Of course, there is a lot of overlap between the two concepts - good communication can often be good marketing, and vice versa. Occasionally good marketting is bad communication - e.g. selling a bad product by lying about it, or even bad communication is good marketting - selling a product by failing to explain what it is.
Whatever the terminology though, open source software is foundationally dependent on lots of communication to link developers and projects. If a project does not have a web page and obtain links from the relevent forums, then projects can never get off the ground.
If open source were to take your advice it would cease to exist.
Moving out of the US would in fact get MS off the breakup hook.
The US would then be limited to imposing restrictions on MS products imported to the US. I imagine this is exactly the approach the EU will adopt if they decide that MS has broken EU antitrust laws and that the US remedies do not address the issues.
I guess the US could also impose arbitrary large monetary fines on any wholly owned US subsidiary, forcing MS to either bow to US decisions or spin off the subsidiary as a separate company.
The Xfree-v4 DRI driver from 3dfx still goes through Glide. Check out this page at 3dfx. You need the new Glide installed before you can install the DRI X-server.
The IIRC the 3.3.6 X-driver used DGA (and so was full screen only). As a result it is entirely possible that it should be as fast as the DRI solution, or faster if the DRI implementation is not yet as well optimised.
So I am quite prepared to believe that the DRI implementation could be *slower* than the older version at the moment.
I agree that it is not an OS issue, (assuming the OS does not allow you to modify system or other users files, i.e. Not Win9x). But I think it is a client, rather than user issue.
Code coming from an unverified source (i.e. not from a trusted installer) should not be allowed to run outside a sandbox. It works fine for Java on the web. The same treatment should apply to anything coming in an email.
Furthermore, any file extracted from an email should be marked non-executable. (The user can chmod it - if they know enough to do that, they can probably understand the risk). Archives are a bit more tricky, but changing umask(2) to 666 before invoking an archive program (such as tar) should do the trick I guess.
The Evolution folks are implementing a Visual Basic clone in their new gnome client. But they are doing it properly, using a Java-like security model.
Yet again I am forced to slap my bald pate and say 'Doh! It was so obvious'.
As obvious as advcating free software as a means to prevent descrimination against the third world in the development of computer skills. We were just looking at the issue from too narrow a viewpoint.
Why can third world countries produce world class scientists? Because the information, which is the fundamental resource, is free to varying extents. (Even when you can't afford the journal, you can always send a postcard to the author of a paper asking for a reprint).
If more information is made available, then there is the potential for third world developer to produce world-class innovations in technology, engineering, and all sorts of other fields. In the long term hopefully correcting the wealth imbalence. Everyone benefits.
OK, that is a naive oversimplification, but the principle is there.
Slashdot doesn't reprint articles it doesn't own from other jornals. It doesn't redistribute mp3's of major pop tracks. Slashdot is not a form of civil disobedience - it is a news site.
So it seems perfectly reasonable that Slashdot should obey the law, and take down the text. (The links can stay though).
I'm all for challenging the DMCA when a good oppertunity arrives. But wait for a case where there is some substance to the defence, and a chance of winning. Fighting an unwinnable case against the DMCA and losing just becuase there are aspects of the DMCA we disagree with is stupid.
This is only an issue becuase it's Microsoft. I dislike Microsoft as much as the next geek, but we can beat Microsoft *legally*.
Besides, I think it is very likely Microsoft was deliberately trying to leverage the DeCSS rulings and publish a secret in such a way that any software produced to interoperate can be considered 'tainted' by illegally obtained information. The Samba people have a chance to fight this - their 'interoperability exception' case is stronger than the DeCSS people, because no one can argue that Samba is a piracy tool. Let them handle this one.
Um. I am extremely embarassed. Never before have I been tempted to post a 'me too'. But this comment is both accurate and important enough that I think it deserves one.
OK, I'll compromise and itemise slightly.
'Me too' to everything except the 'I use napster' sentence, but I have played one set of mp3's, and I decided that I didn't want the albumn, so I deleted them.
Since a good deal of KDE development takes place in Europe, there are rather less barriers to including encryption than there are in more US-centric projects like Mozilla. As a result, the encryption is already there.
Of course recent US legislation has largely removed those barriers.
The KDE html widget was started well before Mozilla was born as an open source project. Gnome took the same approach: have you ever noticed that the Gnome help browser has its own html rendering engine?
When both of these were started, there was no lightweight, modern, open source rendering widget available for the approrpiate language and toolkit, and therefore both projects developed one.
KDE had the head start, and thus their widget was getting pretty capable by the time Moz was OSed. Gnome were a little behind, and therefore it was worth them switching to Mozilla (And recently this work has been taken up by Eazel for Nautilus).
The result: both Konqueror and Nautilus look as though they will be pretty capable. If they maintain some sort of parity, I guess who uses which will depend on which desktop they use (and therefore to some extent which Linux distro).
The competition will encourage developers to provide good standards support, and to fix any deficiencies in their project. Neither side will be able to argue that 'You need 64Mb for a modern web browser' if the competitor runs in 32Mb.
So I am convinced that Konqueror is a good thing, even though I expect to be browsing on Nautilus a year from now.
Perhaps the Gnome team can eventually take the Mozilla engine and build a decent browser for unix with an interface that works, but they are showing little interest in doing that. Why they haven't does puzzle me a lot.
Basically, Nautilus (the Eazle file manager) will now embed the Bonobo mozilla component. In one step Nautilus has leaprogged Konquerer and Opera in standards-compliant web browsing, and provided a native GTK-zilla without any of that skin bloat.
Of course we'll have to wait a few months for a Nautilus beta. But I guess it'll take that long to fix the memory leaks in Mozilla.
I thought I must be the only Linux user who hated skinable apps!
I use Linux all day every day. And I want my display to be as clean and as clear as possible. I want my widgets instantly recognizable, equal in size, and consistently laid out.
Don't get me wrong, I am quite happy to change my GTK theme or Enligtenment theme occasionally, as long as everything remains consistent (and clear - why are most of the Enlightenment themes unusably dark? Too much late night hacking? The Sawmill selection seems better). But I usually end up back at Default, because Netscape 4.x and Xemacs don't use GTK widgets, and Default looks closest to their native appearance.
Gecko I love. But Mozilla and Xmms insist on making their own separate (if mutable) identity and sacrifice usability by smothering themselves in decorations which don't reflect the rest of the UI.
Xemacs has the right idea - you customize not the appearance, but the functionality. I want to add buttons to mozilla which turn off images, or disable cookies, I don't want to change what the buttons look like.
Fortunately they are open source. When someone writes a conventional GTK frontend to gecko, I will be first in the queue to beta test, or maybe even code if some of my other projects let up.
I suppose that the Slashdot readers will know assume that this gives cphack users free reign to copy and use this illegal program.
If the program is illegal, then the GPL does not give anyone any right to distribute it.
However the question is, is the program illegal? This has yet to be demonstrated. The illegal assignment of rights to Mattel which had already been assigned elsewhere does nothing to make the program illegal. The legality of the program is unchanged from before the settlement.
Whether the program is legal or not depends on whether the reverse engineering is considered to be legitimate under any of the reverse-engineering clauses of relevent copyright laws in the relevent countries. This has yet to be proven one way or the other.
What I haven't seen is any laws which might make the essay, which is far more damaging than the program, illegal.
Suppose, now, that the cphack authors had included an extra clause in the license. And suppose this extra clause said that no matter what, they couldn't be held liable for writing it. Do you think such a thing would stand up in court? Sorry, but if you break the law, not even a fine print disclaimer will save you.
Although if UCITA is passed, this sort of ridiculous license cluase may well become legal in some cases.
The agreement also states that Jansson and Skala attest they "are the sole proprietors of all rights" involved with cphack and have "not assigned" them to anyone else.
If it was GPL'd, then they had already assigned the rights to copy, modify and redistribute to other people, and so could not legally sign this agreement.
If you GPL something, you can still sell the code to someone under a non-exclusive license, but you cannot transfer exclusive ownership in the way the settlement appears to have done.
I suspect this was done unintentionally however - presented with a settlement document full of legalese, it wouldn't have occured to me to look for this problem. Neither is it likely to occur to most mainstream lawyers who might not be familiar with the implications of the GPL.
The lesson for the rest of us is, make sure you have the right to distribute your code before you GPL it, and if someone does prosecute, make sure they know it has already been licensed to others.
It seems to me that capitalism is predicated on a free market where prices are dictated by supply and demand, and that this mechanism of price control breaks down in the presence of a monoploy. Thus anti-trust legislation must a pivotal aspect of a capitalist economy. Anti-trust legislation would be redundant in a centralized socialist economy because prices would be determined centrally. Therefore the US use of anti-trust legislation suggest to me that the US is a capitalist, rather than a socialist economy.
I don't see how being offshore would help either. If the US, EU or Japan can't prosecute the company, they can impose punitive duties on its products. Since all three have at least examined anti-trust charges, I don't see any of them kicking up a major stink with the WTO.
Were the Linux cluster users using gcc/g77? It is well known that (at least for most scientific codes) you can get 50-100% speedup by switching from the GNU compilers to commercial ones from Portland or elsewhere.
If there is still a difference, then the next thing to try is the latest dev kernels, which have better SMP (if SMP nodes are used), and significantly faster disk io through the elimination of double caching.
Since most scientific apps should spend most of their eating user CPU cycles, I wouldn't expect there to be very much difference between one OS and another, however node uptime and more established remote admin are points in Linux' favour for big clusters.
I used KDE as my promary environment during early 1999, becuase it gave very convenient access to applications, although WM would have done a very similar job.
Recently I have switched to Gnome for both my home and work machines, mainly for the ability to swallow a graphical pager and task list on the panel. I didn't have the real-estate for KDE's separate task list. A pop-up task list on the desktop is not bad, but I don't find it quite as convenient.
Of course give me a machine with only 16 or 32M RAM and I'll be back to WM straight away. Both the others really need 48, or the paging gets annoying.
I've used SGI's for a number of years, including recently Power Challenges, Octanes, and such like.
Yes, the graphics are good. But the processing stinks. On most of the codes I run (mix of 32-bit fp and integer array index, FFTs and such like), a Pentium II/300 will outperform an R10000 with heaps of cache and memory bandwidth.
The ccNUMA + crossbar stuff is great, but needs a faster processor to go with it.
Still, they are doing great work with Linux. For their software work I hope they do well.
Because MS have just been tub-thumping about how Win2000 includes streaming servers built in with no further licenses. MS could make Linux a second-class web browsing platform if they could capture the streaming server market with a proprietary format for which no Linux player exists. This move is a step towards preventing that.
Hopefully RedHat will get the source code and ensure that the new RealPlayer is actually stable and useable under Linux.
I don't know about Transmeta's mobile-linux, but I can see one significant advantage of using and embedded Linux (or even an embedded DOS): Microwindows.
It is a compact (100Kb) embeddable GUI which implements the bulk of both the Win32 and X11 APIs. That gives you a very simple port of any software from two environments. Why use WinCE, when you can recompile your apps and avoid the royalities?
I presume TM's Mobile Linux is using a cut-down Xfree86. It would be interesting to see how much of the 32Mb ROM they could free up by using a purpose-built embedded GUI. (I think Xfree86 is great, I'm just not convinced it is the optimal choice for embedded applications).
On the whole I agree with you, but the timing of your first point is deliciously ironic.
If you surf on over to the Linux Game Tome, you will find that only today a patch was released for e2fsck (the thing that checks your disk for you if you don't shutdown the machine properly), which allows you to play Tetris while the file system is checked. Of course, because it is open source, your second point kicks in: The feature is optional.
I enjoyed the article. It sounds as though the programmers aren't too different to their open source counterparts (but richer). Presumably all the dodgy tactics come out of the marketting department. I guess one of the strengths of open-source is that we have no marketting department!
Re:Response to Comments
on
A New DeCSS
·
· Score: 2
After I stopped laughing I tried to work out whether DeCSS was a good idea or not. I didn't come to any firm conclusion. So I linked it because it was funny. Good work!
Yes, clearly you are right that there is an implicit license rather than no cpoyright. But that makes the rest of my questions *more* rather than less relevent, because the implicit license, even if it contains an implicit no-warrenty, will certainly be overridden by the UCITA warrenty provision.
The only way I can see of getting out of this is to claim that if the program causes loss of data, the JVM is to blame. But that could only work for content which is supposed to work in a sandbox environment. Plugins and ActiveX are more problematic though.
How does UCITA affect active components in web pages?
Consider, a Java or JavaScript program is downloaded to your computer, may be cached several times on your hard-drive, is loaded and executed. Furthermore, it may be stored and redistributed by proxies and web caches.
So presumably such programs are considered public domain. Does UCITA create license conditions for public domain software? Does it force a warrenty on such software? If so, then you could be liable for a bug in your java applet, unless you want to claim that the fault was with the VM security.
Plugins and ActiveX are even more problematic. I believe IE will download these automatically (but I haven't tried this), in which cases there is no opertunity to agree to a license.
What about interpreted languages? Who is liable, the author of the interpreter or the author of the interpreted code?
What about programs which use libraries?
Despite not being a US citizen (and only rarely paying US taxes) I have relicensed all my scientific software so that its use is illegal if the `No Warrenty' clause is invalidated by local legislation. I've contacted MA and VA officials with this information.
Slashdot Mirror
See the announcement here, and the meeting publicity material on upnp.org.
Not so much marketting as communication.
Of course, there is a lot of overlap between the two concepts - good communication can often be good marketing, and vice versa. Occasionally good marketting is bad communication - e.g. selling a bad product by lying about it, or even bad communication is good marketting - selling a product by failing to explain what it is.
Whatever the terminology though, open source software is foundationally dependent on lots of communication to link developers and projects. If a project does not have a web page and obtain links from the relevent forums, then projects can never get off the ground.
If open source were to take your advice it would cease to exist.
Moving out of the US would in fact get MS off the breakup hook.
The US would then be limited to imposing restrictions on MS products imported to the US. I imagine this is exactly the approach the EU will adopt if they decide that MS has broken EU antitrust laws and that the US remedies do not address the issues.
I guess the US could also impose arbitrary large monetary fines on any wholly owned US subsidiary, forcing MS to either bow to US decisions or spin off the subsidiary as a separate company.
The Xfree-v4 DRI driver from 3dfx still goes through Glide. Check out this page at 3dfx. You need the new Glide installed before you can install the DRI X-server.
The IIRC the 3.3.6 X-driver used DGA (and so was full screen only). As a result it is entirely possible that it should be as fast as the DRI solution, or faster if the DRI implementation is not yet as well optimised.
So I am quite prepared to believe that the DRI implementation could be *slower* than the older version at the moment.
I agree that it is not an OS issue, (assuming the OS does not allow you to modify system or other users files, i.e. Not Win9x). But I think it is a client, rather than user issue.
Code coming from an unverified source (i.e. not from a trusted installer) should not be allowed to run outside a sandbox. It works fine for Java on the web. The same treatment should apply to anything coming in an email.
Furthermore, any file extracted from an email should be marked non-executable. (The user can chmod it - if they know enough to do that, they can probably understand the risk). Archives are a bit more tricky, but changing umask(2) to 666 before invoking an archive program (such as tar) should do the trick I guess.
The Evolution folks are implementing a Visual Basic clone in their new gnome client. But they are doing it properly, using a Java-like security model.
Yet again I am forced to slap my bald pate and say 'Doh! It was so obvious'.
As obvious as advcating free software as a means to prevent descrimination against the third world in the development of computer skills. We were just looking at the issue from too narrow a viewpoint.
Why can third world countries produce world class scientists? Because the information, which is the fundamental resource, is free to varying extents. (Even when you can't afford the journal, you can always send a postcard to the author of a paper asking for a reprint).
If more information is made available, then there is the potential for third world developer to produce world-class innovations in technology, engineering, and all sorts of other fields. In the long term hopefully correcting the wealth imbalence. Everyone benefits.
OK, that is a naive oversimplification, but the principle is there.
Slashdot doesn't reprint articles it doesn't own from other jornals. It doesn't redistribute mp3's of major pop tracks. Slashdot is not a form of civil disobedience - it is a news site.
So it seems perfectly reasonable that Slashdot should obey the law, and take down the text. (The links can stay though).
I'm all for challenging the DMCA when a good oppertunity arrives. But wait for a case where there is some substance to the defence, and a chance of winning. Fighting an unwinnable case against the DMCA and losing just becuase there are aspects of the DMCA we disagree with is stupid.
This is only an issue becuase it's Microsoft. I dislike Microsoft as much as the next geek, but we can beat Microsoft *legally*.
Besides, I think it is very likely Microsoft was deliberately trying to leverage the DeCSS rulings and publish a secret in such a way that any software produced to interoperate can be considered 'tainted' by illegally obtained information. The Samba people have a chance to fight this - their 'interoperability exception' case is stronger than the DeCSS people, because no one can argue that Samba is a piracy tool. Let them handle this one.
Um. I am extremely embarassed. Never before have I been tempted to post a 'me too'. But this comment is both accurate and important enough that I think it deserves one.
OK, I'll compromise and itemise slightly.
'Me too' to everything except the 'I use napster' sentence, but I have played one set of mp3's, and I decided that I didn't want the albumn, so I deleted them.
Since a good deal of KDE development takes place in Europe, there are rather less barriers to including encryption than there are in more US-centric projects like Mozilla. As a result, the encryption is already there.
Of course recent US legislation has largely removed those barriers.
The KDE html widget was started well before Mozilla was born as an open source project. Gnome took the same approach: have you ever noticed that the Gnome help browser has its own html rendering engine?
When both of these were started, there was no lightweight, modern, open source rendering widget available for the approrpiate language and toolkit, and therefore both projects developed one.
KDE had the head start, and thus their widget was getting pretty capable by the time Moz was OSed. Gnome were a little behind, and therefore it was worth them switching to Mozilla (And recently this work has been taken up by Eazel for Nautilus).
The result: both Konqueror and Nautilus look as though they will be pretty capable. If they maintain some sort of parity, I guess who uses which will depend on which desktop they use (and therefore to some extent which Linux distro).
The competition will encourage developers to provide good standards support, and to fix any deficiencies in their project. Neither side will be able to argue that 'You need 64Mb for a modern web browser' if the competitor runs in 32Mb.
So I am convinced that Konqueror is a good thing, even though I expect to be browsing on Nautilus a year from now.
Basically, Nautilus (the Eazle file manager) will now embed the Bonobo mozilla component. In one step Nautilus has leaprogged Konquerer and Opera in standards-compliant web browsing, and provided a native GTK-zilla without any of that skin bloat.
Of course we'll have to wait a few months for a Nautilus beta. But I guess it'll take that long to fix the memory leaks in Mozilla.
I thought I must be the only Linux user who hated skinable apps!
I use Linux all day every day. And I want my display to be as clean and as clear as possible. I want my widgets instantly recognizable, equal in size, and consistently laid out.
Don't get me wrong, I am quite happy to change my GTK theme or Enligtenment theme occasionally, as long as everything remains consistent (and clear - why are most of the Enlightenment themes unusably dark? Too much late night hacking? The Sawmill selection seems better). But I usually end up back at Default, because Netscape 4.x and Xemacs don't use GTK widgets, and Default looks closest to their native appearance.
Gecko I love. But Mozilla and Xmms insist on making their own separate (if mutable) identity and sacrifice usability by smothering themselves in decorations which don't reflect the rest of the UI.
Xemacs has the right idea - you customize not the appearance, but the functionality. I want to add buttons to mozilla which turn off images, or disable cookies, I don't want to change what the buttons look like.
Fortunately they are open source. When someone writes a conventional GTK frontend to gecko, I will be first in the queue to beta test, or maybe even code if some of my other projects let up.
We only get the 10Mb all-in package.
If the program is illegal, then the GPL does not give anyone any right to distribute it.
However the question is, is the program illegal? This has yet to be demonstrated. The illegal assignment of rights to Mattel which had already been assigned elsewhere does nothing to make the program illegal. The legality of the program is unchanged from before the settlement.
Whether the program is legal or not depends on whether the reverse engineering is considered to be legitimate under any of the reverse-engineering clauses of relevent copyright laws in the relevent countries. This has yet to be proven one way or the other.
What I haven't seen is any laws which might make the essay, which is far more damaging than the program, illegal.
Suppose, now, that the cphack authors had included an extra clause in the license. And suppose this extra clause said that no matter what, they couldn't be held liable for writing it. Do you think such a thing would stand up in court? Sorry, but if you break the law, not even a fine print disclaimer will save you.
Although if UCITA is passed, this sort of ridiculous license cluase may well become legal in some cases.
If you GPL something, you can still sell the code to someone under a non-exclusive license, but you cannot transfer exclusive ownership in the way the settlement appears to have done.
I suspect this was done unintentionally however - presented with a settlement document full of legalese, it wouldn't have occured to me to look for this problem. Neither is it likely to occur to most mainstream lawyers who might not be familiar with the implications of the GPL.
The lesson for the rest of us is, make sure you have the right to distribute your code before you GPL it, and if someone does prosecute, make sure they know it has already been licensed to others.
It seems to me that capitalism is predicated on a free market where prices are dictated by supply and demand, and that this mechanism of price control breaks down in the presence of a monoploy. Thus anti-trust legislation must a pivotal aspect of a capitalist economy. Anti-trust legislation would be redundant in a centralized socialist economy because prices would be determined centrally. Therefore the US use of anti-trust legislation suggest to me that the US is a capitalist, rather than a socialist economy.
I don't see how being offshore would help either. If the US, EU or Japan can't prosecute the company, they can impose punitive duties on its products. Since all three have at least examined anti-trust charges, I don't see any of them kicking up a major stink with the WTO.
Interesting.
Were the Linux cluster users using gcc/g77? It is well known that (at least for most scientific codes) you can get 50-100% speedup by switching from the GNU compilers to commercial ones from Portland or elsewhere.
If there is still a difference, then the next thing to try is the latest dev kernels, which have better SMP (if SMP nodes are used), and significantly faster disk io through the elimination of double caching.
Since most scientific apps should spend most of their eating user CPU cycles, I wouldn't expect there to be very much difference between one OS and another, however node uptime and more established remote admin are points in Linux' favour for big clusters.
All day every day.
I used KDE as my promary environment during early 1999, becuase it gave very convenient access to applications, although WM would have done a very similar job.
Recently I have switched to Gnome for both my home and work machines, mainly for the ability to swallow a graphical pager and task list on the panel. I didn't have the real-estate for KDE's separate task list. A pop-up task list on the desktop is not bad, but I don't find it quite as convenient.
Of course give me a machine with only 16 or 32M RAM and I'll be back to WM straight away. Both the others really need 48, or the paging gets annoying.
I've used SGI's for a number of years, including recently Power Challenges, Octanes, and such like.
Yes, the graphics are good. But the processing stinks. On most of the codes I run (mix of 32-bit fp and integer array index, FFTs and such like), a Pentium II/300 will outperform an R10000 with heaps of cache and memory bandwidth.
The ccNUMA + crossbar stuff is great, but needs a faster processor to go with it.
Still, they are doing great work with Linux. For their software work I hope they do well.
Because MS have just been tub-thumping about how Win2000 includes streaming servers built in with no further licenses. MS could make Linux a second-class web browsing platform if they could capture the streaming server market with a proprietary format for which no Linux player exists. This move is a step towards preventing that.
Hopefully RedHat will get the source code and ensure that the new RealPlayer is actually stable and useable under Linux.
I don't know about Transmeta's mobile-linux, but I can see one significant advantage of using and embedded Linux (or even an embedded DOS): Microwindows.
It is a compact (100Kb) embeddable GUI which implements the bulk of both the Win32 and X11 APIs. That gives you a very simple port of any software from two environments. Why use WinCE, when you can recompile your apps and avoid the royalities?
I presume TM's Mobile Linux is using a cut-down Xfree86. It would be interesting to see how much of the 32Mb ROM they could free up by using a purpose-built embedded GUI. (I think Xfree86 is great, I'm just not convinced it is the optimal choice for embedded applications).
On the whole I agree with you, but the timing of your first point is deliciously ironic.
If you surf on over to the Linux Game Tome, you will find that only today a patch was released for e2fsck (the thing that checks your disk for you if you don't shutdown the machine properly), which allows you to play Tetris while the file system is checked. Of course, because it is open source, your second point kicks in: The feature is optional.
I enjoyed the article. It sounds as though the programmers aren't too different to their open source counterparts (but richer). Presumably all the dodgy tactics come out of the marketting department. I guess one of the strengths of open-source is that we have no marketting department!
After I stopped laughing I tried to work out whether DeCSS was a good idea or not. I didn't come to any firm conclusion. So I linked it because it was funny. Good work!
Yes, clearly you are right that there is an implicit license rather than no cpoyright. But that makes the rest of my questions *more* rather than less relevent, because the implicit license, even if it contains an implicit no-warrenty, will certainly be overridden by the UCITA warrenty provision.
The only way I can see of getting out of this is to claim that if the program causes loss of data, the JVM is to blame. But that could only work for content which is supposed to work in a sandbox environment. Plugins and ActiveX are more problematic though.
How does UCITA affect active components in web pages?
Consider, a Java or JavaScript program is downloaded to your computer, may be cached several times on your hard-drive, is loaded and executed. Furthermore, it may be stored and redistributed by proxies and web caches.
So presumably such programs are considered public domain. Does UCITA create license conditions for public domain software? Does it force a warrenty on such software? If so, then you could be liable for a bug in your java applet, unless you want to claim that the fault was with the VM security.
Plugins and ActiveX are even more problematic. I believe IE will download these automatically (but I haven't tried this), in which cases there is no opertunity to agree to a license.
What about interpreted languages? Who is liable, the author of the interpreter or the author of the interpreted code?
What about programs which use libraries?
Despite not being a US citizen (and only rarely paying US taxes) I have relicensed all my scientific software so that its use is illegal if the `No Warrenty' clause is invalidated by local legislation. I've contacted MA and VA officials with this information.