago
I understand that, but couldn't two different files encrypted with two different keys theoretically share some identical blocks. Obviously this is way less likely than two unencrypted files sharing blocks but still it could happen.
100.
by lukesandberg 15y ago
The way the authors quantified it was by taking cross cultural surveys. Basically they surveyed men and women from various cultures about their attitudes around jealousy and determined that in all cultures they studied men appeared to be m
101.
by lukesandberg 15y ago
I should have said that it tends to inhibit ovulation. I don't believe it's a myth though. It's just not a perfect mechanism.
102.
by lukesandberg 15y ago
I read a great book about things like this: http://www.amazon.com/Why-We-Get-Sick-Darwinian/dp/067974674... There are several chapters about the hormonal effects of childbearing and what the evolutionary benefits are for the children. A
103.
by lukesandberg 15y ago
wouldn't deduplicating encrypted data still work but your hit rate would just be a lot lower. you would have to work on a block instead of a file level but in theory you could dedupe file blocks that had been encrypted, because either way
104.
by lukesandberg 15y ago
yeah that is definitely true, but the underlying issue is fairness (which is trickier).
105.
by lukesandberg 15y ago
It's somewhat useful to think of integers this way but it can be dangerous. a lot of operations on integers are 'undefined' on pointers. for example integer subtraction (which is quite common for measuring size) is undefined if the pointe
106.
by lukesandberg 15y ago
>The economic growth since 1978 was, for the most part, produced by the efforts of 10% of the population I don't see how that could be true. productivity gains have been experienced across all levels of the income scale so surely that
107.
by lukesandberg 15y ago
The common view is that 'economic growth' is a tide that lifts all boats. But unfortunately its lifting some boats much faster than others.
108.
by lukesandberg 15y ago
This is the fundamental problem with macro economics. there is no way to isolate variables. So obviously saying it is 100% attributable to Reagan is probably not true. but also (to take some of your points) global competition would effect
109.
by lukesandberg 15y ago
in short ronald reagan tax breaks and labor market deregulation: paul krugman has essentially a whole book on why this occurred ( http://www.amazon.com/Conscience-Liberal-Paul-Krugman/dp/039... ) It is an amazing book that really changed my
110.
by lukesandberg 15y ago
For one thing you are the investor, so you don't have to invest in them if you don't believe in the same causes they do, or if you don't believe that corporations should be engaging in charity at all. As for the question of "why is it more
111.
by lukesandberg 15y ago
using subtraction for comparison can easily be broken due to overflow. This particular case may have been fine... but i wouldn't count on it.
112.
by lukesandberg 15y ago
I think the point is that he needed a comparison function not and equals function. thus the subtraction. Erric Lippert had a nice series on subtle bugs in compareTo implementations, in part3 he talks about a similar version of this bug.
113.
by lukesandberg 15y ago
He also compares jobs to people like Bill Gates and Warren Buffett. The oil executives were being used as a polar example of the kinds of opinions that people can have about the very wealthy. I don't believe the author was trolling. He w
114.
by lukesandberg 15y ago
I believe it's a morality thing.
115.
by lukesandberg 15y ago
I think the major point of the article was that jobs is getting something of a free pass amongst his billionaire peers. I think the last line sums it up pretty well: "[O]ur intuitions about economic desert and fair distribution are...compl
116.
by lukesandberg 15y ago
Exactly if you know who authored a peice of content then you can get a pretty good idea of how authoritative it may be. Google is pretty good about finding content but you want to know that the content is actually good (no content farms) if
117.
by lukesandberg 15y ago
the lambda calculus doesn't talk about concurrency but it does discuss time. like turing machines there is a concept of time inasmuch as a particular algorithm will take so long to complete (complexity). The reason it doesnt talk about co
118.
by lukesandberg 15y ago
The benchmarks in the article are obviously limited. For instance there was no accounting for read performance, or even moderately complex queries. Also given how fast performance was dropping off for the guids this would probably matter
119.
by lukesandberg 15y ago
No need to create a ticket machine, the db can probably do this for you. I've accomplished this with oracle sequences (supports grabbing blocks of ids). I assume there are similar capabilities in other db systems.
120.
by lukesandberg 15y ago
Many regular expression implementations don't deal properly with unicode character classes and collation. So adding something like unicode identifiers to go may be impossible or impractical with regexs whereas with a loop you can rely on m

More ›