As Stephen Jay Gould touched on in The Mismeasure of Man, you can't measure the "intelligence" of a system as a scalar quantity. Yet a lot of AI thought seems to have been wasted on the idea that "If we can get a machine to carry out some task x with some degree of success, we will know it is intelligent." Values of x have included "carrying on a conversation" (i.e. the Turing test), "recognizing objects visually", "responding to natural-language requests for advice or information" (i.e. expert systems), "protecting itself", and so forth.
A lot of these x's have been accomplished... but nobody calls a Quake bot an intelligent being just because it can protect itself. As a more serious example, the "ELIZA effect" (people's tendency to attribute consciousness to anything that emits natural language) has debunked the Turing test in many people's minds.
What attributes of "intelligence" should we be looking for, then?
Information does not want to be free. Information is a non-breathing, non-living entity with no feelings and no desires.
When people have access to easy means of duplicating and transmitting large quantities of information, they will tend to behave in such a way that any particular piece of information, if it is interesting to even a tiny minority, will tend to spread among people to the extent that it is interesting.
In fact, they will desire to behave in this way so much that it will not be possible to stop the spread of any particular piece of information without causing severe harm to the people interested in it -- as by depriving them of privacy, security, and the use of their personal effects.
These people will resist such harm -- through political activism, boycotts, and technical workarounds. They may well also, when their ability to spread interesting information is threatened, react by spreading it more and more quickly. (See the case of Scientology's persecution of Internet-based critics, or the cases of Napster and Gnutella for that matter.) This is what is meant by "The Internet interprets censorship as system damage and routes around it" -- which is not a statement about the technical operations of the Internet (actual routers and hosts) but about the way people use it.
Hence, if you look at the communications-system from the point of view of a piece of information, you will see that information spreading in such a way that it appears to "want to be free".
(No, information does not have opinions. But then, neither does a plant, yet plants "want to" grow towards light sources, due to chemical reactions in their cells. Electricity doesn't have opinions either, but it "wants to" seek ground, and follow the path of lowest resistance. Anthropomorphism is our friend.)
I'm not sure what you mean by saying that HTML has "insufficient structure" on the presentation layer. Maybe some examples would help.
Read upthread. HTML does not specify sections, for instance. It is far too oriented towards the appearance of a document and insufficiently towards rigorous document structure. SGML and XML are all about document structure; that is what industry uses SGML for all the time.
(What do I mean by rigorous structure? In an SGML or XML DTD, you can specify that a document of a particular class must have (for instance) an author, modification date, etc. You can specify (for instance) that headlines must only appear at the beginning of chapters and sections, and that different headlines be used for each. If a file does not have the required features of the document class it claims to be (in the DOCTYPE header) then that is a syntax error. HTML by contrast is very loose and unstructured, oriented towards flexible presentation rather than being able to verify completeness. That may be the Right Thing for playing around, but it is the Wrong Thing for manpages, HOWTOs, or books.)
DocBook is hardly a niche application -- O'Reilly uses it constantly, and it is what Linux HOWTOs and other documentation are kept in by the Linux Documentation Project.
Of course, if a Linux distribution shipped all of the documentation in one format, I'd probably be so happy that I wouldn't care what format that is.
I agree that XHTML is probably a good starting point for unified Linux documentation.
FWIW, XHTML is not the same as XML. XHTML is "a reformulation of HTMLÊ4 as an XML 1.0 application" (to quote the W3C's take on it... and they should know. XHTML can be displayed by ordinary HTML browsers like the one you're using now, but it can also be parsed by an XML parser. It's basically a transitional form... getting people used to writing formally correct XMLish markup while there aren't yet enough XML tools out there.
(In other words, XML is not a markup language; it is a markup metalanguage. XML applications, of which XHTML is one, are markup languages.)
XHTML, because it is HTML, is the Wrong Thing for documentation, because HTML has insufficient structure, and the wrong sorts of what it's got. DocBook may or may not be the Right Thing for manpages, but the Linux Documentation Project folks seem to get along with it for HOWTOs, and they seem to be okay at rendering it into text or HTML or various other formats. DocBook is an SGML system and not XML, but that will be changing with the next major revision, and presumably LDP will be keeping up.
Please, no. HTML does not specify enough structure to adequately represent even a simple manpage. (It doesn't have, for instance, any idea of a section with a section title -- just headlines and paragraphs, not associated with one another.) XML, on the other hand, sounds to me like just the thing, because it makes it possible to specify structure more precisely, and to require that certain kinds of information be present or associated.
The biggest problem I have with the current state of documentation for Linux-based systems is its inconsistency. Debian, for instance, is a great system, but the state of its documentation is a mess. Half of it is in gzipped ASCII and HTML files in/usr/doc, a third in manpages, and the remaining sixth in the accursed GNU Info nonHTML-hypertext system (a plague be upon it).
Anyone building a new help/documentation system should take that problem into account. A new system must subsume all heavily-used existing systems. Either the existing manpages, Info nodes, HOWTOs, and the like must be converted to the new format, distributed as such, and maintained; or the new documentation browser must be able to reference, index, look up, and present the old formats.
I would not mind one bit if DocBook or some other XML DTD were used for all documentation under Linux. I would mind extremely if XML documentation became merely another kind of documentation I had to keep around alongside manpages, Info nodes, zipped ASCII, and HTML.
Every time I hear Stallman talking about freedom, I wonder if he really understands what it means. If I download a piece of GPLed code, I am not allowed to change it, compile it and give away only binaries. Where is the freedom in that?
Under copyright law, the 'default' license is not the public domain. The default license is no license at all -- which is to say, no freedom at all. If I write some code and give or sell you a copy, and I don't tell you specifically that you may copy it and distribute it, then you may not. If you do, you are violating copyright. That is what it means to say that I hold the 'copy-right' (the right to copy) to my work -- you do not, simply because you own a copy of it, have the right to make more copies and distribute them. Only I have that right.
In that context, GPL is significantly freer for you -- it grants you the right to distribute my work (and your derivative work of my work) under certain conditions. When I license my code to you under the GPL, I grant you rights you would not otherwise have -- specifically, the right to copy and distribute it in certain ways. Compared to not being allowed to distribute it at all (the situation normal copyright leaves you in) you have gained significant freedom.
You might suggest that I put my work in the public domain instead, thus permitting anyone to do whatever they want with it. The problem with that is that I want to make sure that anyone who gets a copy of my code -- or any derivative work of my code -- will be able to make further derivatives of it. Under copyright laws, after all, I have certain control over derivative works of my work, and I want to use that control to best ensure the freedom of users of those works. I do this by placing my work under GPL (or a similar license).
By putting my work under GPL, I am not restricting your freedom, because if I did not license it to you, you would have no freedom with regard to it at all, at all. I am not "restricting your freedom in order to ensure someone else's freedom". I am granting both you and the "someone else" freedoms neither of you would otherwise have.
Again, recall the default situation of no license. If I give you a copy of my code without giving you permission to produce derivative works, and you make a derivative work and distribute it, you are violating copyright, just as much as if you ripped a few dozen copies of Windows 2000 and sold them on the streets of New York City. By giving you permission to produce derivative works, I am expanding your freedom -- and by making sure that your derivative works are also free, I am expanding the freedom of their users as well. You have lost nothing; you and the rest of the world have both gained.
Please see this article in c't about Scientology's relation to DisKeeper. Please note that the headline refers only to Windows 2000 being "banned" from government use, not from private use.
Executive Software is not only run by a Scientologist; it is a member of WISE, the World Institute of Scientology Enterprises. WISE member corporations are operated for the benefit of the Church of Scientology, even though the CoS is supposedly a "nonprofit religious organization" and the WISE orgs are for-profit corporations.
You don't have to support the Democrat/Republican regime, you know. The opposition parties in America at present are weak but growing. If you, CiXeL, and those who agree with you join the opposition, it will become strong enough to change America's political reality.
There are several opposition parties, of which the most well-known are the Libertarians, the the Reform Party, the the Greens, and the Socialists. I myself am a Libertarian, but I would rather that you vote for any opposition party rather than voting for the regime (Democrat or Republican) or not voting at all.
Some Americans feel that it is "throwing their vote away" to vote for a candidate who is not likely to win -- and instead vote for the "lesser of two evils" among the Democrat and Republican. The problem with this is that, as you point out, the Dems and Reps really are not very different! Because of this, a vote for either a Democrat or Republican is basically a vote for their combined regime -- which, if you don't support the regime, is worse than throwing your vote away. Voting for even the tiniest of opposition parties, in contrast, registers your opposition to the Dem/Rep regime, and brings us closer to a real political debate in this country -- even if your candidate doesn't win this time.
Americans, please help bring real issues, real debate, and real differences back to our nation's political process. Give up on the Democrat/Republican regime. Vote for the opposition.
Let's suppose I write program foo and release it under the GPL, then release a basic user's manual under the FDL. Because of the viral nature of the FDL, I could not then go write a book (foo In A Nutshell) that expands on the FDL'ed documentation. Strictly interpreted, even quoting a single line of FDL'ed text could render the entire new document FDL'ed. Even paraphrasing the original text might not be enough to get it out from under the FDL, given the translation clause.
No, that's not correct. If you are the copyright holder -- the original author, or someone to whom the original author has given or sold the copyright -- then you can re-license the work under any license you like. Placing a program under the GPL does not restrict you from later licensing it under (for instance) the Artistic License as well (as Wall & co. have done with Perl) -- similarly, placing a document under the GDL does not restrict you from later releasing it under other licenses, even proprietary ones.
The exception would be if you accept GDLed contributions from others and include them in your document. If you write the "User's Guide to the Foobar System" and release it under GDL, and then I write some new sections for that document and give them to you under GDL for inclusion, then you cannot later re-license the whole document (your and my work together) under another license without my permission. (Similarly, Linus alone could not re-license the Linux kernel -- because it has lots of other people's work in it, and they submitted that work to Linus under the GPL.) This means that as the number of contributors to a GPL or GDL project gets larger, it gets much, much harder to re-license it -- and that's how it should be, to make sure nobody's work is made proprietary without his/her consent.
What if a print magazine is doing an article on Widgets, and wants to quote two paragraphs from the GDL'd Widgets Manual. Is it possible? Does the Magazine have to GDL itself? GDL that article? Since the magazine has a circulation of >100 does that have an impact?
Not in my (non-lawyerly) opinion. Such a quote in a review would be covered under the "fair use doctrine" regardless of which license the original text was distributed under.
Some guy wants to take a whole GDL document, modify it with his comments and give it to the 115 people in his lecture class. Does he also have to give them floppies since the distribution is > 100?
Would it be legal for him to photocopy the entire contents of a (proprietary) book and hand them out? No. If he distributes the GDLed document in that kind of quantity, he needs to go to the awful, terrible difficulty of mirroring it on his Web site and handing out the URL along with the document.
Some guy is writing a GDL'd document and wants to include a longish section of a non-GDL'd document. Is this illegal, as it would be with code under GPL?
If you are writing a document to release under GDL, you may include in it anything you could legally include in a document for publication. Without the permission of the author of the included section, you would be restricted to quoting for "fair use" purposes and in "fair use" lengths. With the permission of that author, you can do whatever s/he allows you to.
Suppose I want to quote a large chunk of text that is genuinely public domain. Does the license now infect that text in other places?
No. Nothing can be taken out of the public domain. However, you have no obligation to tell your readers that that chunk of text is in the public domain. If someone used it elsewhere under a proprietary license and you sued them, you would lose if they could prove it was in the public domain before you printed it.
I just looked over some of your posts on this discussion, rambone. It's not "manufacturing consent" when the kinds of stuff you've posted here gets moderated down. It's the self-regulation of the forum. Self-regulation, aka feedback -- one of those processes of liberty that totalitarians never seem to understand.
Insulting the victims of sexual abuse who seek anonymous online forums, and saying that their anonymity only exists so they can "perform porn-related transactions", as you did in this comment, is one of the basest things you could do here. That's truly vile. It's not "banal group-thought" to moderate that down as a troll -- it's beneficial. Virtuous, even.
This is not a forum where that kind of tripe gets equal standing alongside useful and insightful contributions. If you want alt.flame, you know where to find it.
How would your system allows lurkers (another way to put it is people who specialize in reading Slashdot) to moderate?
This is something I hadn't considered. Lurkers should definitely get points!
Most likely the best solution would be to also give points for reading articles, as the current system apparently does. (See the moderation rules.)
I do think that people who post well should get more chance to judge other posts than people whose posts are mediocre or nonexistent -- all else being equal. This is partly because I think they're more likely to know what's good; and partly as a reward for having posted well.
Also, the current system *does not* give moderator priorities to people who have higher karma, but your system does.
No, not exactly. In my system, if you spend your karma by doing moderations, you don't have that karma any more -- you don't keep on getting benefits from karma you've spent, which is what it sounds like you're suggesting. In other words, it gives priority (i.e. a steadier, higher income of points) to consistently well-rated posters, not to people with a strong past posting history (i.e. people who have "high karma" in today's system).
It turns Slashdot into a popularity contest, which I had hoped I'd left behind in high school.
Suppose that the total number of points given out for reading or "by default" exceeds the number of points given out as a result of moderation? Do you still think there's more "popularity contest" than there is now? (Keep in mind how prevalent the accusation of "karma whore!" is right now -- an accusation which is basically equivalent to "popularity-contest-monger!"...)
A lot of the reason I'm proposing this system is to work out the "karma whore!" and "Moderate this up, please!" problems of the current system. But most of it is because it would be interesting to see how a pseudonymous market economy of information works out.
I think that the current system has a lot of social inertia behind it; the current system works moderately well and CmdrTaco has probably spent a lot of time fine-tuning it.
Yes, it has some social inertia, but there are also a lot of complaints about it. Here are a few ways my proposal would deal with these:
Consider the accusation of "karma whore". This accusation arises when someone who can't moderate at the moment (either disqualified, or just doesn't have any points, or already posted in this discussion) sees a post s/he feels is overrated. These people feel powerless to fix a problem, and so they vent instead, making pointless accusations. If they were made to realize instead that by posting well, they can get moderation points to p.d. the posts they think are overrated, they'll do that instead of flaming.
How about the ridiculously small range of potential ratings for a post, and the excess of early +5s? Lots of early posts get pegged all the way +5 for little reason but that a bunch of people push them up at once. By increasing the range of ratings and making it more "expensive" in karma to increase an already-high rating than to bring a post from (say) +1 to +2, moderations would not cluster around a few good posts, but spread out more reasonably.
Consider also the numerous wasted comments asking moderators to moderate a given post up or down. As with the "karma whore" accusations, these come from people not having adequate chance to moderate.
(The current moderation system seems to be based on denying moderation points entirely to people who are suspected of not being "good enough" for them -- i.e. people with low accumulated karma. This causes bitterness due to a lack of representation. My system would dramatically increase the number of moderation points in the system, spread them out a bit so that everyone has at least a few, make each point count less by increasing the "cost" of pushing a score further away from neutral, yet keep the majority of points in the hands of people who are consistently posting well. Wouldn't these be worthwhile changes?)
And perhaps most importantly (and why this discussion is on topic here) Slashdot under a moderation-system like the one I propose would serve as a working, living example of a reputation-enabled, self-policing anonymous/pseudonymous market system. (Sure, those sound like buzzwords -- I'd like them to become buzzwords!) This is a sort of system it would be very useful to develop. It would directly help support pseudonymous freedom of speech, and thus oppose Clintonesque surveillance, by unmasking the "anonymous == irresponsible" red herring. It might also serve as an example for other pseudonymous market systems, such as might be associated with e-cash or the like.
Second, a starting score of 2 allows you to expose your post to the not-insignificant number of people who set their threshold to 2.
In my proposed system above, this purpose would be served by people being able to moderate their own posts. This would let people expend their karma (accumulated goodwill of the community) to "advertise" their posts. It would also, of course, let people blow their karma on flogging a rant that gets lim p.d.'d by everyone... but that's a small price to pay IMHO.
One thing I left out of my system as described above -- there needs to be an influx of karma/moderation points into the system, or else it will run out. (Do the math.) This could be accomplished by giving all participants a small amount of karma every week -- or, better yet, giving everyone one point for every ten posts they make that don't get moderated below zero, to a maximum of (say) 3 points per day.
Again, the idea here is to create a literal marketplace of ideas, more or less -- to let people get "rich" (in points) by posting good comments, and to let people "spend" these "riches" on either promoting comments they like or on advertising their own.
The ultimate test of such a system, of course, would be to let anyone with a lot of karma points (say, 100 -- more than I have) spend them on putting an article on the front page... and, of course, letting people moderate the front-page articles as well as the comments.
We've all heard before that anonymity is essential to free speech, because anything less exposes writers and speakers to fear of retribution -- from government, from criminals (Witness Protection Plan, anyone?), from monied interests, and the like.
The totalitarians' response to this claim almost always has to do with "responsibility" -- generally "making people take responsibility for their speech". This is a really funny use of the word "responsibility", since it seems to actually mean "vulnerability": ensuring that people are vulnerable to assault whenever they speak.
It's true that people sometimes act like assholes when they're anonymous -- see the behavior of a notable subset (but by no means all) of the Anonymous Cowards here. Anonymous forums need a way to protect themselves from abuse without compromising the identities of their participants.
I think that the answer lies in the unhindered creation of pseudonymous identities, coupled with powerful reputation systems for all identities. The Slashdot karma system is a primitive form of a reputation system for pseudonymous identities. There are a few improvements I would make to the current karma system to bring it more in line with how a pseudonymous reputation system should work:
Rather than having the bonus for high karma and the moderation system be separate, make them one system, as such: Any act of moderation costs karma, and gives karma to the one moderated. The exception is when moderating your own posts; boosting them costs you karma, but does not give you any back; reducing them does not cost you karma. This simplifies the system; permits people to more accurately rate the value of their own posts; and makes one's accumulated karma figure more worthwhile.
(Presently, what good is having a lot of karma? It means that you have the freedom to troll a lot before getting harmed by being moderated down. That's not something we want to encourage. Instead, having a lot of karma should give you a positive power, like being able to moderate.)
When you spend karma to change a post's score, you have to spend more points than the amount by which you want to change it. (In other words, karma expenditure is not 100% efficient.) The more you want to raise (or lower) the score, the more it costs (increasing geometrically); also, the further the post is from neutral, the more it costs to push it further away from neutral. These rules help avoid the current surfeit of +5 posts, as well as discouraging bulk transfers of karma among identities.
What these rules would do is create an "economy" of karma/moderation points, similar in many ways to a money economy. In a money economy, if I like your products, I buy them -- which costs me money and enriches you, letting you make more products and advertise them more widely, as well as letting you go buy someone else's products. The end result of the free market is that people produce better and better (or at least more and more marketable) products, driving the increase in technology and productivity; with luck, the end result of this karma economy would be to encourage better and better posts, driving an increase in valuable discussion.
In this way, we can derive a significant benefit from a system of pseudonymous identity, thus demonstrating to the world that responsibility has nothing to do with the vulnerabilities associated with being identifiable.
I'm the primary network administrator for a small college with a ridiculously limited Internet connection. We have ~350 students and 10Mbps Ethernet to all dorm rooms, but we only have a 512Kbps uplink at the moment. As a result, our users often end up contending for time on a rather congested link.
The three or four warez d00dz who think they have to have a dozen MP3s or VCDs downloading at all times don't help. A few months ago we had one or two twits using up well in excess of 50% of our bandwidth, moving traffic we all knew perfectly well was bootlegged media. But we really see this as an excessive-use problem, not a bootlegging problem -- so we put a 200MB/day cap on usage. As soon as any user machine on our network has moved 200MB over the Internet link in one day, it is unceremoniously blocked off until 3AM the next morning. There is a "free period" from 3AM to 6AM during which people may download all they want without limit; also, we grant exceptions for academic use, such as when someone wants to download a new distro CD image. (The funny thing is that the really heavy users don't use the 3AM-6AM window, even though there's plenty of scheduled-download software out there. They just hit 200MB and get blocked -- just about every day.)
We do, actually, have a policy against bootlegging software, music, movies, and the like -- but I'll be the first to admit that's a CYA move, so if RIAA or the like come attack us, we can say we don't tolerate bootlegging. We don't go looking for MP3 servers unless someone raises a fuss. We do block NetBIOS-over-TCP at the firewall, but that's all. (We're planning to block inbound SMTP directed to systems other than our mail server in the near future, but that's to stop spammers, not to limit our users.)
Blocking services by port number is not a solution to excessive use, nor is it a solution to bootlegging or other "contrabandwidth". In a port-blocking situation, the serious abusers tunnel or otherwise route around the censorship; the regular users get stuck with bogus limits on their use; and we sysadmins have to play catch-up maintaining a list of blocked services. If congestion is the problem, ban excessive use, not controversial use.
The manager at "Chevron" has to believe that they are getting something better than the others.
Why? Does the manager at Chevron really need to think that her word processor is "better" than the one the manager at Exxon-Mobil or Shell uses? It seems to me that it's more important to the Chevron manager that her word processor be able to form a component of a more successful business model -- not that the WP itself is somehow "better" (whatever that means).
One way a word processor can improve a business model is to cut expenses. Free software, usually being free[beer] as well as free[speech], can do that. Red Hat claims to be interested in cutting the costs of software industry-wide: even if they charge for software, this accomplishes the same thing.
Another is to cut down on nonproductive uses of time, such as time spent rebooting or re-doing work that's lost in a crash. Free software, being developed to meet needs rather than marketing deadlines, is often more reliable than proprietary software.
In short, the "better"-ness of one word processor over another is not something inherent in the code, but rather something that manifests when you take the word processor and fit it into your business model.
I wrote a bit of a note to the NIPC suggesting that find_ddos be open-sourced, and pointing out some of the advantages which would accrue, including portability, expansion, and increased trust. I also asked that the license under which it is distributed be clarified, so that I could know if I can legally mirror it. Here's the answer I got back:
"The NIPC has determined that it is important not to release the source code publicly. We do, however, have measures in place to help ensure that the executable on our website is not compromised. We will forward your comments to the appropriate personnel for consideration in this matter. Thank you for contacting us."
How's that for null program?
Why not just use the Crusoe as a G4?
on
Darwin on Crusoe?
·
· Score: 4
Instead of "porting" Darwin to x86 and running it on a bog-standard Crusoe, why not talk the Crusoe into being a G4 instead of an x86? I was under the impression that the Crusoe's code-morphing software could be reworked to emulate other instruction sets besides x86...
This seems to be a chronic failing of Governments. Utterly failing to detect and respond to gross understaffing or changing circumstances in anything approaching a resonable timeframe.
Tell me, what incentive does a government have to respond promptly to changing needs?
A business, for instance, has a wonderful incentive to respond to needs: if it doesn't, it will lose out to competitors. If Intel kept pushing first-generation Pentiums while AMD was developing K6 and K7 chips -- and the market needed speed -- Intel would lose out; they would first have become less profitable, and finally would go out of business.
A skilled laborer (for instance, a coder or sysadmin) has a similar incentive. If all you know is Windows NT, but Windows NT is in decline and being replaced by Unix, you have an incentive to learn Unix. If you don't, you will find your skills not worth as much money.
Business and laborers exist in a market relationship with their clients. In order to make more money, businesses have to fill the needs of their customers, and laborers have to fill the needs of their employers. If you're not good at this, you get less money, or (worse yet) you either go out of business (if you're a business) or you lose your job (if you're a laborer).
Governments do not exist in a market relationship with their "clients" -- the residents (citizens and resident aliens) of their territories. A government has a monopoly. Unless it does something seriously bogus and gets kicked out in a revolution, coup, or major political upheaval, it's just gonna sit there and fester.
Hence, governments do not have the kind of incentive to correct shortages that businesses and laborers do. If there's a shortage of beer, the price will go up, causing more businesses to make more beer -- solving the shortage. If there's a shortage of Linux sysadmins, the salaries will go up, attracting more skilled people to the field -- solving the shortage. But if there's a shortage of green cards -- what incentive has INS to act?
INS is not responsible to its nominal clients (the resident aliens and their employers). One could say that it is responsible to the American citizenry through the democratic process, but this responsibility is so very diffuse it is more or less nonexistent. A fussy article in the newspaper does not constitute "being held responsible", sad to say: immigration is not a big-ticket issue in most parts of the country.
When you entrust a particular power to a government bureau -- when you take a certain kind of decision-making out of the hands of individuals in the market -- you will necessarily lose efficiency, because bureaucrats have so little incentive to accurately and speedily meet needs. And in this case, I don't think we get enough benefit out of the process to justify that inefficiency, and these injustices.
End restrictive immigrations laws. Let law-abiding people like Linus Torvalds work wherever people are hiring.
Well, I was thinking that they might find a floppy and not even realise that it might be used for data storage.
This reminds me of the "prayer fans" from Fred Pohl's Heechee Saga (the Gateway series). For decades humans had been finding these little alien artfacts all around the cosmos: crystalline cylinders that open out into a fan shape if you squeeze them right. No clue what they were for, so they called 'em "prayer fans", figuring they had some religious significance.
What Mr. Valenti has done here is what our lawyer friends know as "slander" (when it is spoken aloud in public) or "libel" (when it is published in print).
You have committed slander or libel when you knowingly make a defamatory, false statement about someone. Publicly accusing someone of a crime s/he didn't commit -- and which you know s/he didn't commit -- is a perfect example.
Now, it's possible Valenti is so deluded, so caught up in his own little world, that he is not aware that his victims committed no crime. This would mean that he didn't knowingly make false statements -- because he didn't believe them to be false. But for some reason I doubt this. I suspect he's not an ignorant demagogue, but rather a cynical one: one who knows that he's lying for his own advantage, and revels in it.
I picked these up by doing a search for "Linux security" using the search widget on the bottom of the Slashdot main page. These are just off the first page of results. Doubtless there are several stories about security problems in daemons which weren't turned up by this search (because they didn't contain the string "Linux").
In other words, security holes in Linux (and other free software) are reported on Slashdot. Your statement appears to be a misleading one intended to incite others to fear, be uncertain about, or doubt the honesty of the Slashdot editors. Isn't that what FUD is all about?
Further, keep in mind that while Microsoft thinks itself to be hurt by the reporting of security holes in its products, Linux is not hurt by the reporting of security holes in Linux-related software. Bug-reporting is a threat to the proprietary-software model, but it is an element of the success of the free-software model.
Slashdot Mirror
A lot of these x's have been accomplished ... but nobody calls a Quake bot an intelligent being just because it can protect itself. As a more serious example, the "ELIZA effect" (people's tendency to attribute consciousness to anything that emits natural language) has debunked the Turing test in many people's minds.
What attributes of "intelligence" should we be looking for, then?
When people have access to easy means of duplicating and transmitting large quantities of information, they will tend to behave in such a way that any particular piece of information, if it is interesting to even a tiny minority, will tend to spread among people to the extent that it is interesting.
In fact, they will desire to behave in this way so much that it will not be possible to stop the spread of any particular piece of information without causing severe harm to the people interested in it -- as by depriving them of privacy, security, and the use of their personal effects.
These people will resist such harm -- through political activism, boycotts, and technical workarounds. They may well also, when their ability to spread interesting information is threatened, react by spreading it more and more quickly. (See the case of Scientology's persecution of Internet-based critics, or the cases of Napster and Gnutella for that matter.) This is what is meant by "The Internet interprets censorship as system damage and routes around it" -- which is not a statement about the technical operations of the Internet (actual routers and hosts) but about the way people use it.
Hence, if you look at the communications-system from the point of view of a piece of information, you will see that information spreading in such a way that it appears to "want to be free".
(No, information does not have opinions. But then, neither does a plant, yet plants "want to" grow towards light sources, due to chemical reactions in their cells. Electricity doesn't have opinions either, but it "wants to" seek ground, and follow the path of lowest resistance. Anthropomorphism is our friend.)
Read upthread. HTML does not specify sections, for instance. It is far too oriented towards the appearance of a document and insufficiently towards rigorous document structure. SGML and XML are all about document structure; that is what industry uses SGML for all the time.
(What do I mean by rigorous structure? In an SGML or XML DTD, you can specify that a document of a particular class must have (for instance) an author, modification date, etc. You can specify (for instance) that headlines must only appear at the beginning of chapters and sections, and that different headlines be used for each. If a file does not have the required features of the document class it claims to be (in the DOCTYPE header) then that is a syntax error. HTML by contrast is very loose and unstructured, oriented towards flexible presentation rather than being able to verify completeness. That may be the Right Thing for playing around, but it is the Wrong Thing for manpages, HOWTOs, or books.)
DocBook is hardly a niche application -- O'Reilly uses it constantly, and it is what Linux HOWTOs and other documentation are kept in by the Linux Documentation Project.
That was my point all along, yes.
FWIW, XHTML is not the same as XML. XHTML is "a reformulation of HTMLÊ4 as an XML 1.0 application" (to quote the W3C's take on it ... and they should know. XHTML can be displayed by ordinary HTML browsers like the one you're using now, but it can also be parsed by an XML parser. It's basically a transitional form ... getting people used to writing formally correct XMLish markup while there aren't yet enough XML tools out there.
(In other words, XML is not a markup language; it is a markup metalanguage. XML applications, of which XHTML is one, are markup languages.)
XHTML, because it is HTML, is the Wrong Thing for documentation, because HTML has insufficient structure, and the wrong sorts of what it's got. DocBook may or may not be the Right Thing for manpages, but the Linux Documentation Project folks seem to get along with it for HOWTOs, and they seem to be okay at rendering it into text or HTML or various other formats. DocBook is an SGML system and not XML, but that will be changing with the next major revision, and presumably LDP will be keeping up.
Please, no. HTML does not specify enough structure to adequately represent even a simple manpage. (It doesn't have, for instance, any idea of a section with a section title -- just headlines and paragraphs, not associated with one another.) XML, on the other hand, sounds to me like just the thing, because it makes it possible to specify structure more precisely, and to require that certain kinds of information be present or associated.
The biggest problem I have with the current state of documentation for Linux-based systems is its inconsistency. Debian, for instance, is a great system, but the state of its documentation is a mess. Half of it is in gzipped ASCII and HTML files in /usr/doc, a third in manpages, and the remaining sixth in the accursed GNU Info nonHTML-hypertext system (a plague be upon it).
Anyone building a new help/documentation system should take that problem into account. A new system must subsume all heavily-used existing systems. Either the existing manpages, Info nodes, HOWTOs, and the like must be converted to the new format, distributed as such, and maintained; or the new documentation browser must be able to reference, index, look up, and present the old formats.
I would not mind one bit if DocBook or some other XML DTD were used for all documentation under Linux. I would mind extremely if XML documentation became merely another kind of documentation I had to keep around alongside manpages, Info nodes, zipped ASCII, and HTML.
Under copyright law, the 'default' license is not the public domain. The default license is no license at all -- which is to say, no freedom at all. If I write some code and give or sell you a copy, and I don't tell you specifically that you may copy it and distribute it, then you may not. If you do, you are violating copyright. That is what it means to say that I hold the 'copy-right' (the right to copy) to my work -- you do not, simply because you own a copy of it, have the right to make more copies and distribute them. Only I have that right.
In that context, GPL is significantly freer for you -- it grants you the right to distribute my work (and your derivative work of my work) under certain conditions. When I license my code to you under the GPL, I grant you rights you would not otherwise have -- specifically, the right to copy and distribute it in certain ways. Compared to not being allowed to distribute it at all (the situation normal copyright leaves you in) you have gained significant freedom.
You might suggest that I put my work in the public domain instead, thus permitting anyone to do whatever they want with it. The problem with that is that I want to make sure that anyone who gets a copy of my code -- or any derivative work of my code -- will be able to make further derivatives of it. Under copyright laws, after all, I have certain control over derivative works of my work, and I want to use that control to best ensure the freedom of users of those works. I do this by placing my work under GPL (or a similar license).
By putting my work under GPL, I am not restricting your freedom, because if I did not license it to you, you would have no freedom with regard to it at all, at all. I am not "restricting your freedom in order to ensure someone else's freedom". I am granting both you and the "someone else" freedoms neither of you would otherwise have.
Again, recall the default situation of no license. If I give you a copy of my code without giving you permission to produce derivative works, and you make a derivative work and distribute it, you are violating copyright, just as much as if you ripped a few dozen copies of Windows 2000 and sold them on the streets of New York City. By giving you permission to produce derivative works, I am expanding your freedom -- and by making sure that your derivative works are also free, I am expanding the freedom of their users as well. You have lost nothing; you and the rest of the world have both gained.
Executive Software is not only run by a Scientologist; it is a member of WISE, the World Institute of Scientology Enterprises. WISE member corporations are operated for the benefit of the Church of Scientology, even though the CoS is supposedly a "nonprofit religious organization" and the WISE orgs are for-profit corporations.
There are several opposition parties, of which the most well-known are the Libertarians, the the Reform Party, the the Greens, and the Socialists. I myself am a Libertarian, but I would rather that you vote for any opposition party rather than voting for the regime (Democrat or Republican) or not voting at all.
Some Americans feel that it is "throwing their vote away" to vote for a candidate who is not likely to win -- and instead vote for the "lesser of two evils" among the Democrat and Republican. The problem with this is that, as you point out, the Dems and Reps really are not very different! Because of this, a vote for either a Democrat or Republican is basically a vote for their combined regime -- which, if you don't support the regime, is worse than throwing your vote away. Voting for even the tiniest of opposition parties, in contrast, registers your opposition to the Dem/Rep regime, and brings us closer to a real political debate in this country -- even if your candidate doesn't win this time.
Americans, please help bring real issues, real debate, and real differences back to our nation's political process. Give up on the Democrat/Republican regime. Vote for the opposition.
The exception would be if you accept GDLed contributions from others and include them in your document. If you write the "User's Guide to the Foobar System" and release it under GDL, and then I write some new sections for that document and give them to you under GDL for inclusion, then you cannot later re-license the whole document (your and my work together) under another license without my permission. (Similarly, Linus alone could not re-license the Linux kernel -- because it has lots of other people's work in it, and they submitted that work to Linus under the GPL.) This means that as the number of contributors to a GPL or GDL project gets larger, it gets much, much harder to re-license it -- and that's how it should be, to make sure nobody's work is made proprietary without his/her consent.
I hope this clears things up a bit.
Insulting the victims of sexual abuse who seek anonymous online forums, and saying that their anonymity only exists so they can "perform porn-related transactions", as you did in this comment, is one of the basest things you could do here. That's truly vile. It's not "banal group-thought" to moderate that down as a troll -- it's beneficial. Virtuous, even.
This is not a forum where that kind of tripe gets equal standing alongside useful and insightful contributions. If you want alt.flame, you know where to find it.
Most likely the best solution would be to also give points for reading articles, as the current system apparently does. (See the moderation rules.)
I do think that people who post well should get more chance to judge other posts than people whose posts are mediocre or nonexistent -- all else being equal. This is partly because I think they're more likely to know what's good; and partly as a reward for having posted well.
No, not exactly. In my system, if you spend your karma by doing moderations, you don't have that karma any more -- you don't keep on getting benefits from karma you've spent, which is what it sounds like you're suggesting. In other words, it gives priority (i.e. a steadier, higher income of points) to consistently well-rated posters, not to people with a strong past posting history (i.e. people who have "high karma" in today's system). Suppose that the total number of points given out for reading or "by default" exceeds the number of points given out as a result of moderation? Do you still think there's more "popularity contest" than there is now? (Keep in mind how prevalent the accusation of "karma whore!" is right now -- an accusation which is basically equivalent to "popularity-contest-monger!"A lot of the reason I'm proposing this system is to work out the "karma whore!" and "Moderate this up, please!" problems of the current system. But most of it is because it would be interesting to see how a pseudonymous market economy of information works out.
Consider the accusation of "karma whore". This accusation arises when someone who can't moderate at the moment (either disqualified, or just doesn't have any points, or already posted in this discussion) sees a post s/he feels is overrated. These people feel powerless to fix a problem, and so they vent instead, making pointless accusations. If they were made to realize instead that by posting well, they can get moderation points to p.d. the posts they think are overrated, they'll do that instead of flaming.
How about the ridiculously small range of potential ratings for a post, and the excess of early +5s? Lots of early posts get pegged all the way +5 for little reason but that a bunch of people push them up at once. By increasing the range of ratings and making it more "expensive" in karma to increase an already-high rating than to bring a post from (say) +1 to +2, moderations would not cluster around a few good posts, but spread out more reasonably.
Consider also the numerous wasted comments asking moderators to moderate a given post up or down. As with the "karma whore" accusations, these come from people not having adequate chance to moderate.
(The current moderation system seems to be based on denying moderation points entirely to people who are suspected of not being "good enough" for them -- i.e. people with low accumulated karma. This causes bitterness due to a lack of representation. My system would dramatically increase the number of moderation points in the system, spread them out a bit so that everyone has at least a few, make each point count less by increasing the "cost" of pushing a score further away from neutral, yet keep the majority of points in the hands of people who are consistently posting well. Wouldn't these be worthwhile changes?)
And perhaps most importantly (and why this discussion is on topic here) Slashdot under a moderation-system like the one I propose would serve as a working, living example of a reputation-enabled, self-policing anonymous/pseudonymous market system. (Sure, those sound like buzzwords -- I'd like them to become buzzwords!) This is a sort of system it would be very useful to develop. It would directly help support pseudonymous freedom of speech, and thus oppose Clintonesque surveillance, by unmasking the "anonymous == irresponsible" red herring. It might also serve as an example for other pseudonymous market systems, such as might be associated with e-cash or the like.
One thing I left out of my system as described above -- there needs to be an influx of karma/moderation points into the system, or else it will run out. (Do the math.) This could be accomplished by giving all participants a small amount of karma every week -- or, better yet, giving everyone one point for every ten posts they make that don't get moderated below zero, to a maximum of (say) 3 points per day.
Again, the idea here is to create a literal marketplace of ideas, more or less -- to let people get "rich" (in points) by posting good comments, and to let people "spend" these "riches" on either promoting comments they like or on advertising their own.
The ultimate test of such a system, of course, would be to let anyone with a lot of karma points (say, 100 -- more than I have) spend them on putting an article on the front page ... and, of course, letting people moderate the front-page articles as well as the comments.
We've all heard before that anonymity is essential to free speech, because anything less exposes writers and speakers to fear of retribution -- from government, from criminals (Witness Protection Plan, anyone?), from monied interests, and the like.
The totalitarians' response to this claim almost always has to do with "responsibility" -- generally "making people take responsibility for their speech". This is a really funny use of the word "responsibility", since it seems to actually mean "vulnerability": ensuring that people are vulnerable to assault whenever they speak.
It's true that people sometimes act like assholes when they're anonymous -- see the behavior of a notable subset (but by no means all) of the Anonymous Cowards here. Anonymous forums need a way to protect themselves from abuse without compromising the identities of their participants.
I think that the answer lies in the unhindered creation of pseudonymous identities, coupled with powerful reputation systems for all identities. The Slashdot karma system is a primitive form of a reputation system for pseudonymous identities. There are a few improvements I would make to the current karma system to bring it more in line with how a pseudonymous reputation system should work:
Rather than having the bonus for high karma and the moderation system be separate, make them one system, as such: Any act of moderation costs karma, and gives karma to the one moderated. The exception is when moderating your own posts; boosting them costs you karma, but does not give you any back; reducing them does not cost you karma. This simplifies the system; permits people to more accurately rate the value of their own posts; and makes one's accumulated karma figure more worthwhile.
(Presently, what good is having a lot of karma? It means that you have the freedom to troll a lot before getting harmed by being moderated down. That's not something we want to encourage. Instead, having a lot of karma should give you a positive power, like being able to moderate.)
When you spend karma to change a post's score, you have to spend more points than the amount by which you want to change it. (In other words, karma expenditure is not 100% efficient.) The more you want to raise (or lower) the score, the more it costs (increasing geometrically); also, the further the post is from neutral, the more it costs to push it further away from neutral. These rules help avoid the current surfeit of +5 posts, as well as discouraging bulk transfers of karma among identities.
What these rules would do is create an "economy" of karma/moderation points, similar in many ways to a money economy. In a money economy, if I like your products, I buy them -- which costs me money and enriches you, letting you make more products and advertise them more widely, as well as letting you go buy someone else's products. The end result of the free market is that people produce better and better (or at least more and more marketable) products, driving the increase in technology and productivity; with luck, the end result of this karma economy would be to encourage better and better posts, driving an increase in valuable discussion.
In this way, we can derive a significant benefit from a system of pseudonymous identity, thus demonstrating to the world that responsibility has nothing to do with the vulnerabilities associated with being identifiable.
I'm the primary network administrator for a small college with a ridiculously limited Internet connection. We have ~350 students and 10Mbps Ethernet to all dorm rooms, but we only have a 512Kbps uplink at the moment. As a result, our users often end up contending for time on a rather congested link.
The three or four warez d00dz who think they have to have a dozen MP3s or VCDs downloading at all times don't help. A few months ago we had one or two twits using up well in excess of 50% of our bandwidth, moving traffic we all knew perfectly well was bootlegged media. But we really see this as an excessive-use problem, not a bootlegging problem -- so we put a 200MB/day cap on usage. As soon as any user machine on our network has moved 200MB over the Internet link in one day, it is unceremoniously blocked off until 3AM the next morning. There is a "free period" from 3AM to 6AM during which people may download all they want without limit; also, we grant exceptions for academic use, such as when someone wants to download a new distro CD image. (The funny thing is that the really heavy users don't use the 3AM-6AM window, even though there's plenty of scheduled-download software out there. They just hit 200MB and get blocked -- just about every day.)
We do, actually, have a policy against bootlegging software, music, movies, and the like -- but I'll be the first to admit that's a CYA move, so if RIAA or the like come attack us, we can say we don't tolerate bootlegging. We don't go looking for MP3 servers unless someone raises a fuss. We do block NetBIOS-over-TCP at the firewall, but that's all. (We're planning to block inbound SMTP directed to systems other than our mail server in the near future, but that's to stop spammers, not to limit our users.)
Blocking services by port number is not a solution to excessive use, nor is it a solution to bootlegging or other "contrabandwidth". In a port-blocking situation, the serious abusers tunnel or otherwise route around the censorship; the regular users get stuck with bogus limits on their use; and we sysadmins have to play catch-up maintaining a list of blocked services. If congestion is the problem, ban excessive use, not controversial use.
Why? Does the manager at Chevron really need to think that her word processor is "better" than the one the manager at Exxon-Mobil or Shell uses? It seems to me that it's more important to the Chevron manager that her word processor be able to form a component of a more successful business model -- not that the WP itself is somehow "better" (whatever that means).
One way a word processor can improve a business model is to cut expenses. Free software, usually being free[beer] as well as free[speech], can do that. Red Hat claims to be interested in cutting the costs of software industry-wide: even if they charge for software, this accomplishes the same thing.
Another is to cut down on nonproductive uses of time, such as time spent rebooting or re-doing work that's lost in a crash. Free software, being developed to meet needs rather than marketing deadlines, is often more reliable than proprietary software.
In short, the "better"-ness of one word processor over another is not something inherent in the code, but rather something that manifests when you take the word processor and fit it into your business model.
I wrote a bit of a note to the NIPC suggesting that find_ddos be open-sourced, and pointing out some of the advantages which would accrue, including portability, expansion, and increased trust. I also asked that the license under which it is distributed be clarified, so that I could know if I can legally mirror it. Here's the answer I got back:
"The NIPC has determined that it is important not to release the source code publicly. We do, however, have measures in place to help ensure that the executable on our website is not compromised. We will forward your comments to the appropriate personnel for consideration in this matter. Thank you for contacting us."
How's that for null program?
Instead of "porting" Darwin to x86 and running it on a bog-standard Crusoe, why not talk the Crusoe into being a G4 instead of an x86? I was under the impression that the Crusoe's code-morphing software could be reworked to emulate other instruction sets besides x86 ...
I don't see why not. Elf Sternberg did it.
A business, for instance, has a wonderful incentive to respond to needs: if it doesn't, it will lose out to competitors. If Intel kept pushing first-generation Pentiums while AMD was developing K6 and K7 chips -- and the market needed speed -- Intel would lose out; they would first have become less profitable, and finally would go out of business.
A skilled laborer (for instance, a coder or sysadmin) has a similar incentive. If all you know is Windows NT, but Windows NT is in decline and being replaced by Unix, you have an incentive to learn Unix. If you don't, you will find your skills not worth as much money.
Business and laborers exist in a market relationship with their clients. In order to make more money, businesses have to fill the needs of their customers, and laborers have to fill the needs of their employers. If you're not good at this, you get less money, or (worse yet) you either go out of business (if you're a business) or you lose your job (if you're a laborer).
Governments do not exist in a market relationship with their "clients" -- the residents (citizens and resident aliens) of their territories. A government has a monopoly. Unless it does something seriously bogus and gets kicked out in a revolution, coup, or major political upheaval, it's just gonna sit there and fester.
Hence, governments do not have the kind of incentive to correct shortages that businesses and laborers do. If there's a shortage of beer, the price will go up, causing more businesses to make more beer -- solving the shortage. If there's a shortage of Linux sysadmins, the salaries will go up, attracting more skilled people to the field -- solving the shortage. But if there's a shortage of green cards -- what incentive has INS to act?
INS is not responsible to its nominal clients (the resident aliens and their employers). One could say that it is responsible to the American citizenry through the democratic process, but this responsibility is so very diffuse it is more or less nonexistent. A fussy article in the newspaper does not constitute "being held responsible", sad to say: immigration is not a big-ticket issue in most parts of the country.
When you entrust a particular power to a government bureau -- when you take a certain kind of decision-making out of the hands of individuals in the market -- you will necessarily lose efficiency, because bureaucrats have so little incentive to accurately and speedily meet needs. And in this case, I don't think we get enough benefit out of the process to justify that inefficiency, and these injustices.
End restrictive immigrations laws. Let law-abiding people like Linus Torvalds work wherever people are hiring.
Then they found the disk drive for them.
What Mr. Valenti has done here is what our lawyer friends know as "slander" (when it is spoken aloud in public) or "libel" (when it is published in print).
You have committed slander or libel when you knowingly make a defamatory, false statement about someone. Publicly accusing someone of a crime s/he didn't commit -- and which you know s/he didn't commit -- is a perfect example.
Now, it's possible Valenti is so deluded, so caught up in his own little world, that he is not aware that his victims committed no crime. This would mean that he didn't knowingly make false statements -- because he didn't believe them to be false. But for some reason I doubt this. I suspect he's not an ignorant demagogue, but rather a cynical one: one who knows that he's lying for his own advantage, and revels in it.
(Standard disclaimer: I am not a lawyer.)
Now I've got two solid reasons to think about FreeBSD as an alternative to Debian on new systems here.
Anyone in touch with linux-kernel know if there's been any interest expressed there in bringing anything like these two to Linux?
- Linux 2.1.* Security Hole (February 9, affecting the development-series kernels at the time)
- Bind security problem (April 16, affecting all known distros at the time)
- New Linux Security Holes (January 15, affecting the stable-series kernels at the time)
I picked these up by doing a search for "Linux security" using the search widget on the bottom of the Slashdot main page. These are just off the first page of results. Doubtless there are several stories about security problems in daemons which weren't turned up by this search (because they didn't contain the string "Linux").In other words, security holes in Linux (and other free software) are reported on Slashdot. Your statement appears to be a misleading one intended to incite others to fear, be uncertain about, or doubt the honesty of the Slashdot editors. Isn't that what FUD is all about?
Further, keep in mind that while Microsoft thinks itself to be hurt by the reporting of security holes in its products, Linux is not hurt by the reporting of security holes in Linux-related software. Bug-reporting is a threat to the proprietary-software model, but it is an element of the success of the free-software model.