HP calculators might be more competitive today if Carly hadn't decided that it would be a really good idea to dump the HP calculator engineering team to "save money".
Here's a comment I posted earlier today where I mention the patent reexamination process and suggest a modification. I'll re-print a summary of the data here
It is currently possible to request that a patent be reviewed by the USPTO. This does not require a lawsuit (or technically even a lawyer, though there is a need for a properly-formatted request).
There are two types of re-examination. They differ in several respects. One, inter partes, allows you to basically provide rebuttals to the filer's explanations, and the other ex parte, does not.
My thanks to Thalia for locating the associated fees on the USPTO website: inter partes costs $8800 and ex parte $2500. Both of these costs do not include legal fees, which Thalia estimated (for inter partes) at about $12,000.
The problem is that getting a patent runs about $1,000 (again, not including any legal fees). This tends to slant things towards people acquiring patents, as it is still more expensive to get a patent revoked.
The modification that I'd like to see made would involve *patent owners* having to pay ex parte or inter partes fees if it is determined that their patent was improperly approved. This means that groups like the EFF (and, with some work to make the process particularly easy, perhaps anyone) can initate re-examination requests while supplying prior art examples.
Such a change would encourage patent filers to ensure that their patents really are legitimate when filing (reducing the number of bogus patents), and would not financially penalize someone who knows of prior art and wants to fix the USPTO database (if anything, I'd like to see someone who successfully brings up an example of prior art and gets a patent revoked *paid* a small fee by the patent filer for their time).
This change would involve minimal changes to the system, and not much cost. There might be the issue of collecting from the patent owner, who might be unwilling to pay. I think that an eight-thousand dollar deposit per patent would probably be too weighty, so I'm not sure how to approach that detail yet. However, even if the USPTO needs additional funding to help cover costs of employees needing to review patents where the USPTO cannot collect from the patent filer, I think that we woudl be better off (furthermore, that individual could be barred from being issued future patents until they have paid off their existing dues).
Problem: this change would *have* to grandfather old patents, as companies and individuals would otherwise be liable for masses of money for bogus patents. Irritating as it is that those people were able to get away with such behavior, the system permitted bogus patents, and charging them fees for said patents now is not reasonable (nor would it be acceptable to many people).
I think that this is the only feasible approach to the problem. Trying to ensure that the USPTO never grants invalid patents would require that they maintain a huge staff of PhDs that are up on the bleeding edge in every area of research (and honestly, woudl be better off doing research instead of reviewing patents).
This is only true for GNOME/KDE projects (I have Windows Key-T set up to open an xterm running zsh, and it comes up in, oh, I don't know, a fifth of a second on my P4).
The GNOME and KDE libraries do a huge amount of *crap* when they start up. It's insane that gnome-terminal takes so many CPU cycles to start.
Filesystem journaling does not make the filesystem faster, and it's silly to suggest that it does.
In fact, journaled filesystems are generally noticeably (one might say significantly) slower than non-journaled ones.
Atomic updates of any sort are an additional constraint, do not improve speed, and may hurt it.
However, frequently the techniques used in implementing journalling (e.g. in ext3, where the head keeps moving in the same direction when writing, unlike ext2) can improve performance.
While this could have been done in a non-journalled filesystem, it wasn't, and was less obvious in such a filesystem.
but heaven forbid if "thru","nite", "cuz", "u", or even "hoser" become commonly accepted.
Some of these make sense.
"U" really is the most plausible of these, to my way of thinking. It has a direct parallel -- the first person pronoun is "I", and it seems quite reasonable to make the second person pronoun "u". It is certainly commonly used. It is phonetic (something that English could stand to enjoy a bit more of).
I think that the extensive use of "u" in typed interactive conversation is not surprising. "You" is the most common English word in spoken conversation, and "I" in written conversation (predating chatting). However, while I have not seen a study, it seems reasonable to assume that given the similarities between interactively-typed and spoken conversation, that "you" is the most common word in interactively-typed conversations. Given this, it only makes sense to make "you" one of the shortest words out there.
"Thru" is a more phonetic and shorter variant of "through". I would not mind seeing this becoming part of the standard English language.
I'm a bit more dubious about "nite". It *is* more phonetic and shorter, which is nice. However, it is not a particularly common word, and I'm not sure that it needs to be shortened.
"cuz" I can see being added. "Cuz" is short, and a very common word. There are no homonyms that it might interfere and cause confusion with. It fits with traditional Americanization of words, using "z" rather than "s". This is even more acceptable because it really is a new *word*, even if a shortened version, not an alternate spelling of another word.
"hoser" is slang, and may follow the path of colloquialisms into general use. Just about all of our words have been slang at some point.
They did, back in 1994 - 1997. It just about sunk the company, because Apple is a hardware company that also makes software.
The "cloners" were eating away at Apple's share of the hardware market, and Apple still had to develop the software to run all of it. Less revenue + same expenses = death.
Oh, bullshit. Apple sells their OS and software, and could happily charge cloners licensing fees equal to their own costs (probably charged them more, FWIW).
The problem is that the cloners were simply making *better machines*. They were cheaper, faster, and better designed, and they were eating Apple alive. Apple put a stop to things when it was clear that they weren't going to be able to compete on level ground.
The day Apple killed the cloners and forever locked themselves into a small monopoly market was the day that I swore off buying Apple products. This wasn't done out of ideology -- I'm just not interested in having to deal with a single vendor that overprices everything.
Also, 10 minutes for a reboot?? I reboot my XP machine maybe once a week (for installs usually). I've never had a blue screen (same machine for 8 months so far). It takes MAYBE 2 mintues to reboot.
It takes more than two minutes to get the state of your computer back to where it was -- programs and documents open, web pages back up (unlike Opera, MSIE does not remember the set of open pages in a crash), any passwords re-entered, etc.
Grammar like this is exactly why Project Managers are necessary. You Developers have to be kept away from the clients, be grateful that PMs are there to deal with them for you.
Some people may remember the Macintosh game StuntCopter. When the developer died, his parents released the software into the public domain:
These arcade games were programmed by Duane Blehm. They have all been previously released and are currently offered by most sources of Public Domain software. Duane unexpectedly died a year ago. Cairo ShootOut and Puzz'l required users to send $3.00 to Duane to receive a "Key Code" to unlock all of the features of the program. All of Duane's games contained offers to sell the source code of the programs to programmers who wanted to see how Duane wrote them. Duane's parents have been swamped with Key Code and source code requests that they are unable to supply. These new versions have been altered at the request of Duane's parents. These versions have been unlocked and will allow full access to all of the features. The offers for source code have also been removed. Duane's parents have requested that if you have any of the old versions of Duane's games that you destroy them and replace them with the new versions. Please do not distribute any of the prior versions. Distribution of these new versions is encouraged and requested. Thank You and Have Fun!
Miguel, I don't know whether the Mono Regression Test Suite has performance tests (though I assume that you have some set of code that you use to avoid performance regressions when changing the optimizer). This is just an idea, but it might be kind of neat to keep a script maintaining an up-to-date chart of a comparison of the latest MS CLR and Mono (and maybe DotGNU).
I would imagine that this would provide incentive ("Woohoo, we pulled ahead on the Foo test!") for optimization implementers, besides providing a useful set of (admittedly micro-) benchmarks for people to refer to.
If they currently demonstrate that Mono is slower in some areas, all the better! If I see that "array destruction" is relatively slow, and I'm a CS grad student working on compiler design, I might have an idea or two about what to do.
With closed-source projects, there is little reason to distribute performance data except as a marketing tool. With open source projects, it simply helps distribute information that assists the community in doing development.
"RAND" is generally considered a bad thing by the open source/free world. As another poster has pointed out, it does not mean "royalty-free" -- it means "reasonable". This means that, yes, royalties may be charged for any software using said patent (which obviously creates barriers for anything that is distributed free-as-in-beer and is incompatible with the GPL/LGPL).
The W3C is one of the tech standards organizations that does *not* allow RAND (thanks to outcry from the community when HP, Apple, Microsoft, and a couple others tried to stick royalties into W3C policy) -- it requires that *anyone* with a patent allow royalty-free use.
The W3C's policies are *still* GPL/LGPL-incompatible, though, as the W3C allows royalty-free licenses to disallow derivative implementations that do not implement the exact standard. This means that even the W3C (one of the more open folks out there) will happily endorse a standard that may not be implemented in GPL/LGPL software.
Speaking of which, I'd say that Tucows should ban the inclusion of any software that is spyware-based, and clearly label anything that is adware based.
"Bundling with" adware/spyware counts as use of adware/spyware -- that old loophole, where dishonest software companies used an InstallShield installer and "included and installed by default" Gator or similar should never be accepted by users again.
Well, certainly asking for a password before you install something is one step above the average WinXP setup
Ironically enough, this is not the case.
I have listed several ways that a malicious program can slip past the Mac's password scheme before (grab keystrokes, put up bogus password dialog at a different time that looks like it belongs to a different program, etc).
The problem is that Apple's security design in the Mac OS fails to establish a secure channel between the user and the OS when requesting a password. (This is why you must hit Ctrl-Alt-Del in most versions of Windows before entering a password). Apple's password dialog actually represents a reduction in security over no password, since it (a) passes the password through an insecure environment on a regular basis and (b) does not provide a method to establish that the identity of the requesting program may be trusted -- that the dialog is really a legtimiate password dialog.
I don't think it's unreasonable for people to want to be aware of things that their computer is doing that they might take issue with and still retain a sane level of functionality.
That might be done through legal means, through OS-level functionality restriction, or through malware removal software, but expecting people to either accept spyware or not use their computer is not, I think, reasonable.
The point is that he's not interested in dealing with the problems associated with "video card vendor not caring about small platform". He's got people on the platform that can fix issues, given the source. They don't have it. He wants to know where to get a card that's supported with open source drivers.
Here is the current state of affairs:
NVidia provides only binary drivers. Their Linux drivers are supposed to be of comparable quality to their Windows drivers -- you just have to deal with the fact that the Linux world has zero tradition of binary kernel module compatibility. You will always be fighting an uphill battle to continue using these.
ATI provides binary drivers for their cards. There are *also* 3d open source drivers in DRI, of which the Radeon 9200 is the newest well-supported card. The Radeon 9200 is currently the best choice for a 3d card for a Linux/BSD user. I have heard claims that the Linux drivers are slightly slower than the Windows drivers. Also, keep in mind that the 9200 is *not* the latest-and-greatest from ATI.
Matrox (a) has fallen behind in 3d performance and (b) used to have *excellent* open source support for their cards in the G200 through G450 era. I bought both a G200 and a G450, and the support at the time was better than for any other manufacturers. John Carmack was involved with some of their driver work (John, if you read this, you work was much appreciated). Their Parhelia line has binary-only drivers, supports only Red Hat, and seriously lags kernel releases.
The bottom line is that if you want to get a 3d card for Linux use, the Radeon 9200 with DRI drivers is the way to go. It's not the fastest card ATI makes, but it can currently run all existing 3d software for Linux without speed problems -- Linux may lag Windows in open source 3d support, but also in games, and NWN and Tribes 2 easily run smoothly on the card.
It is possible to request that the USPTO re-examine a patent. This does not require suing (and technically, I believe, does not even require a lawyer).
I believe that there are some fees, but I could not find the cost of a inter partes or ex parte reexamination on the USPTO web site (and would appreciate anyone that knows posting -- I'm talking about the cost sans any associated legal costs, if someone gets a law firm to do this).
The re-examination usually relies on new prior art being brought to light.
I'd like to see this system modified to impose the fee (perhaps with some multiplier) on the *patent filer* if the reexamination finds that the patent is indeed invalid, rather than on the party requesting the reexamination. If the process of requesting patent reexamination was streamlined and made zero-cost (if you're correct), this would effectively eliminate the problem of bogus patents.
It would cost as much as a new computer, and while it might get you in the news media, I don't think the general public has a bad opinion of the patent system. It'd take a *lot* of criticism to do so.
How anybody thought this was worthy of a patent is beyond me.
Probably the patent originators didn't either.
However, a lot of companies pressure their employees (especially those working in a research capacity) to churn out N patents per quarter, regardless of how valuable they are. It's IP portfolio material, and that's valuable. Making actual discoveries is always a bit dicey -- maybe they come up with something, maybe not -- but if your researchers are churning out ammunition for your legal team, at least they're producing something.
The whole thing nicely sums up what's wrong with the patent system.
IMHO, in a sane patent system, there should be no more than maybe 100 patents granted a year. Why? Because there *aren't* all that many major new ideas coming out in an areas that require lots of research.
To mount an NTFS 1.2/3.x (Windows NT4/2000/XP/2003) volume, use the file system type 'ntfs'. The driver currently supports read-only mode (with no fault-tolerance, encryption or journalling) and very limited, but safe, write support
Old-style-clipboards in X == drag-and-drop on Windows/Mac OS. It's just easier to use.
New-style-clipboards n X == copy-and-paste on Windows/Mac OS
Emacs's internal clipboard (the "kill ring") really is an artifact of the fact that Emacs can run in an environment that lacks its own copy/paste support (like a VT200, say). Emacs has a very nice clipboard with a huge history that is generally better than that of the normal operating environment.
It's rather ironic that people are now complaining that X doesn't support clipboards persisting past the life of programs -- it actually *did* at one point with cut buffers, but it's kind of inefficient when you're working with a remote windowing environment to immediately dump every cut to the X server.
Slashdot Mirror
HP calculators might be more competitive today if Carly hadn't decided that it would be a really good idea to dump the HP calculator engineering team to "save money".
Here's a comment I posted earlier today where I mention the patent reexamination process and suggest a modification. I'll re-print a summary of the data here
It is currently possible to request that a patent be reviewed by the USPTO. This does not require a lawsuit (or technically even a lawyer, though there is a need for a properly-formatted request).
There are two types of re-examination. They differ in several respects. One, inter partes, allows you to basically provide rebuttals to the filer's explanations, and the other ex parte, does not.
My thanks to Thalia for locating the associated fees on the USPTO website: inter partes costs $8800 and ex parte $2500. Both of these costs do not include legal fees, which Thalia estimated (for inter partes) at about $12,000.
The problem is that getting a patent runs about $1,000 (again, not including any legal fees). This tends to slant things towards people acquiring patents, as it is still more expensive to get a patent revoked.
The modification that I'd like to see made would involve *patent owners* having to pay ex parte or inter partes fees if it is determined that their patent was improperly approved. This means that groups like the EFF (and, with some work to make the process particularly easy, perhaps anyone) can initate re-examination requests while supplying prior art examples.
Such a change would encourage patent filers to ensure that their patents really are legitimate when filing (reducing the number of bogus patents), and would not financially penalize someone who knows of prior art and wants to fix the USPTO database (if anything, I'd like to see someone who successfully brings up an example of prior art and gets a patent revoked *paid* a small fee by the patent filer for their time).
This change would involve minimal changes to the system, and not much cost. There might be the issue of collecting from the patent owner, who might be unwilling to pay. I think that an eight-thousand dollar deposit per patent would probably be too weighty, so I'm not sure how to approach that detail yet. However, even if the USPTO needs additional funding to help cover costs of employees needing to review patents where the USPTO cannot collect from the patent filer, I think that we woudl be better off (furthermore, that individual could be barred from being issued future patents until they have paid off their existing dues).
Problem: this change would *have* to grandfather old patents, as companies and individuals would otherwise be liable for masses of money for bogus patents. Irritating as it is that those people were able to get away with such behavior, the system permitted bogus patents, and charging them fees for said patents now is not reasonable (nor would it be acceptable to many people).
I think that this is the only feasible approach to the problem. Trying to ensure that the USPTO never grants invalid patents would require that they maintain a huge staff of PhDs that are up on the bleeding edge in every area of research (and honestly, woudl be better off doing research instead of reviewing patents).
Thoughts?
Applications take longer to load in linux.
This is only true for GNOME/KDE projects (I have Windows Key-T set up to open an xterm running zsh, and it comes up in, oh, I don't know, a fifth of a second on my P4).
The GNOME and KDE libraries do a huge amount of *crap* when they start up. It's insane that gnome-terminal takes so many CPU cycles to start.
Filesystem journaling does not make the filesystem faster, and it's silly to suggest that it does.
In fact, journaled filesystems are generally noticeably (one might say significantly) slower than non-journaled ones.
Atomic updates of any sort are an additional constraint, do not improve speed, and may hurt it.
However, frequently the techniques used in implementing journalling (e.g. in ext3, where the head keeps moving in the same direction when writing, unlike ext2) can improve performance.
While this could have been done in a non-journalled filesystem, it wasn't, and was less obvious in such a filesystem.
Given that we're about 60 years short of a full millennium without being invaded,
The Isle of Wight in 1545 doesn't count, eh?
but heaven forbid if "thru","nite", "cuz", "u", or even "hoser" become commonly accepted.
Some of these make sense.
"U" really is the most plausible of these, to my way of thinking. It has a direct parallel -- the first person pronoun is "I", and it seems quite reasonable to make the second person pronoun "u". It is certainly commonly used. It is phonetic (something that English could stand to enjoy a bit more of).
I think that the extensive use of "u" in typed interactive conversation is not surprising. "You" is the most common English word in spoken conversation, and "I" in written conversation (predating chatting). However, while I have not seen a study, it seems reasonable to assume that given the similarities between interactively-typed and spoken conversation, that "you" is the most common word in interactively-typed conversations. Given this, it only makes sense to make "you" one of the shortest words out there.
"Thru" is a more phonetic and shorter variant of "through". I would not mind seeing this becoming part of the standard English language.
I'm a bit more dubious about "nite". It *is* more phonetic and shorter, which is nice. However, it is not a particularly common word, and I'm not sure that it needs to be shortened.
"cuz" I can see being added. "Cuz" is short, and a very common word. There are no homonyms that it might interfere and cause confusion with. It fits with traditional Americanization of words, using "z" rather than "s". This is even more acceptable because it really is a new *word*, even if a shortened version, not an alternate spelling of another word.
"hoser" is slang, and may follow the path of colloquialisms into general use. Just about all of our words have been slang at some point.
They did, back in 1994 - 1997. It just about sunk the company, because Apple is a hardware company that also makes software.
The "cloners" were eating away at Apple's share of the hardware market, and Apple still had to develop the software to run all of it. Less revenue + same expenses = death.
Oh, bullshit. Apple sells their OS and software, and could happily charge cloners licensing fees equal to their own costs (probably charged them more, FWIW).
The problem is that the cloners were simply making *better machines*. They were cheaper, faster, and better designed, and they were eating Apple alive. Apple put a stop to things when it was clear that they weren't going to be able to compete on level ground.
The day Apple killed the cloners and forever locked themselves into a small monopoly market was the day that I swore off buying Apple products. This wasn't done out of ideology -- I'm just not interested in having to deal with a single vendor that overprices everything.
Expose [is] good
Identical functionality is also available on Linux.
Me, I like the more powerful multiple viewport system (as traditional *IX window managers use), but some people just can't get used to it.
Also, 10 minutes for a reboot?? I reboot my XP machine maybe once a week (for installs usually). I've never had a blue screen (same machine for 8 months so far). It takes MAYBE 2 mintues to reboot.
It takes more than two minutes to get the state of your computer back to where it was -- programs and documents open, web pages back up (unlike Opera, MSIE does not remember the set of open pages in a crash), any passwords re-entered, etc.
Grammar like this is exactly why Project Managers are necessary. You Developers have to be kept away from the clients, be grateful that PMs are there to deal with them for you.
Your followup is a run-on sentence.
SSH sessions won't die whether or not you're using hibernation. The only variable involved is how long the computer is down.
Some people may remember the Macintosh game StuntCopter. When the developer died, his parents released the software into the public domain:
These arcade games were programmed by Duane Blehm. They have all been previously released and are currently offered by most sources of Public Domain software. Duane unexpectedly died a year ago. Cairo ShootOut and Puzz'l required users to send $3.00 to Duane to receive a "Key Code" to unlock all of the features of the program. All of Duane's games contained offers to sell the source code of the programs to programmers who wanted to see how Duane wrote them. Duane's parents have been swamped with Key Code and source code requests that they are unable to supply. These new versions have been altered at the request of Duane's parents. These versions have been unlocked and will allow full access to all of the features. The offers for source code have also been removed. Duane's parents have requested that if you have any of the old versions of Duane's games that you destroy them and replace them with the new versions. Please do not distribute any of the prior versions. Distribution of these new versions is encouraged and requested. Thank You and Have Fun!
Miguel, I don't know whether the Mono Regression Test Suite has performance tests (though I assume that you have some set of code that you use to avoid performance regressions when changing the optimizer). This is just an idea, but it might be kind of neat to keep a script maintaining an up-to-date chart of a comparison of the latest MS CLR and Mono (and maybe DotGNU).
I would imagine that this would provide incentive ("Woohoo, we pulled ahead on the Foo test!") for optimization implementers, besides providing a useful set of (admittedly micro-) benchmarks for people to refer to.
If they currently demonstrate that Mono is slower in some areas, all the better! If I see that "array destruction" is relatively slow, and I'm a CS grad student working on compiler design, I might have an idea or two about what to do.
With closed-source projects, there is little reason to distribute performance data except as a marketing tool. With open source projects, it simply helps distribute information that assists the community in doing development.
Take a look here and see if this is what you're looking for.
"RAND" is generally considered a bad thing by the open source/free world. As another poster has pointed out, it does not mean "royalty-free" -- it means "reasonable". This means that, yes, royalties may be charged for any software using said patent (which obviously creates barriers for anything that is distributed free-as-in-beer and is incompatible with the GPL/LGPL).
The W3C is one of the tech standards organizations that does *not* allow RAND (thanks to outcry from the community when HP, Apple, Microsoft, and a couple others tried to stick royalties into W3C policy) -- it requires that *anyone* with a patent allow royalty-free use.
The W3C's policies are *still* GPL/LGPL-incompatible, though, as the W3C allows royalty-free licenses to disallow derivative implementations that do not implement the exact standard. This means that even the W3C (one of the more open folks out there) will happily endorse a standard that may not be implemented in GPL/LGPL software.
Speaking of which, I'd say that Tucows should ban the inclusion of any software that is spyware-based, and clearly label anything that is adware based.
"Bundling with" adware/spyware counts as use of adware/spyware -- that old loophole, where dishonest software companies used an InstallShield installer and "included and installed by default" Gator or similar should never be accepted by users again.
Well, certainly asking for a password before you install something is one step above the average WinXP setup
Ironically enough, this is not the case.
I have listed several ways that a malicious program can slip past the Mac's password scheme before (grab keystrokes, put up bogus password dialog at a different time that looks like it belongs to a different program, etc).
The problem is that Apple's security design in the Mac OS fails to establish a secure channel between the user and the OS when requesting a password. (This is why you must hit Ctrl-Alt-Del in most versions of Windows before entering a password). Apple's password dialog actually represents a reduction in security over no password, since it (a) passes the password through an insecure environment on a regular basis and (b) does not provide a method to establish that the identity of the requesting program may be trusted -- that the dialog is really a legtimiate password dialog.
I don't think it's unreasonable for people to want to be aware of things that their computer is doing that they might take issue with and still retain a sane level of functionality.
That might be done through legal means, through OS-level functionality restriction, or through malware removal software, but expecting people to either accept spyware or not use their computer is not, I think, reasonable.
The point is that he's not interested in dealing with the problems associated with "video card vendor not caring about small platform". He's got people on the platform that can fix issues, given the source. They don't have it. He wants to know where to get a card that's supported with open source drivers.
Here is the current state of affairs:
NVidia provides only binary drivers. Their Linux drivers are supposed to be of comparable quality to their Windows drivers -- you just have to deal with the fact that the Linux world has zero tradition of binary kernel module compatibility. You will always be fighting an uphill battle to continue using these.
ATI provides binary drivers for their cards. There are *also* 3d open source drivers in DRI, of which the Radeon 9200 is the newest well-supported card. The Radeon 9200 is currently the best choice for a 3d card for a Linux/BSD user. I have heard claims that the Linux drivers are slightly slower than the Windows drivers. Also, keep in mind that the 9200 is *not* the latest-and-greatest from ATI.
Matrox (a) has fallen behind in 3d performance and (b) used to have *excellent* open source support for their cards in the G200 through G450 era. I bought both a G200 and a G450, and the support at the time was better than for any other manufacturers. John Carmack was involved with some of their driver work (John, if you read this, you work was much appreciated). Their Parhelia line has binary-only drivers, supports only Red Hat, and seriously lags kernel releases.
The bottom line is that if you want to get a 3d card for Linux use, the Radeon 9200 with DRI drivers is the way to go. It's not the fastest card ATI makes, but it can currently run all existing 3d software for Linux without speed problems -- Linux may lag Windows in open source 3d support, but also in games, and NWN and Tribes 2 easily run smoothly on the card.
It is possible to request that the USPTO re-examine a patent. This does not require suing (and technically, I believe, does not even require a lawyer).
I believe that there are some fees, but I could not find the cost of a inter partes or ex parte reexamination on the USPTO web site (and would appreciate anyone that knows posting -- I'm talking about the cost sans any associated legal costs, if someone gets a law firm to do this).
The re-examination usually relies on new prior art being brought to light.
I'd like to see this system modified to impose the fee (perhaps with some multiplier) on the *patent filer* if the reexamination finds that the patent is indeed invalid, rather than on the party requesting the reexamination. If the process of requesting patent reexamination was streamlined and made zero-cost (if you're correct), this would effectively eliminate the problem of bogus patents.
It would cost as much as a new computer, and while it might get you in the news media, I don't think the general public has a bad opinion of the patent system. It'd take a *lot* of criticism to do so.
Microsoft have an excellent record as far as I know, of never initiating a patent battle. MS' patent portfolio is used purely for defensive purposes.
"We should trust [entity] and give them the legal right, because they haven't yet abused it" is pretty broken WRT real-world usage.
Also, consider what happens when Microsoft starts going under a la SCO. You get lots of desperate people trying to get money from anything...
Sure, defense is great during the good times, but when the bad times come...
How anybody thought this was worthy of a patent is beyond me.
Probably the patent originators didn't either.
However, a lot of companies pressure their employees (especially those working in a research capacity) to churn out N patents per quarter, regardless of how valuable they are. It's IP portfolio material, and that's valuable. Making actual discoveries is always a bit dicey -- maybe they come up with something, maybe not -- but if your researchers are churning out ammunition for your legal team, at least they're producing something.
The whole thing nicely sums up what's wrong with the patent system.
IMHO, in a sane patent system, there should be no more than maybe 100 patents granted a year. Why? Because there *aren't* all that many major new ideas coming out in an areas that require lots of research.
Read /usr/src/linux/Documentation/filesystems/ntfs.txt.
In 2.6.5:
To mount an NTFS 1.2/3.x (Windows NT4/2000/XP/2003) volume, use the file
system type 'ntfs'. The driver currently supports read-only mode (with no
fault-tolerance, encryption or journalling) and very limited, but safe, write
support
Also see the linux-ntfs website.
Old-style-clipboards in X == drag-and-drop on Windows/Mac OS. It's just easier to use.
New-style-clipboards n X == copy-and-paste on Windows/Mac OS
Emacs's internal clipboard (the "kill ring") really is an artifact of the fact that Emacs can run in an environment that lacks its own copy/paste support (like a VT200, say). Emacs has a very nice clipboard with a huge history that is generally better than that of the normal operating environment.
It's rather ironic that people are now complaining that X doesn't support clipboards persisting past the life of programs -- it actually *did* at one point with cut buffers, but it's kind of inefficient when you're working with a remote windowing environment to immediately dump every cut to the X server.