Your metaphors are pointless. Software is not clothing. Non-customized software is useless in a way that non-customized clothing is not. There are perhaps 20-ish common pants sizes for adult men. There are multiple thousands of software configurations for business, one for each business.
So a business can download Apache and install it. So what? Does Apache automatically integrate with accounts receivable? Does it automatically allow the customer to see their pending orders more efficiently then with the competition? Does one upgrade off of a mainframe database by the mere installation of MySQL?
If you insist on a metaphor, free software is a brick for business; it's great that free software drives the materials cost of the "software building" down to zero, but labor always dominated anyhow. The software equivalent of a downloadable house is still non-existent.
So your whole metaphor falls down on the fact that the final product is not freely available; the software equivalent of "commodity jeans" is not available, only the software equivalent of "commodity denim".
If you want to prove I'm wrong, show me the software package I can download off of Sourceforge that all I have to do is install it, and bam, I have a payroll system for a multi-national, multi-thousand employee company, tapping into our existing payroll databases and conversant not just with the laws of multiple countries, but also the multiple union contracts we've negotiated. Hell, show me the commercial package that can do that! Is that an extreme? A little, but you can't even download the equivalent for a four person plumbing company, either.
Failing that, you are pretty much wrong, and there's not much I can say to soften that.
"Software you want" is not, especially for businesses, especially for businesses using the software as a skeleton to hang the rest of the business off of, which is basically every business over 50 employees. Oddly, they just don't sell Major University In A Box, or World-Class Automobile Company In A Box, and you sure as hell can't download it from Sourceforge.
"Heironymous' Law" is more like "Heironymous' Really Idealistic But Wrong Idea"; don't hold your breath waiting for it to appear in the Jargon Dictionary.
No, the benchmarks aren't subjective, that's why this is such a big deal. A benchmark consists of a well-specified objective task a computer can do, with objective measurements of how quickly that task was completed.
That's why, if Apple really tweaked the Intel computers sufficiently to nearly cut their performance in half, it's not a "matter of interpretation", it's not "a thing valid for Apple but not valid for me", it's a lie. A computer that can score a 24 can score a 24, period; if you tweak it until it only scores a 15, it can still score a 24, and your tweaks are lies.
You should be able to assemble identical computers, run identical tests, and get the same numbers within an error factor. That's almost the very definition of objective in a scientific sense.
Now, how you interpret the benchmarks is subjective, because no benchmark can possible match everybody's daily use of the computer, so even though Machine A gets a 24 on an integer test and Machine B gets a 22, Machine B may significantly faster for the real-world tasks I do, whereas Machine A may be even faster then B then the test numbers would indicate for somebody else's tasks.
But the whole point of benchmarks is to provide an objective measurement.
(Which you probably knew, but careless word usage leads to careless thinking; "subjective" is the wrong word here. The phrase is "open to interpretation", but you see, that doesn't let Apple off the hook for lying the way that "subjective" would.)
This particular saying bugs the sh** outta me: "Quietly introduced"
Well, good news: The days of the "quiet introduction" are probably numbered, as evidenced by the fact we picked up this story.
Frankly, the government needs to get moving on passing this sort of legislation; the communication networks that will make this effectively impossible to "quietly introduce" in the future are rapidly forming... the next few years are going to see a massive sea change in politics as at least some of the promises of the Internet are going to start really kicking in...
They dominate... and make no money.... almost no small sites make any money....
Have you ever noticed that he rarely points to sites outside of useit.com and he often is selling his usability reports?
And you aren't the slightest bit interested in the opinions of a person who is running one of the "small sites" and is in all probability making money with it? Think he might just understand a bit?
It sounds to me like you wouldn't be happy no matter what.
Incidentally, figuring out the "blame" for the failure of micropayments is a non-trivial operation; the multi-year stagnation in the browser market (thanks to Netscape's effective disappearence and Microsoft's well-known tendency to not bother with its precious "innovation" unless there's competition) at the same time that the routing market has held a virtual monopoly (ensuring no protocol-level support for micropayments could make any headway), both market conditions and not truly technological conditions, probably had a lot to do with. Despite the fact I'm not holding my breath, they would still solve an awful lot of problems.
You say that bit about "spheres of influence" like it's a bad thing, but it's a necessary part of getting a lot of people to work on one project. You can't get a hundred people to work without them; hell, you can't get four people to work without them. It reduces the communication from n^2 to something more like n log n, which is again why adding people can continue to benefit a project for a long time.
I think those are probably fossil practices left over from pre-Internet days.
It is important to shield the developers from telephone calls and visits from the customer, because otherwise the developers will get nothing done, not necessarily because they have no time (time_in_day - time_dealing_with_customers may still be large), but because they are constantly being interrupted.
As long as the developer has discipline and the customer realizes that the answers won't be instant, though, there can be great value in having email access to the developers, for both parties.
Hopefully more companies will modernize their policies soon.
Two of the three open projects you cite also happen to be bigger than most projects, with more developers working in || at the same time than most projects. Does that help them? Read "the mythical man-month" for the answer. (Hint: no).
You have seriously misread The Mythical Man Month, to the point of absurdity. The Mythical Man Month does emphatically not claim that more people can't do any more then fewer people.
What it says is that programmers are not interchangable, and as a result, adding programmers to an existing project will not get a 1-man-hour to 1-programmer-hour gain. While the old programmers are training the new programmers, and while the communication overhead increases, and while the new programmers are not yet settled in, productivity may fail to increase or may even decrease, over the next few months.
However, once the new guys are settled in and assuming a decent organization, the new organization can indeed move faster.
The Mythical Man Month's main point is to avoid treating programmers, or perhaps more accurately programmer-hours, interchangably. The conclusion based on that is the somewhat more famous part about not adding manpower to a late project because it will make it even more late.
Most strong open-source projects, and probably all the ones in this study, have a strong group of core people who are intimately familiar with what they need to know to continue improving the software. Other people may drift in but will find it difficult to contribute code back. (Seriously, just try to get a patch into Mozilla.) The Mythical Man Month does not apply to steady-state teams of people, only growing ones.
So yes, larger teams help those projects create their large software products. Go look at how much code is in Mozilla; no matter how wizard you are, I don't think a small team of people could even type that much code in anything less then a couple of months. How else do you think they could do it but with a large team?
I find, as a graduate English student, that I can't really think of any generation or era where the intellectual art has really lasted well.
Well, you qualified yourself as an English student but made a claim about "art", so let me take a broad meaning of "art" and mention music.
Up until early in the twentieth century, popular music and academic music were basically one and the same, and the music of the eighteenth and nineteenth century is doing fine. A lot of people (such as myself) listen to it daily, many radio stations remain dedicated to it, and a lot of people still perform. Sure, it's not dominating the music industry, but with so much music, there really isn't any one thing that's dominating any more.
Speaking broadly, it wasn't until the Big Band era, Jazz, Rock and Roll, and eventually almost all music was seperated from academic music... but popular music is actually still largely based on the music theory of the 18th and 19th centuries, though it takes frequent (and justified; I'm not claiming theory uber alles, just mentioning this stuff) liberties with them.
Now academic music is effectively marginalized and even with my rather broad tastes (best I can classify it is as "anything with substance", from classical music to Tubular Bells and even well-done modern video game music; Grandia 2 was superb), I can't stand modern (or perhaps "post-modern" would be better) academic music.
You would recognize many tunes from a hundred years ago, even from the academic tradition of music, even if you couldn't name them. In fact the academic tradition of a hundred years ago has fared better then the popular tradition of that era.
In this post I also kind of assumed that by "intellectual art" I could substitute the word "academic" without loss of meaning.
I recommend getting two; lemur production doesn't work so well with only one lemur machine. Make sure they are compatible; there are two types and you need one of each.
By the way, it's quite likely that local laws will prevent you from owning a lemur-producing machine. Many types of lemur-producing machines are very rare and once they break, they can't be fixed, so if they all break that's it for lemur-machines. As a result many governments are quite protective of them.
Also, operation of the lemur machines in such a manner that they will actually produce lemurs and not just consume resources may be very, very tricky; I'm not certain but I do know a lot of other, similar machines often take many skilled experts to cause to produce.
Anyhow, they're not as user-friendly as you'd like and you can't just pick them up at Best Buy, but you can probably get some lemur machines if you really put your mind to it. Best part of all is there are billions of years behind their development, so they are pretty sophisticated.
It's not so much a "study" as something that just always happens; in the computation domain, it unfortunately means you can never know whether you should shut off the simulation, because you might be one iteration away from a 25% improvement, whereas steady growth would make it easy to know whether you've reached the end of a computation.
Hmmm... there's probably a theorem here waiting to get out.
The difference is that figuring out whether the wheel is slipping is merely a hard problem (it is not an easy problem, and there are a lot of tricky parts, but it is solvable), and it can be done in the domain of "the speed of the wheel".
Deciding whether to stop or not require knowlege of the outside world, which not surprisingly is a lot more complicated then the problem of whether you're slipping. That's a four-dimensional problem, one speed reading on each wheel. Deciding to stop, especially the way we do it with our vision system, is an n-dimensional problem, where "n" is some big number. (Easily thousands.) Extrapolating our success with a four-dimensional problem that doesn't even need heavy AI to solve into thinking that we can equivalently solve a multi-thousand dimensional problem (with only two or three of those dimensions available to our system, no less!) is quite overconfident, to say the least.
Insurance companies are quite pragmatic. If this system was deployed I expect they'll require most drivers to shut it off, or they'll deny coverage for anything they can blame on the system, because I sincerely doubt this system will be a net safety gain.
This is completely different. "Self-powered wagons" are driven by humans. Automating a technology is completely different then simply creating it.
So no, fear that we can't successfully automate something does not apply to all technology, it only applies to attempts to automate things, and we have not had success in automating real-world applications, only relatively "easy" things like factory-line robots, doing the same thing in the same circumstances every time with humans standing by to take over if there's a problem of any kind, and nobody dies if there is a problem.
Concern over automation is quite justified; this kind of AI has been defeating us for 50 years now and frankly the situation hasn't changed much, nor does it seem likely to change terribly soon.
Oh come on... either you are lying to us and you just obviousally told all of us that by making that statement or you dont have enough coffee to think this morning...
No, the problem is you obviously have no experience dealing with real world data. Writing software is a lot harder then "seeing a way to do it". For instance...
1 - simply measure the speed of the car in front of you relative to you.
Not even all the way through your first sentence and you've already hit major problems. First, all measurements have noise in them; you can't know the "speed" with perfect reliability, and how you determine the speed from the incoming data is a non-trivial problem. Since lives are on the line, it's important to get it right. A simple "moving average", for instance, won't work because it will slow the reaction time of the system down unacceptably.
Also, we have a problem with "car in front of you". Now, "thing in front of you" may work, but A: how do you know it's a car? and B: how do you know it's truly in front of you? Maybe you're just turning and that "thing" you see coming at you is just a mailbox.
Pop quiz: You are a radar system. You detect something coming at you at 70 mph from a hundred feet away. You're turning slightly, but not much. As a radar system in a car, this is all you know. Do you slam on the brakes?
Yes? Oops. Turns out that you're on the left lane of a three-lane highway, and somebody's parked their car on the right shoulder due to car problems. The car gets rear-ended by a tailgaiting Mack truck and the driver dies. You get sued because your system was not supposed to hit the brakes then.
No? Oops. There was a Mack truck jackknifed across the road which the driver wasn't looking at and didn't see. In the.4 seconds it takes for the driver to react that you could have been braking, the car continues at 70 mph. By the time the driver reacts, its too late to slow the car down to a survivable speed. The driver hits the cab of the truck and dies. You get sued because your system was supposed to hit the brakes for the driver.
This isn't so easy, is it? The car only has a very limited number of senses; in human terms it's blind except for one whole pixel, deaf, and can only feel limited information about it's own state ("I'm going 50", "I'm turning", etc.). It's not like the car has a picture of the whole road to work with... and of course the reason it doesn't have that is we wouldn't know what to do with it if we did. Computer vision is still a very limited technology; it can not handle the real world to any significant degree yet.
if the car is doing 50mph toward the radar source, hit the breaks!
How hard? What threshold for distance? Does the distance threshold change if the relative velocity changes? How does the brake hitting relate to the relative speed? What if I'm driving down the road at 60mph, I go into a turn, and a mailbox momentarily looks like it's coming at me at 55mph? Do I slam on the brakes then, or wait until I'm sure it's a car? How can I be sure when all I have is a distance and a relative velocity anyhow?
If you're wrong and you "hit the brakes", you may end up killing the user. Lives are on the line; you can't gloss over any of these questions. One wrong answer and people die.
you simply have a threshhold setting for distance to trigger and speed.
That is naive beyond words. The real world is much more complicated then that.
and simply adjust that according to the car's current speedometer reading.
How, exactly? "adjust" is an empty word until you give more specification. It's damned easy to say "simply adjust", but there are hundreds of plausible way to "simply adjust", and most of them are wrong. (It's even possible all of them are wrong; there's no guarentee this system is even possible, you know.)
2 - to hell with the speed of X Y or Z look at the doppler shift. if I detec
You say providing a link is not a burden, but you forget that you have to prominently advertise that link for at least 24 hours.
No, your use of the word "advertising", as if it has to go on the homepage, is an exaggeration not present in the proposal. It says this:
 The reply should be made publicly available in a prominent place for a period of time which is at least equal to the period of time during which the contested information was publicly available, but in any case no less than for 24 hours.
"Prominent place" should probably be interpreted in the light of the very next paragraph:
 So long as the contested information is available on-line, the reply should be attached to it, for example through a clearly visible link.
I would expect that a clearly visible link on the original criticism would meet the criterion of the first paragraph.
Another problem is that, even with links, disagreements are not easily settled by a single reply.
This isn't about settling disagreements; nobody is pretending that is possible.
Do I have the right to reply to your reply, and you to reply in turn... this to be repeated ad infinitum? Or will I be forced to offer a link to your reply even when your reply contains false information and misleading assessments of my own criticism?
The proposal addresses neither issue directly. However, as a practical matter that won't happen, because the link being pointed to can be changed as necessary to update new conditions; new links will not need to be continuously posted so that's a bugaboo.
As for "false information" and "misleading assessments of my own criticism", that's not relevant to this proposal, really; they're already free to do such things and you're already free to sue them under existing libel and slander laws, which is the only recourse available in this country. Since this right-of-reply is symmetric, you're free to reply to them as well, though you will probably do so by modifying the original web page. Since such concerns are an issue whether or not the right of reply exists it can hardly be considered a reason to dismiss it.
Also, I am also glad this is unconstitutional in the US. Do not mistake my saying this isn't that bad with an active desire to have this; please read more closely. (I use "approval" in the sense of "passing muster", not "desirability".) My views are more nuanced then you seem to be expecting.
I'd just like to echo two blog posts I made, an initial comment and a clarification after seeing how excited people were getting about this.
Summarizing, despite the fact I have one of the most stringent definitions of censorship I know of, this doesn't fit as long as it remains as limited in scope as it actually is. There are a lot of ways to screw this up, but the actual proposal (which should have been linked in the Slashdot article!) actually manages to avoid the traps. As such, this can qualify as a bona-fide cultural difference without destroying the world.
Now, be sure you understand that my approval is fragile and the things that people are reading into the proposal, since they didn't RTFProposal, are indeed scary and it's heartening to see people responding to that. But the limited proposal as it stands is not really a threat, until it is expanded.
(If you are against it because you feel it will inevitably expand into unethical extremes, well, I'd say the odds of that are pretty decent too so I would definately respect that view of things.)
Sure, I've "learned" more out of class. It's almost impossible for class to keep up with a motivated learner; it's only a handful of hours a week.
But I've learned different getting my grad degree. I would not have my familiarity with graph theory, algorithmic complexity, internals of compilers, etc. While perhaps none of these things were as in depth as you'd need to actually do work, the framework I've acquired to handle these in two years in grad school would easily have taken me 10 years of my own to learn... because on your own you get very focused on the task(s) at hand and never learn anything more.
Also, unless you're a rare genius with math, having accountability to a professor to really learn the material, and not just skim a book and fool myself into thinking I understood intensely mathematical material, is invaluable. It's a rare person that can truly force themselves to learn material like that... and for things like graph theory that can be important.
Each of those things is paying off, too, in my work. Graph theory in particular, though it's hard to point at a useless class.
(Of course, if you go into the classes assuming that you'll never get anything useful out of them, you won't.)
Note that I did a Masters, and I did not do a thesis; IMHO two years is too small for a thesis, so I actively chose to take the classes instead, which were more valuable to me. (I feel like this was my thesis, since I was writing it the whole time I was in grad school, but they'd never give me a masters in computer science for that.)
But there is certainly no theorem telling us that it's hopeless.
Yes, there is. Rice's theorem proves it impossible in the general case. And in fact, the very fact that you can produce a proof of correctness tends to argue not that your proof is powerful, but that the algorithm, even if it looks fancy on paper, is trivial. The proof system is necessarily less powerful then the computational system; see Godel's theorem.
And all the such proofs I've seen, while very useful in the real world (it's vital to prove your sorting routine is correct) are of effectively trivial routines.
Basically, the existance of a proof is a certificate of triviality. Mercifully many trivial algorithms are still wonderfully useful, like hashtable algorithms and sorting algorithms. But once you want to do something useful, like accept user input and do something, you can't just string together all of the proven correct algorithms and get another proven-correct algorithm; as the algorithm leaves the domain of "sort this list" or "find the shortest path" or even "find the minimal flow from the source of the graph to the sink of the graph" (an interesting problem who's best-known solution has gone from O(n^5) to O(n^3) over the years, which is really cool) into a couple hundred of these things strung together in arbitrary combinations, your proof system necessarily breaks down.
I'd go on, but this is already below the front page fold by five or ten pages, so who'd read it anyway?
You are free to believe that the human mind has more power then a Turing Machine. I do too, in the end. (Many do not.)
However, computers are definately Turing Machines, and generally when one discusses "computability", we really are referring to computers, not Mind. Even if in the end the mind really is just a Turning Machine we still do not have the constructs yet to deal with it directly (hence the whole field of Psychology). That's still the domain of philosophy.
I'd also point out that the original context of the question, about the computer language Z, firmly places the discussion in the domain of actual computing;-)
While analog is different then digital, digital computing can still simulate analog computing to any aarbitrary desired level of accuracy so it's not a fundamentally different kind of computing. (Fortunately, courtesy of quantum mechanics and other noise sources there is a very real limit on how precise analog computers can be, so you do not need arbitrarily detailed simulations; if you'd like a philosophical application of this point, this is often used in support of the serious proposition that we do live in a simulation, since it drastically reduces the amount of computation necessary to correctly simulate our world.)
(If you're thinking of going into serious computer science, make sure you concentrate on what computer science and math means by the word "arbitrary", as in "arbitrarily large"; it's actually the correct way to say what Slashdotters usually use "infinite" (incorrectly) for. A lot of the really philosophically interesting proofs turn on that word.)
Despite some breathless books by over-zealous writers like Penrose, conciousness remains a mystery.
No, you're wrong. Please see my post on this topic. It is impossible to prove all programs correct for any useful definition of "correct".
There exist proof systems which you seem to be referring to that function by examining the complete state space of a given algorithm, and validate it against some criteria. This only works on algorithms that have a finite state space. Since Turing Machines are allowed to have arbitrarily long tapes, the state space for some algorithms can also be arbitrarily large, which to put it colloquially would require an infinite amount of time to validate.
This is still useful; communications protocols are useful programs that have a very easily characterizable finite state space, and have been validated to prove there are no deadlocks in them, but that's because communications protocols are typically very simple things with tight loops and a small handful of state variables; they are very much the exception, not the rule, and even these simple, ideal cases for state space examination spin out of control exponentially quickly as more state is added to them.
Ironically, both my undergrad and grad level Software Engineering classes spent a disproportionate amount of time on these system, which are only useful in such a small handful of cases, where our coverage of "testing" consisted mainly of regurgitating the traditional, and damn near useless outside of a classroom, definitions of black- and white-box testing. (Nearly all real testing is some shade or other of "gray"; this grudgingly admitted and immediately glossed over.) Pity the time was spent that way, it would have been much more useful to practice writing tests, not using those state-space provers; there's an art to test-writing and it would be much more useful for just about everybody except someone who is 100% academic. (I'm not; I'm as close to 50-50 as I've ever seen anyone, so I have virtually no use for the state-space provers; elegant in their niche, but I can't help but see it as a very, very small niche.)
So what happens when someone is programming in some language/dev-system that won't permit non-correct program ta be wrotted? Eh?
Ironically, it's provable that no Turing Complete language can be limited to create only correct programs for any non-trivial definition of correctness. Computer science is full of such fun things.
Proof: In your supposed language, there is a set of "correct" programs for some criterion, and a set of incorrect programs.
(Sub-proof: The program "return 1" is only correct for the problem consisting of "Always return true" and isomorphic problems, and "return 0" is only correct for the problem consisting of "Always return false" and isomorphic problems. Thus, for any given correct behavior there exists at least one incorrect program in your language, OR your language contains only a trivial number of programs and as such is hardly deserving to be called a "language".)
If there does not exist a Turing Machine (read: 'program') that can decide in finite time whether a given program meets a correctness criterion, then we can never decide whether a program is correct, because Turing Machines are the best computational tools we have. Therefore, suppose some Turing Machine M exists that accepts as input some suitable correctness criterion, and a program, and outputs whether that program meets the correctness criterion (in finite time).
By Rice's theorem, no such TM machine can exist. (Indeed, it's fairly trivial to show that any non-trivial definition of "correctness" must certainly include asserting that the program executes in finite time, and therefore a program that could check such a correctness criterion must be able to solve the Halting Problem, if such a machine existed.)
In fact this sort of proof is one of the reasons I seriously wonder why so many people still seem to be pursuing this. (Fortunately, I see many signs that the field is waking up to this fact and good, dare I say useful research has been starting to get done in the past few years; perhaps the empirical successes of unit testing in software engineering has helped prompt this.)
BTW, no offense but I've been on Slashdot for a while, and I direct this at any replier; if you don't even know what Rice's Theorem is, please don't try to "debunk" this post. Many exceedingly smart people have dedicated their lives to computer science. I only wish I were smart enough to craft these tools, rather then simply use them. Rice's theorum has withstood their attack since 1953. The odds of a Slashdot yahoo "debunking" this theorem are exceedingly low. (On the other hand, if you do understand what I was saying I of course welcome correction and comment; still, this is a fairly straightforward application of Rice's theorem and I can't see what could possibly go wrong. It's a pretty simple theorem to apply, it's just the proof that's kinda hairy.)
Also, your determination that grown biological computing is unlikely to ever achieve its ultimate goal, yeah, we prove that, don't we...
I do not believe biological computing is impossible; we are an obvious counterexample. What I don't think is going to happen is that at any given point, the most powerful (man-manufactured) computing device that exists is biological, because biological systems by definition require a life support system to support them, which must consume space, power, and resources better spent directly on the actual computation by a non-livi
For the practical world, unittests seem to be the way to go. A large number of people have been experiencing very great and real success, without having to themselves be absolute geniuses to use.
Provable programs, on the other hand, are typically so complicated for a real-world program that your provably bug-free program still has bugs, because it turns out your model is wrong, because it was written by a human, not a god.
I'm very unimpressed with the whole "provable program" crowd for any but the most trivial proofs; decades of claims, little to show for it, and all based on the rather wrong idea that if we just push the need for perfection back one level, somehow the computer can magically take up the slack and prevent the bugs from ever existing. The unittesting philosphy, that bugs will happen and it's more important to test and fix them, since complete prevention is impossible anyhow, is much more based in reality, and correspondingly works much better in the real world.
(Provably correct computing is one of those branches of technology that may produce something useful, but is unlikely to ever acheive its stated "ultimate goals"; I class pure q-bit based quantum computing and pure biologically-grown computing in here as well.)
This site is reader driven. You know some news, submit it already.
Plenty of free game information has been posted here. It so happens the commercial game community is much more vibrant, for much the same reasons that the commercial movie industry is more vibrant than the non-commercial one. (Looked at all the credits for a recent game lately?)
Slashdot Mirror
Your metaphors are pointless. Software is not clothing. Non-customized software is useless in a way that non-customized clothing is not. There are perhaps 20-ish common pants sizes for adult men. There are multiple thousands of software configurations for business, one for each business.
So a business can download Apache and install it. So what? Does Apache automatically integrate with accounts receivable? Does it automatically allow the customer to see their pending orders more efficiently then with the competition? Does one upgrade off of a mainframe database by the mere installation of MySQL?
If you insist on a metaphor, free software is a brick for business; it's great that free software drives the materials cost of the "software building" down to zero, but labor always dominated anyhow. The software equivalent of a downloadable house is still non-existent.
So your whole metaphor falls down on the fact that the final product is not freely available; the software equivalent of "commodity jeans" is not available, only the software equivalent of "commodity denim".
If you want to prove I'm wrong, show me the software package I can download off of Sourceforge that all I have to do is install it, and bam, I have a payroll system for a multi-national, multi-thousand employee company, tapping into our existing payroll databases and conversant not just with the laws of multiple countries, but also the multiple union contracts we've negotiated. Hell, show me the commercial package that can do that! Is that an extreme? A little, but you can't even download the equivalent for a four person plumbing company, either.
Failing that, you are pretty much wrong, and there's not much I can say to soften that.
"Software" is a commodity.
"Software you want" is not, especially for businesses, especially for businesses using the software as a skeleton to hang the rest of the business off of, which is basically every business over 50 employees. Oddly, they just don't sell Major University In A Box, or World-Class Automobile Company In A Box, and you sure as hell can't download it from Sourceforge.
"Heironymous' Law" is more like "Heironymous' Really Idealistic But Wrong Idea"; don't hold your breath waiting for it to appear in the Jargon Dictionary.
OMG, you mean benchmarks are subjective?
No, the benchmarks aren't subjective, that's why this is such a big deal. A benchmark consists of a well-specified objective task a computer can do, with objective measurements of how quickly that task was completed.
That's why, if Apple really tweaked the Intel computers sufficiently to nearly cut their performance in half, it's not a "matter of interpretation", it's not "a thing valid for Apple but not valid for me", it's a lie. A computer that can score a 24 can score a 24, period; if you tweak it until it only scores a 15, it can still score a 24, and your tweaks are lies.
You should be able to assemble identical computers, run identical tests, and get the same numbers within an error factor. That's almost the very definition of objective in a scientific sense.
Now, how you interpret the benchmarks is subjective, because no benchmark can possible match everybody's daily use of the computer, so even though Machine A gets a 24 on an integer test and Machine B gets a 22, Machine B may significantly faster for the real-world tasks I do, whereas Machine A may be even faster then B then the test numbers would indicate for somebody else's tasks.
But the whole point of benchmarks is to provide an objective measurement.
(Which you probably knew, but careless word usage leads to careless thinking; "subjective" is the wrong word here. The phrase is "open to interpretation", but you see, that doesn't let Apple off the hook for lying the way that "subjective" would.)
But then one has to wonder why you would have to fight sneaky underhanded tactics within your own government.
Because the government is made up of people, and people use sneaky underhanded tactics. Deplorable, yes, but not surprising.
Since I'm not a Bush fan...
The changes I'm talking about aren't coming from Bush, they're coming from the people.
This particular saying bugs the sh** outta me: "Quietly introduced"
Well, good news: The days of the "quiet introduction" are probably numbered, as evidenced by the fact we picked up this story.
Frankly, the government needs to get moving on passing this sort of legislation; the communication networks that will make this effectively impossible to "quietly introduce" in the future are rapidly forming... the next few years are going to see a massive sea change in politics as at least some of the promises of the Internet are going to start really kicking in...
They dominate... and make no money.... almost no small sites make any money....
Have you ever noticed that he rarely points to sites outside of useit.com and he often is selling his usability reports?
And you aren't the slightest bit interested in the opinions of a person who is running one of the "small sites" and is in all probability making money with it? Think he might just understand a bit?
It sounds to me like you wouldn't be happy no matter what.
Incidentally, figuring out the "blame" for the failure of micropayments is a non-trivial operation; the multi-year stagnation in the browser market (thanks to Netscape's effective disappearence and Microsoft's well-known tendency to not bother with its precious "innovation" unless there's competition) at the same time that the routing market has held a virtual monopoly (ensuring no protocol-level support for micropayments could make any headway), both market conditions and not truly technological conditions, probably had a lot to do with. Despite the fact I'm not holding my breath, they would still solve an awful lot of problems.
You say that bit about "spheres of influence" like it's a bad thing, but it's a necessary part of getting a lot of people to work on one project. You can't get a hundred people to work without them; hell, you can't get four people to work without them. It reduces the communication from n^2 to something more like n log n, which is again why adding people can continue to benefit a project for a long time.
I think those are probably fossil practices left over from pre-Internet days.
It is important to shield the developers from telephone calls and visits from the customer, because otherwise the developers will get nothing done, not necessarily because they have no time (time_in_day - time_dealing_with_customers may still be large), but because they are constantly being interrupted.
As long as the developer has discipline and the customer realizes that the answers won't be instant, though, there can be great value in having email access to the developers, for both parties.
Hopefully more companies will modernize their policies soon.
Two of the three open projects you cite also happen to be bigger than most projects, with more developers working in || at the same time than most projects. Does that help them? Read "the mythical man-month" for the answer. (Hint: no).
You have seriously misread The Mythical Man Month, to the point of absurdity. The Mythical Man Month does emphatically not claim that more people can't do any more then fewer people.
What it says is that programmers are not interchangable, and as a result, adding programmers to an existing project will not get a 1-man-hour to 1-programmer-hour gain. While the old programmers are training the new programmers, and while the communication overhead increases, and while the new programmers are not yet settled in, productivity may fail to increase or may even decrease, over the next few months.
However, once the new guys are settled in and assuming a decent organization, the new organization can indeed move faster.
The Mythical Man Month's main point is to avoid treating programmers, or perhaps more accurately programmer-hours, interchangably. The conclusion based on that is the somewhat more famous part about not adding manpower to a late project because it will make it even more late.
Most strong open-source projects, and probably all the ones in this study, have a strong group of core people who are intimately familiar with what they need to know to continue improving the software. Other people may drift in but will find it difficult to contribute code back. (Seriously, just try to get a patch into Mozilla.) The Mythical Man Month does not apply to steady-state teams of people, only growing ones.
So yes, larger teams help those projects create their large software products. Go look at how much code is in Mozilla; no matter how wizard you are, I don't think a small team of people could even type that much code in anything less then a couple of months. How else do you think they could do it but with a large team?
I find, as a graduate English student, that I can't really think of any generation or era where the intellectual art has really lasted well.
Well, you qualified yourself as an English student but made a claim about "art", so let me take a broad meaning of "art" and mention music.
Up until early in the twentieth century, popular music and academic music were basically one and the same, and the music of the eighteenth and nineteenth century is doing fine. A lot of people (such as myself) listen to it daily, many radio stations remain dedicated to it, and a lot of people still perform. Sure, it's not dominating the music industry, but with so much music, there really isn't any one thing that's dominating any more.
Speaking broadly, it wasn't until the Big Band era, Jazz, Rock and Roll, and eventually almost all music was seperated from academic music... but popular music is actually still largely based on the music theory of the 18th and 19th centuries, though it takes frequent (and justified; I'm not claiming theory uber alles, just mentioning this stuff) liberties with them.
Now academic music is effectively marginalized and even with my rather broad tastes (best I can classify it is as "anything with substance", from classical music to Tubular Bells and even well-done modern video game music; Grandia 2 was superb), I can't stand modern (or perhaps "post-modern" would be better) academic music.
You would recognize many tunes from a hundred years ago, even from the academic tradition of music, even if you couldn't name them. In fact the academic tradition of a hundred years ago has fared better then the popular tradition of that era.
In this post I also kind of assumed that by "intellectual art" I could substitute the word "academic" without loss of meaning.
Well, I think these guys can probably help.
I recommend getting two; lemur production doesn't work so well with only one lemur machine. Make sure they are compatible; there are two types and you need one of each.
By the way, it's quite likely that local laws will prevent you from owning a lemur-producing machine. Many types of lemur-producing machines are very rare and once they break, they can't be fixed, so if they all break that's it for lemur-machines. As a result many governments are quite protective of them.
Also, operation of the lemur machines in such a manner that they will actually produce lemurs and not just consume resources may be very, very tricky; I'm not certain but I do know a lot of other, similar machines often take many skilled experts to cause to produce.
Anyhow, they're not as user-friendly as you'd like and you can't just pick them up at Best Buy, but you can probably get some lemur machines if you really put your mind to it. Best part of all is there are billions of years behind their development, so they are pretty sophisticated.
It's not so much a "study" as something that just always happens; in the computation domain, it unfortunately means you can never know whether you should shut off the simulation, because you might be one iteration away from a 25% improvement, whereas steady growth would make it easy to know whether you've reached the end of a computation.
Hmmm... there's probably a theorem here waiting to get out.
The difference is that figuring out whether the wheel is slipping is merely a hard problem (it is not an easy problem, and there are a lot of tricky parts, but it is solvable), and it can be done in the domain of "the speed of the wheel".
Deciding whether to stop or not require knowlege of the outside world, which not surprisingly is a lot more complicated then the problem of whether you're slipping. That's a four-dimensional problem, one speed reading on each wheel. Deciding to stop, especially the way we do it with our vision system, is an n-dimensional problem, where "n" is some big number. (Easily thousands.) Extrapolating our success with a four-dimensional problem that doesn't even need heavy AI to solve into thinking that we can equivalently solve a multi-thousand dimensional problem (with only two or three of those dimensions available to our system, no less!) is quite overconfident, to say the least.
Insurance companies are quite pragmatic. If this system was deployed I expect they'll require most drivers to shut it off, or they'll deny coverage for anything they can blame on the system, because I sincerely doubt this system will be a net safety gain.
This is completely different. "Self-powered wagons" are driven by humans. Automating a technology is completely different then simply creating it.
So no, fear that we can't successfully automate something does not apply to all technology, it only applies to attempts to automate things, and we have not had success in automating real-world applications, only relatively "easy" things like factory-line robots, doing the same thing in the same circumstances every time with humans standing by to take over if there's a problem of any kind, and nobody dies if there is a problem.
Concern over automation is quite justified; this kind of AI has been defeating us for 50 years now and frankly the situation hasn't changed much, nor does it seem likely to change terribly soon.
Oh come on... either you are lying to us and you just obviousally told all of us that by making that statement or you dont have enough coffee to think this morning...
.4 seconds it takes for the driver to react that you could have been braking, the car continues at 70 mph. By the time the driver reacts, its too late to slow the car down to a survivable speed. The driver hits the cab of the truck and dies. You get sued because your system was supposed to hit the brakes for the driver.
No, the problem is you obviously have no experience dealing with real world data. Writing software is a lot harder then "seeing a way to do it". For instance...
1 - simply measure the speed of the car in front of you relative to you.
Not even all the way through your first sentence and you've already hit major problems. First, all measurements have noise in them; you can't know the "speed" with perfect reliability, and how you determine the speed from the incoming data is a non-trivial problem. Since lives are on the line, it's important to get it right. A simple "moving average", for instance, won't work because it will slow the reaction time of the system down unacceptably.
Also, we have a problem with "car in front of you". Now, "thing in front of you" may work, but A: how do you know it's a car? and B: how do you know it's truly in front of you? Maybe you're just turning and that "thing" you see coming at you is just a mailbox.
Pop quiz: You are a radar system. You detect something coming at you at 70 mph from a hundred feet away. You're turning slightly, but not much. As a radar system in a car, this is all you know. Do you slam on the brakes?
Yes? Oops. Turns out that you're on the left lane of a three-lane highway, and somebody's parked their car on the right shoulder due to car problems. The car gets rear-ended by a tailgaiting Mack truck and the driver dies. You get sued because your system was not supposed to hit the brakes then.
No? Oops. There was a Mack truck jackknifed across the road which the driver wasn't looking at and didn't see. In the
This isn't so easy, is it? The car only has a very limited number of senses; in human terms it's blind except for one whole pixel, deaf, and can only feel limited information about it's own state ("I'm going 50", "I'm turning", etc.). It's not like the car has a picture of the whole road to work with... and of course the reason it doesn't have that is we wouldn't know what to do with it if we did. Computer vision is still a very limited technology; it can not handle the real world to any significant degree yet.
if the car is doing 50mph toward the radar source, hit the breaks!
How hard? What threshold for distance? Does the distance threshold change if the relative velocity changes? How does the brake hitting relate to the relative speed? What if I'm driving down the road at 60mph, I go into a turn, and a mailbox momentarily looks like it's coming at me at 55mph? Do I slam on the brakes then, or wait until I'm sure it's a car? How can I be sure when all I have is a distance and a relative velocity anyhow?
If you're wrong and you "hit the brakes", you may end up killing the user. Lives are on the line; you can't gloss over any of these questions. One wrong answer and people die.
you simply have a threshhold setting for distance to trigger and speed.
That is naive beyond words. The real world is much more complicated then that.
and simply adjust that according to the car's current speedometer reading.
How, exactly? "adjust" is an empty word until you give more specification. It's damned easy to say "simply adjust", but there are hundreds of plausible way to "simply adjust", and most of them are wrong. (It's even possible all of them are wrong; there's no guarentee this system is even possible, you know.)
2 - to hell with the speed of X Y or Z look at the doppler shift. if I detec
No, your use of the word "advertising", as if it has to go on the homepage, is an exaggeration not present in the proposal. It says this: "Prominent place" should probably be interpreted in the light of the very next paragraph:I would expect that a clearly visible link on the original criticism would meet the criterion of the first paragraph.
Another problem is that, even with links, disagreements are not easily settled by a single reply.
This isn't about settling disagreements; nobody is pretending that is possible.
Do I have the right to reply to your reply, and you to reply in turn... this to be repeated ad infinitum? Or will I be forced to offer a link to your reply even when your reply contains false information and misleading assessments of my own criticism?
The proposal addresses neither issue directly. However, as a practical matter that won't happen, because the link being pointed to can be changed as necessary to update new conditions; new links will not need to be continuously posted so that's a bugaboo.
As for "false information" and "misleading assessments of my own criticism", that's not relevant to this proposal, really; they're already free to do such things and you're already free to sue them under existing libel and slander laws, which is the only recourse available in this country. Since this right-of-reply is symmetric, you're free to reply to them as well, though you will probably do so by modifying the original web page. Since such concerns are an issue whether or not the right of reply exists it can hardly be considered a reason to dismiss it.
Also, I am also glad this is unconstitutional in the US. Do not mistake my saying this isn't that bad with an active desire to have this; please read more closely. (I use "approval" in the sense of "passing muster", not "desirability".) My views are more nuanced then you seem to be expecting.
I'd just like to echo two blog posts I made, an initial comment and a clarification after seeing how excited people were getting about this.
Summarizing, despite the fact I have one of the most stringent definitions of censorship I know of, this doesn't fit as long as it remains as limited in scope as it actually is. There are a lot of ways to screw this up, but the actual proposal (which should have been linked in the Slashdot article!) actually manages to avoid the traps. As such, this can qualify as a bona-fide cultural difference without destroying the world.
Now, be sure you understand that my approval is fragile and the things that people are reading into the proposal, since they didn't RTFProposal, are indeed scary and it's heartening to see people responding to that. But the limited proposal as it stands is not really a threat, until it is expanded.
(If you are against it because you feel it will inevitably expand into unethical extremes, well, I'd say the odds of that are pretty decent too so I would definately respect that view of things.)
Sure, I've "learned" more out of class. It's almost impossible for class to keep up with a motivated learner; it's only a handful of hours a week.
But I've learned different getting my grad degree. I would not have my familiarity with graph theory, algorithmic complexity, internals of compilers, etc. While perhaps none of these things were as in depth as you'd need to actually do work, the framework I've acquired to handle these in two years in grad school would easily have taken me 10 years of my own to learn... because on your own you get very focused on the task(s) at hand and never learn anything more.
Also, unless you're a rare genius with math, having accountability to a professor to really learn the material, and not just skim a book and fool myself into thinking I understood intensely mathematical material, is invaluable. It's a rare person that can truly force themselves to learn material like that... and for things like graph theory that can be important.
Each of those things is paying off, too, in my work. Graph theory in particular, though it's hard to point at a useless class.
(Of course, if you go into the classes assuming that you'll never get anything useful out of them, you won't.)
Note that I did a Masters, and I did not do a thesis; IMHO two years is too small for a thesis, so I actively chose to take the classes instead, which were more valuable to me. (I feel like this was my thesis, since I was writing it the whole time I was in grad school, but they'd never give me a masters in computer science for that.)
But there is certainly no theorem telling us that it's hopeless.
Yes, there is. Rice's theorem proves it impossible in the general case. And in fact, the very fact that you can produce a proof of correctness tends to argue not that your proof is powerful, but that the algorithm, even if it looks fancy on paper, is trivial. The proof system is necessarily less powerful then the computational system; see Godel's theorem.
And all the such proofs I've seen, while very useful in the real world (it's vital to prove your sorting routine is correct) are of effectively trivial routines.
Basically, the existance of a proof is a certificate of triviality. Mercifully many trivial algorithms are still wonderfully useful, like hashtable algorithms and sorting algorithms. But once you want to do something useful, like accept user input and do something, you can't just string together all of the proven correct algorithms and get another proven-correct algorithm; as the algorithm leaves the domain of "sort this list" or "find the shortest path" or even "find the minimal flow from the source of the graph to the sink of the graph" (an interesting problem who's best-known solution has gone from O(n^5) to O(n^3) over the years, which is really cool) into a couple hundred of these things strung together in arbitrary combinations, your proof system necessarily breaks down.
I'd go on, but this is already below the front page fold by five or ten pages, so who'd read it anyway?
You are free to believe that the human mind has more power then a Turing Machine. I do too, in the end. (Many do not.)
;-)
However, computers are definately Turing Machines, and generally when one discusses "computability", we really are referring to computers, not Mind. Even if in the end the mind really is just a Turning Machine we still do not have the constructs yet to deal with it directly (hence the whole field of Psychology). That's still the domain of philosophy.
I'd also point out that the original context of the question, about the computer language Z, firmly places the discussion in the domain of actual computing
While analog is different then digital, digital computing can still simulate analog computing to any aarbitrary desired level of accuracy so it's not a fundamentally different kind of computing. (Fortunately, courtesy of quantum mechanics and other noise sources there is a very real limit on how precise analog computers can be, so you do not need arbitrarily detailed simulations; if you'd like a philosophical application of this point, this is often used in support of the serious proposition that we do live in a simulation, since it drastically reduces the amount of computation necessary to correctly simulate our world.)
(If you're thinking of going into serious computer science, make sure you concentrate on what computer science and math means by the word "arbitrary", as in "arbitrarily large"; it's actually the correct way to say what Slashdotters usually use "infinite" (incorrectly) for. A lot of the really philosophically interesting proofs turn on that word.)
Despite some breathless books by over-zealous writers like Penrose, conciousness remains a mystery.
No, you're wrong. Please see my post on this topic. It is impossible to prove all programs correct for any useful definition of "correct".
There exist proof systems which you seem to be referring to that function by examining the complete state space of a given algorithm, and validate it against some criteria. This only works on algorithms that have a finite state space. Since Turing Machines are allowed to have arbitrarily long tapes, the state space for some algorithms can also be arbitrarily large, which to put it colloquially would require an infinite amount of time to validate.
This is still useful; communications protocols are useful programs that have a very easily characterizable finite state space, and have been validated to prove there are no deadlocks in them, but that's because communications protocols are typically very simple things with tight loops and a small handful of state variables; they are very much the exception, not the rule, and even these simple, ideal cases for state space examination spin out of control exponentially quickly as more state is added to them.
Ironically, both my undergrad and grad level Software Engineering classes spent a disproportionate amount of time on these system, which are only useful in such a small handful of cases, where our coverage of "testing" consisted mainly of regurgitating the traditional, and damn near useless outside of a classroom, definitions of black- and white-box testing. (Nearly all real testing is some shade or other of "gray"; this grudgingly admitted and immediately glossed over.) Pity the time was spent that way, it would have been much more useful to practice writing tests, not using those state-space provers; there's an art to test-writing and it would be much more useful for just about everybody except someone who is 100% academic. (I'm not; I'm as close to 50-50 as I've ever seen anyone, so I have virtually no use for the state-space provers; elegant in their niche, but I can't help but see it as a very, very small niche.)
So what happens when someone is programming in some language/dev-system that won't permit non-correct program ta be wrotted? Eh?
Ironically, it's provable that no Turing Complete language can be limited to create only correct programs for any non-trivial definition of correctness. Computer science is full of such fun things.
Proof: In your supposed language, there is a set of "correct" programs for some criterion, and a set of incorrect programs.
(Sub-proof: The program "return 1" is only correct for the problem consisting of "Always return true" and isomorphic problems, and "return 0" is only correct for the problem consisting of "Always return false" and isomorphic problems. Thus, for any given correct behavior there exists at least one incorrect program in your language, OR your language contains only a trivial number of programs and as such is hardly deserving to be called a "language".)
If there does not exist a Turing Machine (read: 'program') that can decide in finite time whether a given program meets a correctness criterion, then we can never decide whether a program is correct, because Turing Machines are the best computational tools we have. Therefore, suppose some Turing Machine M exists that accepts as input some suitable correctness criterion, and a program, and outputs whether that program meets the correctness criterion (in finite time).
By Rice's theorem, no such TM machine can exist. (Indeed, it's fairly trivial to show that any non-trivial definition of "correctness" must certainly include asserting that the program executes in finite time, and therefore a program that could check such a correctness criterion must be able to solve the Halting Problem, if such a machine existed.)
Since any language that only allowed one to express correct concepts is isomorphic to a Turing Machine that correctly validates the existance in that language, and no such TM can exist, no such language can exist, unless it is trivial and the correctness criterion is trivial. In which case as I alluded to earlier, it's hardly worth calling a "language" in the computer science sense.
In fact this sort of proof is one of the reasons I seriously wonder why so many people still seem to be pursuing this. (Fortunately, I see many signs that the field is waking up to this fact and good, dare I say useful research has been starting to get done in the past few years; perhaps the empirical successes of unit testing in software engineering has helped prompt this.)
BTW, no offense but I've been on Slashdot for a while, and I direct this at any replier; if you don't even know what Rice's Theorem is, please don't try to "debunk" this post. Many exceedingly smart people have dedicated their lives to computer science. I only wish I were smart enough to craft these tools, rather then simply use them. Rice's theorum has withstood their attack since 1953 . The odds of a Slashdot yahoo "debunking" this theorem are exceedingly low. (On the other hand, if you do understand what I was saying I of course welcome correction and comment; still, this is a fairly straightforward application of Rice's theorem and I can't see what could possibly go wrong. It's a pretty simple theorem to apply, it's just the proof that's kinda hairy.)
Also, your determination that grown biological computing is unlikely to ever achieve its ultimate goal, yeah, we prove that, don't we...
I do not believe biological computing is impossible; we are an obvious counterexample. What I don't think is going to happen is that at any given point, the most powerful (man-manufactured) computing device that exists is biological, because biological systems by definition require a life support system to support them, which must consume space, power, and resources better spent directly on the actual computation by a non-livi
For the practical world, unittests seem to be the way to go. A large number of people have been experiencing very great and real success, without having to themselves be absolute geniuses to use.
Provable programs, on the other hand, are typically so complicated for a real-world program that your provably bug-free program still has bugs, because it turns out your model is wrong, because it was written by a human, not a god.
I'm very unimpressed with the whole "provable program" crowd for any but the most trivial proofs; decades of claims, little to show for it, and all based on the rather wrong idea that if we just push the need for perfection back one level, somehow the computer can magically take up the slack and prevent the bugs from ever existing. The unittesting philosphy, that bugs will happen and it's more important to test and fix them, since complete prevention is impossible anyhow, is much more based in reality, and correspondingly works much better in the real world.
(Provably correct computing is one of those branches of technology that may produce something useful, but is unlikely to ever acheive its stated "ultimate goals"; I class pure q-bit based quantum computing and pure biologically-grown computing in here as well.)
This site is reader driven. You know some news, submit it already.
Plenty of free game information has been posted here. It so happens the commercial game community is much more vibrant, for much the same reasons that the commercial movie industry is more vibrant than the non-commercial one. (Looked at all the credits for a recent game lately?)