_yosefk

74 ms·

571.
by _yosefk 12y ago
A conversational problem indeed... I explicitly said, twice, that it's not a security bug. It's just a feature that makes as much sense as write() interpreting special byte sequences as commands. If such a write() call were exploi
572.
by _yosefk 12y ago
Ahem. Every web server passes "unsanitized user input" to "Turing complete systems". You have a right to expect certain things from subsystems regardless of their "Turning completeness". * A file system should
573.
by _yosefk 12y ago
"Quite humorous"? * In Unix, shell scripts and shell subprocesses are everywhere, and are supposed to be everywhere. * Environment variables are passed across subprocesses by default, you need to explicitly filter the environment
574.
by _yosefk 12y ago
The flip side of "no fragmentation" is, there's no ARM core with hardware multithreading, and there won't be one until ARM decides to make one, which AFAIK may well be "never". You can license a core with hardw
575.
by _yosefk 12y ago
I didn't mean the names matter in themselves but that original visions matter and names actually tend to reflect them.
576.
by _yosefk 12y ago
It's definitely a success, but: * Its success would not save Microsoft if Windows and Office usage sharply declines * Nor would its failure ever break Microsoft * ...both of which resulting from the market being not very large by Micr
577.
by _yosefk 12y ago
Wow! Yardstick Capital at work: "37signals is now a $100 billion dollar company, according to a group of investors who have agreed to purchase 0.000000001% of the company in exchange for $1. Founder Jason Fried informed his employees a
578.
by _yosefk 12y ago
1. From a user's perspective it doesn't matter if it's "constraints" or "poor design". 2. Blaming C does not explain C++'s undecidable grammar, the zillion ABI incompatibilities introduced on top of C
579.
by _yosefk 12y ago
Nothing in C forces the API designer to use -1 as "bad PID" in one place and as "the set of all PIDs" in another, however. Perl's undef isn't that different from returning, gosh, -2 or any other bloody number
580.
by _yosefk 12y ago
If drivers spend much time loading and unloading, and not getting paid, the amount paid per mile required to attract drivers will rise.
581.
by _yosefk 12y ago
A right to be forgotten - that is, not to be mentioned somewhere - looks more like a negative right, not unlike a right not to be mugged or not to be photographed in your own bathroom through an open window. A right to be remembered w
582.
by _yosefk 12y ago
By reducing consumer spending and investment through the taxation necessary for funding said project?
583.
by _yosefk 12y ago
The thing is, compilers can force people to refer to error variables, but they can't prevent people from returning array[0] when i is out of array[]'s bounds, an empty list given an empty JSON string, etc. And forcing the former u
584.
by _yosefk 12y ago
I did cite it - it's from my parent comment's link: http://roscidus.com/blog/blog/2013/06/09/choosing-a-python-r... Are you saying that: * Getenv doesn't return an empty string for a
585.
by _yosefk 12y ago
Incidentally, your link includes a great example of Go's error handling - which is inevitably what actually happens in languages without exceptions: errors are silenced and the program marches on, each step making less sense than the
586.
by _yosefk 12y ago
Why do I miss the point if, like me , you think Go is the new Java? :-) I think it's exactly that, it works and that's fine, and why the JVM doesn't do cheap concurrency I don't know. If it did Java might have been the
587.
by _yosefk 12y ago
First of all - yes they're "just tools" and that's pretty much what I said - Go is for servers and it won't "rule them all", and I wouldn't write the obvious if it weren't for the massive Go hype
588.
by _yosefk 12y ago
Erm... so: * I didn't "find" my use cases - they found me. All are real things I do with Python. * I'd switch to something faster than Python, and finding more errors statically, if it wasn't for the bunch of things
589.
by _yosefk 12y ago
I learned programming & CS at age 18 and I'm on par with people I met who learned it at say 11. I learned English starting at age 5 (I read as almost fluently in English as in my mother tongue, Russian); I learned Hebrew starting a
590.
by _yosefk 12y ago
Yet in reality, I work on chip architecture and hardware accelerators and have never written VB macros in MS Excel. Go figure. (I did write one VB macro in VS 6 though, I think. Perhaps it was that incident that distorted my worldview.)
591.
by _yosefk 12y ago
I agree of course, I just think a scientist taking a more thoughtful approach > a scientist taking a sloppy approach > a "software engineer" taking an overly thoughtful approach. Because the latter could have written ~200K L
592.
by _yosefk 13y ago
Well, let's say you can't use numpy or similar - "C written by someone else" - and further, let's imagine a world where CPython has no GIL. Then there are 2 options: 1. You rewrite things in single-threaded C, getti
593.
by _yosefk 13y ago
Well, "working vs waiting" is the same as "CPU vs I/O", so we're saying the same thing here. As to your "concurrency vs parallelism" distinction - it's one valid one, and pretty widespread; I thi
594.
by _yosefk 13y ago
And why do we need digital literacy any more than machine gun literacy or brain surgery literacy or anything like that? Except Alan Kay, everyone knows perfectly well how useless programming is to anyone but the professional programmer.
595.
by _yosefk 13y ago
Think a few steps further down these lines and you'll be able to separate the people worrying about "teaching everyone to code" into evil (companies looking for cheap labor) and stupid (people thinking there are problems that
596.
by _yosefk 13y ago
None of this is finding the bug. RAM bits almost never flip, and "a single bit" - "the" bit he pointed to - flipping on multiple occasions is a virtual impossibility. As to all those other things: yes, bad stuff, did he
597.
by _yosefk 13y ago
So... what's the actual bug? "Cyclomatic complexity" this, "single point of failure" that sounds very damning, but if you don't see the bug it's sort of less convincing. I mean, I can look at any code and
598.
by _yosefk 13y ago
Sure, if none of these C functions tries to yield, or calls a library function trying to yield. In fact, the problem isn't that they can't be plain C functions, but that they must be plain C functions :) Why would functions ca
599.
by _yosefk 13y ago
No stack though, so you can't have arbitrary code running in several concurrently executing coroutines. As to debugging, I don't think setjmp/longjmp gets in the way of debuggers at all.
600.
by _yosefk 13y ago
The trouble with them is they're stackless, which means you can't have arbitrary functions thoughtlessly coded by someone unaware of threading called inside many different protothreads. Coroutines are easier to combine with code n

More ›