Brunner usually gets credited by the literati as the first cyberpunk author (if a movement by that name can be said to have existed) with the book Shockwave Rider in, I think '76. He has a proto-hacker character who reprograms the global net from a minitel-like access device. He also writes a program reminiscent of the Internet worm.
I have no intention of copying their algorithm - it's interesting but I can see several ways to improve its reponses (as do they - they're not doing exactly what they claim, or at least I can't see how they could be getting the results they do if they haven't done quite a bit of tinkering).
The problem is that I can't even try to improve on their work without fear of infringing their patent. That is how it can kill research. Far more work goes into incrementally improving existing techniques than can ever go into developing new ones.
Obviously, the patent system was intended to provide some profit to innovators, in return for not keeping secrets. It doesn't do that anymore, it has become a way for innovators to protect ownership, instead of simply giving them a chance at some income to encourage innovation.
Alas, I haven't a better system to offer either, although I suspect software would still be profitable even if no algorithmic patents were allowed.
It might make sense to have two kinds of karma: writer's karma and knowledge karma. Knowledgable individuals would be able to review and revise content, but only those with good writer karma get to compose entries. (I'll have to think about the implications of that, this is sort of off the top of my head.) In my experience (I used to write dictionaries before I discovered how little work programming is, a not too dissimilar buisness) the topic experts were usually not the best writers.
Traditionally, material in encyclopedias must come from other printed sources and the encyclopedia editor has to have a long citation list, even if that list isn't printed. That model should remain in place. Even if an expert says something is so-and-so, that doesn't mean it's the accepted wisdom. Encyclopedias are generally about accepted wisdom, not the latest thing.
There still has to be some core group responsible for final editing and meeting deadlines.
This couldn't be like OSS software projects. Some kind of staff has to have a final, near dictatorial say on what goes in and what doesn't. It couldn't work like Everything does. There would have to be standards for citation and academic credibility. It would, in the end, be much like a regular encyclopedia's editing staff, except that articles would be submitted and rewritten by unpaid.
In the end, it might have to be a bit like Mozilla. A paid staff somewhere would have to exist to do revision and editing.
Britannica did try to get into the web directory buisness a while back. They had 1 paragraph reviews of all the listed websites. I'm looking for a URL, since I can't get to britannica.com either. It couldn't really compete with ODP in terms of number of links, but it might have worked in combination with ODP data.
Gutenberg only handles out of copyright and donated material. They pose little threat to the publishing industry.
I haven't seen much come out of the Open Content people. Encyclopedias require a lot of top down organisation - it might work as an open source project, but only one with a full-time editorial staff, at least a small one, that has final editorial say. A little like Mozilla, execpt without the same debugging problems.
Wow! I just submitted the same story. Looks like I was a little late.
I'm very happy to see EB free on the web. Charging people monthly to use an encyclopedia never struck me as a very viable strategy. The fear I have though is that this bodes badly for those who hope to sell information over the web. If EB can't break even doing it, who can?
Encyclopedias are very expensive to maintain. I'm not at all sure an advertising model can pay for adequate work at all. According to the AP wire article, it looks like Microsoft is the main culprit here - they're very nearly giving away Encarta.
Now, Encarta is crap. Before MS, it was the grocery store quality Funk and Wagnall's Encyclopedia. I would hate to see MS take over yet another industry segment.
The death of Britannica would be a travesty. I grew up with one, and I suspect a lot of other/.'ers did. For it to die at Microsoft's hand would be a crime against nature.
Perhaps we need an Open Source general encyclopedia to keep the Redmond minions out of this business?
The article says he asked the FEC for an advisory ruling. The FEC doesn't seem to have known anything about him before that. Also, there's no mention of the FEC trying to enforce any kind of ruling. I suspect they would just as soon not waste time on him.
I too doubt this would stand up to a first amendment challenge. The Supreme Court decided years ago that TV and radio were special exceptions to the first amendment. No such case has been made on the web yet. Second, I can see grounds for challenging the FEC's calculation of a website's costs - $1000 is a lot more than posting political commentary is really worth. If the guy is using his machines from some other purpose on the web, I doubt his real costs for his political sermonising are $1000.
I'm with the ACLU though - public election funding would probably be the best solution. It seems to work well everywhere else.
The patent hasn't been published anywhere where I can find it. It wouldn't surprise me if MS tried to patent any use of the algoritm in decision support. The algorithm has been in use in decision support for years. If MS can keep a patent just for help systems, I can live with that.
It looks like they're trying to patent any use by a computer of methods for generating a belief netowrk or minimising the number of variables in it. (Although the patent is so vague I can't be sure - it's just a bunch of flowcharts. Most of the descriptions apply to standard methods of doing Bayesian learning.)
I'm not sure, but I think this means my CS homework last year was in violation of IBM's patent.
Here, any work you do that uses company resources can be claimed by that company. If you make something outside of your working hours on your own equipment it is yours, no matter what your contract says. That was settled years ago.
Non-competition clauses have no legal force in the state of California. You can quit and go into buisness competing with your former employer anytime you choose. You can also go to work for their clients or their competitors.
Many employers in California still include that kind of language in contracts in hopes of scaring you, but they are invalid.
In other states, most non-competition clauses are illegal, but circumstances vary from state to state. If your investions are your own and were not produced using company resources, I believe they are yours no matter where you live, but I'm not a lawyer - I just know my rights in California.
My interest is specifically in the patentability of software algorithms. There are two specific cases I have in mind: Google's PageRank algorithm, and Microsoft's patent on Bayesian networks.
Google's algorithm is to some extent described in on their homepage. (At least the core of it is - it doesn't actually respond exactly the way it should, but close enough in most cases.) It uses several well known algorithms to rank web pages. Can such a patent actually cover the use of a common algorithm like theirs to rank web pages? Am I thus forbidden to use link networks at all to rank web pages?
Secondly, Microsoft, a number of years ago, apparently independently rediscovered a mathematical construct called a Bayesian network, which now forms the heart of Microsoft's help system. As I understand it, they have patented this usage. Bayesian networks were first described 200 years ago, but can a patent applying them to computer support (or perhaps to other applications - I have not been able to obtain a copy of MS' patent) be valid? Or does the claim of pre-existing art invalidate it?
I'm asking for some informed opinions. In case it's an issue, let me state for the record that I understand no one on/. is giving legal advise - what I do is my own problem.
Would have worked better as a graphic novel
on
Snow Crash
·
· Score: 1
According to the intro to my copy, Snow Crash was originally intended as a graphic novel. In many ways it shows. The action is kind of jerky, we jump from one thing to the next without really developing the story. The imagery is great, but the characterisations leave a lot to be desired. That kind of glossing is necessary in a comic book, but in a novel it doesn't really work.
The political subtext is intruiging too - anti-free trade, anti-corporate, more than a little anti-government and certainly disenchanted with libertarian utopianism. Having lived in a few strip mall towns, certainly I can see the apocalyptic vision of a franchised world all too easily. Stephenson gets full credit for making that vision work without turning it into some "woe is me" end of civilisation sort of story.
Snow Crash treats computing and virtual reality a little more realistically than most, but that too is to be expected nowadays. The linguistics in Snow Crash is pretty bad. My advice to the author: read less Julian Jaynes, more William Calvin. I have yet to see any science fiction author do a really good job of linguistics, so I guess I shouldn't be surprised.
But, in the end, Snow Crash lacks any consistent statement. It's a fun book, but no more than that. I had hopes that the Diamond Age or Cryptonomicon would be more substantial, but DIamond Age is poorly plotted, and Cryptonomicon just stank. No, his best novel remains, IMHO unfortunately, Zodiac.
Does anyone know about US release dates for 23? I only know of it through a friend who saw it in Prague at the Karlovy Vary festival. It certainly isn't available in video yet.
As I understand it, it's the other side of "Cuckoo's Egg" - the story told from the German hacker's point of view. Can anyone confirm that?
I doubt there will ever be a plug-and-play open source e-commerce app. The non-open source turnkey solutions that I've seen either aren't very flexible or aren't very turnkey.
There are lots of tools to implement a web site, both open and closed: Apache, PHP, serverlets, Cold Fusion, whatever you need. There are lots of ways to implement a shopping cart, with or without cookies - you don't need a solution, just a bit of web savvy.
The back end: credit card validation, payment, merchant accounting - these things can never and will never have an open source solution. To do any of these things is not a matter of code, it's a matter of dealing with a bank or credit card organisation.
CyberCash and Bell Emergis (just to name two) are offering for a fee just such backend access. It will probably always be that way. If you have an account or access to someone like that, you can do all the rest using open source solutions.
A fast growing industry - perhaps the next big winner in the so-called information economy - is what is euphemistically called "intelligence services." Many of these are older research firms or consultancies trying to use the web as a new income source. Jane's is one example. Another is Oxford Analytica. To some extent, formerly public media are entering the game, groups like The Economist. And of course, Stratfor made quite a name for itself during the Kosovo War.
This means each of these companies is looking for some new angle that they can impress their customers with. Afterall, they are just repackaging public information. They have no spies, and only rarely get tips from people in the know. It is unsurprising that Jane's might use a forum like/. to get relatively current info. I have my suspicions that others have done the same thing, if not here than elsewhere, without admitting to it. And certainly, all these "intelligence services" know how to use Inktomi, Altavista, and other search services to find relevant material wherever, including/.
What they need to watch out for is that the information on/. does not come with a built-in credibility meter. It is just as possible to pass complete BS off here as it is anywhere else. There's no substitute for actually knowing what your talking about.
I live in Fremont, CA where I have a cable modem (for $50/mo) and I'm moving to Mtn View home of Netscape and the new Microsoft campus) where the best I can do is DSL, with $400 in installation and hardware, and $79/mo for half the speed of my cable modem. It's bizarre that DSL would cost so much in one of the wealthiest parts of Silicon Valley, and AT+T/TCI still haven't got cable modems out there.
In contrast my mother, who lives in Winnipeg (Canada's answer to Cleveland) can get a cable modem for CAN$39.95 or DSL from Manitoba Telephone/Bell Canada for CAN$49.95.
All major urban areas in Canada now have DSL and most have cable modems. A basic unlimited dialup account runs as low as CAN$9/mo., and rarely more than CAN$20 even in isolated rural areas or the north. ISP access in Iqaluit costs less than in Indiana.
I have hopes that this might spur some serious growth in Canadian tech buisness - I'll take Montreal over Palo Alto anyday. But, as I understand it, Finland and Iceland are still the ranking champions for 'Net access.
Social democracy triumphs again. Smoke that, Mike Harris!:^)
The overwhelming majority of the billionaries on Forbes list inherited, if not billions, at least millions. Even the poorest among them started out at least middle-class.
Bill Gates was the child of a lawyer and a Seattle socialite. He was, beyond a doubt, a child of privilege from day one. True, not $100 billion in privilege, but I doubt he's ever missed a meal in his life. Oh he may have worked a bit to get where he is, but I seriously doubt he worked any harder than most single moms or even the average auto mechanic.
1- I'll gladly trade my financial problems for running out of things to buy. After all, there's always next year's top of the line SGI workstation...
2- You may not be able to take money with you, but ironically, you can take your debts with you. When my father died, the hospital threatened to sue repeatedly for unpaid bills. Ultimately, my mother left the USA, in part to escape them.
3- It may not buy love, but it sure as hell improves your bargaining position.
4- Wealth may not make you happy, but I'll take bored angst over wondering where my next meal will come from any day.
No, I sort of speak German (although not very colloquially), and "floating" seemed better than "adrift" (which I think is the Duden definition.) "Betreibend" was the first word that came to mind, but I was pretty sure that wasn't right.
Next time I get to Berlin I'll ask somebody - thanks.
It may not be possible to restricy gambling on the 'Net with laws. Canada is trying to and failing miserably. But, using the credit card companies to limit access is really a very creative solution. After all, Visa and Mastercad have the legal power to limit the activities of their merchants. Getting a merchant account means signing a contract, which can perfecty easily forbid the use of the card for gambling. It's really quite inspired.
BTW, I do code a lot to the Strange Days soundtrack, esp. the Deep Forest songs. That film did have an AMAZING soundtrack. I'm beginning to develop a taste for Japanese pop, I'll have to check out Yellow Monkay, thanks.
Slashdot Mirror
Brunner usually gets credited by the literati as the first cyberpunk author (if a movement by that name can be said to have existed) with the book Shockwave Rider in, I think '76. He has a proto-hacker character who reprograms the global net from a minitel-like access device. He also writes a program reminiscent of the Internet worm.
I have no intention of copying their algorithm - it's interesting but I can see several ways to improve its reponses (as do they - they're not doing exactly what they claim, or at least I can't see how they could be getting the results they do if they haven't done quite a bit of tinkering).
The problem is that I can't even try to improve on their work without fear of infringing their patent. That is how it can kill research. Far more work goes into incrementally improving existing techniques than can ever go into developing new ones.
Obviously, the patent system was intended to provide some profit to innovators, in return for not keeping secrets. It doesn't do that anymore, it has become a way for innovators to protect ownership, instead of simply giving them a chance at some income to encourage innovation.
Alas, I haven't a better system to offer either, although I suspect software would still be profitable even if no algorithmic patents were allowed.
You mean like Linus?
Point well made.
I'd volunteer, but it's not a part time job, and I kinda need my paycheck.
It might make sense to have two kinds of karma: writer's karma and knowledge karma. Knowledgable individuals would be able to review and revise content, but only those with good writer karma get to compose entries. (I'll have to think about the implications of that, this is sort of off the top of my head.) In my experience (I used to write dictionaries before I discovered how little work programming is, a not too dissimilar buisness) the topic experts were usually not the best writers.
Traditionally, material in encyclopedias must come from other printed sources and the encyclopedia editor has to have a long citation list, even if that list isn't printed. That model should remain in place. Even if an expert says something is so-and-so, that doesn't mean it's the accepted wisdom. Encyclopedias are generally about accepted wisdom, not the latest thing.
There still has to be some core group responsible for final editing and meeting deadlines.
This couldn't be like OSS software projects. Some kind of staff has to have a final, near dictatorial say on what goes in and what doesn't. It couldn't work like Everything does. There would have to be standards for citation and academic credibility. It would, in the end, be much like a regular encyclopedia's editing staff, except that articles would be submitted and rewritten by unpaid.
In the end, it might have to be a bit like Mozilla. A paid staff somewhere would have to exist to do revision and editing.
Britannica did try to get into the web directory buisness a while back. They had 1 paragraph reviews of all the listed websites. I'm looking for a URL, since I can't get to britannica.com either. It couldn't really compete with ODP in terms of number of links, but it might have worked in combination with ODP data.
Gutenberg only handles out of copyright and donated material. They pose little threat to the publishing industry.
I haven't seen much come out of the Open Content people. Encyclopedias require a lot of top down organisation - it might work as an open source project, but only one with a full-time editorial staff, at least a small one, that has final editorial say. A little like Mozilla, execpt without the same debugging problems.
Wow! I just submitted the same story. Looks like I was a little late.
/.'ers did. For it to die at Microsoft's hand would be a crime against nature.
I'm very happy to see EB free on the web. Charging people monthly to use an encyclopedia never struck me as a very viable strategy. The fear I have though is that this bodes badly for those who hope to sell information over the web. If EB can't break even doing it, who can?
Encyclopedias are very expensive to maintain. I'm not at all sure an advertising model can pay for adequate work at all. According to the AP wire article, it looks like Microsoft is the main culprit here - they're very nearly giving away Encarta.
Now, Encarta is crap. Before MS, it was the grocery store quality Funk and Wagnall's Encyclopedia. I would hate to see MS take over yet another industry segment.
The death of Britannica would be a travesty. I grew up with one, and I suspect a lot of other
Perhaps we need an Open Source general encyclopedia to keep the Redmond minions out of this business?
The article says he asked the FEC for an advisory ruling. The FEC doesn't seem to have known anything about him before that. Also, there's no mention of the FEC trying to enforce any kind of ruling. I suspect they would just as soon not waste time on him.
I too doubt this would stand up to a first amendment challenge. The Supreme Court decided years ago that TV and radio were special exceptions to the first amendment. No such case has been made on the web yet. Second, I can see grounds for challenging the FEC's calculation of a website's costs - $1000 is a lot more than posting political commentary is really worth. If the guy is using his machines from some other purpose on the web, I doubt his real costs for his political sermonising are $1000.
I'm with the ACLU though - public election funding would probably be the best solution. It seems to work well everywhere else.
The patent hasn't been published anywhere where I can find it. It wouldn't surprise me if MS tried to patent any use of the algoritm in decision support. The algorithm has been in use in decision support for years. If MS can keep a patent just for help systems, I can live with that.
Thanks, that's what I was afraid of. I guess I have to find a different algorithm to do that kind thing.
I remember the good old days, when if something appeared in academic literature, you were safe using it. This is really going to kill research.
I looked at your example.
It looks like they're trying to patent any use by a computer of methods for generating a belief netowrk or minimising the number of variables in it. (Although the patent is so vague I can't be sure - it's just a bunch of flowcharts. Most of the descriptions apply to standard methods of doing Bayesian learning.)
I'm not sure, but I think this means my CS homework last year was in violation of IBM's patent.
That is depressing.
Here, any work you do that uses company resources can be claimed by that company. If you make something outside of your working hours on your own equipment it is yours, no matter what your contract says. That was settled years ago.
Non-competition clauses have no legal force in the state of California. You can quit and go into buisness competing with your former employer anytime you choose. You can also go to work for their clients or their competitors.
Many employers in California still include that kind of language in contracts in hopes of scaring you, but they are invalid.
In other states, most non-competition clauses are illegal, but circumstances vary from state to state. If your investions are your own and were not produced using company resources, I believe they are yours no matter where you live, but I'm not a lawyer - I just know my rights in California.
My interest is specifically in the patentability of software algorithms. There are two specific cases I have in mind: Google's PageRank algorithm, and Microsoft's patent on Bayesian networks.
/. is giving legal advise - what I do is my own problem.
Google's algorithm is to some extent described in on their homepage. (At least the core of it is - it doesn't actually respond exactly the way it should, but close enough in most cases.) It uses several well known algorithms to rank web pages. Can such a patent actually cover the use of a common algorithm like theirs to rank web pages? Am I thus forbidden to use link networks at all to rank web pages?
Secondly, Microsoft, a number of years ago, apparently independently rediscovered a mathematical construct called a Bayesian network, which now forms the heart of Microsoft's help system. As I understand it, they have patented this usage. Bayesian networks were first described 200 years ago, but can a patent applying them to computer support (or perhaps to other applications - I have not been able to obtain a copy of MS' patent) be valid? Or does the claim of pre-existing art invalidate it?
I'm asking for some informed opinions. In case it's an issue, let me state for the record that I understand no one on
According to the intro to my copy, Snow Crash was originally intended as a graphic novel. In many ways it shows. The action is kind of jerky, we jump from one thing to the next without really developing the story. The imagery is great, but the characterisations leave a lot to be desired. That kind of glossing is necessary in a comic book, but in a novel it doesn't really work.
The political subtext is intruiging too - anti-free trade, anti-corporate, more than a little anti-government and certainly disenchanted with libertarian utopianism. Having lived in a few strip mall towns, certainly I can see the apocalyptic vision of a franchised world all too easily. Stephenson gets full credit for making that vision work without turning it into some "woe is me" end of civilisation sort of story.
Snow Crash treats computing and virtual reality a little more realistically than most, but that too is to be expected nowadays. The linguistics in Snow Crash is pretty bad. My advice to the author: read less Julian Jaynes, more William Calvin. I have yet to see any science fiction author do a really good job of linguistics, so I guess I shouldn't be surprised.
But, in the end, Snow Crash lacks any consistent statement. It's a fun book, but no more than that. I had hopes that the Diamond Age or Cryptonomicon would be more substantial, but DIamond Age is poorly plotted, and Cryptonomicon just stank. No, his best novel remains, IMHO unfortunately, Zodiac.
Does anyone know about US release dates for 23? I only know of it through a friend who saw it in Prague at the Karlovy Vary festival. It certainly isn't available in video yet.
As I understand it, it's the other side of "Cuckoo's Egg" - the story told from the German hacker's point of view. Can anyone confirm that?
I doubt there will ever be a plug-and-play open source e-commerce app. The non-open source turnkey solutions that I've seen either aren't very flexible or aren't very turnkey.
There are lots of tools to implement a web site, both open and closed: Apache, PHP, serverlets, Cold Fusion, whatever you need. There are lots of ways to implement a shopping cart, with or without cookies - you don't need a solution, just a bit of web savvy.
The back end: credit card validation, payment, merchant accounting - these things can never and will never have an open source solution. To do any of these things is not a matter of code, it's a matter of dealing with a bank or credit card organisation.
CyberCash and Bell Emergis (just to name two) are offering for a fee just such backend access. It will probably always be that way. If you have an account or access to someone like that, you can do all the rest using open source solutions.
This means each of these companies is looking for some new angle that they can impress their customers with. Afterall, they are just repackaging public information. They have no spies, and only rarely get tips from people in the know. It is unsurprising that Jane's might use a forum like
What they need to watch out for is that the information on
I live in Fremont, CA where I have a cable modem (for $50/mo) and I'm moving to Mtn View home of Netscape and the new Microsoft campus) where the best I can do is DSL, with $400 in installation and hardware, and $79/mo for half the speed of my cable modem. It's bizarre that DSL would cost so much in one of the wealthiest parts of Silicon Valley, and AT+T/TCI still haven't got cable modems out there.
:^)
In contrast my mother, who lives in Winnipeg (Canada's answer to Cleveland) can get a cable modem for CAN$39.95 or DSL from Manitoba Telephone/Bell Canada for CAN$49.95.
All major urban areas in Canada now have DSL and most have cable modems. A basic unlimited dialup account runs as low as CAN$9/mo., and rarely more than CAN$20 even in isolated rural areas or the north. ISP access in Iqaluit costs less than in Indiana.
I have hopes that this might spur some serious growth in Canadian tech buisness - I'll take Montreal over Palo Alto anyday. But, as I understand it, Finland and Iceland are still the ranking champions for 'Net access.
Social democracy triumphs again. Smoke that, Mike Harris!
The overwhelming majority of the billionaries on Forbes list inherited, if not billions, at least millions. Even the poorest among them started out at least middle-class.
Bill Gates was the child of a lawyer and a Seattle socialite. He was, beyond a doubt, a child of privilege from day one. True, not $100 billion in privilege, but I doubt he's ever missed a meal in his life. Oh he may have worked a bit to get where he is, but I seriously doubt he worked any harder than most single moms or even the average auto mechanic.
1- I'll gladly trade my financial problems for running out of things to buy. After all, there's always next year's top of the line SGI workstation...
2- You may not be able to take money with you, but ironically, you can take your debts with you. When my father died, the hospital threatened to sue repeatedly for unpaid bills. Ultimately, my mother left the USA, in part to escape them.
3- It may not buy love, but it sure as hell improves your bargaining position.
4- Wealth may not make you happy, but I'll take bored angst over wondering where my next meal will come from any day.
No, I sort of speak German (although not very colloquially), and "floating" seemed better than "adrift" (which I think is the Duden definition.) "Betreibend" was the first word that came to mind, but I was pretty sure that wasn't right.
Next time I get to Berlin I'll ask somebody - thanks.
It may not be possible to restricy gambling on the 'Net with laws. Canada is trying to and failing miserably. But, using the credit card companies to limit access is really a very creative solution. After all, Visa and Mastercad have the legal power to limit the activities of their merchants. Getting a merchant account means signing a contract, which can perfecty easily forbid the use of the card for gambling. It's really quite inspired.
USGS page
6.6 preliminary mag at 02:46:45 PDT
Looks like it was just south of I-40 near Ludlow, CA.
Location via Mapbast
Is that a colloquialism in German these days?
BTW, I do code a lot to the Strange Days soundtrack, esp. the Deep Forest songs. That film did have an AMAZING soundtrack. I'm beginning to develop a taste for Japanese pop, I'll have to check out Yellow Monkay, thanks.