A customer goes to Dell and looks over the selection. No Linux desktops are available. The customer asks if Linux desktops are available and is told, "No", along with complicated excuses as to why not. The customer goes elsewhere.
"Steven A. Ballmer, Microsoft's chief executive, said, "Everybody in the operating system business wants to be the guy on the bottom," the software that controls the hardware."
"When quizzed on Microsoft's plans, Ballmer replied, "Our view is that virtualization is something that should be built into the operating system."
It seems that Microsoft has decided to bundle their virtualization into Windows in their usual attempt to extend their Windows monopoly into new markets. Also as usual Microsoft is the trailing edge of technology, i.e. the last company into the market.
There are some very real disadvantages to Microsoft bundling their virtualization software into Windows. The first problem is that the virtualization problem begs for a modular solution. The complexity of running a variety of operating systems as application programs begs for the virtualization software to be kept distinct from any of the client operating systems.
In the case of bundling virtualization into Windows Microsoft will run into the same development problem that they had with Longhorn. Windows has ignored modularity for release after release. New features have been consistently bundled into Windows in a deliberately non-modular design. The result is that adding any new feature requires a rewrite of Windows. This problem with Longhorn resulted in huge development costs to produce very little. Now Microsoft is going to rewrite Windows to bundle the logically very complex virtualization feature? I predict that the effort will either fail or take an incredibly long amount of time and an enormous amount of money.
Microsoft may already realize this. This may be one of the reasons that Microsoft is backing Novell's support for the Xen project. Microsoft may be hedging their bet in case the Windows bundled virtualization project fails.
Unfortunately, the open source virtualization projects, Xen and KVM, also are integrated into an operating system, i.e. Linux. I think that the VMware design is superior to either Xen and KVM. One integrated virtualization OS in Linux is fine. KVM should remain that integrated virtualization system and Xen should be developed into a completely stand alone operating system.
I predict that the end result of the virtualization race will be VMware competing with a GPL3 stand alone virtualization OS. I doubt that the bundled Windows virtualization candidate will be finished in time to enter the race.
Software is sold as a stack from operating systems through middleware to application programs. Software companies have to work with the fact that customers often buy a mixed stack from different vendors. Software companies spend a lot of thought on their sales strategy within the stack. They often see sales opportunities elsewhere within the stack as for example Red Hat recently went up the stack by adding Jboss to their product line. Software companies also worry about competing companies which are positioned below them in the stack somehow forcing competitors in the upper portions of the stack into a competitive disadvantage. When Red Hat began offering Jboss in competition to Oracle Oracle countered by offering support for Red Hat Linux mostly as a defense to prevent Red Hat from biasing Red Hat Linux to favor Jboss over Oracle.
I think that worries about competition within the stack played a part in Microsoft's decision to go down the stack and enter the virtualization market. Microsoft is probably worried that Vmware or Xen could place Windows at a disadvantage in a virtualized machine any time that they want. Also if Microsoft eventually manages to dominate the virtualization market then they could begin introducing biases against competing operating systems.
Each Open Source project leader will determine whether or not his project will be released under GPL3. The FSF has said that their projects will move to GPL3. Linus Torvalds and Miguel de Icaza have said that their projects will not move to GPL3. Almost all of the other project leaders have either said nothing or indicated that they would move to GPL3 conditional on waiting to see the final form of GPL3.
The Microsoft-Novell agreement was met with widespread anger within the Open Source community. Many people look at the Microsoft-Novell agreement as a patent attack by Microsoft on Open Source. Part of the Open Source members' reaction was a renewed committment to using GPL3 to break the agreement. So it would be reasonable to predict that if the final wording of GPL3 is satisfactory that a large number of GPL project leaders will move to their projects to GPL3.
Novell and Microsoft can react to GPL3 by changing the Microsoft-Novell agreement to conform to GPL3. Or Novell can try to work around GPL3 by forking all GPL3 projects from a GPL2 base. The cost of doing so could be prohibitive depending on how many projects move to GPL3. Another problem is that some customers may be reluctant to buy a forked distribution.
Everybody seems to agree that whether or not Linus Torvalds moves Linux to GPL3 is extremely important to how strong the Open Source reaction to Microsoft's patent attack will be. Linus Torvalds is on record as disliking GPL3. But he has not said anything about GPL3 and/or the Microsoft-Novell agreement since the agreement was announced. Whether the Microsoft patent attack on Linux changed Linus Torvalds' thinking about GPL3 is unknown.
The parent post seems to be assuming that the bug has been in Vista for a long time. One possibility is that the bug is a major new bug created when a programmer attempted to fix an old minor bug. It happens now and then.
I attended a meeting at the U.S. Patent Office where people presented a series of proposed changes to reform the patent system. Of the various proposals I thought that the one by Beth Noveck was by far the best. I posted comments on the Internet supporting Beth Noveck's proposals.
After further reflection I have a more complicated view of the proposed patent reforms. I think that the reforms are good across all patentable fields but that they will do the least good with the software patent problems faced by Open Source.
The problem with Open Source and patents is that we are disinterested in patents. We have no intention of applying for patents for Open Source software. Applying for software patents is expensive and we have no intention of blocking other people from using our ideas. Similarly we have no interest in watching the parade of new software patent proposals to call foul on the proposed software patents which have prior art. We detest software patent lawsuits and do not want to participate on either side of a software patent lawsuit. However a software patent lawsuit can get us riled up enough to do the work of finding prior art to invalidate the software patent in question.
Our attitude of ignoring and failing to document patentable software causes the Patent Office and patent lawyers a lot of problems. They cannot issue patents for which there is prior art even if the prior art is not patented. Finding undocumented prior art is a monumental problem for the Patent Office and they do a poor job of it. The patent lawyers have responded by trying to get bad patents declared still valid when unpatented prior art is discovered. Two examples are "first to file" and the "obviousness" patent case now before the Supreme Court.
While I think that Beth Noveck's work is all well and good and applicable to all patent fields I think that Open Source would be better off to spent our efforts in trying to abolish software patents once and for all. That would be easier than perpetually being forced to fight a series of software patent lawsuits.
Gardiner seems to be acknowledging that Windows has exceeded its design limitations to the point that it cannot be developed any further. One of Windows basic design flaws is its lack of modularity. Windows needs to be completely redesigned and rewritten in order to have modularity.
Gardiner is talking about reviving Windows by adding virtualization. I see implementing virtualization as the epitome of modularization. The virtualization should be completely separate from any OS that it will run. Gardiner seems to be advocating making virtualization an integral part of Windows which is completely the wrong approach. It is also self defeating as Windows as it currently stands is incapable of accepting new design features.
Microsoft should work on virtualization and salvaging Windows as two completely separate issues.
I think it is obvious to Wall Street that Microsoft is grossly mismanaged from a financial standpoint. This has resulted in very public pressure by some of the larger Microsoft shareholders for Microsoft to use their cash hoard to buy back Microsoft stock. Microsoft has already held one buyback and promises others in the indefinite future. If Microsoft does not placate the shareholders then there will be a shareholder revolt which will turf the current management out of Microsoft.
We are heading into a recession. One result of the recession will be that Microsoft's earnings will decline. It is even possible that Microsoft earnings will turn negative by spring 2007. The fact that Microsoft is sitting on a huge pile of cash while earnings are decreasing will increase the pressure for a shareholder revolt. It would also make conditions easier for a leveraged buyout.
Wall Street is very critical about how Microsoft is spending their development dollars. There are two facets to this criticism. On the one hand it is grossly obvious that Microsoft development spends a log of money to produce very little in the way of commercially viable products. On the other hand pure financial people always want to increase current earnings by reducing development spending. This increase in earnings lasts until the next technical revolution and the cash cow is suddenly caught with unsalable, obsolete products. The new, competitive products were victims of cost-cutting. What Microsoft management is currently doing is spending a lot of development money to develop products that are unsalable, obsolete products on the day they are released.
The solution to Microsoft's problems is new management. As other posters have pointed out the price of a Microsoft leveraged buyout is just too expensive and risky for lenders to risk that kind of money in a leveraged buy out. So the new management is unlikely to come from a leveraged buy out. If new management is forced upon Microsoft it will most likely be from a shareholders' revolt.
If new management is not thrust upon Microsoft in one form or another then Microsoft will die by not being prepared for the next technology revolution, whenever that may be.
This is an extremely good article on the problems faced by Open Source in dealing with patents.
In a way the article is only a snapshot of the current set of problems. It is the nature of lawyers to deliberately introduce more complications into any legal situation in an effort to gain an advantage for their client. So as time marches on patent law will become more convoluted and more of a problem for Open Source.
While fighting this patent case in the Supreme Court is necessary it really is only one bump in a long highway. There will be a continuous series of problems created by innovative lawyers in the future. I think that the best solution that we can find to the software patent problem is to abolish software patents. It would be easier to put the effort into ending the software patent mess forever by abolishing software patents than to be perpetually fighting the latest lawyers' brainstorm.
If Vista is written modularly and has a clean, well documented API then why would an application development team need any help from the Vista development team to get their application working on Vista?
One of the problems that we are having with creating GPL3 is that a license is not a very effective way to solve the DRM and software patent problems. DRM and software patents are embedded in the law. When there is a conflict between the law and a license then the law takes precedence. So GPL3 does not have much maneuvering room to solve the problems that DRM and software patents cause Open Source.
I agree with Richard Stallman's efforts to put clauses in GPL3 to alleviate the DRM and software patent problems. However, I don't have much hope that these clauses will be very effective.
I think that a much more effective course of action is to try to change the laws on DRM and software patents. I think that we should lobby governments all over the world to abolish software patents. In the case of DRM I think that the DRM copyright protection should be legal but that the DRM laws should not contain clauses making it illegal to create software or hardware which can copy DRM protected material. The act of copying copyrighted material should be illegal but the act of creating a copying machine should be legal.
The basic problem with the current European patent law is that it disallows software patents but is vaguely worded to the point that some courts interpret it to allow software patents. Thus in Europe today some countries' court systems allow software patents and some disallow software patents. The pro software patent lobby is trying to create a single Europe wide court that will allow software patents all over Europe. The anti software lobby is trying to block the single Europe wide patent court in an effort to keep software patents from spreading. Part of the problem with fighting the proposed new court is that aside from the software patent issue the single Europe wide patent court is basically a good idea.
I propose that we block the single European patent court as a delaying tactic only. In the meantime we should lobby to get the European patent laws amended to where the anti software patent clause explicitly bans software patents in no uncertain terms. There is a fair bit of support for such amendments among the members of the European Parliament. Once we get the law amended then we could enthusiastically support the creation of a unified European patent court because the new court would disallow software patents all across Europe.
These tumor cells will grow in any dog. It would be interesting to see if they will infect closely related species. Will they grow in wolves, coyotes, jackals, etc.? Are there any breeds of dogs which are immune to these tumor cells? Will they grow in prey bitten by a dog, such as rabbits? One possible use for these tumor cells could be to determine how closely other species are related to dogs.
Slashdot Mirror
then there should be exploits reported as attacking XP but not working on Vista.
------------------
Steve Stites
A customer goes to Dell and looks over the selection. No Linux desktops are available. The customer asks if Linux desktops are available and is told, "No", along with complicated excuses as to why not. The customer goes elsewhere.
-------------
Steve Stites
"Steven A. Ballmer, Microsoft's chief executive, said, "Everybody in the operating system business wants to be the guy on the bottom," the software that controls the hardware."
"When quizzed on Microsoft's plans, Ballmer replied, "Our view is that virtualization is something that should be built into the operating system."
It seems that Microsoft has decided to bundle their virtualization into Windows in their usual attempt to extend their Windows monopoly into new markets. Also as usual Microsoft is the trailing edge of technology, i.e. the last company into the market.
There are some very real disadvantages to Microsoft bundling their virtualization software into Windows. The first problem is that the virtualization problem begs for a modular solution. The complexity of running a variety of operating systems as application programs begs for the virtualization software to be kept distinct from any of the client operating systems.
In the case of bundling virtualization into Windows Microsoft will run into the same development problem that they had with Longhorn. Windows has ignored modularity for release after release. New features have been consistently bundled into Windows in a deliberately non-modular design. The result is that adding any new feature requires a rewrite of Windows. This problem with Longhorn resulted in huge development costs to produce very little. Now Microsoft is going to rewrite Windows to bundle the logically very complex virtualization feature? I predict that the effort will either fail or take an incredibly long amount of time and an enormous amount of money.
Microsoft may already realize this. This may be one of the reasons that Microsoft is backing Novell's support for the Xen project. Microsoft may be hedging their bet in case the Windows bundled virtualization project fails.
Unfortunately, the open source virtualization projects, Xen and KVM, also are integrated into an operating system, i.e. Linux. I think that the VMware design is superior to either Xen and KVM. One integrated virtualization OS in Linux is fine. KVM should remain that integrated virtualization system and Xen should be developed into a completely stand alone operating system.
I predict that the end result of the virtualization race will be VMware competing with a GPL3 stand alone virtualization OS. I doubt that the bundled Windows virtualization candidate will be finished in time to enter the race.
--
Steve Stites
Software is sold as a stack from operating systems through middleware to application programs. Software companies have to work with the fact that customers often buy a mixed stack from different vendors. Software companies spend a lot of thought on their sales strategy within the stack. They often see sales opportunities elsewhere within the stack as for example Red Hat recently went up the stack by adding Jboss to their product line. Software companies also worry about competing companies which are positioned below them in the stack somehow forcing competitors in the upper portions of the stack into a competitive disadvantage. When Red Hat began offering Jboss in competition to Oracle Oracle countered by offering support for Red Hat Linux mostly as a defense to prevent Red Hat from biasing Red Hat Linux to favor Jboss over Oracle.
I think that worries about competition within the stack played a part in Microsoft's decision to go down the stack and enter the virtualization market. Microsoft is probably worried that Vmware or Xen could place Windows at a disadvantage in a virtualized machine any time that they want. Also if Microsoft eventually manages to dominate the virtualization market then they could begin introducing biases against competing operating systems.
---------------------
Steve Stites
Each Open Source project leader will determine whether or not his project will be released under GPL3. The FSF has said that their projects will move to GPL3. Linus Torvalds and Miguel de Icaza have said that their projects will not move to GPL3. Almost all of the other project leaders have either said nothing or indicated that they would move to GPL3 conditional on waiting to see the final form of GPL3.
The Microsoft-Novell agreement was met with widespread anger within the Open Source community. Many people look at the Microsoft-Novell agreement as a patent attack by Microsoft on Open Source. Part of the Open Source members' reaction was a renewed committment to using GPL3 to break the agreement. So it would be reasonable to predict that if the final wording of GPL3 is satisfactory that a large number of GPL project leaders will move to their projects to GPL3.
Novell and Microsoft can react to GPL3 by changing the Microsoft-Novell agreement to conform to GPL3. Or Novell can try to work around GPL3 by forking all GPL3 projects from a GPL2 base. The cost of doing so could be prohibitive depending on how many projects move to GPL3. Another problem is that some customers may be reluctant to buy a forked distribution.
Everybody seems to agree that whether or not Linus Torvalds moves Linux to GPL3 is extremely important to how strong the Open Source reaction to Microsoft's patent attack will be. Linus Torvalds is on record as disliking GPL3. But he has not said anything about GPL3 and/or the Microsoft-Novell agreement since the agreement was announced. Whether the Microsoft patent attack on Linux changed Linus Torvalds' thinking about GPL3 is unknown.
----------------------
Steve Stites
The parent post seems to be assuming that the bug has been in Vista for a long time. One possibility is that the bug is a major new bug created when a programmer attempted to fix an old minor bug. It happens now and then.
-------------------
Steve Stites
I attended a meeting at the U.S. Patent Office where people presented a series of proposed changes to reform the patent system. Of the various proposals I thought that the one by Beth Noveck was by far the best. I posted comments on the Internet supporting Beth Noveck's proposals.
After further reflection I have a more complicated view of the proposed patent reforms. I think that the reforms are good across all patentable fields but that they will do the least good with the software patent problems faced by Open Source.
The problem with Open Source and patents is that we are disinterested in patents. We have no intention of applying for patents for Open Source software. Applying for software patents is expensive and we have no intention of blocking other people from using our ideas. Similarly we have no interest in watching the parade of new software patent proposals to call foul on the proposed software patents which have prior art. We detest software patent lawsuits and do not want to participate on either side of a software patent lawsuit. However a software patent lawsuit can get us riled up enough to do the work of finding prior art to invalidate the software patent in question.
Our attitude of ignoring and failing to document patentable software causes the Patent Office and patent lawyers a lot of problems. They cannot issue patents for which there is prior art even if the prior art is not patented. Finding undocumented prior art is a monumental problem for the Patent Office and they do a poor job of it. The patent lawyers have responded by trying to get bad patents declared still valid when unpatented prior art is discovered. Two examples are "first to file" and the "obviousness" patent case now before the Supreme Court.
While I think that Beth Noveck's work is all well and good and applicable to all patent fields I think that Open Source would be better off to spent our efforts in trying to abolish software patents once and for all. That would be easier than perpetually being forced to fight a series of software patent lawsuits.
---------------------
Steve Stites
Gardiner seems to be acknowledging that Windows has exceeded its design
limitations to the point that it cannot be developed any further. One of Windows
basic design flaws is its lack of modularity. Windows needs to be completely
redesigned and rewritten in order to have modularity.
Gardiner is talking about reviving Windows by adding virtualization. I see
implementing virtualization as the epitome of modularization. The virtualization
should be completely separate from any OS that it will run. Gardiner seems to be
advocating making virtualization an integral part of Windows which is completely
the wrong approach. It is also self defeating as Windows as it currently stands
is incapable of accepting new design features.
Microsoft should work on virtualization and salvaging Windows as two completely
separate issues.
-----------------------
Steve Stites
I think it is obvious to Wall Street that Microsoft is grossly mismanaged from a
financial standpoint. This has resulted in very public pressure by some of the
larger Microsoft shareholders for Microsoft to use their cash hoard to buy back
Microsoft stock. Microsoft has already held one buyback and promises others in
the indefinite future. If Microsoft does not placate the shareholders then
there will be a shareholder revolt which will turf the current management out of
Microsoft.
We are heading into a recession. One result of the recession will be that
Microsoft's earnings will decline. It is even possible that Microsoft earnings
will turn negative by spring 2007. The fact that Microsoft is sitting on a huge
pile of cash while earnings are decreasing will increase the pressure for a
shareholder revolt. It would also make conditions easier for a leveraged
buyout.
Wall Street is very critical about how Microsoft is spending their development
dollars. There are two facets to this criticism. On the one hand it is grossly
obvious that Microsoft development spends a log of money to produce very little
in the way of commercially viable products. On the other hand pure financial
people always want to increase current earnings by reducing development
spending. This increase in earnings lasts until the next technical revolution
and the cash cow is suddenly caught with unsalable, obsolete products. The new,
competitive products were victims of cost-cutting. What Microsoft management is
currently doing is spending a lot of development money to develop products that
are unsalable, obsolete products on the day they are released.
The solution to Microsoft's problems is new management. As other posters have
pointed out the price of a Microsoft leveraged buyout is just too expensive and
risky for lenders to risk that kind of money in a leveraged buy out. So the new
management is unlikely to come from a leveraged buy out. If new management is
forced upon Microsoft it will most likely be from a shareholders' revolt.
If new management is not thrust upon Microsoft in one form or another then
Microsoft will die by not being prepared for the next technology revolution,
whenever that may be.
----------------------
Steve Stites
This is an extremely good article on the problems faced by Open Source
in dealing with patents.
In a way the article is only a snapshot of the current set of problems. It is
the nature of lawyers to deliberately introduce more complications into any
legal situation in an effort to gain an advantage for their client. So as time
marches on patent law will become more convoluted and more of a problem for Open
Source.
While fighting this patent case in the Supreme Court is necessary it really is
only one bump in a long highway. There will be a continuous series of problems
created by innovative lawyers in the future. I think that the best solution
that we can find to the software patent problem is to abolish software patents.
It would be easier to put the effort into ending the software patent mess
forever by abolishing software patents than to be perpetually fighting the
latest lawyers' brainstorm.
----------------------
Steve Stites
If Vista is written modularly and has a clean, well documented API then why would an application development team need any help from the Vista development team to get their application working on Vista?
--------------------
Steve Stites
One of the problems that we are having with creating GPL3 is that a license is not a very effective way to solve the DRM and software patent problems. DRM and software patents are embedded in the law. When there is a conflict between the law and a license then the law takes precedence. So GPL3 does not have much maneuvering room to solve the problems that DRM and software patents cause Open Source.
I agree with Richard Stallman's efforts to put clauses in GPL3 to alleviate the DRM and software patent problems. However, I don't have much hope that these clauses will be very effective.
I think that a much more effective course of action is to try to change the laws on DRM and software patents. I think that we should lobby governments all over the world to abolish software patents. In the case of DRM I think that the DRM copyright protection should be legal but that the DRM laws should not contain clauses making it illegal to create software or hardware which can copy DRM protected material. The act of copying copyrighted material should be illegal but the act of creating a copying machine should be legal.
-----------------------
Steve Stites
The basic problem with the current European patent law is that it disallows
software patents but is vaguely worded to the point that some courts interpret
it to allow software patents. Thus in Europe today some countries' court
systems allow software patents and some disallow software patents. The pro
software patent lobby is trying to create a single Europe wide court that will
allow software patents all over Europe. The anti software lobby is trying to
block the single Europe wide patent court in an effort to keep software patents
from spreading. Part of the problem with fighting the proposed new court is that
aside from the software patent issue the single Europe wide patent court is
basically a good idea.
I propose that we block the single European patent court as a delaying tactic
only. In the meantime we should lobby to get the European patent laws amended
to where the anti software patent clause explicitly bans software patents in no
uncertain terms. There is a fair bit of support for such amendments among the
members of the European Parliament. Once we get the law amended then we could
enthusiastically support the creation of a unified European patent court
because the new court would disallow software patents all across Europe.
------------------
Steve Stites
These tumor cells will grow in any dog. It would be interesting to see if they will infect closely related species. Will they grow in wolves, coyotes, jackals, etc.? Are there any breeds of dogs which are immune to these tumor cells? Will they grow in prey bitten by a dog, such as rabbits? One possible use for these tumor cells could be to determine how closely other species are related to dogs.
----------------
Steve Stites