AST: Who funds it? KB: We have multiple funding sources AST: Is SCO one of them? Is this about the SCO lawsuit? KB: We have multiple funding sources AST: Is Microsoft one of them? KB: We have multiple funding sources
If only Tanenbaum had been a little more clever, there could have been these two extra lines:
There are very powerful forces that are threatened by the development of Linux, and they will fight to the death. Hired character assassins are just the beginning.
Yes, but these hired assassins are more like the Saturday-Night-Live ninjas. If SCO's non-case and Ken Brown's bizarre nonsense are the best they can muster, then they don't have much hope. Their ridiculous opening attacks will tarnish any better-thought-out attacks that might follow. People will become accustomed to the pattern of some Microsoft shill ranting off about theft, communism, and ever-more-insane notions of derivative works and other legal theories followed by an implosion of the accusor when put-up-or-shut-up time finally comes. Microsoft would be better served by cutting off these idiots before Linux's legitimacy prevails in court or mass consciousness.
Who's the greater fool, KB with his million dollars in book revenue or the people who laugh at him on/. all day
I don't think that Ken is going to make very money on his book. I'll bet that it's a net loss (the publishing part, not the shilling part). It's just for vanity purposes. In a few years, I'd bet that a framed share of SCO stock will have more zany-world-that-was collector's value that Brown's book.
Its always been interesting that when somebody (or a group of people) don't want to hear a certain answer, it often goes in one ear and out the other just in time for another "listener" to ask the same basic question phrased slightly differently in hopes of obtaining a reply closer to the desired view.
I really don't understand why he bothered to interview experts. If we assume for one second that Brown isn't a complete idiot, he should have realized that the experts would tell him the truth and then might get a bit uppity when he twisted their words to fit his own agenda. He might also have guessed that they would know how to use that 'inner-net thingy'.
Really, he should have interviewed 'experts' like Rob Enderle or Laura Didio. He wouldn't have had to twist their words and they might have come up with even more creative insults than Brown himself. I guess this a proof by contradiction that Brown is a complete idiot.
I know that XML is "trendy" but it's the wrong tool here - a binary protocol should be used.
If you want a fast XML parser, check out the open-source cwxml library, plug, plug. It is 3x as fast as libxml2 (10x as fast as Xerces) and itt also supports a binary encoding format for XML that is 5-60x faster again and cures the various limitations of WBXML. The W3C is also investigating binary encoding for XML.
In this case, if someone can prove they created a similar combination of program(s) prior to October 17, 1994, that would stand a much better chance of invalidating the patent. A mockup done in 2003 likely will not.
What came first, the problem or the solution?
More and more, it seems that the essential skill in acquiring a patent isn't in the invention because so many patents these day are trivial and should not have been granted on that basis. Instead, it seems that the patent office is rewarding people not for creating a solution to a problem but instead for figuring out a PROBLEM that will occur at some point in the near future.
Configuring a program to launch another program to process files of an identified is obvious thing to do, once you encounter the problem of dealing with zillions of file types on the web. The reward was given for being the first to submit a trivial solution for the problem that hadn't come to fruition yet.
To create an effective base of open-source prior art, we need to be thinking of problems that are just around the corner rather than putting all effort into figuring out better solutions for known problems.
Now can we sue the patent office for not having done its work in the first place, causing all this extra work by other parties?
The answer is reform.
We live in an age of perpetual copyright and overbroad patent protection for trivial ideas. Copyright can be reformed easily. Just make the copyright term 50 years, period. Doesn't matter if it's an individual's or corporations's copyright, it lasts for 50 years from the moment the expression of the idea is fixed in a tangible form.
The USPTO and by infliction and extension most of the patent offices of the world are massively corrupt by virtue of overwhelming incompetence. Perpetual copyrights are bad enough, but bad patents are poisonous to all industries. To avoid copyright problems, you just need to avoid plagiarize the other guy's work, but overbroad and trivial patents are crippling because they are a government-granted monopoly on an idea, and frequently a trivial and obvious idea that many people may have without realizing that it's been patented. It can be very difficult to work around a simple idea.
Fortunately, patent reform is easy. Upon submission and for every alteration, a patent application is published and the public has one year to submit comments and prior art to the patent office. The patent office acts as more of an administrative office, which is what it is improperly and dangerously doing now, rather than a research office. When a patent is granted, it lasts for 10 years from the date of submission. The patent may be extended for another 10 years for a fee of US$1-million, adjusted annually for inflation and different currencies. If the renewal fee is not received by the 10-year date, the invention instantly and irrevocably becomes public domain.
The quality of patents submitted should increase as companies elect submit only patents that they figure are likely to be granted, to avoid giving ideas away that likely aren't good enough for protection.
In my opinion, the real problem with SCO billing people and companies now, before a court has even ruled of the merits of SCO's claims, is that someone probably will pay them.
Hey, a fool and his money. As I see it, this will be the action that lands the SCO executives in jail. Let's hope that they mail pieces of paper through the post office.
First off, SCO would have to pay value added tax (VAT) on every invoice it issued (in the UK this would amount to 17.5%) which is a pretty major disincentive to start sending out invoices you have little chance of collecting.
Does the Canadian GST provide this same disincentive?
Only for a small price, the curriculem must teach C# and the new.NET framework. Thankfully the university did not sell their soul to the devil.
I though that the administrators had merely promised to be much more sneaky about being bought off by Microsoft in the future. Is the MS purchase truly dead or just being snuck in through the back door?
If this were well-supported, as a developer I wouldn't see much reason to use anything else for binary formats.
Exactly! The reason that people have adopted XML format for various purposes is that they got tired of re-inventing the wheel every time they needed a new data format for something. With XML, they get various standard tools to manipulate their format generically and various libraries to access the format within their custom applications. The only problem is that text is bulky and slow to process. But, with the efficiency of binary encoding, there may be no reason left to ever make another ad-hoc data format. The XML structure also provides great flexibility for expanding a format in the future (hence, the 'X'), and the binary format while very efficient also maintains this flexibility.
The ability to translate a structured-binary file into an equivalent structured-text file would be extremely useful.
The library I mentioned includes a program called 'xmlscan' which translates between the formats with various options, such as pretty-formatting text-XML output. The 'xmlscan' program can be used in a similar way to gzip/gunzip to recover a text-XML file, or the library can be used to get direct access to the BXML content without ever translating it into text. The library is in a pre-public-release state until I finialize the initial API.
The library is also 3-5 times faster than other libraries at tokenizing text XML, so other XML parsers could even use it at a low level to speed up their own operations while also getting BXML scanning for free.
They don't mention it in the article, but of course, BXML could be applied to HTML documents, which would greatly speed up transfer and parsing time.
Definitely for HTML versions that are derived from XML. BXML may also be applicable to HTML formats that aren't valid XML, though the 'cwxml' library will of course detect errors in any invalid XML documents.
They include support for adding arbitrary indexes at the end of the file, which if I understand correctly could be used to index image, css, and javascript links, so the browser could submit requests for those files before it's parsed the page.
The indexing mechanism is fairly simple and I have't really pondered how many things it could be applied to.
If W3C is interested in the format, it will likely be undergoing various revisions.
I've already roughed in a framework to support the XL 2003 xml format. However, we've heard the MS Office / XML chant before.
The day that Microsoft makes its file formats truly open will be the day that they apply encryption to all file content so that it can only be decoded or generated by registered software.
It did not see much common usage, the performance hit with non-trivial files was just too large.
They just need to pursue binary encoding of XML. The performance is as good as many binary formats, and much more flexible. W3C is holding a workshop on the subject in a month.
Nope... Nope... the government will step in only after everyone has been fucked and the execs are kicking it in Bermuda on everyone else's retirements.
It's too bad for the SCO employees, but any external investors who invested after it became clear that SCO had no case deserves to loose their money. Anyone who invested before that can sell out now and make a good profit. Anyone who doesn't sell out now deserves to be Darwinized with the other investors. The executives deserve the perp walk, but that can happen after the implosion.
Well then that's excellent. If the upswing is entirely artificial, then it will come crashing back down when the short squeeze is played out. Let's hope for a cascading failure! Monday morning would be an excellent time for some big announcement from IBM, Novell, or class-action Linux copyright holders. It would be an excellent time to short SCO, if there are any shares to short.
Then my grandfather paid for my dad's education, and my dad paid for mine. I've never talked to my dad about the tradition, but when I have kids I definitely want to keep the tradition going.
I went to universities in Canada (New Brunswick, Waterloo) and finished with a BSc(CS), MSc(CS), PhD, and a small profit. I suspect that being smart and taking Computer Science helped.
(There's no levy on students from a different province.)
"Hundreds of customers like and use SCO's Unix products."
This sounds like an outrageous exaggeration. It should say, "Hundreds of customers begrudgingly use SCO's Unix products because they haven't tried Linux yet."
If I copied just the notes in the margins of one of davinci's notebooks into the margins of my copy of 'stranger in a strange land 2: a parody by me' would that be infringement? Is SCO's claim really this weak
You can copy da Vinci as much as you'd like; unfortunately for his heirs, he lived in a time before perpetual copyright (or copyright at all?). I guess he was too busy creating stuff than to have time to grease the politicians.
As for SCO, we can assume that what they showed probably is their strongest claim on direct copyright on Linux, since they would put their strongest simple case forward to show copying. We can also assume that there are only 80 to a few hundred lines of code that they are making these weak claims upon. Since Caldera apparently BSD'd their code just last year, the extent of the copyright infringement, if there is any, is that Caldera isn't mentioned in the copyright notice.
But, the judge in the USL v. BSDI case was quite unimpressed with the importance of comments in the code, realizing that they are not important in a functional sense, and unimpressed that only a small smattering of code was the same in BSD which had I don't know how many hundreds of thousands or millions of lines.
Also, as is discussed elsewhere, the sample that they show appears to have a long and tortured past and has been published in text books many times. And trade secrets contained therein have surely been lost, if not the copyright itself.
SCOs hundreds-of-thousands and millions of lines of code theories are not based on direct copying but are based instead on their theory that they own every line of code ever written by any licensor of the original Unix. This theory is even more wobbly than their claims to be the exclusive owners of the tiny fragments of insignificant code that they claim have been directly copied.
As the GPL requires that if you distribute a modified binary, you distribute the source... the Unix license could require that if you modify the source, you give it back to the owner!
Clearly you need to beware of proprietary licenses. But what does the AT&T Unix license say about exclusivity of the copyright ownership? Does it say that any future code written by any company foolish enough to allow Unix inside their doors is the EXCLUSIVE copyrighted property of AT&T-->SCO? Even the GPL isn't that nasty.
An academic-style license, which is what I assume that AT&T would have been after, would only want to insure that the copy of source code that it has continues to work properly, which would only need a non-exclusive license, which would mean that you must give your derived code back to AT&T, but you can use it for your own purposes as well, such as selling a product.
Please Not I do not believe SCOs claims just showing how they could up the figures without an outright lie.
It seems more to me like they are simply counting every line of code ever contributed by any organization that ever had anything to do with SysV Unix.
Of course, with basic malloc()-type functions, it's also possible that multiple copies spring from some third-party original source, such as a text book entitled "how to write malloc() functions".
Slashdot Mirror
From the k5 article (and Tanebaum's site):
AST: Who funds it?
KB: We have multiple funding sources
AST: Is SCO one of them? Is this about the SCO lawsuit?
KB: We have multiple funding sources
AST: Is Microsoft one of them?
KB: We have multiple funding sources
If only Tanenbaum had been a little more clever, there could have been these two extra lines:
AST: Is K-Mart one of them?
KB: No.
There are very powerful forces that are threatened by the development of Linux, and they will fight to the death. Hired character assassins are just the beginning.
Yes, but these hired assassins are more like the Saturday-Night-Live ninjas. If SCO's non-case and Ken Brown's bizarre nonsense are the best they can muster, then they don't have much hope. Their ridiculous opening attacks will tarnish any better-thought-out attacks that might follow. People will become accustomed to the pattern of some Microsoft shill ranting off about theft, communism, and ever-more-insane notions of derivative works and other legal theories followed by an implosion of the accusor when put-up-or-shut-up time finally comes. Microsoft would be better served by cutting off these idiots before Linux's legitimacy prevails in court or mass consciousness.
Who's the greater fool, KB with his million dollars in book revenue or the people who laugh at him on /. all day
I don't think that Ken is going to make very money on his book. I'll bet that it's a net loss (the publishing part, not the shilling part). It's just for vanity purposes. In a few years, I'd bet that a framed share of SCO stock will have more zany-world-that-was collector's value that Brown's book.
Its always been interesting that when somebody (or a group of people) don't want to hear a certain answer, it often goes in one ear and out the other just in time for another "listener" to ask the same basic question phrased slightly differently in hopes of obtaining a reply closer to the desired view.
I really don't understand why he bothered to interview experts. If we assume for one second that Brown isn't a complete idiot, he should have realized that the experts would tell him the truth and then might get a bit uppity when he twisted their words to fit his own agenda. He might also have guessed that they would know how to use that 'inner-net thingy'.
Really, he should have interviewed 'experts' like Rob Enderle or Laura Didio. He wouldn't have had to twist their words and they might have come up with even more creative insults than Brown himself. I guess this a proof by contradiction that Brown is a complete idiot.
Why didn't someone think of this sooner?
I thought we've had asbestos long johns since the early days of Usenet.
For those of you who have no idea where Orleans is in Ontario
You seem to be assuming that the Merkins would have known what "Ont." means.
"It's free. It works. Doh."
That should be:
"It's free. It works. Duh!" -- Paul Nelson, Riverdale High School (Oregon)
The interjection makes a lot of difference to the meaning.
Google picked up about 700 pages with "that will just be a completely unintentional side effect"...
:-)
Maybe it'll be 7,000 by sundown. I've updated *my* quotes page!
I know that XML is "trendy" but it's the wrong tool here - a binary protocol should be used.
If you want a fast XML parser, check out the open-source cwxml library, plug, plug. It is 3x as fast as libxml2 (10x as fast as Xerces) and itt also supports a binary encoding format for XML that is 5-60x faster again and cures the various limitations of WBXML. The W3C is also investigating binary encoding for XML.
In this case, if someone can prove they created a similar combination of program(s) prior to October 17, 1994, that would stand a much better chance of invalidating the patent. A mockup done in 2003 likely will not.
What came first, the problem or the solution?
More and more, it seems that the essential skill in acquiring a patent isn't in the invention because so many patents these day are trivial and should not have been granted on that basis. Instead, it seems that the patent office is rewarding people not for creating a solution to a problem but instead for figuring out a PROBLEM that will occur at some point in the near future.
Configuring a program to launch another program to process files of an identified is obvious thing to do, once you encounter the problem of dealing with zillions of file types on the web. The reward was given for being the first to submit a trivial solution for the problem that hadn't come to fruition yet.
To create an effective base of open-source prior art, we need to be thinking of problems that are just around the corner rather than putting all effort into figuring out better solutions for known problems.
Now can we sue the patent office for not having done its work in the first place, causing all this extra work by other parties?
The answer is reform.
We live in an age of perpetual copyright and overbroad patent protection for trivial ideas. Copyright can be reformed easily. Just make the copyright term 50 years, period. Doesn't matter if it's an individual's or corporations's copyright, it lasts for 50 years from the moment the expression of the idea is fixed in a tangible form.
The USPTO and by infliction and extension most of the patent offices of the world are massively corrupt by virtue of overwhelming incompetence. Perpetual copyrights are bad enough, but bad patents are poisonous to all industries. To avoid copyright problems, you just need to avoid plagiarize the other guy's work, but overbroad and trivial patents are crippling because they are a government-granted monopoly on an idea, and frequently a trivial and obvious idea that many people may have without realizing that it's been patented. It can be very difficult to work around a simple idea.
Fortunately, patent reform is easy. Upon submission and for every alteration, a patent application is published and the public has one year to submit comments and prior art to the patent office. The patent office acts as more of an administrative office, which is what it is improperly and dangerously doing now, rather than a research office. When a patent is granted, it lasts for 10 years from the date of submission. The patent may be extended for another 10 years for a fee of US$1-million, adjusted annually for inflation and different currencies. If the renewal fee is not received by the 10-year date, the invention instantly and irrevocably becomes public domain.
The quality of patents submitted should increase as companies elect submit only patents that they figure are likely to be granted, to avoid giving ideas away that likely aren't good enough for protection.
In my opinion, the real problem with SCO billing people and companies now, before a court has even ruled of the merits of SCO's claims, is that someone probably will pay them.
Hey, a fool and his money. As I see it, this will be the action that lands the SCO executives in jail. Let's hope that they mail pieces of paper through the post office.
First off, SCO would have to pay value added tax (VAT) on every invoice it issued (in the UK this would amount to 17.5%) which is a pretty major disincentive to start sending out invoices you have little chance of collecting.
Does the Canadian GST provide this same disincentive?
Only for a small price, the curriculem must teach C# and the new .NET framework. Thankfully the university did not sell their soul to the devil.
I though that the administrators had merely promised to be much more sneaky about being bought off by Microsoft in the future. Is the MS purchase truly dead or just being snuck in through the back door?
If this were well-supported, as a developer I wouldn't see much reason to use anything else for binary formats.
Exactly! The reason that people have adopted XML format for various purposes is that they got tired of re-inventing the wheel every time they needed a new data format for something. With XML, they get various standard tools to manipulate their format generically and various libraries to access the format within their custom applications. The only problem is that text is bulky and slow to process. But, with the efficiency of binary encoding, there may be no reason left to ever make another ad-hoc data format. The XML structure also provides great flexibility for expanding a format in the future (hence, the 'X'), and the binary format while very efficient also maintains this flexibility.
The ability to translate a structured-binary file into an equivalent structured-text file would be extremely useful.
The library I mentioned includes a program called 'xmlscan' which translates between the formats with various options, such as pretty-formatting text-XML output. The 'xmlscan' program can be used in a similar way to gzip/gunzip to recover a text-XML file, or the library can be used to get direct access to the BXML content without ever translating it into text. The library is in a pre-public-release state until I finialize the initial API.
The library is also 3-5 times faster than other libraries at tokenizing text XML, so other XML parsers could even use it at a low level to speed up their own operations while also getting BXML scanning for free.
They don't mention it in the article, but of course, BXML could be applied to HTML documents, which would greatly speed up transfer and parsing time.
Definitely for HTML versions that are derived from XML. BXML may also be applicable to HTML formats that aren't valid XML, though the 'cwxml' library will of course detect errors in any invalid XML documents.
They include support for adding arbitrary indexes at the end of the file, which if I understand correctly could be used to index image, css, and javascript links, so the browser could submit requests for those files before it's parsed the page.
The indexing mechanism is fairly simple and I have't really pondered how many things it could be applied to.
If W3C is interested in the format, it will likely be undergoing various revisions.
I've already roughed in a framework to support the XL 2003 xml format. However, we've heard the MS Office / XML chant before.
The day that Microsoft makes its file formats truly open will be the day that they apply encryption to all file content so that it can only be decoded or generated by registered software.
It did not see much common usage, the performance hit with non-trivial files was just too large.
They just need to pursue binary encoding of XML. The performance is as good as many binary formats, and much more flexible. W3C is holding a workshop on the subject in a month.
Nope... Nope... the government will step in only after everyone has been fucked and the execs are kicking it in Bermuda on everyone else's retirements.
It's too bad for the SCO employees, but any external investors who invested after it became clear that SCO had no case deserves to loose their money. Anyone who invested before that can sell out now and make a good profit. Anyone who doesn't sell out now deserves to be Darwinized with the other investors. The executives deserve the perp walk, but that can happen after the implosion.
I believe it was a short squeeze.
Well then that's excellent. If the upswing is entirely artificial, then it will come crashing back down when the short squeeze is played out. Let's hope for a cascading failure! Monday morning would be an excellent time for some big announcement from IBM, Novell, or class-action Linux copyright holders. It would be an excellent time to short SCO, if there are any shares to short.
Then my grandfather paid for my dad's education, and my dad paid for mine. I've never talked to my dad about the tradition, but when I have kids I definitely want to keep the tradition going.
Pay It Forward.
Loans.
Lots of loans.
I went to universities in Canada (New Brunswick, Waterloo) and finished with a BSc(CS), MSc(CS), PhD, and a small profit. I suspect that being smart and taking Computer Science helped.
(There's no levy on students from a different province.)
"Hundreds of customers like and use SCO's Unix products."
This sounds like an outrageous exaggeration. It should say, "Hundreds of customers begrudgingly use SCO's Unix products because they haven't tried Linux yet."
Note: * 6 and 7 are just for a bit of karma whoring :)
1. Confess to being a karma whore.
2. ???
3. Karma!
If I copied just the notes in the margins of one of davinci's notebooks into the margins of my copy of 'stranger in a strange land 2: a parody by me' would that be infringement? Is SCO's claim really this weak
You can copy da Vinci as much as you'd like; unfortunately for his heirs, he lived in a time before perpetual copyright (or copyright at all?). I guess he was too busy creating stuff than to have time to grease the politicians.
As for SCO, we can assume that what they showed probably is their strongest claim on direct copyright on Linux, since they would put their strongest simple case forward to show copying. We can also assume that there are only 80 to a few hundred lines of code that they are making these weak claims upon. Since Caldera apparently BSD'd their code just last year, the extent of the copyright infringement, if there is any, is that Caldera isn't mentioned in the copyright notice.
But, the judge in the USL v. BSDI case was quite unimpressed with the importance of comments in the code, realizing that they are not important in a functional sense, and unimpressed that only a small smattering of code was the same in BSD which had I don't know how many hundreds of thousands or millions of lines.
Also, as is discussed elsewhere, the sample that they show appears to have a long and tortured past and has been published in text books many times. And trade secrets contained therein have surely been lost, if not the copyright itself.
SCOs hundreds-of-thousands and millions of lines of code theories are not based on direct copying but are based instead on their theory that they own every line of code ever written by any licensor of the original Unix. This theory is even more wobbly than their claims to be the exclusive owners of the tiny fragments of insignificant code that they claim have been directly copied.
As the GPL requires that if you distribute a modified binary, you distribute the source ... the Unix license could require that if you modify the source, you give it back to the owner!
Clearly you need to beware of proprietary licenses. But what does the AT&T Unix license say about exclusivity of the copyright ownership? Does it say that any future code written by any company foolish enough to allow Unix inside their doors is the EXCLUSIVE copyrighted property of AT&T-->SCO? Even the GPL isn't that nasty.
An academic-style license, which is what I assume that AT&T would have been after, would only want to insure that the copy of source code that it has continues to work properly, which would only need a non-exclusive license, which would mean that you must give your derived code back to AT&T, but you can use it for your own purposes as well, such as selling a product.
Please Not I do not believe SCOs claims just showing how they could up the figures without an outright lie.
It seems more to me like they are simply counting every line of code ever contributed by any organization that ever had anything to do with SysV Unix.
Of course, with basic malloc()-type functions, it's also possible that multiple copies spring from some third-party original source, such as a text book entitled "how to write malloc() functions".