>"I wouldn't mind getting the service for a discount if I consented to my viewing habits being sold, or for free if they actually attached my name to it."
Since Tivo is not making a profit (and is well under water, except for the AOL bail-out), you can assume that all Tivo users are currently "getting the service at a discount".
Difficult upgrade ...if you're upgrading, you must have
Windows 98 or above. In fact, Microsoft says you'll have the best
experience only if you buy a brand-new system with XP preinstalled. That's
a pretty hefty investment in an untried operating system. Unless
you're planning to buy a new system when Windows XP hits the shelves,
your XP experience likely won't be as smooth as Microsoft
hopes.... it's not an upgrade you should consider lightly.
The provisional patent is intended to allow you to sell your idea to a company withing twelve months, then let them go thru the expensive patent process.
It does have it's drawbacks (many):
For example, you loose your provisional patent filing date if you don't file for patent the within the 12 month period.
I don't think this is the proper solution. Letting it lapse gives you no claim to prior art.
Wouldn't the USPTO be the best registry for software?
A commercial venture, or even a non-profit rganization may collapse, and isn't controlled by regulation, and may or may not meet standards (current and future), and can't guarantee that their database hasn't been comprimised.
This is the job of the USPTO, and they should provide the registration service for free. It would help them get out of their software patent mess.
You need to understand that increasing the volume of software and it's complexity may logically cover the assumed problem space more thoroughly, BUT, it simultaneously increases the probability of indirect program failure. Furthermore, you may think you understand the problem space thoroughly, but you can't guarantee it (how often do we see an exactly accurate computation of the wrong problem, i.e. the Hubble Space Telescope mirror was ground within exacting tolerances of the wrong shape).
>"The fact that you are espousing the view that we shouldn't even try in the software industry sounds to me that you're [name calling omitted]"
I don't mean to sound like a totally haphazard approach is best. I do believe in standards and requirements (etc...), but they have diminishing returns: For example: if you spend longer in the design phase, you might save some time in the integration and debugging phases, but it's returns diminish radically as you increase the time spent on design (or requirements). Youthful programmers want to spend eternity in the requirements, design, and coding phase... and have a distaste for integration and debugging. The tools provided for software design analysis are a joke. They create more problems than they solve, and are only cost effective when you really have to show due diligence with software that might kill someone (I know: I've both written and used such software)
In the analog world, a thorough design is measurable, in software, it becomes a black hole.
>"...and software engineers are not the be-all-end-all of the engineering world."
I never said, nor believe, otherwise.
>"And if you're curious as to what I consider as the most complex thing ever developed by Man, it's language."
Isn't it curious how there can be such a wide range of speech that's still intelligible by even the most ignorant human? That's the great thing about the analog world: tolerances.
Any attempt we make at computer based speech recognition (AI or neural nets or even patter matching) becomes bogged down in miles of code when we even begin to tolerate variations. While the increase in code seemingly makes software tolerant, it simultaneously increases the complexity and probability of indirect failure.
When we truly understand analog logic, we'll probably find speech recognition very complex, but not as complex as we made it out to be with digital logic.
Damn! I thought this was a good example. I recall at the time I thought the only reason IBM did it was for proprietary lock-in... but I don't have any evidence (and, in general, I don't recall anything well).
Does anybody remember back before the MS monopoly to an example of IBM making proprietary protocols/standards/software to lock-in customers? The PS/2 hardware and it's proprietary bus (I forget the name of the bus) is a great example of them trying to shutdown the Open PC Hardware revolution and regain their monopoly -- but that tactic didn't work.
It's like we turned the clock back a year+... I thought crazy ideas were out, in favor of profitable ideas?
Oh well, the crazy ideas were a lot of fun.
It does sound like it would be economical to replace all those cell phone towers with one high-flying antenna... but a baloon sounds more efficient than a plane.
The article said "Solar powered" -- I don't read any mention of Solar Powered on their web page.
The appeals court main concern in the Microsoft remedy was that we're just replacing one monopoly for another.
The prevailing software business model is unique with it's "proprietary lock-in". Assuring lack of competition in software applications.
Proprietary formats, standards, protocols, etc... lead to the "software vendor lock-in". Microsoft didn't invent this... they were doing ASCII along with everyone else when IBM was doing EBCDIC (Later, of course, MS with their monopoly would do TrueType while everybody else had already standardized to Postscript). They're also not the only software vendor to do this: every Unix consortium is comprised of vendors using one hand to shake, the other to stab in the back (trying to get others to abide by the standard, while they proprietarily create modifications to the standard).
Open source has a chance to create a level, standards based, playing field. Allowing true, balanced, competition.
When Compaq, and others, reverse engineered the IBM PC Bios, the PC revolution was ignited: anybody could make a motherboard or an adapter card, or memory (The MS OS and the Intel CPU were the only proprietary pieces left atop this open hardware platform). The prices came down, the innovation moved at record pace.
But, even if the current "proprietary lock-in" paradigm ends, the patent office, in granting frivilous software patents, may be providing the next lock-in paradigm.
Can and will congress legislate and mandate the proper regulations aimed at the software industry to end the "proprietary lock-in" paradigm (this also means modifying the DMCA to omit those portions that protect the lock-in) and assure that there are sufficient restrictions in the patent laws to limit the number and scope of software patents?
>"At least, I hope so.. otherwise, we're stuck with this ad hockery that we do now..."
We've actually been stuck here for quite some time. Long enough to know that the analogy to the analog world makes perfect logical sense, but is nonsense when implemented on an application of modest complexity.
It's the simplicity of the software that allows full path testing. It takes years to test such simple code, and the user pays through the nose; and rightfully so.
Once the software gets complex, pilots & civilians die. There's plenty of examples of software requirements bloat projects in the military and aerospace that get quitely get swepped under the rug.
Was it the Mars Polar Lander that interprteted its' own parachute deploying as having landed, therby turning off the brake thrusters and making a deep dull thud? Or, was it the one that got english/metric units fouled up causing catastrophic failure?
I assure you, the complexity of the software has increased beyond the point of rigorous testing, and that's made all the difference in the world
>"I'm not sure that I completely agree with you. It seems that you're saying that digital devices are incapable of having any degree of tolerance. That's not true."
Tell me that "char foo[20]" is +/- 10% of twenty bytes. You step one byte beyond 20, and the OS could come down; as if a 747 would fall from the sky if the hinge on an overhead bin failed.
When an engine fails on a 747, there's three others to take it's place. When Word fails, does it have a backup editor that catches your file and allows you to continue editing? Sometimes we have signal handlers to clean up our mess -- but they often fail in a crisis too.
Software has zero tolerances. It either does what it was told to do, or it fails. You can cover more cases to try to hedge failure, but you simultaneously increase the complexity and the chances for failure.
If you're building software that means life or death, then you keep it small and you do go to the effort of understanding all failure modes, and you don't launch it atop Windows (the FDA wouldn't allow it), and you charge a mint for it. But this isn't typical software, and the effort necessary to apply these concepts to moderately complex applications requires that the effort grow exponentially.
You call me arrogant when you claim you're so much smarter than the rest of the industry and know what the solution is. Reply back in 20 years, once you've found that you don't know all the answers.
>"It's also a function of how software is developed. The process by which a car (or a toaster, or an airplane) is designed, built and assembled is 50 or more years mature."
We've been comparing software development to analog-world development like this for quite a long time, still, no "silver bullet" solutions.
Someday, some conversion of consciousness will make us laugh that we thought the two were relatable. For now, all we can do is take the heat.
Airplanes are analog devices... they can survive within tolerances.
(You have to read the whole text before yelling "bullshit")
Furthermore, you obviously don't understand complex software. Engineers with the required Basic/Fortran class often mistake their simple assignements for actual, useable, integrated, software.
The software on your PC has more components than the hardware on your 747.
You can still build a "failure mode tree" for a 747... we realized long ago that such analysis of software was futile (you can do it for firmware, control systems, etc... but not integrated applications).
If builders built buildings the way programmers wrote programs, the first woodpecker to come along would destroy civilization!"
I first heard this user mantra in `82 with my first programming job -- and they said that was an old adage.
The problem isn't programmers lack of responsiveness to users, as has been suggested for the last 40 years. If that were true, it would have been solved by now.
The true problem is the inhearent complexity of software, where any useful integrated program enters the realm of chaos, and exhibits behavior "as if at random".
It's digital nature makes it more susceptible. While you can plumb a toilet within wide tolerances, software must be exact. Furthermore, a broken toilet doesn't take the city's sewer system down with it.
It's ease of modification makes it even more susceptible. A problem in hardware will be there for years, we'll learn to work around it, and it may become the standard. But with software, the fix (and the next set of bugs) will come with the next upgrade or patch.
The fellows suggestion that "speech recognition will cure this" is another example of how requirements bloat, to solve "the problems of software usability", exacerbates the problem.
Some problems need to be blamed on the programmers and management: the Window's kernel hung around much too long. Microsoft kept adding mounds of complexity with small doses of functionality to keep the ever faster processors busy; it was no wonder you couldn't keep it up.
Open source has been the best solution so far. If it has a problem, the "open hood" policy allows your local mechanic can fix it, or determine what the original programmer wanted the user to do in the first place.
Actually, this patent is probably never going to be litigated. Those website's who've already paid their $30K-$80K extortion to these guys should have held back.
The patent office is reviewing this patent, and it's expected to be thrown out upon review (which is why TechSearch has offered web sites to pay "just" $10K now, and $60K after the completed review doesn't reject the patent, to save web sites the huge license fee they're going to charge if the patent isn't overturned). Even the thugs at TechSearch know they've lost this one, and they're just trying to squeeze a few more dollars before the end.
During review, 64% of patents have their claims changed -- this patent is expected to be completely overturned by prior art that goes back to the 50's.
Digital Cable and WebTV better watch out...
Not to give these fools any more ideas on who they can sue... but this patent seems to be extremely broad!
This
patent seems to cover Digital Cable and WebTV.
>"Secondly, the remote query and data retrieval system can be connected
to the remote host via an auxiliary digital I/O line 37. An example of such
a line is derived from a commercial FM broadcast station and uses one of
its two sidebands to transmit digital data (so called "silent radio"). This
is a low bandwidth (under 2400 baud) channel of the type used to provide
announcements on "Times Square" like displays."
The Atvef-A data, used by Microsoft's WebTV gets data from TV broadcasts
in just this manner, and that data is used to retrieve AV data.
>"Third, as noted above, the EUS 10 can interface the host 11 via a
video input line 26a. This type of signal is derived, for example, from the
Cable TV (CATV) Adapter's (i.e. Sprucer) 22 identified above. In an operational
example, a remote query could be initiated via the adapter remote control
24 which would then be transmitted to the local cable service center 26 which,
by prearrangement, will forward it to the server 11. The query, and its associated
response, may be preformulated and prestored at the server 11 or even at
an EUS requiring only periodic updates via a packeted query answer (i.e.
PQA; a Prestored Query Answer). After being cleared via an appropriate security
check (e.g. subscriber password), the server 11 may initiate an AV (Audio-Visual)
response to the EUS 10. The channel through which the specified AV response
is sent to the EUS 10, and as also designated by pre-arrangement, is tuned
via the CATV adapter."
This is how Digital Cable works -- which all cable companies are migrating
to.
This patent seems to center around Audio/Video realtime compression and decompression over phone lines to a personal computer.
This may have been theoretically possible at the time, but even RealVideo (TM), today, couldn't do this on a 1991 state-of-the-art '286 over a 9600 bps modem!
I thought the patent office required that the idea be producible?
I wish Sony wouldn't do proprietary memory sticks.
I thougth they'd tried that (years ago) before and found nobody would buy. I know I've been burnt by proprietary memory (10x the cost of regular memory), and have no desire to buy any product that doesn't use a standard memory chip.
Slashdot Mirror
>"I wouldn't mind getting the service for a discount if I consented to my viewing habits being sold, or for free if they actually attached my name to it."
Since Tivo is not making a profit (and is well under water, except for the AOL bail-out), you can assume that all Tivo users are currently "getting the service at a discount".
Bull.
I spent a week installing Win98 on an old P200 w/ no existing OS.
In fact, Microsoft says you'll have the best experience only if you buy a
brand-new system with XP preinstalled. That's a pretty hefty
investment in an untried operating system. Unless you're planning to
buy a new system when Windows XP hits the shelves, your XP
experience likely won't be as smooth as Microsoft hopes.
an upgrade you should consider lightly.
We keep hearing that Linux is hard to install.
User's think it's a problem with the OS, but that's totally misguided. The real issue is that OS's are tough to install and integrate on raw Open Hardware systems (search for "Compaq" to see how they reverse engineered the PC bios, igniting the Open Hardware revolution), and Microsoft doesn't allow the major OEM's to install anything but Windows (See the "Findings of Fact" [section V.C.4 for a most interesting study] in the Microsoft antitrust trial).
Try installing Windows on a raw system (with no OS or other OS). It is just as difficult.
Now they come up with an OS that only installs on very specific hardware, and only atop W98.
My guess is: Windows' lemmings won't complain a bit (they never do; they just get their brother-in-law to fix it for free).
He expressly gave the right for all to use ssh in his original headers many years ago.
I wish he didn't desire to withdraw that now, as he's done a great job for all of us.
At least the publicity, even in loosing, lets everybody know who started this and who sells it commercially.
I bet /. could sell Karma to the whores!
The provisional patent is intended to allow you to sell your idea to a company withing twelve months, then let them go thru the expensive patent process.
It does have it's drawbacks (many):
For example, you loose your provisional patent filing date if you don't file for patent the within the 12 month period.
I don't think this is the proper solution. Letting it lapse gives you no claim to prior art.
Wouldn't the USPTO be the best registry for software?
A commercial venture, or even a non-profit rganization may collapse, and isn't controlled by regulation, and may or may not meet standards (current and future), and can't guarantee that their database hasn't been comprimised.
This is the job of the USPTO, and they should provide the registration service for free. It would help them get out of their software patent mess.
>You can build "tolerance" into software
You need to understand that increasing the volume of software and it's complexity may logically cover the assumed problem space more thoroughly, BUT, it simultaneously increases the probability of indirect program failure. Furthermore, you may think you understand the problem space thoroughly, but you can't guarantee it (how often do we see an exactly accurate computation of the wrong problem, i.e. the Hubble Space Telescope mirror was ground within exacting tolerances of the wrong shape).
>"The fact that you are espousing the view that we shouldn't even try in the software industry sounds to me that you're [name calling omitted]"
I don't mean to sound like a totally haphazard approach is best. I do believe in standards and requirements (etc...), but they have diminishing returns: For example: if you spend longer in the design phase, you might save some time in the integration and debugging phases, but it's returns diminish radically as you increase the time spent on design (or requirements). Youthful programmers want to spend eternity in the requirements, design, and coding phase... and have a distaste for integration and debugging. The tools provided for software design analysis are a joke. They create more problems than they solve, and are only cost effective when you really have to show due diligence with software that might kill someone (I know: I've both written and used such software)
In the analog world, a thorough design is measurable, in software, it becomes a black hole.
>"...and software engineers are not the be-all-end-all of the engineering world."
I never said, nor believe, otherwise.
>"And if you're curious as to what I consider as the most complex thing ever developed by Man, it's language."
Isn't it curious how there can be such a wide range of speech that's still intelligible by even the most ignorant human? That's the great thing about the analog world: tolerances.
Any attempt we make at computer based speech recognition (AI or neural nets or even patter matching) becomes bogged down in miles of code when we even begin to tolerate variations. While the increase in code seemingly makes software tolerant, it simultaneously increases the complexity and probability of indirect failure.
When we truly understand analog logic, we'll probably find speech recognition very complex, but not as complex as we made it out to be with digital logic.
Damn! I thought this was a good example. I recall at the time I thought the only reason IBM did it was for proprietary lock-in... but I don't have any evidence (and, in general, I don't recall anything well).
Does anybody remember back before the MS monopoly to an example of IBM making proprietary protocols/standards/software to lock-in customers? The PS/2 hardware and it's proprietary bus (I forget the name of the bus) is a great example of them trying to shutdown the Open PC Hardware revolution and regain their monopoly -- but that tactic didn't work.
I need a good example.
It's like we turned the clock back a year+... I thought crazy ideas were out, in favor of profitable ideas?
Oh well, the crazy ideas were a lot of fun.
It does sound like it would be economical to replace all those cell phone towers with one high-flying antenna... but a baloon sounds more efficient than a plane.
The article said "Solar powered" -- I don't read any mention of Solar Powered on their web page.
The appeals court main concern in the Microsoft remedy was that we're just replacing one monopoly for another.
The prevailing software business model is unique with it's "proprietary lock-in". Assuring lack of competition in software applications.
Proprietary formats, standards, protocols, etc... lead to the "software vendor lock-in". Microsoft didn't invent this... they were doing ASCII along with everyone else when IBM was doing EBCDIC (Later, of course, MS with their monopoly would do TrueType while everybody else had already standardized to Postscript). They're also not the only software vendor to do this: every Unix consortium is comprised of vendors using one hand to shake, the other to stab in the back (trying to get others to abide by the standard, while they proprietarily create modifications to the standard).
Open source has a chance to create a level, standards based, playing field. Allowing true, balanced, competition.
When Compaq, and others, reverse engineered the IBM PC Bios, the PC revolution was ignited: anybody could make a motherboard or an adapter card, or memory (The MS OS and the Intel CPU were the only proprietary pieces left atop this open hardware platform). The prices came down, the innovation moved at record pace.
But, even if the current "proprietary lock-in" paradigm ends, the patent office, in granting frivilous software patents, may be providing the next lock-in paradigm.
Can and will congress legislate and mandate the proper regulations aimed at the software industry to end the "proprietary lock-in" paradigm (this also means modifying the DMCA to omit those portions that protect the lock-in) and assure that there are sufficient restrictions in the patent laws to limit the number and scope of software patents?
>"At least, I hope so .. otherwise, we're stuck with this ad hockery that we do now..."
We've actually been stuck here for quite some time. Long enough to know that the analogy to the analog world makes perfect logical sense, but is nonsense when implemented on an application of modest complexity.
>"Exactly. As an ex-Boeing engineer"
Obviously not a software engineer.
It's the simplicity of the software that allows full path testing. It takes years to test such simple code, and the user pays through the nose; and rightfully so.
Once the software gets complex, pilots & civilians die. There's plenty of examples of software requirements bloat projects in the military and aerospace that get quitely get swepped under the rug.
>but because of the robustness of the systems
...and the simplicity of the software.
Was it the Mars Polar Lander that interprteted its' own parachute deploying as having landed, therby turning off the brake thrusters and making a deep dull thud? Or, was it the one that got english/metric units fouled up causing catastrophic failure?
I assure you, the complexity of the software has increased beyond the point of rigorous testing, and that's made all the difference in the world
>"I'm not sure that I completely agree with you. It seems that you're saying that digital devices are incapable of having any degree of tolerance. That's not true."
Tell me that "char foo[20]" is +/- 10% of twenty bytes. You step one byte beyond 20, and the OS could come down; as if a 747 would fall from the sky if the hinge on an overhead bin failed.
When an engine fails on a 747, there's three others to take it's place. When Word fails, does it have a backup editor that catches your file and allows you to continue editing? Sometimes we have signal handlers to clean up our mess -- but they often fail in a crisis too.
Software has zero tolerances. It either does what it was told to do, or it fails. You can cover more cases to try to hedge failure, but you simultaneously increase the complexity and the chances for failure.
If you're building software that means life or death, then you keep it small and you do go to the effort of understanding all failure modes, and you don't launch it atop Windows (the FDA wouldn't allow it), and you charge a mint for it. But this isn't typical software, and the effort necessary to apply these concepts to moderately complex applications requires that the effort grow exponentially.
You call me arrogant when you claim you're so much smarter than the rest of the industry and know what the solution is. Reply back in 20 years, once you've found that you don't know all the answers.
>"It's also a function of how software is developed. The process by which a car (or a toaster, or an airplane) is designed, built and assembled is 50 or more years mature."
We've been comparing software development to analog-world development like this for quite a long time, still, no "silver bullet" solutions.
Someday, some conversion of consciousness will make us laugh that we thought the two were relatable. For now, all we can do is take the heat.
Airplanes are analog devices... they can survive within tolerances.
(You have to read the whole text before yelling "bullshit")
Furthermore, you obviously don't understand complex software. Engineers with the required Basic/Fortran class often mistake their simple assignements for actual, useable, integrated, software.
The software on your PC has more components than the hardware on your 747.
You can still build a "failure mode tree" for a 747... we realized long ago that such analysis of software was futile (you can do it for firmware, control systems, etc... but not integrated applications).
If builders built buildings the way programmers wrote programs, the first woodpecker to come along would destroy civilization!"
I first heard this user mantra in `82 with my first programming job -- and they said that was an old adage.
The problem isn't programmers lack of responsiveness to users, as has been suggested for the last 40 years. If that were true, it would have been solved by now.
The true problem is the inhearent complexity of software, where any useful integrated program enters the realm of chaos, and exhibits behavior "as if at random".
It's digital nature makes it more susceptible. While you can plumb a toilet within wide tolerances, software must be exact. Furthermore, a broken toilet doesn't take the city's sewer system down with it.
It's ease of modification makes it even more susceptible. A problem in hardware will be there for years, we'll learn to work around it, and it may become the standard. But with software, the fix (and the next set of bugs) will come with the next upgrade or patch.
The fellows suggestion that "speech recognition will cure this" is another example of how requirements bloat, to solve "the problems of software usability", exacerbates the problem.
Some problems need to be blamed on the programmers and management: the Window's kernel hung around much too long. Microsoft kept adding mounds of complexity with small doses of functionality to keep the ever faster processors busy; it was no wonder you couldn't keep it up.
Open source has been the best solution so far. If it has a problem, the "open hood" policy allows your local mechanic can fix it, or determine what the original programmer wanted the user to do in the first place.
Actually, this patent is probably never going to be litigated. Those website's who've already paid their $30K-$80K extortion to these guys should have held back.
The patent office is reviewing this patent, and it's expected to be thrown out upon review (which is why TechSearch has offered web sites to pay "just" $10K now, and $60K after the completed review doesn't reject the patent, to save web sites the huge license fee they're going to charge if the patent isn't overturned). Even the thugs at TechSearch know they've lost this one, and they're just trying to squeeze a few more dollars before the end.
During review, 64% of patents have their claims changed -- this patent is expected to be completely overturned by prior art that goes back to the 50's.
Digital Cable and WebTV better watch out... Not to give these fools any more ideas on who they can sue... but this patent seems to be extremely broad!
This patent seems to cover Digital Cable and WebTV.
>"Secondly, the remote query and data retrieval system can be connected to the remote host via an auxiliary digital I/O line 37. An example of such a line is derived from a commercial FM broadcast station and uses one of its two sidebands to transmit digital data (so called "silent radio"). This is a low bandwidth (under 2400 baud) channel of the type used to provide announcements on "Times Square" like displays."
The Atvef-A data, used by Microsoft's WebTV gets data from TV broadcasts in just this manner, and that data is used to retrieve AV data.
>"Third, as noted above, the EUS 10 can interface the host 11 via a video input line 26a. This type of signal is derived, for example, from the Cable TV (CATV) Adapter's (i.e. Sprucer) 22 identified above. In an operational example, a remote query could be initiated via the adapter remote control 24 which would then be transmitted to the local cable service center 26 which, by prearrangement, will forward it to the server 11. The query, and its associated response, may be preformulated and prestored at the server 11 or even at an EUS requiring only periodic updates via a packeted query answer (i.e. PQA; a Prestored Query Answer). After being cleared via an appropriate security check (e.g. subscriber password), the server 11 may initiate an AV (Audio-Visual) response to the EUS 10. The channel through which the specified AV response is sent to the EUS 10, and as also designated by pre-arrangement, is tuned via the CATV adapter."
This is how Digital Cable works -- which all cable companies are migrating to.
This patent seems to center around Audio/Video realtime compression and decompression over phone lines to a personal computer.
This may have been theoretically possible at the time, but even RealVideo (TM), today, couldn't do this on a 1991 state-of-the-art '286 over a 9600 bps modem!
I thought the patent office required that the idea be producible?
[XYZ]modem all predate this patent.
They were specifically designed for compressing data across a slow communications link.
In Unix, ftp sites would use "compress" (.Z files)... long before '91.
I wish Sony wouldn't do proprietary memory sticks.
I thougth they'd tried that (years ago) before and found nobody would buy. I know I've been burnt by proprietary memory (10x the cost of regular memory), and have no desire to buy any product that doesn't use a standard memory chip.
There are not enough distributions.
For example, who's distribution is aimed at the average luser (now that Corel is dead)?