> All he did was spew a cultlike agenda the way all good cult leaders do.
Let's review the facts here.
From the skeptic's dictionary: "Three ideas seem essential to the concept of a cult. One is thinking in terms of us/them with total alienation from them."
RMS has publicly admitted that ESR/OSS does some good things, thereby creating a gray area. Every member of FSF knows and regularly associates with people who are not members of the FSF.
"The second is the intense, though often subtle, indoctrination techniques used to recruit and hold members."
Text is not intense, no matter how big the font.
"And the third is the charismatic cult leader."
RMS is not very charismatic. ESR is more charismatic than RMS. And Linus is a nicer guy than either one, though I think ESR is more charismatic, and has a more dangerous agenda than RMS, though that's a different post.
What makes some people prefer GPL over OSS and proprietary is not that they like the guy that drafted it, but that it appeals to the morals and ideals that they already had even before learning about GNU.
Simply because my ideals are not a subset of your ideals, assuming perhaps prematurely that you have any at all, doesn't make me a 'lemming'. Maybe instead of calling me foolish without justification or qualification, you should raise a question on the specific way in which you think RMS is duping people.
> There's more to a board position in chess than the layout of the pieces.
I have to agree, but if I'm not mistaken a second time, the extra information you're referring to should be storable separately from the graph, and would be at most some constant amount. We can count to 50 with just one byte, for example. Only the graph risks becoming very large.
This approach wouldn't reduce the number of traversals needed, of course, so time would be just as much a factor.
The problem is that the potential exists for an ASP to circumvent the GPL. If they don't distribute, then they can take GPL code, modify it, close the source, and use it all day long without even infringing on the GPL. Unless I'm wrong, of course.
> The search tree for chess has more nodes than there are atoms of the universe.
Then maybe it's not appropriate to use a tree. Why not use a graph? Each node could represent one unique layout for the board - with a graph, there would be no need for this kind of redundancy.
So, uh, anyone know how many ways you can put chess pieces on a chess board?
If you had read the earlier posts, you would know that certain sections of the tree can be discarded. Specifically, those sections that lead to an unfavorable outcome. The set of winning games is smaller than the set of possible games.
The point isn't to construct the whole tree, but rather the solution as expressed by the tree. If part of the tree doesn't express a part of the solution, why do you need it?
Sure, but newspeak was distinctive not because it made communication more efficient, but because it made certain kinds of communication impossible. You couldn't say something like 'independant thinking' in newspeak, because it didn't exist. The very closest thing would be thoughtcrime. Newspeak didn't shorten or abbreviate words, it only killed them.
> Latin would require so many word-imports it would end up looking much like net-English anyway.
There was a story about how the Pope was getting into a helicopter, and they had to bless it or something, but obviously there wasn't a word for helicopters in Latin, so they made one up on the spot. Of course, the word was 'helicopterum.'
I think this is critical. There's a tendency to conceptually separate languages, assume that they can't crossbreed and merge. Yet they do - English merges Latin with German and comes out with something which is uniquely more appropriate than either one alone.
> would such a language be a niche language, or do you think it could come to rival even English for dominance?
It's a moot point, because a large subset of English will be a subset of the world language, and perhaps all of it. Same for other languages.
Eventually we will just realize that the world language already exists, as a superset of all existing languages, not only including the multitude of words and phrases for each semantic meaning (and the endless shades of meaning), but multiple syntaxes and grammars, which can carry their own meaning and flavor.
> They DON'T relate any more than Slackware v.X relates to Debian v.Y.
Slackware and Debian don't use the same package file format (.deb vs.tgz), whereas Red Hat and Mandrake both use rpms. What would be more accurate is to say that Red Hat and Mandrake are no more alike than Caldera and SuSE, which both also use rpm.
What does the future hold for the HURD if things keep going the way they're going? What if acceptance for the HURD explodes, like it did for Linux? In either case, will it at some point replace Linux, or move into a niche alongside it, say, running the big iron while Linux scales down into all the little embedded devices? Do you or other GNU folks actually use it for anything now? If so, what? Do you feel like Linux has created interest in the HURD, or just stolen its thunder?
Netscape opened up a failing project in order to save it. Microsoft has more coders than Netscape, but the internet has more coders than Microsoft. That sort of thing. I think grey wants to know what would happen if a popular, successful project was opened, like Windows, or one of Apple's expensive authoring tools.
> What the studios are afraid of in this scenario > is that their scriptwriters accidently (or maybe > not-so-accidently) use a plot identical to one > found in a fanfic story.
Then why does George Lucas only go after erotic fanfic? Is he planning on having Obi-Wan and Jar Jar getting a little close in Episode Two?
This may be a reason why slash and erotic fanfic in general is so popular - there's zero possibility of the work being stolen, because it would be inconceivable for the studios to actually make television and movies out of such things.
On the other hand, I imagine most fanfic writers wouldn't mind their writing being used by the studios, even if they didn't get credit, because at least they and their friends would know about it. But maybe I'm an optimist.
One of the things I hate most of all about the Windows interface is how frequently my irrational clicking and mouse jiggling leads to 'losing the taskbar' - resizing it to zero. Not only am I losing real-estate which I need for getting ahold of those icons, I've actually got a minor booby-trap built into the interface! What you describe might be nice for some, but all I want is a way to turn that option off, since I hardly ever manually resize my panel anyway (except to hide it).
> It's hard to put spacers in the panel.
That's true - the panel is either unnecessarially long, or exactly as long as the sum length of your applets, launchers, menus and drawers. It's not possible to put extra space in a corner panel (except for Panel/Right Click -> Global Properties -> Misc -> Applet Padding, which isn't the same thing), or to make the panel window only show a part of the panel's contents. Another thing that might be nice is floating panels, that is, panels which can be positioned arbitrarially anywhere on the desktop. I have two vertical corner panels, one in each upper corner of the desktop, and if I could, I'd probably put more up there in between them.
Well, there was the interview on IBM last week, but that wasn't a Slashdot Interview. The last Gnome developer to be interviewed on Slashdot was Havoc Pennington, I believe, and that was a long time ago - several months or more. So I'm not sure, but it sounds to me like you're trollin' for Jesus.
Why Miguel, well, because his company just started doing things, and we all want to know when the.debs for Helix Code's gear are going to be available. (What do you mean, I'm the only one?)
That's right. Unless Napster is grossly inefficient, there's no reason why it would take up any more than an irc client and an ftp server. Why don't they ban other obvious web hogs? Graphical web browsers? AllAdvantage-type software? Remote X sessions? MUDs and MMORPGs? Everquest isn't 'real work'.
A better reason for net admins to ban Napster is simply that they can't handle it all and need to cut something, anything, but at that point there should be enough tuition money to buy more gear.
In particular, executable files and similar things, like shell scripts and Word documents could be used to disseminate viruses. The 'instant mirror' ability sharing community software creates does present one solution to the slashdot effect, though.
Also, is there anything like ID3 tags for image files?
I agree that it is very unlikely. I think the most evil classmate I had in high school was the one who stole my code. Wasn't very bright, though... he left my name on the listing.
I'll agree that an autonomous mind requres a goal-finding mechanism, but I won't agree that an AI has to stick with the one that it started out with. The parent goal can be respecified each time the AI runs, and it can be much more specific than the basic human desires. I think that's the separation, really. Humans have a very coarse set of desires - full stomach, human contact, that kind of thing... whereas AI is important, in particular, for fine-grained, specialized application of intelligence.
BTW, Carl Jung was a raving nutter. Not that it has anything to do with your other comments.
The problem isn't specifically that we don't understand the human mind, but that we don't have enough of the right kinds of evidence to agree on how it works. For all we know, we may already understand intelligence well enough to design AI, but we just haven't applied the right set of ideas to the right hardware. There's no shortage of ideas floating around.
> Basically he points out that if we create a machine that is smarter than ourselves, it will do the same with respect to itself.
Assuming we give it autonomy. That's important to note - intelligence doesn't require such a thing as desire. Presumably we would be making these powerful thinking machines to determine how to more perfectly express our desires, so they'd be more like extensions of ourselves than like children or aliens.
Your argument seems to be, I understand it and can express it mathematically, therefore it isn't inteligence, and isn't what's going on in the brain, but this doesn't address the challenge at all.
You don't know what's going on in the brain, and you don't know what intelligence is. If intuition, or primary-process thinking, isn't understandable and expressable mathematically, then the goal of AI is literally impossible, and Turing-machine completeness is a crock.
> They can stare at their opponent to try and see if he's bluffing.
This is not a measure of intelligence, unless you think that a polygraph is intelligent.
> Priorities vs. Wants, etc., etc., etc. I have yet to see a machine that can make these types of decisions appropriately.
My operating system doesn't run a distributed.net client if other programs are taking up all the CPU. That's a decision based on a priority.
If what you want is a program that can make decisions that are human enough and complex enough for a human to fret about, well, there's a lot of work in that, and pretending that the incremental steps don't count just puts you that much farther from the goal. They do count.
> Take the example of something more fast-paced than Chess like Soccer.
Uhh... I think it's pretty clear that the problem here has nothing to do with intelligence. It's a question of motor coordination and perception. Reliance on intelligence may actually make the game harder.
Slashdot Mirror
> All he did was spew a cultlike agenda the way all good cult leaders do.
:)
Let's review the facts here.
From the skeptic's dictionary:
"Three ideas seem essential to the concept of a cult. One is thinking in terms of us/them with total alienation from them."
RMS has publicly admitted that ESR/OSS does some good things, thereby creating a gray area. Every member of FSF knows and regularly associates with people who are not members of the FSF.
"The second is the intense, though often subtle, indoctrination techniques used to recruit and hold members."
Text is not intense, no matter how big the font.
"And the third is the charismatic cult leader."
RMS is not very charismatic. ESR is more charismatic than RMS. And Linus is a nicer guy than either one, though I think ESR is more charismatic, and has a more dangerous agenda than RMS, though that's a different post.
What makes some people prefer GPL over OSS and proprietary is not that they like the guy that drafted it, but that it appeals to the morals and ideals that they already had even before learning about GNU.
Simply because my ideals are not a subset of your ideals, assuming perhaps prematurely that you have any at all, doesn't make me a 'lemming'. Maybe instead of calling me foolish without justification or qualification, you should raise a question on the specific way in which you think RMS is duping people.
Or maybe you should go troll somewhere else.
> There's more to a board position in chess than the layout of the pieces.
I have to agree, but if I'm not mistaken a second time, the extra information you're referring to should be storable separately from the graph, and would be at most some constant amount. We can count to 50 with just one byte, for example. Only the graph risks becoming very large.
This approach wouldn't reduce the number of traversals needed, of course, so time would be just as much a factor.
The problem is that the potential exists for an ASP to circumvent the GPL. If they don't distribute, then they can take GPL code, modify it, close the source, and use it all day long without even infringing on the GPL. Unless I'm wrong, of course.
> The search tree for chess has more nodes than there are atoms of the universe.
Then maybe it's not appropriate to use a tree. Why not use a graph? Each node could represent one unique layout for the board - with a graph, there would be no need for this kind of redundancy.
So, uh, anyone know how many ways you can put chess pieces on a chess board?
If you had read the earlier posts, you would know that certain sections of the tree can be discarded. Specifically, those sections that lead to an unfavorable outcome. The set of winning games is smaller than the set of possible games.
The point isn't to construct the whole tree, but rather the solution as expressed by the tree. If part of the tree doesn't express a part of the solution, why do you need it?
Sure, but newspeak was distinctive not because it made communication more efficient, but because it made certain kinds of communication impossible. You couldn't say something like 'independant thinking' in newspeak, because it didn't exist. The very closest thing would be thoughtcrime. Newspeak didn't shorten or abbreviate words, it only killed them.
> Latin would require so many word-imports it would end up looking much like net-English anyway.
There was a story about how the Pope was getting into a helicopter, and they had to bless it or something, but obviously there wasn't a word for helicopters in Latin, so they made one up on the spot. Of course, the word was 'helicopterum.'
> English is an amalgum of languages,
I think this is critical. There's a tendency to conceptually separate languages, assume that they can't crossbreed and merge. Yet they do - English merges Latin with German and comes out with something which is uniquely more appropriate than either one alone.
> would such a language be a niche language, or do you think it could come to rival even English for dominance?
It's a moot point, because a large subset of English will be a subset of the world language, and perhaps all of it. Same for other languages.
Eventually we will just realize that the world language already exists, as a superset of all existing languages, not only including the multitude of words and phrases for each semantic meaning (and the endless shades of meaning), but multiple syntaxes and grammars, which can carry their own meaning and flavor.
> 1. No good Browser.
Don't forget that AOL owns Netscape. By the time AOL finishes fscking Be, inc, Netscape 6 for the BeOS could be ready to fly.
> They DON'T relate any more than Slackware v.X relates to Debian v.Y.
.tgz), whereas Red Hat and Mandrake both use rpms. What would be more accurate is to say that Red Hat and Mandrake are no more alike than Caldera and SuSE, which both also use rpm.
Slackware and Debian don't use the same package file format (.deb vs
I'd say that the server is the killer app for the desktop.
What does the future hold for the HURD if things keep going the way they're going? What if acceptance for the HURD explodes, like it did for Linux? In either case, will it at some point replace Linux, or move into a niche alongside it, say, running the big iron while Linux scales down into all the little embedded devices? Do you or other GNU folks actually use it for anything now? If so, what? Do you feel like Linux has created interest in the HURD, or just stolen its thunder?
Netscape opened up a failing project in order to save it. Microsoft has more coders than Netscape, but the internet has more coders than Microsoft. That sort of thing. I think grey wants to know what would happen if a popular, successful project was opened, like Windows, or one of Apple's expensive authoring tools.
> What the studios are afraid of in this scenario
> is that their scriptwriters accidently (or maybe
> not-so-accidently) use a plot identical to one
> found in a fanfic story.
Then why does George Lucas only go after erotic fanfic? Is he planning on having Obi-Wan and Jar Jar getting a little close in Episode Two?
This may be a reason why slash and erotic fanfic in general is so popular - there's zero possibility of the work being stolen, because it would be inconceivable for the studios to actually make television and movies out of such things.
On the other hand, I imagine most fanfic writers wouldn't mind their writing being used by the studios, even if they didn't get credit, because at least they and their friends would know about it. But maybe I'm an optimist.
> you should be able to grap an edge and move it.
One of the things I hate most of all about the Windows interface is how frequently my irrational clicking and mouse jiggling leads to 'losing the taskbar' - resizing it to zero. Not only am I losing real-estate which I need for getting ahold of those icons, I've actually got a minor booby-trap built into the interface! What you describe might be nice for some, but all I want is a way to turn that option off, since I hardly ever manually resize my panel anyway (except to hide it).
> It's hard to put spacers in the panel.
That's true - the panel is either unnecessarially long, or exactly as long as the sum length of your applets, launchers, menus and drawers. It's not possible to put extra space in a corner panel (except for Panel/Right Click -> Global Properties -> Misc -> Applet Padding, which isn't the same thing), or to make the panel window only show a part of the panel's contents. Another thing that might be nice is floating panels, that is, panels which can be positioned arbitrarially anywhere on the desktop. I have two vertical corner panels, one in each upper corner of the desktop, and if I could, I'd probably put more up there in between them.
Well, there was the interview on IBM last week, but that wasn't a Slashdot Interview. The last Gnome developer to be interviewed on Slashdot was Havoc Pennington, I believe, and that was a long time ago - several months or more. So I'm not sure, but it sounds to me like you're trollin' for Jesus.
.debs for Helix Code's gear are going to be available. (What do you mean, I'm the only one?)
Why Miguel, well, because his company just started doing things, and we all want to know when the
The Matrix was the Wachowski brothers saying "how cool!" Pi was Aronofsky saying "watch out." ExistenZ was Cronenberg saying "you suck."
That's right. Unless Napster is grossly inefficient, there's no reason why it would take up any more than an irc client and an ftp server. Why don't they ban other obvious web hogs? Graphical web browsers? AllAdvantage-type software? Remote X sessions? MUDs and MMORPGs? Everquest isn't 'real work'.
A better reason for net admins to ban Napster is simply that they can't handle it all and need to cut something, anything, but at that point there should be enough tuition money to buy more gear.
In particular, executable files and similar things, like shell scripts and Word documents could be used to disseminate viruses. The 'instant mirror' ability sharing community software creates does present one solution to the slashdot effect, though.
Also, is there anything like ID3 tags for image files?
I agree that it is very unlikely. I think the most evil classmate I had in high school was the one who stole my code. Wasn't very bright, though ... he left my name on the listing.
I'll agree that an autonomous mind requres a goal-finding mechanism, but I won't agree that an AI has to stick with the one that it started out with. The parent goal can be respecified each time the AI runs, and it can be much more specific than the basic human desires. I think that's the separation, really. Humans have a very coarse set of desires - full stomach, human contact, that kind of thing ... whereas AI is important, in particular, for fine-grained, specialized application of intelligence.
BTW, Carl Jung was a raving nutter. Not that it has anything to do with your other comments.
The problem isn't specifically that we don't understand the human mind, but that we don't have enough of the right kinds of evidence to agree on how it works. For all we know, we may already understand intelligence well enough to design AI, but we just haven't applied the right set of ideas to the right hardware. There's no shortage of ideas floating around.
> Basically he points out that if we create a machine that is smarter than ourselves, it will do the same with respect to itself.
Assuming we give it autonomy. That's important to note - intelligence doesn't require such a thing as desire. Presumably we would be making these powerful thinking machines to determine how to more perfectly express our desires, so they'd be more like extensions of ourselves than like children or aliens.
We like a challenge.
Your argument seems to be, I understand it and can express it mathematically, therefore it isn't inteligence, and isn't what's going on in the brain, but this doesn't address the challenge at all.
... I think it's pretty clear that the problem here has nothing to do with intelligence. It's a question of motor coordination and perception. Reliance on intelligence may actually make the game harder.
You don't know what's going on in the brain, and you don't know what intelligence is. If intuition, or primary-process thinking, isn't understandable and expressable mathematically, then the goal of AI is literally impossible, and Turing-machine completeness is a crock.
> They can stare at their opponent to try and see if he's bluffing.
This is not a measure of intelligence, unless you think that a polygraph is intelligent.
> Priorities vs. Wants, etc., etc., etc. I have yet to see a machine that can make these types of decisions appropriately.
My operating system doesn't run a distributed.net client if other programs are taking up all the CPU. That's a decision based on a priority.
If what you want is a program that can make decisions that are human enough and complex enough for a human to fret about, well, there's a lot of work in that, and pretending that the incremental steps don't count just puts you that much farther from the goal. They do count.
> Take the example of something more fast-paced than Chess like Soccer.
Uhh