The gun control arguments are a strong case that indicates US citicens think violence is an answer. It isn't. The Taliban almost won a civil war. I don't think they did really "outlaw" guns, they just collected them to use them against their remaining enemies. They just think that violence is an answer. Also note that it's impossible to collect guns in a country where most people live in villages and have a million holes in the mountains to hide their own guns if they did want to keep them. The Taliban certainly got only guns from supporters.
Also, the argument "strict gun control means that only criminal have guns" is wrong. Europe generally has what is consideret "strict gun controls", yet million people here have access to firearms: soldiers and police, security services, hunters, people practicing firearm sports, people who convinced the authorities that they need guns to protect themselves. Sometimes, people arm themselves even though they aren't allowed to (e.g. in Munich, a gas station's owner shot a policeman, who robbed gas stations in his spare time, using his police gun).
However, the overall picture of Europe is that the general crime level doesn't differ from the crime level in USA that much (though we don't put 3% of our adults into jail), while murder, especially using firearms, is way down. Between Washington DC and Brussel, there's a factor of 170! And that's though Washington DC by the letter even has firearm control (doesn't help - it's just on paper, and nobody checks at the "borders").
I don't think that strict gun control (in our, European sense) would bring down US murder rate to a civilized level. Guns don't kill people, people kill people. And only severely deranged people are able to kill other people. However, the arguments exchanged in gun control discussion show that too many US people are somewhat deranged. They actually considder using deadly forces to protect them.
The final point however is, that if you think using guns against oppressive governments is ok, you agree that terrorism is ok. Well, you say "freedom fighter" instead of terrorism, but that's it.
Guns do not equal encryption. Encryption is just an envelope that keeps others from reading your mail. Encrypting is completely free of any violence, and still provides a mean of security (it allows secure transactions of whatever you think off: TANs for online banking, love letters you write within the company network and don't want your supervisor to be read, company secrets you don't want to expose to industry spies like Echelon, and so on). Encrypting secures your basic rights.
Yes, you can use encryptiong to coordinate efforts that do cause harm, but it's not the encrypted e-mail that crashed the WTC, it was carpet knifes, planes, and kerosine. Nobody would want to outlaw these, because carpet knifes, planes, and kerosine obviously are almost always used for good.
I've written a web server in 200 lines of Forth (runs on GNU Forth, is part of the gforth-0.5.0 package). It's not an overfeatured web-server like Apache, it doesn't do SSL, doesn't allow any other language for server-side scripting than Forth, but I don't need that. If you want Apache, you can have Apache. Read more about it in http://jwdt.com/~paysan/httpd-en.html
Tip 1: To break the rules you must know them. It's much less fun to break rules you aren't aware off. And tell people when you broke the rules - if nobody's aware of you breaking rules, WTF are the rules for then?!
Tip 2: Don't use other code - it's bad, since you are the master programmer, and the others are just morons, and you'd have to understand it to use it. You know: if you want it done right, you have to do it yourself.
Tip 3: Don't listen to customer, they are stupid morons, anyway. Write the program as you want to use it. Then brag about how c00l and 3l11t it is, then the customers will buy it anyway. If not, release it under GPL, and go IPO as new open source company.
Tip 4: Don't write comments into your programs. Your code is obvious. Since you assume the other people know about rule 2, they won't read your code anyway.
Tip 5: If you write comments to break rule 4, write them as fast as possible and with ltos of tyops, so that oepple kan now that you are busy duing codink.
Tip 6: Don't read mainstream best-seller coding practice books reviewed by/.. You know, mainstream sucks, it's for Windows-programming morons only. Read out-of-print books like "Thinking Forth" (really a gem, no fun!).
Hey, this topic comes up in/. once in a while, when another fool gets his ToE (Theory of Everything) into the light of/.. These sort of theory is easy to make a fool out of you. A lot of people tried, and all of them weren't rewarded, even Einstein failed.
To me, each of the ToEs, including the string theory, sounds a lot like crackpot stuff. 11 dimensional vibrations of strings/membranes? Believers quoted with "The big bang is a fact, not a theory" - did they participate in that event? It's cool (and intellectualy chalanging) math, but does it reflect or model reality?
The final ToE will very likely look like crackpot stuff at the first sight, too. But unlike the crackpot stuff we have today, it will be possible to make experimental proofs.
I personally believe that it will look much less spectacular than cold fusion, supersymmetric particles and such like. There are already a lot of self-contradictory results of current theories that a good ToE will get rid off, like black holes (no light escapes - how does gravity escape?).
So the theories we have now are bad enough, but at least they are somewhat in consistency to our observations. I prefer the problems of QM and GR to nutty "star trek" phy"sick"s any day. Too many people trekkify the classic theories, anyway (FTL travel, quantum teleportation, using black holes for time travels, etc... Cut their funding!).
I don't make money with free software (at least not up to now). I make money with bloody proprietary software (and with bloody casino capitalism stock market speculations).
I give back to the community by writing free software. Don't tax me, I've already given more than 10% of my working time back to the community. Church taxing monks doesn't make sense, eh?
And if you spent this money just on lawyers to defend free software, there's something terribly wrong. Laywers are utterly unproductive. I know, they are part of the Golgafrinchina's load of unproductive people who crashed here, but they still are unproductive.
I know, in the US, you can get almost any result out of a trial, if you pay enough to the right lawyer (good price/performance: OJ Simpson's lawyer, bad price/performance: Bill Gates' "we have our own render farm to generate faked videos" DoJ trial lawyers). I don't understand why you need a legal system if it just manages to keep an anarchistic state (the right of the stronger). Well, perhaps because the lawyers are strong enough;-).
Well, Columbus was a crackpot. He read Marco Polo's story about China (or more likely from a second source), and took Polo's venetian miles of foot march as sea miles of direct distance, and concluded that China ("India" - he really must have been a crackpot) was some few thousand sea miles in the west. All the sane people knew that China is 20000 sea miles in the west, and noone could get there with 15th century technology. If it wasn't for the west-indian islands, Columbus won't have survived.
To the population of these west-indian islands, and to whole Amerika, this discovery by religious fanatists turned out as a real disaster. Columbus and his serfs killed most of the inhabitants of the Isabella island, and the gold-drunken conquistadores didn't do any better to the Aztecs and Incas.
This is really an example for an exploration that never should have happend that way. BTW: America was discouverd before by several expeditions, which all concluded that it wasn't worth the risk of crossing the ocean, and then again forgotten.
I really don't understand this. The whole point of life is to give life, to create new life. Usually of our own kind, but since we do sexual breeding, it is a new combination. It's risky. The parents of Hitler and Stalin created life that killed many million peoples.
Some religions actually discourage breeding, at least for their most "holy" members. Many tell you that sex is something bad. Some even tell you that their most holy person wasn't originated by sex, but created by a god.
Come on, they are not experts. They have no idea what life is about. Many don't want to accept this world, and wait for a better world after death.
Scientists have to calculate the risks themselves. It's already risky to create new life by interbreeding, you might create killer bees or killer dogs or killer bakterias when you do that.
It's more risky when you purposly modify plants to create poisonous substances. You know they are evil, but you don't know where they will get.
When you create new life forms, do that with care. I don't think a toy life form with 300 genes will likely survive long in a world where lots of other life wants to eat, but I can't know.
There's a humorous film where a scientist creates a bacteria that composts dry paper. Well, it escapes, and does compost paper. In fact, all paper available. The result is a bit like the After-Y2K comic.
The GPL doesn't steal the 'IP' of the original authors; they still can take the sources, close the improvements and sell the product (or sell it, if the use of the product is limited in a GPL distribution, like the one of libraries or worse, incremental compilers).
The GPL regards a closed source fork by a third party (i.e. not the original authors) as theft. It doesn't differ from Bill Gates view here. It is quite clear to everyone that taking public commons is theft, too.
And finally, I can't believe that people still question whether the other software could be written open sourced, when the greatest things that require the most effort already are open sourced. Get out through the doors, don't let your view be filtered by windows!
Grammar+vocabulary based translation is doomed. The main problem is that natural language neither have strict grammar nor real vocables. I.e. what you have is a somewhat loose grammar, and ambiguous words that often have a lot of different meanings, depending on context.
The problem of translation is that the classes of meanings are different in different languages. If people would only write unambiguous sentences, translation would be easy, but often enough, people use ambiguousities deliberately (jokes, oracles, poems, etc.). So it isn't enough to deduce the meaning of a sentence/paragraph/text by looking at the context, but also you have to carry the remaining ambiguousity into the target language. And that's one of the harder parts of translation.
I'm talking about translation by humans, not by machines here, that's what is difficult enough for humans. I don't expect a good machine translation any time soon, as often enough human translations are too bad. Literature, with rhymes and tone has much more problems than just meaning, and here good translations are often impossible.
So IMHO translation can't be done using string-based rules. You have to collect "meanings" from the words, the grammar rules and the vocabulary files can help you, but after all, you have to search for a sentence in the target language that matches the remaining set of meanings of the source language sentence best.
Well, the math books would still be full of the same letters as before, since math doesn't tell you anything about this world. Math is perfect, because it makes up its own assumptions (axioms).
BTW evolution: "evolution" isn't just one theory, it is a set of theories. Darwin's theory is just one evolution theory, it bases evolution on mutation and selection by survival of the fitest. Lamark's theory is another evolution theory, and it's based on something like a "genetic memory".
Actually, we know a lot more about genetics now than both Darwin and Lamark knew. We know that Lamark isn't completely off, since there are ways to exchange genetic informations, including gene ferries (mostly for bacterias and plants), and gene editors. We don't know if these gene editors are just deactivated retro virii, or if they are used once upon a time, but one thing we can say with pretty high accuracy: this live as it is now wasn't created, but evolved. Even the pope accepts this.
There is no observation whatsoever that any species ever was created. All creationist arguments I know of focus around "proofs" that this or that can't have evolved, mostly by taking something complex as the eye, and leaving out all the steps that led to that eye as it is now, and even failing to explain why the human eye has the sensors behind supply and sense, instead of having it in front of it like octopus eyes do.
Creationism is a "last resort explanation" - if everything else fails, assume it was created. It's also only shifting the problem, since it explains something lesser (the world) by being created by something higher (God), and completely misses to explain how this higher (God) came into being. He for sure hasn't evolved? So who created God?
In all my school days, only one teacher told us things as if they were facts, and that one was the religion teacher of elementary school. After all, her facts weren't based on observation or by formal proof, but on reading of just one really old book.
And so we are back on topic: why do teachers tell things from old books as if it was truth and no new insight has been gained later? Well, it's because teachers never left school. They haven't seen the real world. They aren't "authentic". For them, all knowledge is like a fairy tale, something that goes from generation to generation. It isn't. Knowledge is but a tool to get around with the real world.
What pupils really need, apart from meta-tools like letters and math, is to learn how to solve real world problems, problems they'll face in their future life. Today, the most important thing is to learn how to learn. I mean: to learn yourself, without being driven by a teacher. This isn't a budget problem, it's a cultural problem. Few, if any teachers like it if pupils know more than them. Many even dislike if they knew more than they have been teached in school yet. This must stop!
The GPL simply disallows to restrict the right of redistribution. Thus if you get a Corel Linux CD, you can redistribute the GPL'd programs including source (you can get the source CD from Corel for shipping&handling - if they fail to do so, sue them out of this Linux business). The CD itself is a mere aggregation, and therefore not bound under GPL. They just can't forbit you to dissect it and redistribute the parts that are under GPL (if they do so, their rights under GPL stop and they go out of this Linux business). Anyone can create a "free Corel CD" out of it by omitting all the software that isn't redistributable under an OSS license, and give that to everybody.
This is just a stupid, associal EULA. It's a reason to boycott Corel, but not one for legal actions. They have associal clueless lawyers (who's more associal than someone who disciminates against children?) who should be fired.
Please, people at Corel, wake up. Read the EULAs the jerks in the legal department put out. They do harm to you. This is not in your interest.
I work quite constantly a 40 hour week. I don't think overtime is anything good, since you don't work concentrated anymore. IMHO 40 hours work per week is already too much, you can't work concentrated for more than 4 hours a day on a day per day basis - and working unconcentrated while reading slashdot is just a waste of time. I could do that at home, too.
Waiting for the machine is a nuisance, and should be avoided by getting faster hardware or faster software. Meetings should not dominate your workload, either.
I've seen code written in 60 hour weeks. It took us a lot of weeks to get the bugs out of it. The human brain isn't a ship's diesel, which has a standard output and keeps that forever - the brain tires, quite fast so. To paraphrase Brooks "Throwing overtime at a late project makes it later".
Therefore I prefer salary, it gives me no advantage on overtime, so I care about not accumulating overtime myself. The amount of work done beyound 50 hours per week is negative (yes, *negative*), even in non-IT studies (I've one of WW II England, working on ammution). It's much worse in the IT field, the amount of work isn't only negative, it's also of much worse quality.
Most of the quality of free software comes IMHO from the fact that the volunteers working on such projects do naturally make breaks when they get unconcentrated, because it's no longer fun to continue coding. And most of the miserable quality of commercial software is resulted from 70 hours weeks. Hey, it looks all like work of cafeine junkies and has to be thrown away (Mozilla, anyone?). It ain't done until it's ready, and overwork harms both you and the project.
Well, I've solved my geek-gift problem quite simply: I buy the geekish stuff for me myself (books before christmas, hardware afterwards), and let the gifter only choose the wrapping. A wish-list just produces questions like "What's that? Where am I supposed to buy such a thing?" etc. If the gifter insists on a surprise present, let him/her choose something non-geeky, i.e. winter clothes, sweats, and such.
All I saw on the discussion is perfectly on that list. A elvish ring that is supposed to be ordered via Internet? A special swiss army knife that is sold-out? Books that have "Computer" *and* "Programming" in the title, but neither "Dummy" nor "Windows", and therefore isn't available at the local book store? Come on, no non-geek can get such a present for you.
BTW: What about an original, 100% Y2K proof chinese abacus;-) (100% Y2K proof, because the chinese calendar has year 4696 now - cycle 78, year 16 to be precise).
The observer mystery is a common myth about QM, and comes from the problem to understand the wave-particle dualism. Things can show particle or wave behaviour, and when you force them to show particle behaviour, the wave function collapses.
What really does that is an "observation" device (like a CRT screen, or the fovia of the eye), not the "observer". You can put such a device anywhere in your experiment, and you can prove that it really does the same as any other "observer" does - reduce the accuracy of the measurement according to the uncertaincy relation.
For Schroedinger's cat that means, the alpha or beta particle is the "observer", because it already turns the possibility to radiate into a fact (and the scintillation counter does the rest).
My father worked once in fusion research (the hot fusion one), that was back around 1970. While first, people though it won't take too long to make a working fusion power plant, they quickly raised the time estimation to "50 years", and that remained constant over the last 30 years (we are still 50 years off a working fusion power plant).
So if any physicist says to you "in 50 years", he means "I have no idea how to achieve that", or worse "According to all informations I have, the effort is doomed". For fusion, that means "There's an already burning fusion device just 8 light minutes southwest, try to make use of that energy source first".
I've some of my own ideas about how to "unify" forces, and I think the main reason why the visible efforts are doomed is that they don't want to abandon the incosistent old theories QM and GR, or any of their consequences. A new model must break predictions made by both of them, not try to merge two inconsistent theories in one framework. Well, anyone knows that all theories are wrong, and better theories are just better approximations, but still wrong. Thinking that one time one find the ultimative theory that represents the world by 100% is wishfull thinking.
We are already at a point where experimental difference to the current theories is difficult to come by, and most of the questions left are rather theoretical. When those theoretical questions are resolved, we may have a theory that - while still being wrong - predicts all observations with enough accuracy that no experiment can be thought of that will show the weak points. But I doubt that. This sort of thinking just prevents people to actually do and publish experiments that differ from theory. A scholastic physics that doesn't look out and do experiments to question theories isn't physics anymore.
I think the chinese government is too nervous about demonstrations, free speach and such like. We have demonstrations here, too (more in the past, when it was about nuclear technology and such things), and some were definitely bigger than the one on tiananmen square.
The sky doesn't fall if many people demonstrate. The Kohl government in Germany was quite stable (in power for 16 years) though there were quite some demonstrations, and Kohl never promised something to the demonstraters. The previous government under Schmidt had some large unsuccessful demonstrations, too, and wasn't overthrown by election but by a small party who left the coalition. The only government in Germany that was ever overthrown by demonstration was the one of the democratic republic of Germany. And what they really wanted was bananas and free journeys to the west. Open up when it's still time is better than waiting until it is too late and then get swept away.
I think the main problem is a cultural one, it's the definition of "losing your face". Here, in Germany, someone who starts shouting or fighting, when he was critisized, is losing his face, because he doesn't keep cool (i.e. admitting a fault or rejecting the critics as void is keeping cool). In China, it seems to be the opposite.
I also think that US people don't know much about China. Deng might not look as a reformer if you compare today's China with the USA, but if you compare it with Mao's China, you definitely see the improvement. Deng killed thousands, Mao killed millions.
The overall human right trend in China is positive, although it's a slow progress. The USA has more of a trend backwards with death penalty and weak children rights, and a political elite that is quite close to christian fundamentalism.
I often hear from China that human rights are a western idea. I beg to differ. They came here about 250 years ago, in a time where China was en vogue, and it isn't surprising that a lot of those idea(l)s can be found in Confuzius and Mencius writings. The people in Taiwan realized that, and are proud of it.
I think the judge took Microsofts own FUD ("Linux Myth page") as arguments against their defeat;-). I think that makes live easy: Microsoft can't argue in court that Linux is a thread while outside court they tell everybody that Linux isn't worth a try.
Since the punishment isn't clear yet, I propose to deny them all the lock-out actions; i.e. force them to give out good API docs (and if they fail to deliver, they'll have to release the source), file formats/network specs and future API plans; keep also device manufacturers responsible for publishing informations (if they fail, but provide a Windows driver, they are aiding illegal actions - if they really want to keep their driver secret, they have to provide it for several popular alternative OSes, too).
I'm not a fan of ESR's "let the market decide". The market already has decided to make M$ a monopoly. As long as I remember, M$ software was worse than the competition, and I remember the days when M$ primarily sold Microsoft BASIC. People choosed uninformed, and they still do. Adam Smith's marked requires high competence on both sides (seller and buyer), in the software industry, we have low competence on both sides.
I'm a coder. I test my programs under all conditions. They crash. That's the purpose of testing them; as it is written in TAOCP: "Errs and errs and errs, but less and less and less". If your faith doesn't crash, you don't test hard enough (especially if you don't adjust it, you are obviously adjusting the test set).
I'm not a natural born atheist. I believed in Santa Claus and the easter bunny, in angels and so on (and certainly in God). I stopped believing in the easter bunny at the age of three, when I found out that actually the parents are doing that. I found out later that the parents also were doing what was supposed Santa Claus' and the angel's work. When I watched a very interesting exposition about Echn'aton, I even realized that this God, the christian one, was just an invention of an egyptean pharao, which got dragged over the mists of time. It took me some time to realize this.
I studied other religions, and found out that there are some without gods (buddhism, taoism), and even further, some without faith (confucianism). Instead of using some mysterious sources for knowing what is good and bad, these religions argue, or even use *rational* arguments to prove their point. They still may be wrong on the premises/results, or don't go deep enough, but a moral that is explained by showing how ethic behaviour is better for you in *this* live on the long run is much more "modern" than christianity with it's pesimistic human picture ("all men are sinner") and it's questionable moral source (a god originally invented by a pharao who had political problems with his priesthood, and still carrying all the legathy of the egyptean after-life believe).
It is much easier to follow Confuzius telling "Do unto others as you want others do unto you", because he explains this sort of optimistic tit-for-tat game. He doesn't explain it in 20th century game theory words, but it's close enough. And this Jesus Christ, 500 years later, has no better reason to explain the same rule by telling you that God loves you?
I'm much happier with Don being a god, and TAOCP being the bible, as with him talking about God and the Bible.
And now let's stop this Atheists slam FUD-spraying Christians (if you don't believe, you'll go to hell) and go back to the Windows Loser slamming!
Oh, come on, the "end user" can't compile Linux today, because he doesn't find the "compile kernel" push-button. The current distributions come with a set of kernels, so you don't have to configure and compile one yourself.
It's like cars that became popular when the end user didn't have to tinker with them (although the Ford model T came with a big manual "How To Drive and *Repair* a Ford Model T"), but up to now, noone welds the hood. You still can tinker if you want. It's the "you don't have to" that is important.
And BTW: the DOS command line isn't powerful, and never was. And I don't care about the preferred tools of fools, since if it's really made for fools, only fools want to use it.
This does not make any sense. Linux was open source from almost the very beginning, and it wasn't very useful back then. It was a tinker-toy, nothing serious (hey, read the announcement! Linus even says it won't be anything big like HURD!). One must look at history, why we are using Linux now.
When Linux was born, the free software community lacked a kernel. Most of the important hacker things were available, compilers, shells, editors, TeX.
The GNU project was slowly developing HURD; it was slow, since all serious hackers already had machines with a kernel on it; may it be SunOS, HP-UX, or whatever; even though it wasn't free, they didn't pay for it. Only 386 machines had such a sucking OS that there was something to do about.
At the same time Linux came around, there was the 386BSD project (which forked into the *BSDs later). At that time, it was at least as exciting as Linux (and more complete), but it went into troubles when AT&T claimed that they had rights on it. Since both systems weren't ready at that time, people rather contributed to Linux instead of 386BSD, which could have been a dead end. When the AT&T struggles were resolved, it was a bit too late. The BSD teams choosed also a more elitist development model, which lead to the well-known splits... Linus most important invention wasn't the Linux kernel, it was the bazaar development model.
If Linux hadn't been free software, noone would have noticed. If it wasn't developed bazaar-style, BSD would have a lead.
I laugh every time I see a brief explanation what Linux is, and it always reads as "Linux was developed early in 1991 by an undergrad student in Helsinki". No, it wasn't. Linux 0.1 was, the latter versions were developed by a team of volunteers all over the world. It's like saying that the Saturn V was developed by Hermann Oberth in his backyard. And we all use computers developed by Konrad Zuse in the living room of his parents.
Please, people, do *not* confuse the probability wave of a particle with the particle itself. And second: Quantum mechanics itself is not very accurate, take quantum electro dynamics (QED) instead. For those who can read, but can't do math, there's a very fine book about QED written by Richard Feynman himself ("QED: The Strange Theory of Light and Matter", for $27.65 at amazon.com; if you are interested in that matter, get it or lend if from your public library soon).
Basically, the "slow spherical wave" creeping out of a radioactive nucleus is the probability function - does it emit now or later? The actual process of emiting the particle is completely unknown to the observer, and he has no means to determine it in advance. The current theories, QED including, completely side-step this issue. They can tell you with high accuracy about the likelyhood of radioactive processes, and they also can tell you with high accuracy about the energy and the path such an emitted alpha particle has, but they can't tell you more in advance. All these theories even have problems with rapidely changing settings of the experiments, because they simply don't talk about that.
Time is no more an illusion than space, energy, or force.
You mean something like the Direct Rendering Infrastructure from Precision Insight (http://www.precisioninsight.com/piinsights.html), included in the forthcoming XFree86 release 4.0? That's much closer to have the application directly talk to the graphic card (through an OpenGL driver, certainly) than Direct3D.
Direct3D doesn't talk directly to the graphic card, it is a layered approach just like everything else - the fact that it also sidesteps GDI doesn't make it better, if anything, it shows how badly designed GDI actually is. Direct3D is even more layered than OpenGL, at least earlier versions (you put commands into a buffer and have them executed instead of executing them right when you need them). Later versions look more like OpenGL, for DirectX 8.0 SGI and Microsoft promised to unify OpenGL and Direct3D.
I mean, how M$-brainwashed can one be to ask for Direct3D on Linux, when top game programmers like John Carmack say "trash Direct3D, take OpenGL" with rage, and even volunteer to make Mesa faster (for G200/G400) than the Windows ICD? M$ is a marketing driven company. They tell you DirectX is the best thing they invented since sliced bread. And rev. 2.0 is a lightyear ahead of 1.0. And now we have rev. 7.0. If it was a honest GNU project, the revision would be very likely 0.7, and the programmers would tell you that it's half-baken and they will add a lot of cool things before it becomes 1.0.
Well, for the suits: There's a common trend in how to wear "formal" dresses (aka "suits"), and that's do what students some time ago weared. The common suit with jacked and tie is something students in Germany invented about 1830 (well, they invented the tie, the short jacked with no sign of silk was because they simply were poor). Expect to be served by a waiter wearing jeans and T-shirt any time soon (I mean in an expensive restaurant).
On the Labour shortage: once the underqualified people go, the pressure will be reduced. It's simply wrong to assume that much helps much. Just as a hundret monkeys won't write Shakespeare in the entire livetime of the universe, a hundret code monkeys can only write a lot of bad code. Call me elitist, but this is the only point I go with the bible: if your hand is ill, you better chop it off than have it infect your whole body. Don't let bad code infect your project.
Tight control of what gets into a project is near to impossible in the typical commercial development process. First, someone is assigned to do a task, and to be a good team player, you have to let him go until the failure is obvious, and if at all, only make small suggestions to overcome the worst mistakes (this lengthens the time to diagnose failure, too).
Then, reviews are often limited to formal ones, i.e. "did we meet all ISO 9001 documentation requirements?". Noone even dares to look at someone else's code, left alone to critizise it. And so on (could write a whole book about it, but someone else already did, see http://www.dilbert.com/).
I really don't buy this. Yes, it's obvious that you are discouraged as geekgirl. But aren't you discouraged as geekboy, too? Isn't it that girls turn away from you when they hear that you are a geek? Isn't that what makes geekboys cry "OH MY GAWD THERE IS A GIRL HERE!!!"? As geek, you have the Dilbert stigma, face it. So telling me that geekgirls don't happen (or happen very rarely) because they are stigmatized... I don't buy that. We all are.
One of the most common prejustice I had to face was that I was "sitting alone before the computer all night long". I wasn't sitting alone there - there were thousands of people I could (and do) talk to. Once the chicks find out, they develop the same (or worse) obsession in sending e-mails and chat and so on, but they don't want to look what's behind all that and learn more than how to operate their favourite programs. Ok, most guys don't want to do more, either. But the minority that wants to know more is much smaller among chicks.
My impression is that this is a hunter/collector genomic thing. Young males have the genomic role as hunter, and hunting is only successful if you are an obsessed hunter; so the ability to become obsessed by hunting something down (e.g. bugs) was an advantage. Collectors (female role) don't need and can't afford to be obsessed; they must always balance effort and result - it's exactly how you describe your feelings: day work is enough. Collecting also allows social activities, because fruits don't run away when you chat while collecting them. This is deeply wired in most of us (nature always allows some variation), so that's a selection for males to become hackers*.
There's also the "reputation game" that's really a boy-only thing. Hackers, according to ESR, are driven by getting peer reputation. Well, that's genetically wired into us to get chicks, but in this case, it goes completely to/dev/null, because the chicks don't know about this game. This, and the case that peer reputation never did help chicks to mate keeps them off. There is no motivation. So we don't see just a 10:1 relation we see in the money-motivated CS industry, we see a 100:1 or a 1000:1 relation between male and female hackers.
So we have at least three factors that limits female penetration in the hacker community: less personality to face stigmatization, less prone to become obsessed by programming, and finally less motivated, because the reputation game doesn't make sense (genetically) for chicks.
*) well, same for hierarchical structures. From our ape ancestor we "learned" (by genetic memory) a whole lot of evil Goa'uld-like tactics to get on the top, like insider relationships, intrigues, and so on. With we, I mean, we males. Female apes stay in their group and have to accept their position (within the female hierarchy); they only can rise by death of a higher rank. Human society has much more flexibility and isn't fixed like ape "societies", but some things are "legacy code" in our DNA we can't get rid of. That's why professors are mostly male, even of professions where students are mostly female. You don't become professor because you are good. You become professor because you are good enough, have insider relationships and can intrigue against your competition.
> Are there any programs like "Mavis Beacon Teaches Typing" that will teach Dvorak?
I did a short search and find a program for the Mac (I don't have one), and a tcl/tk program KP (ftp://stampede.cs.berkeley.edu/pub/kp/kp-0.97.tar .gz), the keyboard practicer. Well, KP doesn't have lesson files, and therefore is pretty worthless. Learning to type without prepared lessons is hard to do. KP also is a bit annoying, since it doesn't simply map keysyms to dvorak keycodes, but depends on a US keyboard layout.
I've tried about half an hour, and got to 10-15 wpm on one of my own (english, technical) texts. I didn't feel all too comfortable (ok, first half hour;-), because I had to move away from the home row quite often - not nearly as often as with QWERTY, but more often than it deemed me neccesary.
I decided to do a histogram over my texts, because that would reveal which letters *I* really need on the home row. The result was (most frequent on middle finger, next on index finger, then ring finger, small finger, index finger one off, and for right hand small finger one off):
My english texts: oaeiu cstnrl My german texts: uaeio hrntsd My programs: iaeou drtsnl Hitchhikers guide: ioeau rnthsd Original Dvorak: aoeui dhtns-
I used a copy of "The Hitchhikers Guide to the Galaxy" to check against a real native speaker, and a non-technical text (my german texts are non-technical, too). I miss the r and l keys on the home row, but I think I could go with standard Dvorak, though. The other question is, do I have the time to do the training? Uhm, I doubt.
Slashdot Mirror
The gun control arguments are a strong case that indicates US citicens think violence is an answer. It isn't. The Taliban almost won a civil war. I don't think they did really "outlaw" guns, they just collected them to use them against their remaining enemies. They just think that violence is an answer. Also note that it's impossible to collect guns in a country where most people live in villages and have a million holes in the mountains to hide their own guns if they did want to keep them. The Taliban certainly got only guns from supporters.
Also, the argument "strict gun control means that only criminal have guns" is wrong. Europe generally has what is consideret "strict gun controls", yet million people here have access to firearms: soldiers and police, security services, hunters, people practicing firearm sports, people who convinced the authorities that they need guns to protect themselves. Sometimes, people arm themselves even though they aren't allowed to (e.g. in Munich, a gas station's owner shot a policeman, who robbed gas stations in his spare time, using his police gun).
However, the overall picture of Europe is that the general crime level doesn't differ from the crime level in USA that much (though we don't put 3% of our adults into jail), while murder, especially using firearms, is way down. Between Washington DC and Brussel, there's a factor of 170! And that's though Washington DC by the letter even has firearm control (doesn't help - it's just on paper, and nobody checks at the "borders").
I don't think that strict gun control (in our, European sense) would bring down US murder rate to a civilized level. Guns don't kill people, people kill people. And only severely deranged people are able to kill other people. However, the arguments exchanged in gun control discussion show that too many US people are somewhat deranged. They actually considder using deadly forces to protect them.
The final point however is, that if you think using guns against oppressive governments is ok, you agree that terrorism is ok. Well, you say "freedom fighter" instead of terrorism, but that's it.
Guns do not equal encryption. Encryption is just an envelope that keeps others from reading your mail. Encrypting is completely free of any violence, and still provides a mean of security (it allows secure transactions of whatever you think off: TANs for online banking, love letters you write within the company network and don't want your supervisor to be read, company secrets you don't want to expose to industry spies like Echelon, and so on). Encrypting secures your basic rights.
Yes, you can use encryptiong to coordinate efforts that do cause harm, but it's not the encrypted e-mail that crashed the WTC, it was carpet knifes, planes, and kerosine. Nobody would want to outlaw these, because carpet knifes, planes, and kerosine obviously are almost always used for good.
I've written a web server in 200 lines of Forth (runs on GNU Forth, is part of the gforth-0.5.0 package). It's not an overfeatured web-server like Apache, it doesn't do SSL, doesn't allow any other language for server-side scripting than Forth, but I don't need that. If you want Apache, you can have Apache. Read more about it in http://jwdt.com/~paysan/httpd-en.html
Tip 1: To break the rules you must know them. It's much less fun to break rules you aren't aware off. And tell people when you broke the rules - if nobody's aware of you breaking rules, WTF are the rules for then?!
/.. You know, mainstream sucks, it's for Windows-programming morons only. Read out-of-print books like "Thinking Forth" (really a gem, no fun!).
Tip 2: Don't use other code - it's bad, since you are the master programmer, and the others are just morons, and you'd have to understand it to use it. You know: if you want it done right, you have to do it yourself.
Tip 3: Don't listen to customer, they are stupid morons, anyway. Write the program as you want to use it. Then brag about how c00l and 3l11t it is, then the customers will buy it anyway. If not, release it under GPL, and go IPO as new open source company.
Tip 4: Don't write comments into your programs. Your code is obvious. Since you assume the other people know about rule 2, they won't read your code anyway.
Tip 5: If you write comments to break rule 4, write them as fast as possible and with ltos of tyops, so that oepple kan now that you are busy duing codink.
Tip 6: Don't read mainstream best-seller coding practice books reviewed by
Hey, this topic comes up in /. once in a while, when another fool gets his ToE (Theory of Everything) into the light of /.. These sort of theory is easy to make a fool out of you. A lot of people tried, and all of them weren't rewarded, even Einstein failed.
To me, each of the ToEs, including the string theory, sounds a lot like crackpot stuff. 11 dimensional vibrations of strings/membranes? Believers quoted with "The big bang is a fact, not a theory" - did they participate in that event? It's cool (and intellectualy chalanging) math, but does it reflect or model reality?
The final ToE will very likely look like crackpot stuff at the first sight, too. But unlike the crackpot stuff we have today, it will be possible to make experimental proofs.
I personally believe that it will look much less spectacular than cold fusion, supersymmetric particles and such like. There are already a lot of self-contradictory results of current theories that a good ToE will get rid off, like black holes (no light escapes - how does gravity escape?).
So the theories we have now are bad enough, but at least they are somewhat in consistency to our observations. I prefer the problems of QM and GR to nutty "star trek" phy"sick"s any day. Too many people trekkify the classic theories, anyway (FTL travel, quantum teleportation, using black holes for time travels, etc... Cut their funding!).
I don't make money with free software (at least not up to now). I make money with bloody proprietary software (and with bloody casino capitalism stock market speculations).
;-).
I give back to the community by writing free software. Don't tax me, I've already given more than 10% of my working time back to the community. Church taxing monks doesn't make sense, eh?
And if you spent this money just on lawyers to defend free software, there's something terribly wrong. Laywers are utterly unproductive. I know, they are part of the Golgafrinchina's load of unproductive people who crashed here, but they still are unproductive.
I know, in the US, you can get almost any result out of a trial, if you pay enough to the right lawyer (good price/performance: OJ Simpson's lawyer, bad price/performance: Bill Gates' "we have our own render farm to generate faked videos" DoJ trial lawyers). I don't understand why you need a legal system if it just manages to keep an anarchistic state (the right of the stronger). Well, perhaps because the lawyers are strong enough
Well, Columbus was a crackpot. He read Marco Polo's story about China (or more likely from a second source), and took Polo's venetian miles of foot march as sea miles of direct distance, and concluded that China ("India" - he really must have been a crackpot) was some few thousand sea miles in the west. All the sane people knew that China is 20000 sea miles in the west, and noone could get there with 15th century technology. If it wasn't for the west-indian islands, Columbus won't have survived.
To the population of these west-indian islands, and to whole Amerika, this discovery by religious fanatists turned out as a real disaster. Columbus and his serfs killed most of the inhabitants of the Isabella island, and the gold-drunken conquistadores didn't do any better to the Aztecs and Incas.
This is really an example for an exploration that never should have happend that way. BTW: America was discouverd before by several expeditions, which all concluded that it wasn't worth the risk of crossing the ocean, and then again forgotten.
I really don't understand this. The whole point of life is to give life, to create new life. Usually of our own kind, but since we do sexual breeding, it is a new combination. It's risky. The parents of Hitler and Stalin created life that killed many million peoples.
Some religions actually discourage breeding, at least for their most "holy" members. Many tell you that sex is something bad. Some even tell you that their most holy person wasn't originated by sex, but created by a god.
Come on, they are not experts. They have no idea what life is about. Many don't want to accept this world, and wait for a better world after death.
Scientists have to calculate the risks themselves. It's already risky to create new life by interbreeding, you might create killer bees or killer dogs or killer bakterias when you do that.
It's more risky when you purposly modify plants to create poisonous substances. You know they are evil, but you don't know where they will get.
When you create new life forms, do that with care. I don't think a toy life form with 300 genes will likely survive long in a world where lots of other life wants to eat, but I can't know.
There's a humorous film where a scientist creates a bacteria that composts dry paper. Well, it escapes, and does compost paper. In fact, all paper available. The result is a bit like the After-Y2K comic.
The GPL doesn't steal the 'IP' of the original authors; they still can take the sources, close the improvements and sell the product (or sell it, if the use of the product is limited in a GPL distribution, like the one of libraries or worse, incremental compilers).
The GPL regards a closed source fork by a third party (i.e. not the original authors) as theft. It doesn't differ from Bill Gates view here. It is quite clear to everyone that taking public commons is theft, too.
And finally, I can't believe that people still question whether the other software could be written open sourced, when the greatest things that require the most effort already are open sourced. Get out through the doors, don't let your view be filtered by windows!
Grammar+vocabulary based translation is doomed. The main problem is that natural language neither have strict grammar nor real vocables. I.e. what you have is a somewhat loose grammar, and ambiguous words that often have a lot of different meanings, depending on context.
The problem of translation is that the classes of meanings are different in different languages. If people would only write unambiguous sentences, translation would be easy, but often enough, people use ambiguousities deliberately (jokes, oracles, poems, etc.). So it isn't enough to deduce the meaning of a sentence/paragraph/text by looking at the context, but also you have to carry the remaining ambiguousity into the target language. And that's one of the harder parts of translation.
I'm talking about translation by humans, not by machines here, that's what is difficult enough for humans. I don't expect a good machine translation any time soon, as often enough human translations are too bad. Literature, with rhymes and tone has much more problems than just meaning, and here good translations are often impossible.
So IMHO translation can't be done using string-based rules. You have to collect "meanings" from the words, the grammar rules and the vocabulary files can help you, but after all, you have to search for a sentence in the target language that matches the remaining set of meanings of the source language sentence best.
Well, the math books would still be full of the same letters as before, since math doesn't tell you anything about this world. Math is perfect, because it makes up its own assumptions (axioms).
BTW evolution: "evolution" isn't just one theory, it is a set of theories. Darwin's theory is just one evolution theory, it bases evolution on mutation and selection by survival of the fitest. Lamark's theory is another evolution theory, and it's based on something like a "genetic memory".
Actually, we know a lot more about genetics now than both Darwin and Lamark knew. We know that Lamark isn't completely off, since there are ways to exchange genetic informations, including gene ferries (mostly for bacterias and plants), and gene editors. We don't know if these gene editors are just deactivated retro virii, or if they are used once upon a time, but one thing we can say with pretty high accuracy: this live as it is now wasn't created, but evolved. Even the pope accepts this.
There is no observation whatsoever that any species ever was created. All creationist arguments I know of focus around "proofs" that this or that can't have evolved, mostly by taking something complex as the eye, and leaving out all the steps that led to that eye as it is now, and even failing to explain why the human eye has the sensors behind supply and sense, instead of having it in front of it like octopus eyes do.
Creationism is a "last resort explanation" - if everything else fails, assume it was created. It's also only shifting the problem, since it explains something lesser (the world) by being created by something higher (God), and completely misses to explain how this higher (God) came into being. He for sure hasn't evolved? So who created God?
In all my school days, only one teacher told us things as if they were facts, and that one was the religion teacher of elementary school. After all, her facts weren't based on observation or by formal proof, but on reading of just one really old book.
And so we are back on topic: why do teachers tell things from old books as if it was truth and no new insight has been gained later? Well, it's because teachers never left school. They haven't seen the real world. They aren't "authentic". For them, all knowledge is like a fairy tale, something that goes from generation to generation. It isn't. Knowledge is but a tool to get around with the real world.
What pupils really need, apart from meta-tools like letters and math, is to learn how to solve real world problems, problems they'll face in their future life. Today, the most important thing is to learn how to learn. I mean: to learn yourself, without being driven by a teacher. This isn't a budget problem, it's a cultural problem. Few, if any teachers like it if pupils know more than them. Many even dislike if they knew more than they have been teached in school yet. This must stop!
The GPL simply disallows to restrict the right of redistribution. Thus if you get a Corel Linux CD, you can redistribute the GPL'd programs including source (you can get the source CD from Corel for shipping&handling - if they fail to do so, sue them out of this Linux business). The CD itself is a mere aggregation, and therefore not bound under GPL. They just can't forbit you to dissect it and redistribute the parts that are under GPL (if they do so, their rights under GPL stop and they go out of this Linux business). Anyone can create a "free Corel CD" out of it by omitting all the software that isn't redistributable under an OSS license, and give that to everybody.
This is just a stupid, associal EULA. It's a reason to boycott Corel, but not one for legal actions. They have associal clueless lawyers (who's more associal than someone who disciminates against children?) who should be fired.
Please, people at Corel, wake up. Read the EULAs the jerks in the legal department put out. They do harm to you. This is not in your interest.
I work quite constantly a 40 hour week. I don't think overtime is anything good, since you don't work concentrated anymore. IMHO 40 hours work per week is already too much, you can't work concentrated for more than 4 hours a day on a day per day basis - and working unconcentrated while reading slashdot is just a waste of time. I could do that at home, too.
Waiting for the machine is a nuisance, and should be avoided by getting faster hardware or faster software. Meetings should not dominate your workload, either.
I've seen code written in 60 hour weeks. It took us a lot of weeks to get the bugs out of it. The human brain isn't a ship's diesel, which has a standard output and keeps that forever - the brain tires, quite fast so. To paraphrase Brooks "Throwing overtime at a late project makes it later".
Therefore I prefer salary, it gives me no advantage on overtime, so I care about not accumulating overtime myself. The amount of work done beyound 50 hours per week is negative (yes, *negative*), even in non-IT studies (I've one of WW II England, working on ammution). It's much worse in the IT field, the amount of work isn't only negative, it's also of much worse quality.
Most of the quality of free software comes IMHO from the fact that the volunteers working on such projects do naturally make breaks when they get unconcentrated, because it's no longer fun to continue coding. And most of the miserable quality of commercial software is resulted from 70 hours weeks. Hey, it looks all like work of cafeine junkies and has to be thrown away (Mozilla, anyone?). It ain't done until it's ready, and overwork harms both you and the project.
Well, I've solved my geek-gift problem quite simply: I buy the geekish stuff for me myself (books before christmas, hardware afterwards), and let the gifter only choose the wrapping. A wish-list just produces questions like "What's that? Where am I supposed to buy such a thing?" etc. If the gifter insists on a surprise present, let him/her choose something non-geeky, i.e. winter clothes, sweats, and such.
;-) (100% Y2K proof, because the chinese calendar has year 4696 now - cycle 78, year 16 to be precise).
All I saw on the discussion is perfectly on that list. A elvish ring that is supposed to be ordered via Internet? A special swiss army knife that is sold-out? Books that have "Computer" *and* "Programming" in the title, but neither "Dummy" nor "Windows", and therefore isn't available at the local book store? Come on, no non-geek can get such a present for you.
BTW: What about an original, 100% Y2K proof chinese abacus
The observer mystery is a common myth about QM, and comes from the problem to understand the wave-particle dualism. Things can show particle or wave behaviour, and when you force them to show particle behaviour, the wave function collapses.
What really does that is an "observation" device (like a CRT screen, or the fovia of the eye), not the "observer". You can put such a device anywhere in your experiment, and you can prove that it really does the same as any other "observer" does - reduce the accuracy of the measurement according to the uncertaincy relation.
For Schroedinger's cat that means, the alpha or beta particle is the "observer", because it already turns the possibility to radiate into a fact (and the scintillation counter does the rest).
My father worked once in fusion research (the hot fusion one), that was back around 1970. While first, people though it won't take too long to make a working fusion power plant, they quickly raised the time estimation to "50 years", and that remained constant over the last 30 years (we are still 50 years off a working fusion power plant).
So if any physicist says to you "in 50 years", he means "I have no idea how to achieve that", or worse "According to all informations I have, the effort is doomed". For fusion, that means "There's an already burning fusion device just 8 light minutes southwest, try to make use of that energy source first".
I've some of my own ideas about how to "unify" forces, and I think the main reason why the visible efforts are doomed is that they don't want to abandon the incosistent old theories QM and GR, or any of their consequences. A new model must break predictions made by both of them, not try to merge two inconsistent theories in one framework. Well, anyone knows that all theories are wrong, and better theories are just better approximations, but still wrong. Thinking that one time one find the ultimative theory that represents the world by 100% is wishfull thinking.
We are already at a point where experimental difference to the current theories is difficult to come by, and most of the questions left are rather theoretical. When those theoretical questions are resolved, we may have a theory that - while still being wrong - predicts all observations with enough accuracy that no experiment can be thought of that will show the weak points. But I doubt that. This sort of thinking just prevents people to actually do and publish experiments that differ from theory. A scholastic physics that doesn't look out and do experiments to question theories isn't physics anymore.
I think the chinese government is too nervous about demonstrations, free speach and such like. We have demonstrations here, too (more in the past, when it was about nuclear technology and such things), and some were definitely bigger than the one on tiananmen square.
The sky doesn't fall if many people demonstrate. The Kohl government in Germany was quite stable (in power for 16 years) though there were quite some demonstrations, and Kohl never promised something to the demonstraters. The previous government under Schmidt had some large unsuccessful demonstrations, too, and wasn't overthrown by election but by a small party who left the coalition. The only government in Germany that was ever overthrown by demonstration was the one of the democratic republic of Germany. And what they really wanted was bananas and free journeys to the west. Open up when it's still time is better than waiting until it is too late and then get swept away.
I think the main problem is a cultural one, it's the definition of "losing your face". Here, in Germany, someone who starts shouting or fighting, when he was critisized, is losing his face, because he doesn't keep cool (i.e. admitting a fault or rejecting the critics as void is keeping cool). In China, it seems to be the opposite.
I also think that US people don't know much about China. Deng might not look as a reformer if you compare today's China with the USA, but if you compare it with Mao's China, you definitely see the improvement. Deng killed thousands, Mao killed millions.
The overall human right trend in China is positive, although it's a slow progress. The USA has more of a trend backwards with death penalty and weak children rights, and a political elite that is quite close to christian fundamentalism.
I often hear from China that human rights are a western idea. I beg to differ. They came here about 250 years ago, in a time where China was en vogue, and it isn't surprising that a lot of those idea(l)s can be found in Confuzius and Mencius writings. The people in Taiwan realized that, and are proud of it.
I think the judge took Microsofts own FUD ("Linux Myth page") as arguments against their defeat ;-). I think that makes live easy: Microsoft can't argue in court that Linux is a thread while outside court they tell everybody that Linux isn't worth a try.
Since the punishment isn't clear yet, I propose to deny them all the lock-out actions; i.e. force them to give out good API docs (and if they fail to deliver, they'll have to release the source), file formats/network specs and future API plans; keep also device manufacturers responsible for publishing informations (if they fail, but provide a Windows driver, they are aiding illegal actions - if they really want to keep their driver secret, they have to provide it for several popular alternative OSes, too).
I'm not a fan of ESR's "let the market decide". The market already has decided to make M$ a monopoly. As long as I remember, M$ software was worse than the competition, and I remember the days when M$ primarily sold Microsoft BASIC. People choosed uninformed, and they still do. Adam Smith's marked requires high competence on both sides (seller and buyer), in the software industry, we have low competence on both sides.
I'm a coder. I test my programs under all conditions. They crash. That's the purpose of testing them; as it is written in TAOCP: "Errs and errs and errs, but less and less and less". If your faith doesn't crash, you don't test hard enough (especially if you don't adjust it, you are obviously adjusting the test set).
I'm not a natural born atheist. I believed in Santa Claus and the easter bunny, in angels and so on (and certainly in God). I stopped believing in the easter bunny at the age of three, when I found out that actually the parents are doing that. I found out later that the parents also were doing what was supposed Santa Claus' and the angel's work. When I watched a very interesting exposition about Echn'aton, I even realized that this God, the christian one, was just an invention of an egyptean pharao, which got dragged over the mists of time. It took me some time to realize this.
I studied other religions, and found out that there are some without gods (buddhism, taoism), and even further, some without faith (confucianism). Instead of using some mysterious sources for knowing what is good and bad, these religions argue, or even use *rational* arguments to prove their point. They still may be wrong on the premises/results, or don't go deep enough, but a moral that is explained by showing how ethic behaviour is better for you in *this* live on the long run is much more "modern" than christianity with it's pesimistic human picture ("all men are sinner") and it's questionable moral source (a god originally invented by a pharao who had political problems with his priesthood, and still carrying all the legathy of the egyptean after-life believe).
It is much easier to follow Confuzius telling "Do unto others as you want others do unto you", because he explains this sort of optimistic tit-for-tat game. He doesn't explain it in 20th century game theory words, but it's close enough. And this Jesus Christ, 500 years later, has no better reason to explain the same rule by telling you that God loves you?
I'm much happier with Don being a god, and TAOCP being the bible, as with him talking about God and the Bible.
And now let's stop this Atheists slam FUD-spraying Christians (if you don't believe, you'll go to hell) and go back to the Windows Loser slamming!
Oh, come on, the "end user" can't compile Linux today, because he doesn't find the "compile kernel" push-button. The current distributions come with a set of kernels, so you don't have to configure and compile one yourself.
It's like cars that became popular when the end user didn't have to tinker with them (although the Ford model T came with a big manual "How To Drive and *Repair* a Ford Model T"), but up to now, noone welds the hood. You still can tinker if you want. It's the "you don't have to" that is important.
And BTW: the DOS command line isn't powerful, and never was. And I don't care about the preferred tools of fools, since if it's really made for fools, only fools want to use it.
This does not make any sense. Linux was open source from almost the very beginning, and it wasn't very useful back then. It was a tinker-toy, nothing serious (hey, read the announcement! Linus even says it won't be anything big like HURD!). One must look at history, why we are using Linux now.
When Linux was born, the free software community lacked a kernel. Most of the important hacker things were available, compilers, shells, editors, TeX.
The GNU project was slowly developing HURD; it was slow, since all serious hackers already had machines with a kernel on it; may it be SunOS, HP-UX, or whatever; even though it wasn't free, they didn't pay for it. Only 386 machines had such a sucking OS that there was something to do about.
At the same time Linux came around, there was the 386BSD project (which forked into the *BSDs later). At that time, it was at least as exciting as Linux (and more complete), but it went into troubles when AT&T claimed that they had rights on it. Since both systems weren't ready at that time, people rather contributed to Linux instead of 386BSD, which could have been a dead end. When the AT&T struggles were resolved, it was a bit too late. The BSD teams choosed also a more elitist development model, which lead to the well-known splits... Linus most important invention wasn't the Linux kernel, it was the bazaar development model.
If Linux hadn't been free software, noone would have noticed. If it wasn't developed bazaar-style, BSD would have a lead.
I laugh every time I see a brief explanation what Linux is, and it always reads as "Linux was developed early in 1991 by an undergrad student in Helsinki". No, it wasn't. Linux 0.1 was, the latter versions were developed by a team of volunteers all over the world. It's like saying that the Saturn V was developed by Hermann Oberth in his backyard. And we all use computers developed by Konrad Zuse in the living room of his parents.
Please, people, do *not* confuse the probability wave of a particle with the particle itself. And second: Quantum mechanics itself is not very accurate, take quantum electro dynamics (QED) instead. For those who can read, but can't do math, there's a very fine book about QED written by Richard Feynman himself ("QED: The Strange Theory of Light and Matter", for $27.65 at amazon.com; if you are interested in that matter, get it or lend if from your public library soon).
Basically, the "slow spherical wave" creeping out of a radioactive nucleus is the probability function - does it emit now or later? The actual process of emiting the particle is completely unknown to the observer, and he has no means to determine it in advance. The current theories, QED including, completely side-step this issue. They can tell you with high accuracy about the likelyhood of radioactive processes, and they also can tell you with high accuracy about the energy and the path such an emitted alpha particle has, but they can't tell you more in advance. All these theories even have problems with rapidely changing settings of the experiments, because they simply don't talk about that.
Time is no more an illusion than space, energy, or force.
You mean something like the Direct Rendering Infrastructure from Precision Insight (http://www.precisioninsight.com/piinsights.html), included in the forthcoming XFree86 release 4.0? That's much closer to have the application directly talk to the graphic card (through an OpenGL driver, certainly) than Direct3D.
Direct3D doesn't talk directly to the graphic card, it is a layered approach just like everything else - the fact that it also sidesteps GDI doesn't make it better, if anything, it shows how badly designed GDI actually is. Direct3D is even more layered than OpenGL, at least earlier versions (you put commands into a buffer and have them executed instead of executing them right when you need them). Later versions look more like OpenGL, for DirectX 8.0 SGI and Microsoft promised to unify OpenGL and Direct3D.
I mean, how M$-brainwashed can one be to ask for Direct3D on Linux, when top game programmers like John Carmack say "trash Direct3D, take OpenGL" with rage, and even volunteer to make Mesa faster (for G200/G400) than the Windows ICD? M$ is a marketing driven company. They tell you DirectX is the best thing they invented since sliced bread. And rev. 2.0 is a lightyear ahead of 1.0. And now we have rev. 7.0. If it was a honest GNU project, the revision would be very likely 0.7, and the programmers would tell you that it's half-baken and they will add a lot of cool things before it becomes 1.0.
Well, for the suits: There's a common trend in how to wear "formal" dresses (aka "suits"), and that's do what students some time ago weared. The common suit with jacked and tie is something students in Germany invented about 1830 (well, they invented the tie, the short jacked with no sign of silk was because they simply were poor). Expect to be served by a waiter wearing jeans and T-shirt any time soon (I mean in an expensive restaurant).
On the Labour shortage: once the underqualified people go, the pressure will be reduced. It's simply wrong to assume that much helps much. Just as a hundret monkeys won't write Shakespeare in the entire livetime of the universe, a hundret code monkeys can only write a lot of bad code. Call me elitist, but this is the only point I go with the bible: if your hand is ill, you better chop it off than have it infect your whole body. Don't let bad code infect your project.
Tight control of what gets into a project is near to impossible in the typical commercial development process. First, someone is assigned to do a task, and to be a good team player, you have to let him go until the failure is obvious, and if at all, only make small suggestions to overcome the worst mistakes (this lengthens the time to diagnose failure, too).
Then, reviews are often limited to formal ones, i.e. "did we meet all ISO 9001 documentation requirements?". Noone even dares to look at someone else's code, left alone to critizise it. And so on (could write a whole book about it, but someone else already did, see http://www.dilbert.com/).
I really don't buy this. Yes, it's obvious that you are discouraged as geekgirl. But aren't you discouraged as geekboy, too? Isn't it that girls turn away from you when they hear that you are a geek? Isn't that what makes geekboys cry "OH MY GAWD THERE IS A GIRL HERE!!!"? As geek, you have the Dilbert stigma, face it. So telling me that geekgirls don't happen (or happen very rarely) because they are stigmatized... I don't buy that. We all are.
/dev/null, because the chicks don't know about this game. This, and the case that peer reputation never did help chicks to mate keeps them off. There is no motivation. So we don't see just a 10:1 relation we see in the money-motivated CS industry, we see a 100:1 or a 1000:1 relation between male and female hackers.
One of the most common prejustice I had to face was that I was "sitting alone before the computer all night long". I wasn't sitting alone there - there were thousands of people I could (and do) talk to. Once the chicks find out, they develop the same (or worse) obsession in sending e-mails and chat and so on, but they don't want to look what's behind all that and learn more than how to operate their favourite programs. Ok, most guys don't want to do more, either. But the minority that wants to know more is much smaller among chicks.
My impression is that this is a hunter/collector genomic thing. Young males have the genomic role as hunter, and hunting is only successful if you are an obsessed hunter; so the ability to become obsessed by hunting something down (e.g. bugs) was an advantage. Collectors (female role) don't need and can't afford to be obsessed; they must always balance effort and result - it's exactly how you describe your feelings: day work is enough. Collecting also allows social activities, because fruits don't run away when you chat while collecting them. This is deeply wired in most of us (nature always allows some variation), so that's a selection for males to become hackers*.
There's also the "reputation game" that's really a boy-only thing. Hackers, according to ESR, are driven by getting peer reputation. Well, that's genetically wired into us to get chicks, but in this case, it goes completely to
So we have at least three factors that limits female penetration in the hacker community: less personality to face stigmatization, less prone to become obsessed by programming, and finally less motivated, because the reputation game doesn't make sense (genetically) for chicks.
*) well, same for hierarchical structures. From our ape ancestor we "learned" (by genetic memory) a whole lot of evil Goa'uld-like tactics to get on the top, like insider relationships, intrigues, and so on. With we, I mean, we males. Female apes stay in their group and have to accept their position (within the female hierarchy); they only can rise by death of a higher rank. Human society has much more flexibility and isn't fixed like ape "societies", but some things are "legacy code" in our DNA we can't get rid of. That's why professors are mostly male, even of professions where students are mostly female. You don't become professor because you are good. You become professor because you are good enough, have insider relationships and can intrigue against your competition.
> Are there any programs like "Mavis Beacon Teaches Typing" that will teach Dvorak?
r .gz), the keyboard practicer. Well, KP doesn't have lesson files, and therefore is pretty worthless. Learning to type without prepared lessons is hard to do. KP also is a bit annoying, since it doesn't simply map keysyms to dvorak keycodes, but depends on a US keyboard layout.
;-), because I had to move away from the home row quite often - not nearly as often as with QWERTY, but more often than it deemed me neccesary.
I did a short search and find a program for the Mac (I don't have one), and a tcl/tk program KP (ftp://stampede.cs.berkeley.edu/pub/kp/kp-0.97.ta
I've tried about half an hour, and got to 10-15 wpm on one of my own (english, technical) texts. I didn't feel all too comfortable (ok, first half hour
I decided to do a histogram over my texts, because that would reveal which letters *I* really need on the home row. The result was (most frequent on middle finger, next on index finger, then ring finger, small finger, index finger one off, and for right hand small finger one off):
My english texts: oaeiu cstnrl
My german texts: uaeio hrntsd
My programs: iaeou drtsnl
Hitchhikers guide: ioeau rnthsd
Original Dvorak: aoeui dhtns-
I used a copy of "The Hitchhikers Guide to the Galaxy" to check against a real native speaker, and a non-technical text (my german texts are non-technical, too). I miss the r and l keys on the home row, but I think I could go with standard Dvorak, though. The other question is, do I have the time to do the training? Uhm, I doubt.