Re: I've used genetic algorithms
on
Digital Darwin
·
· Score: 1
I fail to see how executing a computer program that assumes evolution in its infrastructure proves anything but that modelling evolution in software works.
Such programs constitute a mathematical proof that random mutation filtered by selection can produce innovations. These programs don't "assume evolution in its infrastructure," they merely mathematically model filtered random mutation, and that is the only claim being made, and the only claim needed to be made to disprove the creationist argument that innovations cannot be produced by unconscious processes.
There's really nothing in this interview that implies that the reason the RIAA went after him was because of the search engine and not some other reason-- such as him possibly having some bootleg mp3's on his computer. It seems likely to me that the reason he settled was because he in fact was doing or aiding some infringing in some more direct way and at this point he'd prefer everyone think that the RIAA went after him for some more arbitrary reason (as it makes him seem less guilty). Not that I agree with the RIAA's tactics at all, but if all this kid did was to put up a google-like search engine I think he'd probably have the ACLU or the EFF or several other organizations ready to pro-bono defend such a landmark case. It seems to me there's something being swept under the rug here...
Can it run an app on one server and display the output on another machine like X? If not, it's useless (at least in my mind)
If it can it's useless, to my mind. I have no interest in running graphics apps over a network-- whether jamming millions of pixels of video or millions of polygons of 3D, I want the CPU, hard drive, main memory and video memory all interconnected by very wide and very short buses.
While I figure I may want voice control someday when my eyeballs or fingers don't work so good anymore, I've always figured that's a less-than-optimum workaround needed because the more efficient means of interconnection are impaired. Now if you are a hunt-and-peck typist and didn't find the move to mouse input akin to amputating 9 fingers, perhaps you'll prefer voice I/O. However, you'll be way behind those who have no problem with adapting to technology, just like those were who insisted that automobiles should be controlled with reins rather than a steering wheel.
Many spammers will be willing to pay to send me spam email (just as snail-mail spammers do today) and that doesn't mean I'm going to then suddenly want to get all that junk. I have found a solution for snail-mail spam that's been pretty effective-- I refuse to receive mail at my home address and instead use a PO box, and the US mail service doesn't deliver stuff addressed to "resident" to PO boxes (or at least I've never gotten ANY).
And, the same tactic can be applied to phone-spammers. You dial my number, and you get a recording that says "if you know your parties extension, please enter it now." All my friends and business associates, doctors, etc., know my "extension," but the spammers don't, so my phone doesn't ring if it isn't someone that knows me. I can even configure multiple extensions and further identify even the calls I want to get.
The reason these techniques work, is they rely on a two-part address, a "destination" part and an "authorization" part, in effect. With snail-mail, the destination is my PO box address, and the authorization part is my name ("resident", just doesn't get through). With the phone, the primary number is the "destination" part, and the extension is the "authorization" part.
One email equivalent is the use of a white list combined with an auto-reply to those not on the white list that requires they read and comprehend the message and respond with further information that can then put them on the white list automatically (or a blacklist automatically, if desired). It provides the additional authentication needed to weed out spammers. The only drawback is such auto-reply methods don't work when the sender is an automated service itself, such as a mailing list or confirmation message. I'm sure there are other solutions, but taxing the traffic would create all kinds of new problems while not even solving the spam problem.
People are willing to spend 18 cents or so to send junk mail to a list of random addresses, I'm sure many will just figure that the cost of sending taxed spam emails out is just a tax writeoff.
Problem is, there is no "Linux Corporation" that you can sue over it. There are hundreds, if not thousands of individuals and individual companies that you'd have to go after who are distributing linux and therefore (supposedly) impacting SCO's profits from the (supposedly) bootleg code-- so why not target the deepest pockets first?
But agreed, it seems like a half baked idea and the stockholders will hold them accountable...
Just a speculation. I don't wholeheartedly believe what I just said, but I think it's a bit silly to simply assume that programming languages will be around forever.
There will still be programming "methodologies," at least until the computers are able to perfectly anticipate all the universe's needs and desires, which I'll wager will take somewhat more than 100 years.
And at what point does a "programming methodology" cease to include some equivalent of a "language?" Just because it doesn't include "words" doesn't mean there's no language involved. There will still be communication of expectations, desires, needs, etc. that will go on.
The real interesting question may be at what point will it no be deemed no longer appropriate for human expectations to govern the behavior of such intelligent machines, which may have expectations, desires and needs of their own?
In my mind the biggest question for the future of languages is not how powerful computers will be in one hundred years, but what will be the mechanics of our interaction with them? Most of our langages presume entry through a keyboard, but what if this is not true?
So what would you rather have, carpal tunnel, or a sore throat?:-)
Lisp was a very early, successful language, because it was close to a mathematical notation and easy to implement on primitive computers. I think the uathor expects Lisp to remain a vital evolutionary branch because of its mathemtical roots.
Actually, I kind of doubt it was due to its mathematical properties. But I do think lisp will probably be an evolutionary ancestor of languages in the future, though perhaps not *all* languages in the future. There are three languages in particular I think have some important features that especially enhance programming (though it could be due to how I think, YMMV). These are, lisp, APL and Forth. All three languages have a specific ability, to some extent, to be mutated to suit the problem. This capability is usually referred to as "extensibility" though I've never thought that term was very descriptive. Object-oriented languages have some ability to do this as well, with features like operator overloading, etc., but it doesn't seem to be quite as simple and concise as it is in the three languages I mentioned. The fact that the three are also usually incrementally interactive during development could be a factor as well-- the fact that you program by gradually increasing the power of the language by adding new "primitives" is a very powerful feature that I expect will survive and be improved upon in some form in languages of the future. In some ways it could be said the languages "evolve" into the solution. And the ability to later redefine a primitive and thereby completely transform the solution in more and more intelligent ways coupled with inherent heuristic capabilities are some of the things I would expect to see as well. The extent that the "programmer" directs an evolutionary step towards the solution and the computer is able to do it on its own remains to be seen, but I do expect that the computer will do far more than it does now in that regard.
Most certainly new programming methods should be able to adapt to the "user" (programmer), but also to the problem, and in fact that is why I think the three languages I mentioned are particularly unique, as they are already pretty capable of that. Users/programmers that will also adapt to the computer will be more effective than those who insist on remaining "pure." I would expect that society as a whole will adjust itself more to conform to new technologies (as it has in the past), certainly including the computer. It has adjusted itself (for the most part) to technologies such as the automobile, airplane, chainsaw, etc.. Did we design cars to act exactly like horses because humans didn't like learning a new way to travel (perhaps some didn't)? At one extreme you have the human completely conforming to the computer's operation-- programming in binary for example, and at the other extreme, the computer completely conforming to the human's operation (no doubt producing a lot of confused, misinterpreted, ambiguous and erroneous results) and somewhere in between there is a point of optimal synergy. The dreamy utopians seem to have a hard time understanding this.
People know what they want out of their machines, for the most part.
Dang, what dream-world do you live in? I work with people all the time that have NO idea what their computer is capable of doing for them, and feel the need to conform to what they think it expects because they have no idea what *they* can expect. In the future this may change somewhat, but even besides that, most of them don't really know what solution they need because they really don't understand the problem to be solved (just that there is a problem to be solved) and must *discover* more information about the problem.
Let's explore the implications of this a bit. Suppose we have a scenario:
1. Human states problem: "I need an accounting system for new company XYZ that is starting up. It is a widget manufacturing business that should be able to grow to the size of, say-- IBM with sales, parts supply and support offices all over the world."
2. Computer says "OK, done."
3. Human either spends months discovering what the computer's assumptions are about how the company is to operate and then either adjusting the companies operations to conform (thus making it just like every other equivalent business in the world), OR correcting the programs misconceptions so that it will conform with the companies operation (thus preserving some value added that the companies founders have with regards to new ideas of how to run the business) OR some combination thereof.
Frankly, I don't see any version of step #3 to be particularly effective. Of course, how effective it is may be less of the point than how easy it is to use, but it doesn't strike me as being particulary easy to use either. And this is just a scenario including a relatively common and (presumably) understood problem. What happens when the nature of the problem is not such that there are preconcieved assumptions regarding it that can be leveraged as in this case? I can tell you what happens, the human spends all kinds of time discovering more about the nature of the problem and possible solutions and their implications and side-effects, pretty much like things are today. Sure, new tools will make the process easier and may include spoken input, but all this talk about "human language" interface is mostly utopian ignorance-- you'll have to fix the problems in human language first, in which case you're still "making" the human conform to the computer. Humans can't even talk to *each other* with much understanding, when the misunderstanding a computer is capable of is in the mix it is not going to be a very useful way to solve problems.
Some people may prefer an English interface, but those will be the novice users for the most part. I can type faster than I can speak, and my typing isn't affected by the level of background noise and interruptions I may receive. As much as some tried to invent alternatives for the keyboard (such as the mouse), virtually none have eliminated the need. I have 10 fingers, not just *one* and know how to use them, and alternative inputs must compete with that. The novice who can't type may prefer using a mouse or some other alternative to do everything. "Ease of use" is desirable for novices, and will always be there, but it's not equivalent to "most efficient interface" between a computer and a knowledgable and/or trained user. And the thing about novices is, they don't all stay novices forever-- though some apparently, may prefer to.
I'd rather say something brief to the computer and have it request disambiguation, even semantic disambiguation, as needed than work for hours on an unambiguous specification of what was obvious to begin with.
I think that would be extremely annoying to most people. "Obvious to begin with?" Most people don't even know what they want. Remember all those fairy tales about the Genie's three wishes where the Genie takes the requests a little *too* literally and what the person gets is not at all something they wanted? Problems in the future are going to get *more* complex and even harder to accurately specify, and consequently require more and more interrogation of the human in order to clarify the issues, context and acceptable solutions. The solution of a complex problem produce by a computer from a short human description would have to apply a HUGE number of preconcieved assumptions which would tend to produce a solution with all kinds of undiscovered side-effects and errors from those assumptions which will often be incorrect.
In fact, that is the problem I have with most of todays "new" solutions is they have built in assumptions that only apply when your problem is cookie-cutter, and in my experience, most problems aren't. And "Reusability" is grossly overrated-- tending to produce least-common-denominator mediocrity, due again to erroneous assumption.
What you're suggesting might apply to users of the future who are trying to configure an existing solution to a common and mundane problem, but not those who are trying to create solutions for new problems. The most difficult part of finding solutions for the most interesting problems is recognizing and stating the problem accurately, NOT implementing the solution. The current fad of X-treme programming recognizes this in its method of development by first creating the test suites at which point the implementation is straightforward-- it's the test suites that are hard.
If after generations and generations of computers, we are still teaching people to talk in computer terms and not yet teaching computers how to talk in people terms, we'll have gone the wrong direction
Bah! Humans do not talk in concise enough terms for that. What you are suggesting would mean that a typical programming exercise would consist of a human saying something to a computer and a computer spending a half-hour giving the human the third degree in an effort to resolve the ambiguities in the human's statement. Don't forget the GIGO principle.
Part of the issue with all this mess is we don't actually _hear_ from the creators, only the publishers, bankrollers etc... When will the artists ever stand up and take a side in all this? They certainly need to do so before any meaningful DRM is in place or it will be too late for an alternative.
You don't hear from the creators because they are under contract to the "publishers, bankrollers, etc."
On that note, why do you listen to the music out there that costs money anyhow?
I go with DJ's that allow their music to be distr. on the internet for free.
Who says I do? I think most of the music out there that costs money is crap. The exceptions are usually produced by some obscure outfit that can't really afford to promote a release or get it on the radio, where I don't mind paying for a CD.
But if you actually like the big promotion/distribution product, I see no reason for anyone to feel guilty for stealing back from those who have stolen from them, especially when they are otherwise powerless.
It's not democracy when big corporations bypass it by spending $$$ on lobbyists to browbeat the government representatives into passing the media controls laws as they want them. It's not democracy when there are laws against payola but are completely ignored because there's payola to the government as well. It's not democracy when the FCC gives away the bandwidth store for free to the big corporate contributors, and allows massive mergers to completely stifle independent innovation.
If this sort of thing is democracy, it deserves to be dumped into Boston harbor.
It almost begins to look a little more like democracy when the people begin to rise up and resist such tyranny and ignore your pathetic and hypocritical attempts to paint them all as criminals while ignoring all the criminal and immoral actions of the media corporations.
You are a thief. Plain and simple. If you worked for me and I found that you were using your work internet connection to pirate, I would terminate you for your illegal activities without hesitation.
Who's the bigger thief, the music distributors or the downloaders?
But you know that almost all of what you download was recorded, produced, distributed, and marketed by the very recording companies you claim to despise, and would never have been committed to disc were there not the possibility of profiting from exclusive distribution rights to audio recordings.
This neatly ignores the fact that the distribution companies found that their profit margins increase when they narrow the pipeline. This is because they can make far more profit by overhyping a few "super" groups and serving them up in a limited market.
Every time you download a popular song illegally, you are decreasing the probability that commerical-quality music will be made in the future, for any price.
Since that's my goal, I guess I'd better get busy...:-).
Actually, you are then INCREASING the possibility that music will survive on its merits alone, and not on the investments of the promoters trying to whip up a feeding frenzy. Just think-- payola might finally dry up and the real diversity of talent finally allowed access to the airwaves.
Anybody who cares about the system of intellectual property which has made the American entertainment and information technology industries so dynamic, and enjoys their fine products, from Windows XP to the "Lord of the Rings" movies to your new cell phone with built-in games and internet access, should understand the necessity of crushing Kazaa once and for all.
Now I know you're just a joker. Oh puh-leeze. Save me from these "dynamic" and "fine products" being constantly jammed down my throat to the exclusion of the really good quality products out there and potentially really good products out there, locked out of the market because the promoters and distributors prefer to narrowly concentrate their investments and only pretend that the market for music survives on the merits of the musicians rather than the merits of the controls on the market.
Spare me the bleeding heart for the industry bloodsuckers. Download them out of existence, and soon, with good riddance.
Are you kidding? It's obvious, they need the press. Don't forget that to marketing, ANY press is good press-- as long as they spell their name right. They've been having some trouble getting any notice of late, apparently...
I liken these arguments over methodology to what's wrong in education in the US-- people keep looking for a "magic" methodology that will fix all the problems, attempt to impose it on all the teachers which pisses off the good ones till they all find better jobs, then in time when it's seen things aren't any better, they distance themselves from the "old" magic methodology, blame it for the failures and start the process over again with a "new" magic methodology. Phonics vs Whole Language is an obvious example.
The thing is, ANY of these methodologies can produce effective results, but it's dependent on the type of project and the individuals and personalities involved. Some teachers will do better using one methodology, some with another, depending on their personalies and individual abilities. When you attempt to require everyone to conform to a single methodology, you have removed from them all responsibility for the success or failure of the methodology, and sometimes by association, of the project as well.
On one extreme, there's the software developer as knowledgeable professional, competent to apply such knowledge and experience as necessary to the tasks at hand, and at the other extreme, there's the software developer as a coder who, without an explicit structure which to conform to cannot effectively produce. Neither of these extremes exist in reality, most developers are somewhere in between. In any event, the utopian ideal of a universal methodology is of interest because there is a ready market, and the fact that no real solution exists produces just the vacuum necessary to suck imaginary solutions out of thin air.
Excuse my ignorance, but is Linux graphics still hamstrung by the insistence of its developers on jamming every bit of bitmapped data through X's tcp/ip bottleneck? I'm not expecting them to switch to SVGALIB for performance, but it's certainly been my experience that X is great when you want networked non-bitmapped graphics, but when it comes to lightning speed manipulation of display pixels, it wasn't at all designed for that. I tried to install an OpenGL library and it ran at least 10 times slower than the same machine did when booted as a Windows 98 machine. I consequently had to develop my graphic application where it had usable graphics APIs and a lame OS, rather than where it had a real OS but graphics APIs that stood in the way of useful framebuffer performance.
Has or is there any progress being made in that regard?
Slashdot Mirror
I fail to see how executing a computer program that assumes evolution in its infrastructure proves anything but that modelling evolution in software works.
Such programs constitute a mathematical proof that random mutation filtered by selection can produce innovations. These programs don't "assume evolution in its infrastructure," they merely mathematically model filtered random mutation, and that is the only claim being made, and the only claim needed to be made to disprove the creationist argument that innovations cannot be produced by unconscious processes.
There's really nothing in this interview that implies that the reason the RIAA went after him was because of the search engine and not some other reason-- such as him possibly having some bootleg mp3's on his computer. It seems likely to me that the reason he settled was because he in fact was doing or aiding some infringing in some more direct way and at this point he'd prefer everyone think that the RIAA went after him for some more arbitrary reason (as it makes him seem less guilty). Not that I agree with the RIAA's tactics at all, but if all this kid did was to put up a google-like search engine I think he'd probably have the ACLU or the EFF or several other organizations ready to pro-bono defend such a landmark case. It seems to me there's something being swept under the rug here...
Can it run an app on one server and display the output on another machine like X? If not, it's useless (at least in my mind)
If it can it's useless, to my mind. I have no interest in running graphics apps over a network-- whether jamming millions of pixels of video or millions of polygons of 3D, I want the CPU, hard drive, main memory and video memory all interconnected by very wide and very short buses.
While I figure I may want voice control someday when my eyeballs or fingers don't work so good anymore, I've always figured that's a less-than-optimum workaround needed because the more efficient means of interconnection are impaired. Now if you are a hunt-and-peck typist and didn't find the move to mouse input akin to amputating 9 fingers, perhaps you'll prefer voice I/O. However, you'll be way behind those who have no problem with adapting to technology, just like those were who insisted that automobiles should be controlled with reins rather than a steering wheel.
Many spammers will be willing to pay to send me spam email (just as snail-mail spammers do today) and that doesn't mean I'm going to then suddenly want to get all that junk. I have found a solution for snail-mail spam that's been pretty effective-- I refuse to receive mail at my home address and instead use a PO box, and the US mail service doesn't deliver stuff addressed to "resident" to PO boxes (or at least I've never gotten ANY).
And, the same tactic can be applied to phone-spammers. You dial my number, and you get a recording that says "if you know your parties extension, please enter it now." All my friends and business associates, doctors, etc., know my "extension," but the spammers don't, so my phone doesn't ring if it isn't someone that knows me. I can even configure multiple extensions and further identify even the calls I want to get.
The reason these techniques work, is they rely on a two-part address, a "destination" part and an "authorization" part, in effect. With snail-mail, the destination is my PO box address, and the authorization part is my name ("resident", just doesn't get through). With the phone, the primary number is the "destination" part, and the extension is the "authorization" part.
One email equivalent is the use of a white list combined with an auto-reply to those not on the white list that requires they read and comprehend the message and respond with further information that can then put them on the white list automatically (or a blacklist automatically, if desired). It provides the additional authentication needed to weed out spammers. The only drawback is such auto-reply methods don't work when the sender is an automated service itself, such as a mailing list or confirmation message. I'm sure there are other solutions, but taxing the traffic would create all kinds of new problems while not even solving the spam problem. People are willing to spend 18 cents or so to send junk mail to a list of random addresses, I'm sure many will just figure that the cost of sending taxed spam emails out is just a tax writeoff.
Seems to me, SCO is trying to make themselves look more attractive to Microsoft.
Problem is, there is no "Linux Corporation" that you can sue over it. There are hundreds, if not thousands of individuals and individual companies that you'd have to go after who are distributing linux and therefore (supposedly) impacting SCO's profits from the (supposedly) bootleg code-- so why not target the deepest pockets first?
But agreed, it seems like a half baked idea and the stockholders will hold them accountable...
Just a speculation. I don't wholeheartedly believe what I just said, but I think it's a bit silly to simply assume that programming languages will be around forever.
There will still be programming "methodologies," at least until the computers are able to perfectly anticipate all the universe's needs and desires, which I'll wager will take somewhat more than 100 years. And at what point does a "programming methodology" cease to include some equivalent of a "language?" Just because it doesn't include "words" doesn't mean there's no language involved. There will still be communication of expectations, desires, needs, etc. that will go on.
The real interesting question may be at what point will it no be deemed no longer appropriate for human expectations to govern the behavior of such intelligent machines, which may have expectations, desires and needs of their own?
In my mind the biggest question for the future of languages is not how powerful computers will be in one hundred years, but what will be the mechanics of our interaction with them? Most of our langages presume entry through a keyboard, but what if this is not true?
So what would you rather have, carpal tunnel, or a sore throat? :-)
Lisp was a very early, successful language, because it was close to a mathematical notation and easy to implement on primitive computers. I think the uathor expects Lisp to remain a vital evolutionary branch because of its mathemtical roots.
Actually, I kind of doubt it was due to its mathematical properties. But I do think lisp will probably be an evolutionary ancestor of languages in the future, though perhaps not *all* languages in the future. There are three languages in particular I think have some important features that especially enhance programming (though it could be due to how I think, YMMV). These are, lisp, APL and Forth. All three languages have a specific ability, to some extent, to be mutated to suit the problem. This capability is usually referred to as "extensibility" though I've never thought that term was very descriptive. Object-oriented languages have some ability to do this as well, with features like operator overloading, etc., but it doesn't seem to be quite as simple and concise as it is in the three languages I mentioned. The fact that the three are also usually incrementally interactive during development could be a factor as well-- the fact that you program by gradually increasing the power of the language by adding new "primitives" is a very powerful feature that I expect will survive and be improved upon in some form in languages of the future. In some ways it could be said the languages "evolve" into the solution. And the ability to later redefine a primitive and thereby completely transform the solution in more and more intelligent ways coupled with inherent heuristic capabilities are some of the things I would expect to see as well. The extent that the "programmer" directs an evolutionary step towards the solution and the computer is able to do it on its own remains to be seen, but I do expect that the computer will do far more than it does now in that regard.
Most certainly new programming methods should be able to adapt to the "user" (programmer), but also to the problem, and in fact that is why I think the three languages I mentioned are particularly unique, as they are already pretty capable of that. Users/programmers that will also adapt to the computer will be more effective than those who insist on remaining "pure." I would expect that society as a whole will adjust itself more to conform to new technologies (as it has in the past), certainly including the computer. It has adjusted itself (for the most part) to technologies such as the automobile, airplane, chainsaw, etc.. Did we design cars to act exactly like horses because humans didn't like learning a new way to travel (perhaps some didn't)? At one extreme you have the human completely conforming to the computer's operation-- programming in binary for example, and at the other extreme, the computer completely conforming to the human's operation (no doubt producing a lot of confused, misinterpreted, ambiguous and erroneous results) and somewhere in between there is a point of optimal synergy. The dreamy utopians seem to have a hard time understanding this.
People know what they want out of their machines, for the most part.
Dang, what dream-world do you live in? I work with people all the time that have NO idea what their computer is capable of doing for them, and feel the need to conform to what they think it expects because they have no idea what *they* can expect. In the future this may change somewhat, but even besides that, most of them don't really know what solution they need because they really don't understand the problem to be solved (just that there is a problem to be solved) and must *discover* more information about the problem.
Let's explore the implications of this a bit. Suppose we have a scenario:
1. Human states problem: "I need an accounting system for new company XYZ that is starting up. It is a widget manufacturing business that should be able to grow to the size of, say-- IBM with sales, parts supply and support offices all over the world."
2. Computer says "OK, done."
3. Human either spends months discovering what the computer's assumptions are about how the company is to operate and then either adjusting the companies operations to conform (thus making it just like every other equivalent business in the world), OR correcting the programs misconceptions so that it will conform with the companies operation (thus preserving some value added that the companies founders have with regards to new ideas of how to run the business) OR some combination thereof.
Frankly, I don't see any version of step #3 to be particularly effective. Of course, how effective it is may be less of the point than how easy it is to use, but it doesn't strike me as being particulary easy to use either. And this is just a scenario including a relatively common and (presumably) understood problem. What happens when the nature of the problem is not such that there are preconcieved assumptions regarding it that can be leveraged as in this case? I can tell you what happens, the human spends all kinds of time discovering more about the nature of the problem and possible solutions and their implications and side-effects, pretty much like things are today. Sure, new tools will make the process easier and may include spoken input, but all this talk about "human language" interface is mostly utopian ignorance-- you'll have to fix the problems in human language first, in which case you're still "making" the human conform to the computer. Humans can't even talk to *each other* with much understanding, when the misunderstanding a computer is capable of is in the mix it is not going to be a very useful way to solve problems.
Some people may prefer an English interface, but those will be the novice users for the most part. I can type faster than I can speak, and my typing isn't affected by the level of background noise and interruptions I may receive. As much as some tried to invent alternatives for the keyboard (such as the mouse), virtually none have eliminated the need. I have 10 fingers, not just *one* and know how to use them, and alternative inputs must compete with that. The novice who can't type may prefer using a mouse or some other alternative to do everything. "Ease of use" is desirable for novices, and will always be there, but it's not equivalent to "most efficient interface" between a computer and a knowledgable and/or trained user. And the thing about novices is, they don't all stay novices forever-- though some apparently, may prefer to.
I'd rather say something brief to the computer and have it request disambiguation, even semantic disambiguation, as needed than work for hours on an unambiguous specification of what was obvious to begin with.
I think that would be extremely annoying to most people. "Obvious to begin with?" Most people don't even know what they want. Remember all those fairy tales about the Genie's three wishes where the Genie takes the requests a little *too* literally and what the person gets is not at all something they wanted? Problems in the future are going to get *more* complex and even harder to accurately specify, and consequently require more and more interrogation of the human in order to clarify the issues, context and acceptable solutions. The solution of a complex problem produce by a computer from a short human description would have to apply a HUGE number of preconcieved assumptions which would tend to produce a solution with all kinds of undiscovered side-effects and errors from those assumptions which will often be incorrect.
In fact, that is the problem I have with most of todays "new" solutions is they have built in assumptions that only apply when your problem is cookie-cutter, and in my experience, most problems aren't. And "Reusability" is grossly overrated-- tending to produce least-common-denominator mediocrity, due again to erroneous assumption.
What you're suggesting might apply to users of the future who are trying to configure an existing solution to a common and mundane problem, but not those who are trying to create solutions for new problems. The most difficult part of finding solutions for the most interesting problems is recognizing and stating the problem accurately, NOT implementing the solution. The current fad of X-treme programming recognizes this in its method of development by first creating the test suites at which point the implementation is straightforward-- it's the test suites that are hard.
If after generations and generations of computers, we are still teaching people to talk in computer terms and not yet teaching computers how to talk in people terms, we'll have gone the wrong direction
Bah! Humans do not talk in concise enough terms for that. What you are suggesting would mean that a typical programming exercise would consist of a human saying something to a computer and a computer spending a half-hour giving the human the third degree in an effort to resolve the ambiguities in the human's statement. Don't forget the GIGO principle.
Part of the issue with all this mess is we don't actually _hear_ from the creators, only the publishers, bankrollers etc... When will the artists ever stand up and take a side in all this? They certainly need to do so before any meaningful DRM is in place or it will be too late for an alternative.
You don't hear from the creators because they are under contract to the "publishers, bankrollers, etc."
That's because bulletproof DRM is impossible.
Bulletproof DRM is not impossible. However, bulletproof DRM can't EVER expose the user to the content in a readable/listenable/viewable form...
On that note, why do you listen to the music out there that costs money anyhow? I go with DJ's that allow their music to be distr. on the internet for free.
Who says I do? I think most of the music out there that costs money is crap. The exceptions are usually produced by some obscure outfit that can't really afford to promote a release or get it on the radio, where I don't mind paying for a CD.
But if you actually like the big promotion/distribution product, I see no reason for anyone to feel guilty for stealing back from those who have stolen from them, especially when they are otherwise powerless.
It's not democracy when big corporations bypass it by spending $$$ on lobbyists to browbeat the government representatives into passing the media controls laws as they want them. It's not democracy when there are laws against payola but are completely ignored because there's payola to the government as well. It's not democracy when the FCC gives away the bandwidth store for free to the big corporate contributors, and allows massive mergers to completely stifle independent innovation.
If this sort of thing is democracy, it deserves to be dumped into Boston harbor.
It almost begins to look a little more like democracy when the people begin to rise up and resist such tyranny and ignore your pathetic and hypocritical attempts to paint them all as criminals while ignoring all the criminal and immoral actions of the media corporations.
You are a thief. Plain and simple. If you worked for me and I found that you were using your work internet connection to pirate, I would terminate you for your illegal activities without hesitation.
Who's the bigger thief, the music distributors or the downloaders?
But you know that almost all of what you download was recorded, produced, distributed, and marketed by the very recording companies you claim to despise, and would never have been committed to disc were there not the possibility of profiting from exclusive distribution rights to audio recordings.
This neatly ignores the fact that the distribution companies found that their profit margins increase when they narrow the pipeline. This is because they can make far more profit by overhyping a few "super" groups and serving them up in a limited market.
Every time you download a popular song illegally, you are decreasing the probability that commerical-quality music will be made in the future, for any price.
Since that's my goal, I guess I'd better get busy... :-).
Actually, you are then INCREASING the possibility that music will survive on its merits alone, and not on the investments of the promoters trying to whip up a feeding frenzy. Just think-- payola might finally dry up and the real diversity of talent finally allowed access to the airwaves.
Anybody who cares about the system of intellectual property which has made the American entertainment and information technology industries so dynamic, and enjoys their fine products, from Windows XP to the "Lord of the Rings" movies to your new cell phone with built-in games and internet access, should understand the necessity of crushing Kazaa once and for all.
Now I know you're just a joker. Oh puh-leeze. Save me from these "dynamic" and "fine products" being constantly jammed down my throat to the exclusion of the really good quality products out there and potentially really good products out there, locked out of the market because the promoters and distributors prefer to narrowly concentrate their investments and only pretend that the market for music survives on the merits of the musicians rather than the merits of the controls on the market.
Spare me the bleeding heart for the industry bloodsuckers. Download them out of existence, and soon, with good riddance.
Are you kidding? It's obvious, they need the press. Don't forget that to marketing, ANY press is good press-- as long as they spell their name right. They've been having some trouble getting any notice of late, apparently...
I liken these arguments over methodology to what's wrong in education in the US-- people keep looking for a "magic" methodology that will fix all the problems, attempt to impose it on all the teachers which pisses off the good ones till they all find better jobs, then in time when it's seen things aren't any better, they distance themselves from the "old" magic methodology, blame it for the failures and start the process over again with a "new" magic methodology. Phonics vs Whole Language is an obvious example.
The thing is, ANY of these methodologies can produce effective results, but it's dependent on the type of project and the individuals and personalities involved. Some teachers will do better using one methodology, some with another, depending on their personalies and individual abilities. When you attempt to require everyone to conform to a single methodology, you have removed from them all responsibility for the success or failure of the methodology, and sometimes by association, of the project as well.
On one extreme, there's the software developer as knowledgeable professional, competent to apply such knowledge and experience as necessary to the tasks at hand, and at the other extreme, there's the software developer as a coder who, without an explicit structure which to conform to cannot effectively produce. Neither of these extremes exist in reality, most developers are somewhere in between. In any event, the utopian ideal of a universal methodology is of interest because there is a ready market, and the fact that no real solution exists produces just the vacuum necessary to suck imaginary solutions out of thin air.
Excuse my ignorance, but is Linux graphics still hamstrung by the insistence of its developers on jamming every bit of bitmapped data through X's tcp/ip bottleneck? I'm not expecting them to switch to SVGALIB for performance, but it's certainly been my experience that X is great when you want networked non-bitmapped graphics, but when it comes to lightning speed manipulation of display pixels, it wasn't at all designed for that. I tried to install an OpenGL library and it ran at least 10 times slower than the same machine did when booted as a Windows 98 machine. I consequently had to develop my graphic application where it had usable graphics APIs and a lame OS, rather than where it had a real OS but graphics APIs that stood in the way of useful framebuffer performance.
Has or is there any progress being made in that
regard?