Lots of people will recommend 'Learning Perl', but I strongly recommend not buying that book. Unless you have problems falling asleep, then this book will be useful.
Well, that book put me to sleep, but I then handed it to a colleague who needed to learn perl, and never did get it back. For what it's worth, that person was an intelligent non-programmer. Apparently that book ended up on a Great Journey and has landed on the desks and shelves of many people just like my so-called, book-stealing friend.
I've more than 15 Perl and Perl related books in my bookcase, but there are two I've actually found useful: Mastering Regular Expressions, and The Perl Cookbook.
I think the only perl books I don't have now but might want (I have these two) are Perl in a Nutshell (looks like a really handy reference) and possibly Object Oriented Perl, because it was written by Damian Conway, whose papers and projects I have found very useful in the past.
C and C++ share a declaration syntax that, I think it is fair to say, is an acquired taste at best. Damian Conway, an OO programming researcher in Australia has published papers on a Modest proposal to resyntax C++ with a graduate student of his, B. M. Werther.
The result was a language with C++'s semantics (at that time) and a much cleaner syntax. Which, of course, went nowhere. Do you have any comments on SPECS itself or any other attempts to improve what might be called the "human factors" aspects of the C++ language?
These words are wonderful, yet they are falling on deaf ears. Yes, of course WE know that this must be stopped - but what does the average consumer know about this? Nothing. Guess who's voting these people into office: the average consumer. Net result: we can clamor all we want, but until public opinion changes en masse, we will have accomplished nothing.
I think your sentiment is simultaneously too pessimistic and too idealistic at the same time.
I don't think that the history of legislation is primarily made up of cases where laws are passed or not passed due to a broad-based public consensus on the content of the laws themselves. The beautiful thing about representative democracy is precisely the fact that change can occur due to the actions of very small numbers of people. In otherwords, because some groups of people can gain influence disproportionate to their actual numbers.
And this, of course, is exactly why lack of change is the rule: when there is no very broad consensus, politicians are loathe to side with one interest group or another unless they can see a clear, one-sided benefit for themselves and their own careers. Donations to campaigns are one obvious benefit, as are winning (or keeping) the allegiance of an entire block of voters. But getting money for casting a vote isn't a great plan if the vote itself will later be used as the basis for negative campaigning by an opponent. In this case, the attack ads for voting in favor of shrink-wrap licenses practically write themselves.
Moreover, I expect that, to the average politician, angering even a small minority of net-savvy techno-geeks who make it clear they will back up what they say would be viewed as a really suspect thing to do unless it delivered a truly huge number of votes or dollars. Your state senator might not know what a EULA is, but if he or she knows that voting to strengthen them is going to lead to direct political attacks amplified Lord knows how much by that Internet thing, you can be sure that doing nothing will look much more attractive.
So, yes, action is needed, and education is always a good thing, but I'm far more optimistic that an action on something like this can protect the status quo without anything like a majority of citizens, or even the internet-aware citizens, knowing about it. Vocal minorities are way louder than you would think.
For example, in Perl (like most scripting languages) there is a single data type to represent everything from strings to characters to all sorts of numbers.
There are many reasons why Perl may or may not be a good first language, but this can't be one of them, since it just isn't true. TCL used to be the textbook example of a language that used the string as its representation for everything, but TCL isn't string only anymore, either.
Really, I don't know where people get this idea about Perl, which certainly isn't the most typeful language on the block, but is hardly short of interesting types. I suppose it comes from the fact that Perl does provide a lot of automagical operators and conversions (although the fact that conversions are involved should make it clear that Perl has more than one underlying type).
But one thing I'd like to point out in particular:
Java/C++/C/Fortran/Basic etc...have different data types (short, long, char, String etc...) for very good reasons.
Says somebody who clearly hasn't ever had to justify the difference between a short and long to a beginning (or non-)programmer. I would actually argue that the short/long difference is exactly the kind of "people working for the computer" stuff that doesn't belong anywhere near somebody's first exposure to programming.
If you want to teach the importance of types at the same time you teach programming, you would do better to use a language that has a more interesting and flexible type system; one obvious candidate would be Haskell, or it's interpreted cousin Hugs.
As for the assertion that neither of these teams has helped the local economies, that's clearly absurd.
It is? Tell me: do you have any numbers to back this up? I ask, because people who have looked at the econmic impact of publicly funded stadiums in particular, or the presence of sports teams in general, have found essentially no evidence of any net benefit. To be sure, there are people who make more money due to the presence of a heavily-subsidized sports franchise, but there are others who do worse, and there are also huge opportunity costs as well.
And even more than that, getting to a Super Bowl unites a city and makes it exciting to live in (if St. Louis can ever be called that...but that's another story) in important if not economically quantifiable games.
s/games/ways/?
But, actually, this is just the problem: the cash that St. Louis just basically gave to the Rams most certainly was economically quantifiable, and the other very pressing needs of St. Louis are also economically quantifiable, so the idea of spending that much cash on something just to "make it exciting to live in" goes way past stupid and borders on the immoral.
I'm really sorry, but I'm curious why it has a Score 5 (I'm not bitching, I just want to know why it's so funny:P)
It's very simple. In this case, 3 moderators decided at roughly the same time that the post in questions deserved to be a 'Score:2, Informative', while a fourth decided an instant later that it should be a 'Score:2, funny'; pow: it's a 'Score:5, funny'.
Or, in other words, Slashdot's RDBMS backend doesn't use place moderation updates inside a transaction. This should be simple enough to fix unless they're silly enough to use a database that doesn't support transactions. But who would do something silly like that?:-)
If I'm right about this, the Postgresql folks are entitled to one collective "nyeah, nyeah, we told you so!" on this topic.
Perhaps a more important effect of the internet is the way news and rumors get around. I've heard talk show hosts bring up rumors and facts they've gotten from the internet often enough. We're a minority, but we're an influential one.
But in fact, it's hardly just one "we" anymore, at least compared to the Net culture of old (and not to imply that the old days were one big happy family or anything...).
What I find much more potentially interesting about certain internet forums (Usenet groups, things like Slashdot, other chunks of the web) is that they are automatically archived, and reasonably searchable. In that kind of environment, it is
easier to cite some kinds of evidence directly (it makes more sense to ask somebody for their sources)
easier to find what you think could be relevent in a political message
easier to point out that somebody really did say something that they deny having said.
All in all, it's very possibly a politician's worst nightmare.:-)
Microsoft did *not* lose an appeal here; the Supreme Court simply refused to hear the case--as it does with 99% of the cases that it is offered.
This generally gives *no* indication as to how the court would rule on the issue.
Well, my inner geek then asks: what is the proportion of cases that the court hears where it does overturn an appeals court ruling? I guess I did know that the 9th circuit is the one most likely to get whacked, so I wonder whether this actually bodes well for the ruling, a copy of which is here.
In other words, if the Supremes thought the 9th was really over-reaching here, might they not have taken the case anyway?
I make no appologies for this statement: I hope people in my country (England) never resort to whining and whinging about their employee situation and resort to sueing the way Americans do.
There are certainly situations where the presence of lawyers have lead to worse outcomes for most of the people concerned. And it does make me wince to see what could be seen by somebody as a personal injury worthy of a law suit.
But the current Microsoft case isn't anything like that. This wasn't just a workmen's compensation case run amok or a frivolous suit about the lack of access to the employee's washroom. This was a case filed on the behalf of thousands of people who were systematically and illegally classified as temps when the way their jobs and compensation was structured legally made them employees. This legal fact cannot be changed even by an employee's willingness to sign it away, and, in the case of Microsoft, many of the employees treated as temps or contractors emphatically wanted Microsoft to recognize the employment status they already had, which is why there was a lawsuit in the first place.
If people can't wake up to their current job situation, whether they are being abused, used, misused, overworked, etc, then that's their problem. Not a judges, not the companies (although if people start leaving one team in droves the company should look at its managers).
The point is that people in this case did wake up to their current job situation. They were being treated illegally, not just badly. At any of several points, Microsoft could have recognized this, and either restructured the way these people were employed (and eliminating the whole reason for the complaint), or officially granted them the status they legally had. Instead, they basically just acted as if being Microsoft meant they were beyond the law.
Here's a similar case that might make the point. Suppose you take a job at a company, and that company asks you to sign a legally indefensible and unenforcable clause in a contract. If everything else about the job is great, you might well sign off on the bogus clause, with the knowledge that if it ever becomes an issue, it would legally become a non-issue. Yes, a lawyer or even a judge may have to become involved, but standing up for your legal rights is not whining.
One other point about cases like this which may not be immediately clear to a non-US citizen is the infuriating correlations between employment status and things like health insurance, pensions, and other tax-advantaged benefits. It might be nice if you really did have freedom of action to choose among all these things no matter how you were employed, but you can't. Pretending you can might make you feel better morally or intellectually superior, but it won't help you get treatment at a hospital when you need it if you are uninsured because insurance companies decided that you, the temp, were uninsurable due to some previously existing condition.
OS X is unlikely to have support for large numbers of users
Like what, like just two bits for the uid?
Well, you can't say that would byte... Seriously, you could argue that's two bits more than they've traditionally had, so Steve Jobs could definitely sell that as an amazing innovation.:-)
I mean, hey, the last time Jobs introduced a big new OS (on the original NeXT cube), the big breakthrough was that the interface wasn't a one-bit monochrome like a Mac; nosiree, we had a full, luxurious 2-bit grayscale...
The whole point of the Mac interface has always been a consistent look and feel. User customization is not consistent with this concept.
Huh? I think the consistent look-and-feel point is all about how applications look and feel, and is a concept that implies consistent placement of menus, the use of standard UI components, and the like. What those components look like when rendered on the screen should be a lot less important, as long as they have an equivalent level of consistency. And, given that you'll be logging into a Mac OSX box in any case, your look and feel will be consistent and idiosyncratic to you, down to the frosted chrome penguin close buttons.
Another problem with your statement is that it flies in the face of history. Since the time of the original Mac, users have been allowed to individualize their computers to a considerable extent: think placement of icons on the desk top (no icon dock needed), the ability to choose background patterns (even dynamic ones, like the "fish tank" background), edit icons, and you name it. Dramatic changes in the apppearance of the interface is as Mac-like as the one-button mouse.:-)
They shut out the MVS users, and I did not speak up because I was not an MVS user. They shut out the VMS users, and I did not speak up because I wasn't a VMS user. They shut out the Sun users, and I did not speak up because I was not a Sun user. They shut out the BSD users, and I did not speak up because I was not a BSD user. Then they shut me out, and there was no one left to speak up for me.
Of course, you're correct. But maybe even more correct than you imply.:-)
I think the problem is even more chilling than this. It isn't just open systems or open source systems, but almost anything besides Windows.
Probably the biggest threat to the apparently resurgent Macintosh (for example) is the deliberate lack of driver support or documentation for things like multifunction devices. In the example of Apple, or course, I know that some people would see a delicious irony in all of this, but that's just the problem.
A bad attitude doesn't really matter any more. You or I or anybody might not be a fan of FooOS for the best of reasons, but the only thing that might keep any of us in business is the realization that nobody has the luxury of accepting "special favors", or thinking "serves 'em right" when another OS goes down for the simple lack of information about how to tweak which registers to make the dang thing work.
While all this is quite legal, I've never understood how the statute would create a 'right' to make copies, which ESR seems to think is being infringed. That is, there is nothing here that can be construed as forcing authors to provide their material in easy-to-copy format. It simply declares that if you can do it, it's not a crime.
What you say is true. What is interesting here, of course, is that it was the US government that basically forced the DVD makers to use the weak encryption scheme that was cracked, and that will, barring a new encryption scheme, make DVDs much easier to copy.
But I think the interesting part of the law, as it concerns the DVD makers, is in the bit about archival purposes, since the "altered form" would appear to guarantee the interoperability criterion by itself. Since I can make an archival copy of something, and consult the data in the archive rather than the original, once I have that right, I need not ever be concerned with the strictures that the DVD people want to place on the devices that can "officially" play the original copy of the work.
But, on the other hand, if I give even a single copy of the work away, I've obviously violated the copyright unless I'm licensed to do that. The whole thing is so fair it makes me incredulous that it was a work of the US legislative process.:-)
Specifically that RSI is caused by people who use Keyboards and other systems with insufficient resistance. Notably RSI begins to show up at the same time journalists moved away from mechanical typwrites en masse. Something about he lack of resistance ecouraging/allowing bad hand posture?
First of all, it's pretty unlikely that there is just one cause of typing-related injuries, any more than there is just one way to hyperextend/twist/maim any of your tendons.
But I do honestly think that a lot of RSI-type problems are caused by crappy hand posture while typing, and those are most certainly somewhat with bad keyboard designs. I learned to type back in the day on a manual (I always used to say "acoustic":-)) typewriter, and had practice on several electric typewriters as well. Typing on those suckers was a whole different ballgame. You had to hold your wrists a certain way, and you couldn't do some of the stupid keyboard tricks (like pound on the backspace key). And there ain't no such thing as CTRL, ALT, or ESC on a selectric.:-) Heck, even CAPS LOCK had a function back then, since, unless you were a whiz at changing typing balls or daisy wheels, CAPS was the only way to get a distinctive look, and you wouldn't want to do the "shift key bounce" for very many strokes, since it would ruin your speed.
And let's talk about speed for a moment. I was an okay manual typist, but I couldn't type anything near the speed I can these days on a decent keyboard, which is upwards of 70 words per minute (usually around 75, but often in the vicinity of 100). This just wasn't possible for most people. And if it was possible, it was also the case that you had to take a break, no doutb about it. My guess is that these days that most people are:
Typing more
Typing faster
Typing with worse posture
Typing after fatigue has set in
Typing on poorly designed keyboards
It would be miraculous if typing injuries didn't occur. Typing used to be a profession, and one that was best left to the pros. Not anymore! And just as amateur athletes appear to injure themselves more often than you would expect, the same goes with amateur typists.
And how are my wrists doing? Pretty damned well, thank you, and I'm an EMACS user to boot.:-)
If you'd like an online tutorial, you might want to check out The CGI Resource Index, which is made by the same guy as Matt's Script Archive. Between the tutorials on the Resource Index, looking at the source of Matt's script, and reading the O'Reilly books, you can learn just about anything you want to know about Perl.
Others have responded that Matt's scripts aren't really a good source of information on how to program perl idiomatically, or maybe even correctly.:-)
But to add something to this thread, I'd like to point out that you really, truly can learn much more about Perl than you'd ever get from Matt's Script Archive by going to the Perl home page at www.perl.com
After you've spent a couple of weeks digesting that, it is possible that you would want to know even more, so you can go to Tom Christiansen's Far More Than Everything You've Ever Wanted to Know About... web page, if you haven't already. Oh yeah, and you can buy O'Reilly books, too.:-)
It's been reported that the foot-mouse idea doesn't really work. Perhaps it just takes more getting used to. Maybe one should take organ lessons first, or something.:-)
More accurately, a foot-mouse is probably as tricky as playing organ for most people.:-)
Seriously, one of the few common interfaces that uses the feet is the automobile. And there, what you do with your feet is not generate "events" in the sense of moving, clicking, selecting, etc. but maintain a state, either strongly decelerating (braking) or controlling acceleration. Note that even trying to add a clutch pedal in there is probably pushing it (heh, heh).
But even for simple clicking, there are problems. Apparently the neural pathways between toe and brain take longer to traverse than to the hand and back. There could be a simple issue of time-delay effects here.
Yes, they are slower, but you'd be surprised how slow the pathways are between the motor cortex and the hand. According the Card, Moran, & Newell camp, there's probably nothing too horrible about the delay or timing you could achieve with a foot activated switch in light use (rather than trying to type with your feet); there are many machines in factories that take advantage of this.
Another matter is the granularity of control. Your brain has a whole lot of grey matter devoted to hand use, and much much less for the foot. You just aren't going to be as nimble.
This almost certainly plays a big role in why foot interfaces that do exist don't generally require fine control. The nimbleness issue, however, also has a lot to do with the size and shape of the effector in question. And there are individual differences: I can pick up small objects pretty easily with my feet, and so can my 4-year-old son, but my wife and daughter cannot.
But another really big issue with the foot control of anything has to do with the fact that your feet are already doing something for you when you are just there reading slashdot. You don't just sit on your butt; your feet contribute a lot to your posture and balance while sitting. If you move your trunk a degree or three to the left or right, you can notice that muscles in your back, legs, and, yes, even feet, will change their level of "stretch". Similarly, if you move your foot, you will definitely notice that you have to compensate for that by using muscles in your other leg, too.
But, fortunately, all of this is reflexive, so you really can sit down, chew gum and read slashdot at the same time without falling off your chair, unless you feel a sudden need to practice some organ solo...
This is a simple trivia quiz, not an estimation quiz.
I have to disagree. There are specific facts that can be helpful to you to get exact answers, but half of the battle is in knowing what to do with information that is more shaky. And we certainly agree that people are not likely to know the exact answers here (that would be trivia).
Estimation is the process of making reasonable assumptions based on incomplete or limited data. You need to have some data to base your estimates on, however.
Yup. But it's surprising how close you can come with very little data. And if you know you're not going to be close at all, you should learn how to widen your confidence intervals.
I could perhaps estimate the length of the Golden Gate Bridge or the weight of a Boeing 747, but only if I had a frame of reference. As I have only seen the GGB in artistic visuals not intended to give a sense of scale, and I have only seen a 747 in Hollyweird movies, I couldn't provide a worthwhile estimate for those.
The estimate doesn't have to be especially worthwhile. It just has to have confidence bounds that include the true value. Yes, smaller intervals mean something, too, but that wasn't the explicit point of the exercise. Interestingly, I fell into that trap a bit myself, which is why I suggested that people try this out.
So, let's do the Golden Gate Bridge. I've only seen it once or twice myself; how in the heck should I know how long the main span is? On the other hand, I do know that the big deal about that bridge is that it does have a very impressive main span. So how long does a bridge have to be to be impressive for this? A mile, within a factor of 2? That's a confidence interval of like 2500 to 10000 feet. Yes, it's pretty broad, but it's also correct. Or maybe you know that 10000 feet is right out, because that would mean the main towers would have to be too tall to have the nice shape they have. Fine; then reduce that end of the estimate, to, oh, 6000 feet? That's still a covering interval, and a shorter one. Maybe that's the best that you can do, but give yourself credit for getting that far from almost nothing.
Again, let's do the weight of a 747. Like I have any idea what that is. Fine; the guess will be pretty wild, then. I think a 747 holds a lot of people. At least 300, probably less than 600? Fine. I can work with that. How much does a plane weigh? Well, how much does a car weigh? A big SUV might weigh 6000 pounds and carries 6 people, so that's 1000 pounds per person. Oh yeah: the person weighs 200 pounds, and they have 50 pounds of baggage. So what does that give us? 375,000 to 750,000 pounds. Anything I'm leaving out? Ah: fuel. The SUV gets 15 miles per gallon to haul those 6 people, or like.01 gallons per person mile. I suppose a jet aircraft could be worse, but I'm not sure how much worse. Now, a 747 can go at least across the Pacific without refueling; I bet that's oh, 6000 miles. 6000 miles *.01 gallons/person/mile*600 people = 36000 gallons of fuel. I know water weighs 8 pounds a gallon, and gasoline (or jet fuel) is rather lighter; maybe 5 pounds per gallon? So that's 90,000 to 180,000 pounds of fuel, for a total wild ass guess of 465,000 to 930,000 pounds. I'm pretty sure that the low end is below, but the high end might not be high enough; I'd be willing to go to 1.3 million or so to be comfortable.
And, as it turns out, the actual weight is in both ranges. In both cases, you could have treated the questions as trivia questions, but then you would have missed the point of the quiz.
I happened to know that the Shuttle orbits in about 90 minutes, and that roughly 50 people signed the Declaration of Independence, so I did well on those. But the significant factor here my knowledge of trivia, not estimation skills.
But one person's trivia is another person's estimation problem.
Take the Declaration of Independence signing. You know that representatives of 13 colonies signed the things, and that each state had one or more delegates. So the minimum number has to be greater than 13. For the maximum number, I suppose you might choose low. I figured 5 would be generous, so my estimate was [20,65], which includes the real value.
Now take the space shuttle. That's trivia for you, but I had to figure it out. First, I took out my trusty notepad, pretended it was a decent pendulum, and estimated its period. About 1 second; how convenient. The pad is 11/12 of a foot, and there are 5280 feet in a mile. I guessed that the shuttle orbits like 200 miles above the surface, or about 4200 miles above the center of the earth. Now, I know that the period of a pendulum depends only on the acceleration due to gravity, and the length of the bob; in fact it's proportional to (a/l)^.5. Gravity is (4000/4200)^2 weaker in orbit than on earth, so putting all this together, I get an estimate of 86 minutes, which I'll gather is good to within plus or minus 20%. So my interval is [70,102] minutes. Got it again... And this is the whole point: the exact answer is a trivia question, but the ability to get the answer to within an order of magnitude or a factor of 2, and to know how loose your confidence interval is, depends on your estimation ability.
I just have to wonder, with an "increase of 1%" what the margin of error is - does this represent a real increase?
Others have mentioned possible problems in interpreting such data which include (but are not limited to) the following:
This is a population, not a sample
Even if it were a sample, it wouldn't be a randomly selected sample.
Even if it were a randomly selected sample, it would be a random sample of domain names, not host machines.
And many, many, more.:-)
OK, having said that, it might be useful to pretend that none of these were concerns, and that we really did want to know whether a 1% increase in the number of domains served by Apache meant anything. Here's the short answer:
I can't tell you that.
This is especially true if the domains surveyed in some sense are the population. In that case, whether or not you care that Apache added 500,000 domain names to the population while IIS added 125,000 is basically up to you. There are many explanations for why this could have happened, and not all of them are very interesting. (Again, others have pointed out why.)
Personally, I would have been more interested in certain kinds of longitudinal breakdowns rather than the overall numbers. Some of those questions would include:
How many sites went from IIS to Apache vs. vice versa?
How many "new to the survey" sites went with one solution or another?
How many "new to the survey" sites are actually new to the web?
How many sites running Apache (or IIS) closed down during the last month?
Call me a geek, but these are questions I think could be more interesting to ask. And, yes, some of the answers to these are given or hinted at on the netcraft website.
But there is one more question, which is the one the original poster asked:
But what if this really were a sampling question; is a 1% difference likely to be reliable?
If all N of the netcraft domains were independently and randomly sampled from the total population of domains, then a 95% confidence interval for a given market share, M, where M is between 0 and 1 is:
[M - 1.96*(M*(1-M)/N)^.5, M + 1.96*(M*(1-M)/N)^.5]
For Apache's market share in November, we would get the interval [.5479,.5483]. For the October share, the interval is something like [.5365,.5369]. Those are pretty tight intervals, but the sample size is over 8 million...
And this is the real point: when you have random samples this huge, error bars are pretty danged small. So it's too bad these really aren't random samples...
While I was checking out the website for Programming Pearls, second edition, I stumbled across what turns out to be the complete text of Appendix 2, which is a quiz that tests your estimation ability.
Whether or not you buy the book, and whether or not you think of yourself as a programmer, if you consider yourself a thinking person, you should take this quiz. If you can't do back-of-the-envelope calculations, you should find this out as soon as possible, and fix the problem while you still can.
What browser are you using? With my browser of choice, KFM, the global Reply button for a story doesn't appear in the story header. I have to switch to Netscape to see the button. Maybe we should submit a report to our Cmdr. about this.;)
Now, this is too weird. Either something just got fixed, or button visability depends on aspects of your customized view at the moment...or maybe I really wasn't ever looking in the right place. But the "reply" button is certainly there now, when I just looked a second ago. I think I'll personally wait before assuming it's a bug...
Of course, now the question is whether I would have gotten my post moderated up to 5 (first time ever, although I've made better posts in the past...) if I had been able to find the button and start a new thread off of the main article.
Yes, this is slighly off-topic, but Slashdot won't let me start a new main thread, and this is a space-related hack.
Once upon the time, the military decided it would be really great to know exactly where you were anywhere in the world, say by just pressing a button on a hand-held unit. The geeks in the backroom found out a way to do this, using satellites (this alone was quite a hack, actually...) Now, lo and behold, we can all use GPS to find out exactly where we are.
Well, not exactly. The military realized it would not be a great idea to let just anybody have such nice positioning information. It would suck if Saddam Hussein knew exactly where all his tanks were during a battle, too. So the GPS system also has a built-in method to screw up the signal to a greater or lesser extent depending on who you are and whether or not we're fighting a war.
Now comes the real hack: a bunch of geeky geoscientists (or is that redundant?) decided that they could track tectonic plate movements using GPS...if only they could obtain more accuracy than the generals would be comfortable with. So what they did was design a method that all but ignored the "for the public" tracking information you could get from the GPS system, and instead focused on analyzing the inevitable phase distortions of the carrier frequency itself to achieve better than 1 cm location accuracy, after lots of post-processing. A crude analogy here would be to come up with a system that would do something useful with TCP/IP packets by ignoring the "useful" contents of the packets themselves, but concentrating on the quirky bits (like the TCP finger-printing people) or the weird statistics of packet arrival times.
None of this is exactly what the military had in mind, but this is (so far) only useful for surveying applications, an most notably the study and identification of known and unknown faults in tectoncially active regions of the world. You can look at some of the more recent data at this JPL site put together by Michael Heflin. The next time somebody asks you how we know that plate tectonics really works, just send them here.:-)
You make your own heaven, your own hell. It's true that you can never return to a state of MS-virginity any more than can a heroin addict return to that unblemished state of never having shot up,
But, but...it was years ago; surely you must have pity? I was young and foolish, and a Macintosh was involved, for heavens sake. There must be some way, right?
I mean, is there any chance you might start selling indulgences, or trading them for code contributions to a favored project or something? It seems only fair...
Arnold Schwartzenegger is a very interesting example, since he is a native german speaker (being from Austria originally). He should be able to very easily dub his own dialog into German, so you should be able to get his real voice. It is really kinda sad if they don't take the german speaking market seriously enough to do it that way.
On the other hand, I've heard from some (admittedly pretentious) Germans I know that Arnold's German accent in English is far more impressive than his just speaking German. Speaking English, he can sound downright frightening with very little effort; to do the same in German would require a lot more. And he is, by his own admission, not the greatest actor in the world, so it's very possible that Arnold Schwartenegger isn't a good enough actor to do his own voice in his native language.
Plus, I'm sure that gig doesn't pay nearly enough to be very interesting to him except possibly as a point of pride or vanity.
Slashdot Mirror
Well, that book put me to sleep, but I then handed it to a colleague who needed to learn perl, and never did get it back. For what it's worth, that person was an intelligent non-programmer. Apparently that book ended up on a Great Journey and has landed on the desks and shelves of many people just like my so-called, book-stealing friend.
I think the only perl books I don't have now but might want (I have these two) are Perl in a Nutshell (looks like a really handy reference) and possibly Object Oriented Perl, because it was written by Damian Conway, whose papers and projects I have found very useful in the past.
The result was a language with C++'s semantics (at that time) and a much cleaner syntax. Which, of course, went nowhere. Do you have any comments on SPECS itself or any other attempts to improve what might be called the "human factors" aspects of the C++ language?
I think your sentiment is simultaneously too pessimistic and too idealistic at the same time.
I don't think that the history of legislation is primarily made up of cases where laws are passed or not passed due to a broad-based public consensus on the content of the laws themselves. The beautiful thing about representative democracy is precisely the fact that change can occur due to the actions of very small numbers of people. In otherwords, because some groups of people can gain influence disproportionate to their actual numbers.
And this, of course, is exactly why lack of change is the rule: when there is no very broad consensus, politicians are loathe to side with one interest group or another unless they can see a clear, one-sided benefit for themselves and their own careers. Donations to campaigns are one obvious benefit, as are winning (or keeping) the allegiance of an entire block of voters. But getting money for casting a vote isn't a great plan if the vote itself will later be used as the basis for negative campaigning by an opponent. In this case, the attack ads for voting in favor of shrink-wrap licenses practically write themselves.
Moreover, I expect that, to the average politician, angering even a small minority of net-savvy techno-geeks who make it clear they will back up what they say would be viewed as a really suspect thing to do unless it delivered a truly huge number of votes or dollars. Your state senator might not know what a EULA is, but if he or she knows that voting to strengthen them is going to lead to direct political attacks amplified Lord knows how much by that Internet thing, you can be sure that doing nothing will look much more attractive.
So, yes, action is needed, and education is always a good thing, but I'm far more optimistic that an action on something like this can protect the status quo without anything like a majority of citizens, or even the internet-aware citizens, knowing about it. Vocal minorities are way louder than you would think.
There are many reasons why Perl may or may not be a good first language, but this can't be one of them, since it just isn't true. TCL used to be the textbook example of a language that used the string as its representation for everything, but TCL isn't string only anymore, either.
Really, I don't know where people get this idea about Perl, which certainly isn't the most typeful language on the block, but is hardly short of interesting types. I suppose it comes from the fact that Perl does provide a lot of automagical operators and conversions (although the fact that conversions are involved should make it clear that Perl has more than one underlying type).
But one thing I'd like to point out in particular:
Says somebody who clearly hasn't ever had to justify the difference between a short and long to a beginning (or non-)programmer. I would actually argue that the short/long difference is exactly the kind of "people working for the computer" stuff that doesn't belong anywhere near somebody's first exposure to programming.
If you want to teach the importance of types at the same time you teach programming, you would do better to use a language that has a more interesting and flexible type system; one obvious candidate would be Haskell, or it's interpreted cousin Hugs.
It is? Tell me: do you have any numbers to back this up? I ask, because people who have looked at the econmic impact of publicly funded stadiums in particular, or the presence of sports teams in general, have found essentially no evidence of any net benefit. To be sure, there are people who make more money due to the presence of a heavily-subsidized sports franchise, but there are others who do worse, and there are also huge opportunity costs as well.
s/games/ways/?
But, actually, this is just the problem: the cash that St. Louis just basically gave to the Rams most certainly was economically quantifiable, and the other very pressing needs of St. Louis are also economically quantifiable, so the idea of spending that much cash on something just to "make it exciting to live in" goes way past stupid and borders on the immoral.
Or, in other words, Slashdot's RDBMS backend doesn't use place moderation updates inside a transaction. This should be simple enough to fix unless they're silly enough to use a database that doesn't support transactions. But who would do something silly like that? :-)
If I'm right about this, the Postgresql folks are entitled to one collective "nyeah, nyeah, we told you so!" on this topic.
But in fact, it's hardly just one "we" anymore, at least compared to the Net culture of old (and not to imply that the old days were one big happy family or anything...).
What I find much more potentially interesting about certain internet forums (Usenet groups, things like Slashdot, other chunks of the web) is that they are automatically archived, and reasonably searchable. In that kind of environment, it is
All in all, it's very possibly a politician's worst nightmare. :-)
Well, my inner geek then asks: what is the proportion of cases that the court hears where it does overturn an appeals court ruling? I guess I did know that the 9th circuit is the one most likely to get whacked, so I wonder whether this actually bodes well for the ruling, a copy of which is here.
In other words, if the Supremes thought the 9th was really over-reaching here, might they not have taken the case anyway?
There are certainly situations where the presence of lawyers have lead to worse outcomes for most of the people concerned. And it does make me wince to see what could be seen by somebody as a personal injury worthy of a law suit.
But the current Microsoft case isn't anything like that. This wasn't just a workmen's compensation case run amok or a frivolous suit about the lack of access to the employee's washroom. This was a case filed on the behalf of thousands of people who were systematically and illegally classified as temps when the way their jobs and compensation was structured legally made them employees. This legal fact cannot be changed even by an employee's willingness to sign it away, and, in the case of Microsoft, many of the employees treated as temps or contractors emphatically wanted Microsoft to recognize the employment status they already had, which is why there was a lawsuit in the first place.
The point is that people in this case did wake up to their current job situation. They were being treated illegally, not just badly. At any of several points, Microsoft could have recognized this, and either restructured the way these people were employed (and eliminating the whole reason for the complaint), or officially granted them the status they legally had. Instead, they basically just acted as if being Microsoft meant they were beyond the law.
Here's a similar case that might make the point. Suppose you take a job at a company, and that company asks you to sign a legally indefensible and unenforcable clause in a contract. If everything else about the job is great, you might well sign off on the bogus clause, with the knowledge that if it ever becomes an issue, it would legally become a non-issue. Yes, a lawyer or even a judge may have to become involved, but standing up for your legal rights is not whining.
One other point about cases like this which may not be immediately clear to a non-US citizen is the infuriating correlations between employment status and things like health insurance, pensions, and other tax-advantaged benefits. It might be nice if you really did have freedom of action to choose among all these things no matter how you were employed, but you can't. Pretending you can might make you feel better morally or intellectually superior, but it won't help you get treatment at a hospital when you need it if you are uninsured because insurance companies decided that you, the temp, were uninsurable due to some previously existing condition.
Well, you can't say that would byte... Seriously, you could argue that's two bits more than they've traditionally had, so Steve Jobs could definitely sell that as an amazing innovation. :-)
I mean, hey, the last time Jobs introduced a big new OS (on the original NeXT cube), the big breakthrough was that the interface wasn't a one-bit monochrome like a Mac; nosiree, we had a full, luxurious 2-bit grayscale...
Huh? I think the consistent look-and-feel point is all about how applications look and feel, and is a concept that implies consistent placement of menus, the use of standard UI components, and the like. What those components look like when rendered on the screen should be a lot less important, as long as they have an equivalent level of consistency. And, given that you'll be logging into a Mac OSX box in any case, your look and feel will be consistent and idiosyncratic to you, down to the frosted chrome penguin close buttons.
Another problem with your statement is that it flies in the face of history. Since the time of the original Mac, users have been allowed to individualize their computers to a considerable extent: think placement of icons on the desk top (no icon dock needed), the ability to choose background patterns (even dynamic ones, like the "fish tank" background), edit icons, and you name it. Dramatic changes in the apppearance of the interface is as Mac-like as the one-button mouse. :-)
Of course, you're correct. But maybe even more correct than you imply. :-)
I think the problem is even more chilling than this. It isn't just open systems or open source systems, but almost anything besides Windows.
Probably the biggest threat to the apparently resurgent Macintosh (for example) is the deliberate lack of driver support or documentation for things like multifunction devices. In the example of Apple, or course, I know that some people would see a delicious irony in all of this, but that's just the problem.
A bad attitude doesn't really matter any more. You or I or anybody might not be a fan of FooOS for the best of reasons, but the only thing that might keep any of us in business is the realization that nobody has the luxury of accepting "special favors", or thinking "serves 'em right" when another OS goes down for the simple lack of information about how to tweak which registers to make the dang thing work.
What you say is true. What is interesting here, of course, is that it was the US government that basically forced the DVD makers to use the weak encryption scheme that was cracked, and that will, barring a new encryption scheme, make DVDs much easier to copy.
But I think the interesting part of the law, as it concerns the DVD makers, is in the bit about archival purposes, since the "altered form" would appear to guarantee the interoperability criterion by itself. Since I can make an archival copy of something, and consult the data in the archive rather than the original, once I have that right, I need not ever be concerned with the strictures that the DVD people want to place on the devices that can "officially" play the original copy of the work.
But, on the other hand, if I give even a single copy of the work away, I've obviously violated the copyright unless I'm licensed to do that. The whole thing is so fair it makes me incredulous that it was a work of the US legislative process. :-)
First of all, it's pretty unlikely that there is just one cause of typing-related injuries, any more than there is just one way to hyperextend/twist/maim any of your tendons.
But I do honestly think that a lot of RSI-type problems are caused by crappy hand posture while typing, and those are most certainly somewhat with bad keyboard designs. I learned to type back in the day on a manual (I always used to say "acoustic" :-)) typewriter, and had practice on several electric typewriters as well. Typing on those suckers was a whole different ballgame. You had to hold your wrists a certain way, and you couldn't do some of the stupid keyboard tricks (like pound on the backspace key). And there ain't no such thing as CTRL, ALT, or ESC on a selectric. :-) Heck, even CAPS LOCK had a function back then, since, unless you were a whiz at changing typing balls or daisy wheels, CAPS was the only way to get a distinctive look, and you wouldn't want to do the "shift key bounce" for very many strokes, since it would ruin your speed.
And let's talk about speed for a moment. I was an okay manual typist, but I couldn't type anything near the speed I can these days on a decent keyboard, which is upwards of 70 words per minute (usually around 75, but often in the vicinity of 100). This just wasn't possible for most people. And if it was possible, it was also the case that you had to take a break, no doutb about it. My guess is that these days that most people are:
It would be miraculous if typing injuries didn't occur. Typing used to be a profession, and one that was best left to the pros. Not anymore! And just as amateur athletes appear to injure themselves more often than you would expect, the same goes with amateur typists.
And how are my wrists doing? Pretty damned well, thank you, and I'm an EMACS user to boot. :-)
Others have responded that Matt's scripts aren't really a good source of information on how to program perl idiomatically, or maybe even correctly. :-)
But to add something to this thread, I'd like to point out that you really, truly can learn much more about Perl than you'd ever get from Matt's Script Archive by going to the Perl home page at www.perl.com
After you've spent a couple of weeks digesting that, it is possible that you would want to know even more, so you can go to Tom Christiansen's Far More Than Everything You've Ever Wanted to Know About... web page, if you haven't already. Oh yeah, and you can buy O'Reilly books, too. :-)
More accurately, a foot-mouse is probably as tricky as playing organ for most people. :-)
Seriously, one of the few common interfaces that uses the feet is the automobile. And there, what you do with your feet is not generate "events" in the sense of moving, clicking, selecting, etc. but maintain a state, either strongly decelerating (braking) or controlling acceleration. Note that even trying to add a clutch pedal in there is probably pushing it (heh, heh).
Yes, they are slower, but you'd be surprised how slow the pathways are between the motor cortex and the hand. According the Card, Moran, & Newell camp, there's probably nothing too horrible about the delay or timing you could achieve with a foot activated switch in light use (rather than trying to type with your feet); there are many machines in factories that take advantage of this.
This almost certainly plays a big role in why foot interfaces that do exist don't generally require fine control. The nimbleness issue, however, also has a lot to do with the size and shape of the effector in question. And there are individual differences: I can pick up small objects pretty easily with my feet, and so can my 4-year-old son, but my wife and daughter cannot.
But another really big issue with the foot control of anything has to do with the fact that your feet are already doing something for you when you are just there reading slashdot. You don't just sit on your butt; your feet contribute a lot to your posture and balance while sitting. If you move your trunk a degree or three to the left or right, you can notice that muscles in your back, legs, and, yes, even feet, will change their level of "stretch". Similarly, if you move your foot, you will definitely notice that you have to compensate for that by using muscles in your other leg, too.
But, fortunately, all of this is reflexive, so you really can sit down, chew gum and read slashdot at the same time without falling off your chair, unless you feel a sudden need to practice some organ solo...
I have to disagree. There are specific facts that can be helpful to you to get exact answers, but half of the battle is in knowing what to do with information that is more shaky. And we certainly agree that people are not likely to know the exact answers here (that would be trivia).
Yup. But it's surprising how close you can come with very little data. And if you know you're not going to be close at all, you should learn how to widen your confidence intervals.
The estimate doesn't have to be especially worthwhile. It just has to have confidence bounds that include the true value. Yes, smaller intervals mean something, too, but that wasn't the explicit point of the exercise. Interestingly, I fell into that trap a bit myself, which is why I suggested that people try this out.
So, let's do the Golden Gate Bridge. I've only seen it once or twice myself; how in the heck should I know how long the main span is? On the other hand, I do know that the big deal about that bridge is that it does have a very impressive main span. So how long does a bridge have to be to be impressive for this? A mile, within a factor of 2? That's a confidence interval of like 2500 to 10000 feet. Yes, it's pretty broad, but it's also correct. Or maybe you know that 10000 feet is right out, because that would mean the main towers would have to be too tall to have the nice shape they have. Fine; then reduce that end of the estimate, to, oh, 6000 feet? That's still a covering interval, and a shorter one. Maybe that's the best that you can do, but give yourself credit for getting that far from almost nothing.
Again, let's do the weight of a 747. Like I have any idea what that is. Fine; the guess will be pretty wild, then. I think a 747 holds a lot of people. At least 300, probably less than 600? Fine. I can work with that. How much does a plane weigh? Well, how much does a car weigh? A big SUV might weigh 6000 pounds and carries 6 people, so that's 1000 pounds per person. Oh yeah: the person weighs 200 pounds, and they have 50 pounds of baggage. So what does that give us? 375,000 to 750,000 pounds. Anything I'm leaving out? Ah: fuel. The SUV gets 15 miles per gallon to haul those 6 people, or like .01 gallons per person mile. I suppose a jet aircraft could be worse, but I'm not sure how much worse. Now, a 747 can go at least across the Pacific without refueling; I bet that's oh, 6000 miles. 6000 miles * .01 gallons/person/mile*600 people = 36000 gallons of fuel. I know water weighs 8 pounds a gallon, and gasoline (or jet fuel) is rather lighter; maybe 5 pounds per gallon? So that's 90,000 to 180,000 pounds of fuel, for a total wild ass guess of 465,000 to 930,000 pounds. I'm pretty sure that the low end is below, but the high end might not be high enough; I'd be willing to go to 1.3 million or so to be comfortable.
And, as it turns out, the actual weight is in both ranges. In both cases, you could have treated the questions as trivia questions, but then you would have missed the point of the quiz.
But one person's trivia is another person's estimation problem.
Take the Declaration of Independence signing. You know that representatives of 13 colonies signed the things, and that each state had one or more delegates. So the minimum number has to be greater than 13. For the maximum number, I suppose you might choose low. I figured 5 would be generous, so my estimate was [20,65], which includes the real value.
Now take the space shuttle. That's trivia for you, but I had to figure it out. First, I took out my trusty notepad, pretended it was a decent pendulum, and estimated its period. About 1 second; how convenient. The pad is 11/12 of a foot, and there are 5280 feet in a mile. I guessed that the shuttle orbits like 200 miles above the surface, or about 4200 miles above the center of the earth. Now, I know that the period of a pendulum depends only on the acceleration due to gravity, and the length of the bob; in fact it's proportional to (a/l)^.5. Gravity is (4000/4200)^2 weaker in orbit than on earth, so putting all this together, I get an estimate of 86 minutes, which I'll gather is good to within plus or minus 20%. So my interval is [70,102] minutes. Got it again... And this is the whole point: the exact answer is a trivia question, but the ability to get the answer to within an order of magnitude or a factor of 2, and to know how loose your confidence interval is, depends on your estimation ability.
Others have mentioned possible problems in interpreting such data which include (but are not limited to) the following:
OK, having said that, it might be useful to pretend that none of these were concerns, and that we really did want to know whether a 1% increase in the number of domains served by Apache meant anything. Here's the short answer:
I can't tell you that.
This is especially true if the domains surveyed in some sense are the population. In that case, whether or not you care that Apache added 500,000 domain names to the population while IIS added 125,000 is basically up to you. There are many explanations for why this could have happened, and not all of them are very interesting. (Again, others have pointed out why.)
Personally, I would have been more interested in certain kinds of longitudinal breakdowns rather than the overall numbers. Some of those questions would include:
Call me a geek, but these are questions I think could be more interesting to ask. And, yes, some of the answers to these are given or hinted at on the netcraft website.
But there is one more question, which is the one the original poster asked:
But what if this really were a sampling question; is a 1% difference likely to be reliable?
If all N of the netcraft domains were independently and randomly sampled from the total population of domains, then a 95% confidence interval for a given market share, M, where M is between 0 and 1 is:
[M - 1.96*(M*(1-M)/N)^.5, M + 1.96*(M*(1-M)/N)^.5]
For Apache's market share in November, we would get the interval [.5479, .5483]. For the October share, the interval is something like [.5365, .5369]. Those are pretty tight intervals, but the sample size is over 8 million...
And this is the real point: when you have random samples this huge, error bars are pretty danged small. So it's too bad these really aren't random samples...
Whether or not you buy the book, and whether or not you think of yourself as a programmer, if you consider yourself a thinking person, you should take this quiz. If you can't do back-of-the-envelope calculations, you should find this out as soon as possible, and fix the problem while you still can.
Now, this is too weird. Either something just got fixed, or button visability depends on aspects of your customized view at the moment...or maybe I really wasn't ever looking in the right place. But the "reply" button is certainly there now, when I just looked a second ago. I think I'll personally wait before assuming it's a bug...
Of course, now the question is whether I would have gotten my post moderated up to 5 (first time ever, although I've made better posts in the past...) if I had been able to find the button and start a new thread off of the main article.
Yes, this is slighly off-topic, but Slashdot won't let me start a new main thread, and this is a space-related hack.
Once upon the time, the military decided it would be really great to know exactly where you were anywhere in the world, say by just pressing a button on a hand-held unit. The geeks in the backroom found out a way to do this, using satellites (this alone was quite a hack, actually...) Now, lo and behold, we can all use GPS to find out exactly where we are.
Well, not exactly. The military realized it would not be a great idea to let just anybody have such nice positioning information. It would suck if Saddam Hussein knew exactly where all his tanks were during a battle, too. So the GPS system also has a built-in method to screw up the signal to a greater or lesser extent depending on who you are and whether or not we're fighting a war.
Now comes the real hack: a bunch of geeky geoscientists (or is that redundant?) decided that they could track tectonic plate movements using GPS...if only they could obtain more accuracy than the generals would be comfortable with. So what they did was design a method that all but ignored the "for the public" tracking information you could get from the GPS system, and instead focused on analyzing the inevitable phase distortions of the carrier frequency itself to achieve better than 1 cm location accuracy, after lots of post-processing. A crude analogy here would be to come up with a system that would do something useful with TCP/IP packets by ignoring the "useful" contents of the packets themselves, but concentrating on the quirky bits (like the TCP finger-printing people) or the weird statistics of packet arrival times.
None of this is exactly what the military had in mind, but this is (so far) only useful for surveying applications, an most notably the study and identification of known and unknown faults in tectoncially active regions of the world. You can look at some of the more recent data at this JPL site put together by Michael Heflin. The next time somebody asks you how we know that plate tectonics really works, just send them here. :-)
Hmm...so how many lice would fit on the head of a pinhead.
Ooh! You're missing one: according to AltaVista, there at least 445 places where you can read about the millienium problem.
I'm not exactly sure what that is, but I'm guessing it's a million times dicier than the millenium problem. :-)
But, but...it was years ago; surely you must have pity? I was young and foolish, and a Macintosh was involved, for heavens sake. There must be some way, right?
I mean, is there any chance you might start selling indulgences, or trading them for code contributions to a favored project or something? It seems only fair...
On the other hand, I've heard from some (admittedly pretentious) Germans I know that Arnold's German accent in English is far more impressive than his just speaking German. Speaking English, he can sound downright frightening with very little effort; to do the same in German would require a lot more. And he is, by his own admission, not the greatest actor in the world, so it's very possible that Arnold Schwartenegger isn't a good enough actor to do his own voice in his native language.
Plus, I'm sure that gig doesn't pay nearly enough to be very interesting to him except possibly as a point of pride or vanity.