He might indeed, though I don't see how compulsory licensing is a complete solution.
Unless ISP's will open every private packet (including encrypted packets), the ISP cannot know what the usage level is. People will have motivation to simply hide their usage unless the usage fee is unconditionally paid up front (as a part of joining the ISP). Once the fee is paid, there is no motivation to hide usage since no more money expenditure is caused by usage, and in fact there is motivation to expose usage since that will direct the already-paid money to people you like.
I don't claim to be an authority on this, but my understanding is that BMI/ASCAP licenses work this way, and I assume this is what is being used as a model. An organization pays based on its size (e.g., a square dance caller pays less than a radio station), but not necessarily based on its usage. Fees might be collected up front just for being who you are and all detailed tracking is only for apportionment of where the up-front fee goes.
I'm not even sure I listen to CD's as often as once a year. And even then, the only music I do listen to is on CD's that I actually bought from a store paying real money already. Am I going to have to pay this compulsory tax on my machine(s)?:(
What about other vices that some people have and others don't? Like Internet porn... Hmmm. Maybe a similar payment scheme for that industry would simplify things as well. A simple tax on everyone who uses an ISP since many people use such materials. Then the money could just be divvied up among those whose pictures were being used and deposited into a public kitty (hey, I didn't make up the term) for safekeeping. Then -- voila' -- justice and administrative simplicity in one tight little package.
I hold the best computer languages for actually programming the computer will still be analytic, logical languages.
I guess this is the fundamental point on which we just have to agree to disagree. I think that analysis and logic are critical operations, but I hope to find that the computer languages of the future will cease to be pedantic about the specific mode of expression, perhaps building in a sense of redundancy of expression so that no matter what language you express the idea in, it ends up with effectively the same internal representation.
One of the biggest differences right now about how computers do things and how people do things is that computers do not "degrade gracefully" when you go outside the ordinary way they expect to receive things. They tend to "fail catastrophically" on the least little deviation from the expected. In a hundred years, I hope they learn to be more laid back about what doesn't really matter (the manner of expression) and to focus more on what really does matter (the goal of the expression).
It might be that they will fail at the goal for the first few revs. But if we don't at least deploy them with the intent of trying, we won't get there.
Sometimes the path toward the future is a crooked one. For a further illustration of this phenomenon, search for (and read) 'A Personal Footnote' at the end of my 2001 paper on error handling in Lisp.
What other languages can use one word to make an entire sentence ?
I believed most of the other parts of the prior message, but not this sentence. English is pretty maleable. For example, the oft-cited ability to verb almost any noun.;)
A counterexample to the claim about one word for a whole sentence, though, is this: In Spanish, "Como como como." ("I eat how I eat.") is a complete and grammatically correct sentence. I'm pretty sure other languages have these, too.
... Written languages are too vague: *I* heard her say you were an idiot i *HEARD* her say you were an idiot ... different meanings depending on the inflection....
You seem to have adequately case-marked the tonality in your text, therefore defeating your own point.
Incidentally, all modalities trade off certain expressiveness for others. Although speech adds tonality, it loses the difference between homonyms.
The book, once red, had turned blue. The book, once read, had turned blue.
I think 100 years is enough time to find out that English is a poor language... even for communication among humans.
In the last several hundred, that hasn't turned out to be shown. Why different now, as the entire world of natural-language speakers seems to be adopting English as the de facto interchange language for international communication?
Is my context "father of family in house", "animal on planet earth", "logician working on problem"... why it depends on the context of the problem!
We haven't mentioned ubiquity of computing in this discussion, but if you assume that in 100 years, computing will be built into lots of things (see movies like The Sixth Day, where the refrigerator talks). But it's the continuity of contexts (sometimes overlapping) and the understanding of roles (we expect more of our teachers than our refrigerators, for example) that allows the disambiguation. Will there be sometimes questions to ask? Sure. Will programs sometimes require a release 2.0 (or 202.0) to correct a flaw in its resolution? Sure.
But we're not talking about whether or not the handling of every situation is perfect--or I'm not meaning to talk about that. Compilers today are imperfect but a discussion of what we program in today wouldn't deny their existence because they aren't perfect. In a hundred years, we might have different programming languages that are buggy in different ways, but it doesn't mean we won't still have them.
You seem to be saying "won't happen because it can't be done perfectly". I am trying to say "will happen, even if not perfectly. an iterative process."
Incidentally, on Star Trek, if Kirk said "computer, blow up the ship", it wouldn't do it. It requires him to say (paraphrasing) "Computer, as Captain of the ship I exercise my specific authority and here is the code..." Further, the holodecks of later shows don't respond to mere requests to do a safety override or emergency shutdown; you have to identify the invocation of a command privilege. Doctors and engineers sometimes surprise us by exercising privileges we ordinarily did not assume they had, because those roles are not the default and must be named. So there are again modes of disambiguation of role and modes of establishment of credential actively in play when there a person is serving in more than one role.
Yes, but we're supposed to be talking about what we will program in after 100 years, not for the next 100 years getting to that point. I'm saying that 100 years is enough time to assure that things like English language understanding at some level are subroutine-callable. Surely not full AI. But enough to be able to take a simple command in English rather than in a formal language and to construct a model of what is being asked for, to be able to query about ambiguities, and to develop a basic plan (or to be able to identify it as beyond the skill of the processor).
One thing I'd like not to see in 100 years is people comparing programming languages by writing "Hello World". The basic capability to start up and print a message should be in there already. Likewise with factorial, fibonacci, and plenty of other "common knowledge" that any person is expected to know before beginning the activity of programming. Any self-respecting computer of 100 years from now had better see a request for such a program and say "You're joking, right?"... and if it has the good sense to complain that the joke is getting old and tired, so much the better.
There's a school of thought that says you should use neural nets for everything. If you do, then you'd have to teach language just as it's taught to a child. That would still be programming. Kids are programmed. It just takes 18 years and is very labor intensive.
But I thought the original discussion wasn't "how will we program tools for the next 100 years" but rather "what will programming be like in 100 years".
In fact, as a kind of intermediate approach to the question, I have often talked about a division between "languages that express a goal" and "languages that tell you how to accomplish a goal". The former group contains languages that can express algorithms, but that do not busy themselves with all manner of details. The languages of the AI community (Lisp, Smalltalk, Prolog, ML, Haskell, etc.) tend to be like this. Their focus is on abstract behaviors, not on machine-level layout details. A Lisp programmer may have no idea how the memory of his program is laid out. The latter group contains only details--you can never really climb out of the details--you are always right at the machine. Assembly code, C, and Java are such languages. You can't program assembly, C, and Java without continually confronting how memory is laid out.
So, even if you don't believe in the neural nets appraoch, I'd say Lisp and friends are the first step in an evolution away from the machine and on to grander ideas... just as the first step in the evolution of a (good/successful) manager's career is learning to let go of "micromanaging" and to start to think Big Picture, eschewing obsessions with micro-optimizations that are below the abstraction horizon they are assigned to care about.
The problem is that English is *not* very precise... When lawyers attempt to make English be more precise, look at the messes they make... I don't know about you, but I don't want that.
But it's what we've got. Human language is, alas, imprecise. But we have more than 50 years of experience with that and we know nothing better is on the horizon. I think you'll be lucky if between now and a hundred years from now, you can teach 10% of the world's population the meaning of the world algorithm, much less the use of an algorithm.
But take heart -- while the computer has been called a relentless judge of incompleteness, the fact is that some of that incompleteness is just due to their bad schooling. Lack of common sense. Lack of context. If we can add that stuff in, maybe the kinds of problems computers give us won't sound like the whinings of a small child, ill-informed about the things in the world that we collectively agree should be 'obvious'. That won't fix everything, but it will fix some things.
For example, most non-computer people are able to take showers in finite time even with "Lather, Rinse, Repeat" written on shampoo bottles. They don't loop infinitely. Maybe in a hundred years, computers won't either because someone will have filled them in on the joke.
And legalese is not inherently required to be expressed as badly as it commonly is, that's just a fashion. Like doctors having bad handwriting. Social pressure would fix that if people were willing to tell their lawyers to go back and rewrite a text in prettier form. (Some probably are too cheap to pay by the hour to have that happen. Then again, if they did, they could perhaps read the result. The lawyer is probably just as happy you can't, just like a high paid computer consultant is often just as happy his clients can't understand the script he's written them, so they'll have to call him for upgrades. Again, not a technical problem, but a social one.)
If there is contextual information enough that a skilled human (like Scotty or Geordi) would know what was being referred to, so had better computers. Furthermore, I think "within the parameters of what we can actually do" would again go without saying.
It is this kind of "common sense" that is the subject of heavy investigation by Doug Lenat's Cyc group. OpenCyc is available now. I don't doubt that there will be some evolution in these ideas over the next 100 years.
As to Asimov's laws, they only become necessary if one has full AI, so it's reasonable to suppose that the problem originally posed ('non need for humans') presupposed AI. I answered in that context. I wasn't advocating AI or not, just working with the given context. One can't both presuppose something and tell me that such a presupposition is unreasonable. Not and have a civil conversation. This will be my last reply to you on this subthread.
Talking to computers is an interface issue. I'm sure you could point me to a Star Trek episode where someone does something that today could only be done with a programming language, but I suspect those things are the exception.
In Wolf In the Fold in Star Trek TOS, the computer is instructed to "compute to the last digit, the value of pi". It is not told "Computer, new program. Name it use_up_all_memory. It will have no arguments. Let papprox equal three. Loop doing... End program. Call use_up_all_memory, aggressively using as much computer memory as you need.'
In some languages (Basic, APL, Teco, HyperTalk, JavaScript, Lisp, Prolog), the line between interaction and program are blurred.
Also, as we've seen from the web, and the frequency of "This page under construction." pages, it's often useful to first start your program and then write it. Again, a blurring of the interaction between interaction and program. With a sufficiently forgiving debugger, the line between "user interface" and "debugger" is likewise blurred. Most languages haven't spent a lot of effort on enhancing the end-user-friendlieness of debuggers, but Lisp and Dylan have.
A Wolf In The Fold also makes use of interactive data sifting, actually. As does Conscience of the King.
I also recall episodes of Next Generation where Worf has saved holodeck programs or modified their parameters. ("Increase difficulty level.", etc.) He doesn't declare datatypes for things that have obvious types. He doesn't fuss about what directory to save files in. He uses simple file names that are presumably part of a working directory. The focus is on simplicity of naming and interface.
The programming language MOO (and integrated part of the text-based virtual reality also called MOO) offers this kind of instance-based, interactive construction that I might expect to be part of the smart environment of the future. Objects are named relative to their physical (or virtual) proximity. Rather than "initialization methods", one just makes an object and initializes it by hand. But don't be confused into thinking there is no programming language underlying that. Rather, the acts of programming and of using a program are blurred, as they are in database tools and languages.
If you ever listen to the types of commands they give to their computers in star trek, they are subjective and ambiguous.
Modern programming environments intervene earlier and earlier to tell us about syntax errors. Why not semantic errors as a logical extension of this? 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've been watching the progression of Microsoft Word from Spell Check to Grammar Check. I'm assuming future releases will continue the progression with Fact Check and, inevitably, Politics Check. (When it gets to that you're gonna wish it would allow you some ambiguity.;)
Any computer capable of understanding such commands would have no need for the crew (as it would quickly realize).
I think Asimov addressed this in his 3 (or sometimes 4) laws of robotics. Whether computers have no need of us will depend on how well we've programmed them (regardless of which side you're cheering for.)
I don't think it's the first order functional nature of Lisp that has allowed it to survive, but rather the "late binding" nature of it.
Static, strongly-typed languages, make the assumption that everything that needs to be known about the world is knowable at compile time. Such programs need to be recompiled (at least) and rewritten (often) because the world changes and either the source program itself or its compiled form needs to accomodate that change.
Lisp, because it delays many decisions until runtime, and because its runtime tagging accomodates datatypes that are not among the set that was declared at compile time, naturally accomodates changes in the environment around it, and naturally survives well during transitions between old and new ways to do things.
Static languages often breed static ways of thinking, and often need new static specifications at regular intervals to accomodate the mismatch with how the world really is. Dynamic languages breed dynamic thinking, which (I claim) is more robust over time.
I def. think that a new languange based on quantum computing will be at the forefront.
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.
It doesn't matter if quantum technology is used or not, for the same reason as it doesn't matter whether a brain is a parallel or single threaded machine, whether it's made of carbon-based or silicon-based technology, etc. What matters is that it can talk to you, can understand you, and can improve life.
If you want to know what computer languages should and hopefully will look like in the future, you have only to watch Star Trek. I'm not kidding. The desire to pack computer use into a short TV program has led the authors of that show and shows like it to pare out all but the absolute essentials of describing what you want the computer to do. That is what computer programming should be like, since that's what people programming is like. People don't put up with excess verbiage, and neither should computers.
This kind of reminds me of how the size of candy bars shrinks for a while, then finally a "new bigger size" is introduced at a higher price, then the new bigger one starts to shrink.
If a problem as important as candy bars hasn't been fixed after years of this "market manipulation", how could we possibly expect a problem as unimportant as bandwidth to ever be fixed?
Maybe I'm selling the world short, but I sorta think that consumer education (usually by competitors with a better product) is perhaps the best we can hope for.
The feedback is "the market share of the person doing something better went up and the market share of the person doing something less good went down".
The market is a feedback loop.
The feedback should not be "you didn't let me into this site" but rather "you aren't telling me what sites you're letting me into".
Suppose I alleged that this _was_ a compilation consisting merely of a list of websites to be blocked by filtering software applications? How would I know whether or not I was allowed to violate the prohibition if the person selling it to me was not forthcoming about that fact? How can the nature of the software be used as a reason to block you from investigating the nature of the software?
I think it was a mistake for the court to make a claim based on rights. A better ruling would have said that there was no damage to be had--that the decision to purchase this software was voluntary, and that people not happy with this software should decide not to purchase it. If people would voluntarily decide not to purchase software that did not expose its mechanisms, then such software would not exist. If people want such software, and therefore pay for it, then the court might reasonably presume it's something they shouldn't intervene in.
Jesse Ventura said something I really liked a couple years back which went something like 'government has no business meddling in things that people are able to solve for themselves'. Where we need intervention is in problems that we have no mechanisms as individuals to address. The free market is capable of sorting out this issue.
Teaching often oversimplifies issues. A lot of the job of ethics is to remind people that the real world is more complicated than classroom situations. We divide up classes into arithmetic, history, science, etc. as if these things can really be separated one from another. Ethics reminds us that all of these things can come together in messy ways. If the student doesn't go out asking more questions than when they came in, the attempt to teach ethics has probably failed.
But even then, it doesn't suffice to ask questions in the absence of domain knowledge. The questions you ask will frame the discussion, and if you don't understand how to properly frame the debate, you're going to do a bad job. Recently, for example, news agencies (especially in the US) have substituted one person pro and one person con on nearly any issue in order to appear to achieve balance. But not all issues are best framed that way... Some are not binary, some are not things with equal weight on both sides, and so on...
Citing "Freedom of speach", the first ammendment, etc, there still seems to be an ignorant crowd that thinks that we shouldn't have any legal means to curb spam. [...]
I think the reason for this is that the US Founding Fathers blew it when they talked about freedom of speech in the First Amendment. Speech is communication between speaker and listener, and in case of a dispute, this appears to say that the speaker gets the tie broken in their favor.
I think of it a different way--as the freedom to hear [pointer to my personal web site where I expand on this in more detail].
That gives the final say on appropriateness back to the listener, where it belongs.
When you view it in that mode, it's easy to see that spam has nothing to do with free speech.
Free speech is just a way of guaranteeing that we in a democracy have access to a free flow of ideas. It isn't supposed to be a way of forcing us to endure a free flow of anything. That's not freedom, it's slavery.
Would paying really BE more expensive?
on
IETF to Look at Spam
·
· Score: 2, Insightful
I've been using email on the net for 20+ years and have been as grateful as anyone that it hasn't cost.
However, of late, with other people making such gross abuse of the world's mail systems that I feel I am (and all of us are) paying anyway, it might be worth revisiting this question. I'm not 100% convinced that paying per piece of mail sent would be more expensive than, effectively, paying (in both time and in dollars spent on ISP infrastructure) per piece of mail received. I send a lot less mail than I receive, and I bet most people do.
If you build a building with floors that are too weak, people do not say that the cause of it falling in is "people walked on the floor".
If you send someone out into a boxing arena with no training and the guy gets knocked out, is the cause "boxing gloves" or "lack of training"?
Why would you say that if you build a programmer body with equally little training (whether or not we knew such training was required early enough to fix it), and then the fact was that ordinary use of arms for typing in the way we want programmers to type is the "cause" of Carpal Tunnel. The cause seems to me to be plainly "inadequate prep" not "failure to define programming as a task which does not require typing".
Your mileage may, of course, vary. But my point was not to be arrogant, it was to say that I think programming requires certain preparation. Preparation maybe neither you nor I knew, and that I got lucky enough to get. But still, I think teaching people who didn't get the training to type carefully is a bad plan. It looks to me like it'd be better for sports jocks to learn to program, and then program to their heart's -- or fingers' -- content.
Slashdot Mirror
He might indeed, though I don't see how compulsory licensing is a complete solution.
Unless ISP's will open every private packet (including encrypted packets), the ISP cannot know what the usage level is. People will have motivation to simply hide their usage unless the usage fee is unconditionally paid up front (as a part of joining the ISP). Once the fee is paid, there is no motivation to hide usage since no more money expenditure is caused by usage, and in fact there is motivation to expose usage since that will direct the already-paid money to people you like.
I don't claim to be an authority on this, but my understanding is that BMI/ASCAP licenses work this way, and I assume this is what is being used as a model. An organization pays based on its size (e.g., a square dance caller pays less than a radio station), but not necessarily based on its usage. Fees might be collected up front just for being who you are and all detailed tracking is only for apportionment of where the up-front fee goes.
I'm not even sure I listen to CD's as often as once a year. And even then, the only music I do listen to is on CD's that I actually bought from a store paying real money already. Am I going to have to pay this compulsory tax on my machine(s)? :(
What about other vices that some people have and others don't? Like Internet porn... Hmmm. Maybe a similar payment scheme for that industry would simplify things as well. A simple tax on everyone who uses an ISP since many people use such materials. Then the money could just be divvied up among those whose pictures were being used and deposited into a public kitty (hey, I didn't make up the term) for safekeeping. Then -- voila' -- justice and administrative simplicity in one tight little package.
I hold the best computer languages for actually programming the computer will still be analytic, logical languages.
I guess this is the fundamental point on which we just have to agree to disagree. I think that analysis and logic are critical operations, but I hope to find that the computer languages of the future will cease to be pedantic about the specific mode of expression, perhaps building in a sense of redundancy of expression so that no matter what language you express the idea in, it ends up with effectively the same internal representation.
One of the biggest differences right now about how computers do things and how people do things is that computers do not "degrade gracefully" when you go outside the ordinary way they expect to receive things. They tend to "fail catastrophically" on the least little deviation from the expected. In a hundred years, I hope they learn to be more laid back about what doesn't really matter (the manner of expression) and to focus more on what really does matter (the goal of the expression).
It might be that they will fail at the goal for the first few revs. But if we don't at least deploy them with the intent of trying, we won't get there.
Sometimes the path toward the future is a crooked one. For a further illustration of this phenomenon, search for (and read) 'A Personal Footnote' at the end of my 2001 paper on error handling in Lisp.
What other languages can use one word to make an entire sentence ?
;)
I believed most of the other parts of the prior message, but not this sentence. English is pretty maleable. For example, the oft-cited ability to verb almost any noun.
A counterexample to the claim about one word for a whole sentence, though, is this: In Spanish, "Como como como." ("I eat how I eat.") is a complete and grammatically correct sentence. I'm pretty sure other languages have these, too.
A bit like valgol, only worse.
... Written languages are too vague: ...
*I* heard her say you were an idiot
i *HEARD* her say you were an idiot
... different meanings depending on the inflection.
You seem to have adequately case-marked the tonality in your text, therefore defeating your own point.
Incidentally, all modalities trade off certain expressiveness for others. Although speech adds tonality, it loses the difference between homonyms.
The book, once red, had turned blue.
The book, once read, had turned blue.
I think 100 years is enough time to find out that English is a poor language... even for communication among humans.
In the last several hundred, that hasn't turned out to be shown. Why different now, as the entire world of natural-language speakers seems to be adopting English as the de facto interchange language for international communication?
Is my context "father of family in house", "animal on planet earth", "logician working on problem"... why it depends on the context of the problem!
We haven't mentioned ubiquity of computing in this discussion, but if you assume that in 100 years, computing will be built into lots of things (see movies like The Sixth Day, where the refrigerator talks). But it's the continuity of contexts (sometimes overlapping) and the understanding of roles (we expect more of our teachers than our refrigerators, for example) that allows the disambiguation. Will there be sometimes questions to ask? Sure. Will programs sometimes require a release 2.0 (or 202.0) to correct a flaw in its resolution? Sure.
But we're not talking about whether or not the handling of every situation is perfect--or I'm not meaning to talk about that. Compilers today are imperfect but a discussion of what we program in today wouldn't deny their existence because they aren't perfect. In a hundred years, we might have different programming languages that are buggy in different ways, but it doesn't mean we won't still have them.
You seem to be saying "won't happen because it can't be done perfectly". I am trying to say "will happen, even if not perfectly. an iterative process."
Incidentally, on Star Trek, if Kirk said "computer, blow up the ship", it wouldn't do it. It requires him to say (paraphrasing) "Computer, as Captain of the ship I exercise my specific authority and here is the code..." Further, the holodecks of later shows don't respond to mere requests to do a safety override or emergency shutdown; you have to identify the invocation of a command privilege. Doctors and engineers sometimes surprise us by exercising privileges we ordinarily did not assume they had, because those roles are not the default and must be named. So there are again modes of disambiguation of role and modes of establishment of credential actively in play when there a person is serving in more than one role.
Yes, but we're supposed to be talking about what we will program in after 100 years, not for the next 100 years getting to that point. I'm saying that 100 years is enough time to assure that things like English language understanding at some level are subroutine-callable. Surely not full AI. But enough to be able to take a simple command in English rather than in a formal language and to construct a model of what is being asked for, to be able to query about ambiguities, and to develop a basic plan (or to be able to identify it as beyond the skill of the processor).
... and if it has the good sense to complain that the joke is getting old and tired, so much the better.
One thing I'd like not to see in 100 years is people comparing programming languages by writing "Hello World". The basic capability to start up and print a message should be in there already. Likewise with factorial, fibonacci, and plenty of other "common knowledge" that any person is expected to know before beginning the activity of programming. Any self-respecting computer of 100 years from now had better see a request for such a program and say "You're joking, right?"
There's a school of thought that says you should use neural nets for everything. If you do, then you'd have to teach language just as it's taught to a child. That would still be programming. Kids are programmed. It just takes 18 years and is very labor intensive.
But I thought the original discussion wasn't "how will we program tools for the next 100 years" but rather "what will programming be like in 100 years".
In fact, as a kind of intermediate approach to the question, I have often talked about a division between "languages that express a goal" and "languages that tell you how to accomplish a goal". The former group contains languages that can express algorithms, but that do not busy themselves with all manner of details. The languages of the AI community (Lisp, Smalltalk, Prolog, ML, Haskell, etc.) tend to be like this. Their focus is on abstract behaviors, not on machine-level layout details. A Lisp programmer may have no idea how the memory of his program is laid out. The latter group contains only details--you can never really climb out of the details--you are always right at the machine. Assembly code, C, and Java are such languages. You can't program assembly, C, and Java without continually confronting how memory is laid out.
So, even if you don't believe in the neural nets appraoch, I'd say Lisp and friends are the first step in an evolution away from the machine and on to grander ideas... just as the first step in the evolution of a (good/successful) manager's career is learning to let go of "micromanaging" and to start to think Big Picture, eschewing obsessions with micro-optimizations that are below the abstraction horizon they are assigned to care about.
The problem is that English is *not* very precise... When lawyers attempt to make English be more precise, look at the messes they make ... I don't know about you, but I don't want that.
But it's what we've got. Human language is, alas, imprecise. But we have more than 50 years of experience with that and we know nothing better is on the horizon. I think you'll be lucky if between now and a hundred years from now, you can teach 10% of the world's population the meaning of the world algorithm, much less the use of an algorithm.
But take heart -- while the computer has been called a relentless judge of incompleteness, the fact is that some of that incompleteness is just due to their bad schooling. Lack of common sense. Lack of context. If we can add that stuff in, maybe the kinds of problems computers give us won't sound like the whinings of a small child, ill-informed about the things in the world that we collectively agree should be 'obvious'. That won't fix everything, but it will fix some things.
For example, most non-computer people are able to take showers in finite time even with "Lather, Rinse, Repeat" written on shampoo bottles. They don't loop infinitely. Maybe in a hundred years, computers won't either because someone will have filled them in on the joke.
And legalese is not inherently required to be expressed as badly as it commonly is, that's just a fashion. Like doctors having bad handwriting. Social pressure would fix that if people were willing to tell their lawyers to go back and rewrite a text in prettier form. (Some probably are too cheap to pay by the hour to have that happen. Then again, if they did, they could perhaps read the result. The lawyer is probably just as happy you can't, just like a high paid computer consultant is often just as happy his clients can't understand the script he's written them, so they'll have to call him for upgrades. Again, not a technical problem, but a social one.)
If there is contextual information enough that a skilled human (like Scotty or Geordi) would know what was being referred to, so had better computers. Furthermore, I think "within the parameters of what we can actually do" would again go without saying.
It is this kind of "common sense" that is the subject of heavy investigation by Doug Lenat's Cyc group. OpenCyc is available now. I don't doubt that there will be some evolution in these ideas over the next 100 years.
As to Asimov's laws, they only become necessary if one has full AI, so it's reasonable to suppose that the problem originally posed ('non need for humans') presupposed AI. I answered in that context. I wasn't advocating AI or not, just working with the given context. One can't both presuppose something and tell me that such a presupposition is unreasonable. Not and have a civil conversation. This will be my last reply to you on this subthread.
Talking to computers is an interface issue. I'm sure you could point me to a Star Trek episode where someone does something that today could only be done with a programming language, but I suspect those things are the exception.
In Wolf In the Fold in Star Trek TOS, the computer is instructed to "compute to the last digit, the value of pi". It is not told "Computer, new program. Name it use_up_all_memory. It will have no arguments. Let papprox equal three. Loop doing ... End program. Call use_up_all_memory, aggressively using as much computer memory as you need.'
In some languages (Basic, APL, Teco, HyperTalk, JavaScript, Lisp, Prolog), the line between interaction and program are blurred.
Also, as we've seen from the web, and the frequency of "This page under construction." pages, it's often useful to first start your program and then write it. Again, a blurring of the interaction between interaction and program. With a sufficiently forgiving debugger, the line between "user interface" and "debugger" is likewise blurred. Most languages haven't spent a lot of effort on enhancing the end-user-friendlieness of debuggers, but Lisp and Dylan have.
A Wolf In The Fold also makes use of interactive data sifting, actually. As does Conscience of the King.
I also recall episodes of Next Generation where Worf has saved holodeck programs or modified their parameters. ("Increase difficulty level.", etc.) He doesn't declare datatypes for things that have obvious types. He doesn't fuss about what directory to save files in. He uses simple file names that are presumably part of a working directory. The focus is on simplicity of naming and interface.The programming language MOO (and integrated part of the text-based virtual reality also called MOO) offers this kind of instance-based, interactive construction that I might expect to be part of the smart environment of the future. Objects are named relative to their physical (or virtual) proximity. Rather than "initialization methods", one just makes an object and initializes it by hand. But don't be confused into thinking there is no programming language underlying that. Rather, the acts of programming and of using a program are blurred, as they are in database tools and languages.
If you ever listen to the types of commands they give to their computers in star trek, they are subjective and ambiguous.
;)
Modern programming environments intervene earlier and earlier to tell us about syntax errors. Why not semantic errors as a logical extension of this? 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've been watching the progression of Microsoft Word from Spell Check to Grammar Check. I'm assuming future releases will continue the progression with Fact Check and, inevitably, Politics Check. (When it gets to that you're gonna wish it would allow you some ambiguity.
Any computer capable of understanding such commands would have no need for the crew (as it would quickly realize).
I think Asimov addressed this in his 3 (or sometimes 4) laws of robotics. Whether computers have no need of us will depend on how well we've programmed them (regardless of which side you're cheering for.)
I don't think it's the first order functional nature of Lisp that has allowed it to survive, but rather the "late binding" nature of it.
Static, strongly-typed languages, make the assumption that everything that needs to be known about the world is knowable at compile time. Such programs need to be recompiled (at least) and rewritten (often) because the world changes and either the source program itself or its compiled form needs to accomodate that change.
Lisp, because it delays many decisions until runtime, and because its runtime tagging accomodates datatypes that are not among the set that was declared at compile time, naturally accomodates changes in the environment around it, and naturally survives well during transitions between old and new ways to do things.
Static languages often breed static ways of thinking, and often need new static specifications at regular intervals to accomodate the mismatch with how the world really is. Dynamic languages breed dynamic thinking, which (I claim) is more robust over time.
I def. think that a new languange based on quantum computing will be at the forefront.
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.
It doesn't matter if quantum technology is used or not, for the same reason as it doesn't matter whether a brain is a parallel or single threaded machine, whether it's made of carbon-based or silicon-based technology, etc. What matters is that it can talk to you, can understand you, and can improve life.
If you want to know what computer languages should and hopefully will look like in the future, you have only to watch Star Trek. I'm not kidding. The desire to pack computer use into a short TV program has led the authors of that show and shows like it to pare out all but the absolute essentials of describing what you want the computer to do. That is what computer programming should be like, since that's what people programming is like. People don't put up with excess verbiage, and neither should computers.
This kind of reminds me of how the size of candy bars shrinks for a while, then finally a "new bigger size" is introduced at a higher price, then the new bigger one starts to shrink.
If a problem as important as candy bars hasn't been fixed after years of this "market manipulation", how could we possibly expect a problem as unimportant as bandwidth to ever be fixed?
Maybe I'm selling the world short, but I sorta think that consumer education (usually by competitors with a better product) is perhaps the best we can hope for.
The feedback is "the market share of the person doing something better went up and the market share of the person doing something less good went down".
The market is a feedback loop.
The feedback should not be "you didn't let me into this site" but rather "you aren't telling me what sites you're letting me into".
Suppose I alleged that this _was_ a compilation consisting merely of a list of websites to be blocked by filtering software applications? How would I know whether or not I was allowed to violate the prohibition if the person selling it to me was not forthcoming about that fact? How can the nature of the software be used as a reason to block you from investigating the nature of the software?
I think it was a mistake for the court to make a claim based on rights. A better ruling would have said that there was no damage to be had--that the decision to purchase this software was voluntary, and that people not happy with this software should decide not to purchase it. If people would voluntarily decide not to purchase software that did not expose its mechanisms, then such software would not exist. If people want such software, and therefore pay for it, then the court might reasonably presume it's something they shouldn't intervene in.
Jesse Ventura said something I really liked a couple years back which went something like 'government has no business meddling in things that people are able to solve for themselves'. Where we need intervention is in problems that we have no mechanisms as individuals to address. The free market is capable of sorting out this issue.
Teaching often oversimplifies issues. A lot of the job of ethics is to remind people that the real world is more complicated than classroom situations. We divide up classes into arithmetic, history, science, etc. as if these things can really be separated one from another. Ethics reminds us that all of these things can come together in messy ways. If the student doesn't go out asking more questions than when they came in, the attempt to teach ethics has probably failed.
But even then, it doesn't suffice to ask questions in the absence of domain knowledge. The questions you ask will frame the discussion, and if you don't understand how to properly frame the debate, you're going to do a bad job. Recently, for example, news agencies (especially in the US) have substituted one person pro and one person con on nearly any issue in order to appear to achieve balance. But not all issues are best framed that way... Some are not binary, some are not things with equal weight on both sides, and so on...
Citing "Freedom of speach", the first ammendment, etc, there still seems to be an ignorant crowd that thinks that we shouldn't have any legal means to curb spam. [...]
I think the reason for this is that the US Founding Fathers blew it when they talked about freedom of speech in the First Amendment. Speech is communication between speaker and listener, and in case of a dispute, this appears to say that the speaker gets the tie broken in their favor.
I think of it a different way--as the freedom to hear [pointer to my personal web site where I expand on this in more detail]. That gives the final say on appropriateness back to the listener, where it belongs. When you view it in that mode, it's easy to see that spam has nothing to do with free speech.
Free speech is just a way of guaranteeing that we in a democracy have access to a free flow of ideas. It isn't supposed to be a way of forcing us to endure a free flow of anything. That's not freedom, it's slavery.
I've been using email on the net for 20+ years and have been as grateful as anyone that it hasn't cost.
However , of late, with other people making such gross abuse of the world's mail systems that I feel I am (and all of us are) paying anyway, it might be worth revisiting this question. I'm not 100% convinced that paying per piece of mail sent would be more expensive than, effectively, paying (in both time and in dollars spent on ISP infrastructure) per piece of mail received. I send a lot less mail than I receive, and I bet most people do.
Hmm... I was more imagining
Scruffy, the Vampire Spayer's dog
Perhaps it's
Snuffy, the Vampire Killer Killer ?
If you build a building with floors that are too weak, people do not say that the cause of it falling in is "people walked on the floor".
If you send someone out into a boxing arena with no training and the guy gets knocked out, is the cause "boxing gloves" or "lack of training"?
Why would you say that if you build a programmer body with equally little training (whether or not we knew such training was required early enough to fix it), and then the fact was that ordinary use of arms for typing in the way we want programmers to type is the "cause" of Carpal Tunnel. The cause seems to me to be plainly "inadequate prep" not "failure to define programming as a task which does not require typing".
Your mileage may, of course, vary. But my point was not to be arrogant, it was to say that I think programming requires certain preparation. Preparation maybe neither you nor I knew, and that I got lucky enough to get. But still, I think teaching people who didn't get the training to type carefully is a bad plan. It looks to me like it'd be better for sports jocks to learn to program, and then program to their heart's -- or fingers' -- content.