leif

42 ms·

361.
by leif 15y ago
see also cuckoo hashing
362.
by leif 15y ago
of course, only years later would I understand the proof of why: M/B-way merge sort is optimal in external memory
363.
by leif 15y ago
funny you should bring up merge sort: in my first google phone screen, I was asked "why is merge sort good" and I gave "it's asymptotically optimal and stable" and couldn't come up with anything else, the interviewer eventually conceded tha
364.
by leif 15y ago
Too bad this guy never took a data structures course. Binary search on an array is O(log N) in the DAM model, while search in a B-tree is optimal: O(log N/log B). Note that if your elements are english words, let's be extremely generous an
365.
by leif 15y ago
need an option for "have an invite but they're over capacity and won't let me sign up yet" I have no opinion yet, I was backpacking for a few days during the announcement, just got back and had these invites. To me, it's nothing more than
366.
by leif 15y ago
I didn't read the parent, thanks.
367.
by leif 15y ago
squashfs plus unionmount commonly used for livecds, I compress /usr, saves me about 5GB and some I/O and therefore some battery
368.
by leif 15y ago
you misunderstand me. I have a working solution, using fuse. if he thinks that's bad, it's on him to provide a correct in-kernel solution, then show me it's faster
369.
by leif 15y ago
I don't care what he says, I'm still waiting for an inclusion of some kind of union mount functionality in mainline. I don't need anything fancy, just one ro dir and one rw dir. If he can't even get me that in-kernel, I'll stick with un
370.
by leif 15y ago
That was my point. valgrind is very user friendly. Keep using it!
371.
by leif 15y ago
There's just a timeout on replies that grows with depth. Why not fuzz your own code then?
372.
by leif 15y ago
don't worry, I understand, I pasted it for others Yeah, it's not perfect, but nothing we have is. I don't know what more to say. It works pretty well for most things I've seen. Do you have any examples of projects which can't use somethi
373.
by leif 15y ago
Sorry, I don't mean to be patronizing. You asked whether the compiler catches it, it didn't, but valgrind did so I showed that. Of course valgrind and friends have false positives, but in other languages where these errors are compile erro
374.
by leif 15y ago
gcc 4.6 warns about returning the address of a local variable by default (you don't even need -Wall): % cat foo.c #include <stdio.h> int *bad(void) { int a = 42; return &a; } int main(void) { int *p = bad(); pr
375.
by leif 15y ago
C++ compilers will actually warn you about plenty of memory issues if you ask them to, by writing type-safe code and turning on lots of warnings. Static analyzers today are very good, and will catch a surprisingly good amount of unsafe mem
376.
by leif 15y ago
I think you need a font where asterisk is not superscripted. Try Inconsolata.
377.
by leif 15y ago
I read the title and half expected an article about Yahoo! asking Hulu to save them.
378.
by leif 15y ago
the absolute overhead is linear in the number of elements, sub-linear in the size of input
379.
by leif 15y ago
I keep /usr in a squashfs, mounted with aufs over top of it, for the 0.01% speedup I probably get (I grew up on gentoo, forgive me). Periodically, I need to rebuild the squashfs to reclaim space in the aufs writable directory. I once had a
380.
by leif 15y ago
8 MB is constant, but the number of bits in the input grows like log of the largest input value, so the 8mb overhead contribution goes down as the input grows. Of course, the kernel isn't going to support more than 64 bits of sleep times an
381.
by leif 15y ago
higher resolution, but I don't think precision (especially in the face of I/O)
382.
by leif 15y ago
constant in the number of elements, sub-constant in the size of the input
383.
by leif 15y ago
yeah I thought it was base64 and that was annoying
384.
by leif 15y ago
That's not how their market works. They need to keep enterprise (which requires backwards compatibility) and consumer unified if they want to throw their weight around the way they need to (particularly with hardware vendors). This may no
385.
by leif 15y ago
The problem of deflation is worse. There is a finite supply of btc. Read that again. Finite. in fifty years, our whenever the mills run dry, are we just going to stop increasing population? Are we going to stop creating more things of value
386.
by leif 15y ago
I think they are going after this seamlessness. Remember that microsoft is obsessed with home networking.
387.
by leif 15y ago
Other tiling wms can "split the screen". Of course, that's where the similarities end. The problem with tiling wms on linux is that apps are not built for that model: to be stretched and squeezed like that. Notice how the video got real sma
388.
by leif 15y ago
It's not finished yet. Maybe they will eliminate the windows 7-style desktop you saw before the end. Chill ouuuuut. But this isn't starting off on the right foot Where did you expect them to start from? Build the kernel from scratch?
389.
by leif 15y ago
I was an intern on the windows user experience team last summer, so I saw a lot of this new ui. While I will never personally enjoy it (emacs is my ui), I'm proud of them for it and I think it will do well. The first thing to remember is t
390.
by leif 15y ago
While it would've been better to just ignore him, I must say, fighting back with working code lends Zed some amount of class.

More ›