I was astonished in my computer architecture class as an undergraduate to find out that the C language specification requires that all single-precision floating point numbers be converted to double-precision before any operations are done on them. So working with float's rather than double's in C is not only more vulnerable to underflow and what-not, but it's slower. Since then I've never declared a variable to be a float again.
I'm sorry, but this reply doesn't make any sense to me. If applications have to cache something, they should use the file system, which would not affect their resident memory at all. Maybe I'm just old school, but where I come from operating systems handle the memory hierarchy, not applications.
Depends on the compiler. When I took my "hardware for CS students" class as an undergrad, our big project for the semester was to write a CPU simulator in c, and the campus labs had just rolled out the upgrade from gcc from 2.x to 3.x. I had a bug in my program that I just couldn't isolate, and after about eight hours of chasing it, I realized that the compiler had allocated space for an integer variable right in the middle of an already-allocated array, so the two variables were stomping on each other. I changed the name of the integer variable and my problems went away.
Apparently I wasn't the only one, because within a week all the labs were back on GCC 2.95.
I agree with everything you've said, however I take issue with the broad idea that the purpose of democracy is to pick the best leaders. It's not, the purpose is to prevent any single leader from doing too much damage. A certain amount of fickleness and inconstancy on the part of the voters is actually a good thing, because the elected leaders know they have to stay on their toes.
Well, this is actually the old way of parsing a genome. What you've described is essentially a Hidden Markov Model with discrete states, which is the bread and butter of genome sequencing.
In addition to the other factors listed here, Postgres's default configuration was tuned to a dramatically underpowered machine for many years. Yes, that meant that the occasional user who did have that kind of machine saw acceptable performance out of the box, but the other 9/10 systems burned a lot of the DBA's time tuning the system. I think that's a big reason why Postgres has a reputation for difficulty in some quarters.
Are probabilistic techniques still not considered conventional? It's not like the mathematical foundations of probability haven't been laid hundreds of years ago (I'm talking about the rules of inference, not the axiomatization).
I have a friend who was an accountant, found that he liked writing VBA scripts in Excel, and eventually wound up at a company big enough to need an IT department but small enough not to afford a big one, and now he's basically got a dual CFO/CIO role. Every six months or so he tells me about the new mobile app he's got in development for the salesmen to place orders, or for manufacturing line inspectors to check inventory, or something. I always point out that it would be cheaper, faster, and more general to just build a wep application, and he always rolls his eyes and tells me I don't understand the marketing power of mobile apps.
Which is true, I don't. But it got me thinking about all the "browser is the new OS" and "the Internet is OS-agnostic" and "this is the end of Microsoft" talk from the late 90s. Anyone else remember those days? Because Steve Jobs and the IPhone pretty much undid it all. It's 15 years later and everyone cares about the OS again. Weird.
If it's a core question, it's not "bike shedding." If the problem is really bike shedding, a good leader will get both sides to realize they're being ridiculous and wasting everyone's time over something trivial.
One of my favorite tidbits from "The Best And The Brightest" is that Dean Rusk, President Kennedy's nominee for Secretary of State, had to fill out a security questionnaire like this one, and to the question, "Have any of your relatives been part of an organization whose purpose was to subvert the government of the United States?" he answered "yes" because both his grandfathers had fought for the Confederacy in the Civil War.
Tax breaks. A surprisingly large portion of the book is in praise (what's the opposite of a screed?) of having a company you control pay for your meals, car, and house so you avoid personal income tax.
> In fact, your average NASA employee doted on astronauts like a star struck little girl. What the crew wanted, the crew got.
This is the question I've always had about the Challenger, but never heard addressed: How much of a factor was the crew's opinion considered to be? I strongly suspect that the astronauts themselves exerted informal but real pressure to prefer flying, since it would be their moment in the sun.
I couldn't agree more. Once upon a time I spent about a week refactoring a really ugly build system (produced a combination of .NET assemblies, C++ libraries, Java applets, and a hacked version of Apache with bash embedded - don't ask) and got everything just right with this approach, then within a week it was all broken again. I removed the rule-generation from the Makefiles and everything was fine after that. Since then my policy has always been that copy-pasta is OK in the build system, and not to worry about it.
(I'm not saying generating the rules from code in a real build system wouldn't be a good idea, jut that it isn't with make.)
> The problem with buses for transport is they use the same infrastructure as cars.
I've always thought that was their #1 selling point. Light rails are just insanely expensive. When you've already got roads in place, buses become very attractive.
It's possible to structure "quantum communication" (meaning passing qubits back and forth, basically) so that if someone in the middle reads them at all, the recipient is aware of it. It's not really possible to prevent the interception but it is possible always to detect it, at least in theory.
While I wholeheartedly agree with your overall sentiment that the baby boomers poisoned the well of liberalism for the next few generations, I feel compelled to point out that Nixon's landslide was in 1972. 1968 was one of the closest elections of the century.
Uhm, a 95% re-election rate is not a 5% chance of getting fired ever. The chance of being re-elected N times is (1-0.05)^N, so the chance of not being re-elected N times is 1 - (1 - 0.05)^N. For N = 10 (a nice twenty year run in the House), that's only ~ 60% chance of being re-elected each time. For N = 15, it drops to 46%. I think that's actually pretty reasonable, you've got enough security at the beginning to allow you some latitude but not so much that you stop caring.
Wow, I had no idea haxe was still around. I looked at using it about 4-5 years ago and dismissed it as not ready for prime time, but that looks like a reasonably active community.
If you're talking about the sharp decline in computer science degrees granted to women around 1984, I have a couple of guesses.
First, no one really knew what kind of career path getting a computer science degree entailed until around then anyway, and once women had figured it out they decided they didn't want it.
Second, in my opinion women are much more sensitive to their career's stability and predictability than men are, for the simple reason that they're the ones who have to plan out when to have children. You'll notice that the drop-off more or less coincided with the first (or second, or maybe third - I forget which) big bust in the PC industry in the 80s. Mainframes and even microcomputers did not have boom-and-bust cycles like PCs did in the 80s, although individual companies might blow up the industry as a whole was relatively stable. College students looking for stability around that time might sensibly decide "computing" wasn't for them any more.
My favorite reference in this discussion (well, about women in science and engineering as a whole) is Philip Greenspun's article (http://philip.greenspun.com/careers/women-in-science) who basically makes the argument that the median scientist who doesn't win Noble Prizes actually has a pretty crappy career and that women are smart enough to stay away.
I wasn't arguing the "women aren't good at programming" position.
My personal opinion is that the most important reason there are few women programmers, next to which any other factor fades into insignificance, is that women are simply less interested in doing it. Now that I think about it, the fact that there used to be many more women programmers when women's career options were more limited is pretty strong evidence that this is true: Now that women have more options, there are fewer women programmers.
Back in the day I think the popular perception of programmers was kind of an extension of a secretary (it's just typing on toggle panels instead of a typewriter, right?) so there were lots of women stuck in secretarial positions who tried it out.
I was under the impression that most people had accepted that the reason more men than women do software development is that more men find it interesting than women do, not because women are less intelligent or because there's a conspiracy to keep them out. Too bad, I guess not.
As an aside, my boss at the moment is a woman, and a few years ago she had to staff up significantly and hired several young women at the junior level. The interesting thing is that her team hasn't really changed because of "the diversity" since it turns out that women who like programming act pretty much like the men who like programming: they're kind of socially awkward, have sometimes unreasonable expectations that the world will be a meritocracy, make a fetish out of liking cartoons and movies that most people don't, and so on.