It seem to me more and more patents are being ruled as invalid, If this is indeed the case why are they being assigned in the first place ?...
I wonder how many patents would stand up to a further examination.
There are those who would see this as a vindication of the current system.
What's the harm in issuing bad patents when they are inevitably invalidated?
I do not share this veiw.
Palm has obviously been damaged by Xerox's patent aggression: the cost to license Jot, the R&D and marketing to incorporate it into the product, and the consumer confusion, to say nothing of the effort to actually invalidate the patent.
Until its incentive structure changes, the patent office will continue to hand out patents like leis at a luau.
A recent Slashdot article notes that Microsoft is applying for ten a day.
That's called working the system (or, as those who have already worked the system say, catching up).
I'm willing to believe that a few tens of thousands of the world's highest paid engineers can come up with a few novel ideas, but I also believe that Edison was too liberal in his assessment that genius is only 99% perspiration.
it's horrifyingly expensive to [dial overseas] using stock long distance service
No kidding!
My wife called me in Moscow twice, for the princely sum of $500.
(I think it was a little over $4 per minute).
I hadn't expensed anything else on the trip, so the company paid for the calls without objection.
I later found out that AT&T had an international plan that would have lowered the cost to about $5 per month plus twenty-five cents per minute.
Would you put Gamebyro (nee NetImmerse) in the same category?
I was interested to see that Mad Doc is using it for Empire Earth 2, even as Stainless Steel Studios is introducing an engine licensing program of its own.
take this register-starved, CISC-mired turkey out back and give it a proper burial
This aging architecture has maintained an incredible price/performance ratio.
At this price level, the only thing that compares is the G5.
A comparable UltraSPARC, Itanium, POWER, or PA-RISC system will cost much more.
As for registers, AMD64 doubled the number of general-purpose registers, which are already subject to register renaming.
IMO, it would have made sense to put a checkbox for it in File Management Preferences.
But come on, dismiss an entire desktop because of the lack of one checkbox?
Outrageous!
I haven't looked at GNOME 2.6, yet, but I've been troubled by the "have it our way" philosophy of the likes of Havoc Pennington and Marco Pesenti.
There was a discussion on the Epiphany list about abolishing the font dialog, because the user may not understand "serif" and "monospace."
Petreley's point is that you ignore the power user at your own peril.
The absence of configuration options (and don't kid yourself that the debate is about just one option) serves to polarize the user base into the "daft simpletons" and the gurus.
There's a healthy middle class of users who have no patience for crap like GConf, when it comes to a simple preference like not littering the desktop with innumerable Nautilus windows.
Re:id's games are not entirely designed to make mo
on
DOOM III This Summer
·
· Score: 2, Informative
Keep in mind that client are only part of id's business.
The other part, and possibly the greater part, is licensing their game engines to other developers.
Their game are not strictly intended to be money makers, they are also working models and advertisements for the engines.
I used to think the same, but I read an interview in which an id executive claimed it was a surprisingly small portion of their revenue.
I don't have the link handy, but Google found a quote from Todd Hollenshead: "engine licensing accounts for 20% of id Software's revenue."
20% is a big piece of the pie, but it seems the games are still the moneymakers.
why wait nearly 10 years?...
can you really [s]ue if no money is being made?
Kodak only acquired the patents in 1997.
Wang didn't know or didn't care about the alleged infringement.
"A patentee has the exclusive right to make, use, or sell the invention...
Anyone who, without permission, makes, uses or sells the patented invention is a direct infringer of the patent" [Intellectual Property in a Nutshell].
What if SCO never fired the first shot in this IP war?...
Would it be possible for SCO to continue to exist supplying to main stream folk or would they have simply continued to fade away?
We could consider numerous "what-if" scenarios.
Most of the good ones would require smarter, more visionary executives than Caldera or SCO ever employed.
Part of Red Hat's continued success is due to strategic investments in research and development, especially the kernel improvements by Alan Cox, Stephen Tweedie, and Ingo Molnar.
SuSE has made similar investments, garnering the attention of IBM and Novell.
I don't think OpenServer and UnixWare were good enough, especially under the care of such a small company, to compete with Linux or BSD.
System V was a significant, though largely symbolic, asset.
Caldera could have parlayed it into success in the Linux market.
(Who knows: UnitedLinux might have grown to 1.1 or 2.0.)
Instead, it was eaten from the inside out by old-SCO opportunists.
I'd like to have either a 2.5-3lb subnotebook with a nice 12" screen (and preferably below $1k, like the Servelinux), or a ~4lb notebook that gets a much longer battery life than anything else on the market (besides maybe a Mac)
I'm looking for a light notebook with long battery life.
I thought about going without a CD-ROM drive, and the ThinkPad X40 looks good in this category.
However, I'd like to watch movies and play the occasional copy-protected game, so I'm waffling between an iBook and the ThinkPad T41.
The iBook advertises an improbable six hours of battery life.
The T41 can swap the CD-ROM drive for a battery.
can't anyone see the appeal of using chips like these in a ultra-quiet desktop model?
Home builders clamored for the K6-2+ and K6-III+ processors for the same reason, but they were difficult to obtain.
I happily ran a Celeron 850 (~20 W, IIRC) without a CPU fan.
Interestingly, Motorola markets the G4 (7457 and 7447A) used in the iBook, Powerbook, and iMac as an embedded processor.
Now, if only Apple would market a headless iMac, like the old cube...
I'd love to have a Linux version which would be a straight recompile, but that's not possible yet.
I'm aware of and considering GNUStep, but I really would like a straight recompile.
[While I don't love the term GNU/Linux, I'll use it here to distinguish the O/S from the kernel.]
If you paste the (proprietary) OS X GUI onto GNU/Linux, you haven't improved GNU/Linux; you've forked OS X.
This isn't semantics or zealotry: if your app. runs on OSX/Linux, then you aren't really gaining GNU/Linux users.
A significant reason to use GNU/Linux is that it's free (beer and speech), so a free solution like GNUstep is probably your best bet for an easy port.
That said, I'm not completely sold on the idea of a cross-platform GUI.
If you care about the user experience, then it's important to be native, meaning that in addition to using native widgets, you observe the conventions.
For now, the easiest way to do this is to actually use the target O/S's framework.
Recognize this early and architect your app. to make it less painful.
[CIOs] don't want more source code...
if you write to the Red Hat distribution, you can't go and run on Debian...
open source does not mean open standards
The basic problem is that video games are a poor format for story telling.
It's a new field with no masters, but I I hope you're wrong.
Consider the Kuleshov effect, described in Hamlet on the Holodeck, as follows:
Lev Kuleshov demonstrated that audiences will take the same footage of an actor's face as signifying appetite, grief, or affection, depending on whether it is juxtaposed with images of a bowl of soup, a dead woman, or a little girl playing with a teddy bear.
I seem to recall one of the Mapplethorpe photos playing a similar trick with the words "fond" and "fondle."
In sufficiently skilled hands, I think this can be a powerful technique to prune the exponential branches a thorougly non-linear story would require.
Now, I'm not saying that there shouldn't be any story in a game, far from it, I like stories in games, but the most memorable story is the one that you create by yourself while playing the game.
Amen.
Story telling is fine, insofar as it goes, but the real promise of the medium is interactivity.
The most critically acclaimed games have told decent stories (System Shock 2, Half-Life, Max Payne), but were almost completely linear.
I keep hearing that the game industry doesn't nead any more idea people; if that's true, then why does each year bring more of the same?
Video games will take an important step when they match Choose Your Own Adventure and Clue for interactivity.
If you're a game designer, please read the following books: Janet Murray's Hamlet on the Holodeck, Chris Crawford's Chris Crawford on Game Design, and Andrew Rollings's Game Architecture and Design.
I know everyone does it, but it's always a jolt to see a product advertised with a stock photograph I've already seen.
The picture of the woman displaying a "thumbs up" sign is used by a local copy shop (without the slingshot).
That's like moving into someone's summer home while they're away at their winter home, and bitching when you have to move out when they come back in June....
Okay, bad analogy
Yes, the analogy is bad, as are most comparisons of "intellectual property" to real property.
After ten years of open and notorious use, including an ISO standard, a speculator buys a patent on which JPEG possibly infringes.
Consider that some states allow real property to be stripped by adverse possession in that amount of time.
It's more like an out-of-state lawyer bought your neighbor's lot after discovering that your house overlaps it by a foot.
I let this transpire for almost 18 years then BAM! i sue everyone who is using my patent with only a few months left on my patent.
The defendants will claim noninfringement, implied license, estoppel by laches, and that the patent is invalid.
At worst, they should be barred from continuing to infringe the patent.
Some of them may be able to live six months without it, depending on how much stock is already in the retail channel.
The EFF page links to FTC and NAS recommendations, which are much too mild.
The overriding problem is patent law--everything else is quibbling over details.
The fight against software patents in Europe may be the most important battle, right now.
This is one area in which the rest of the world should not "harmonize" with the U.S.
As for the details, the most important reform is a change in the incentive structure.
It seems like it's easier to get a patent than a municipal parking permit, because the office subsists on application fees.
If there were a reapplication fee for every rejected application, the office would change overnight from a Walmart greeter to a Viper Club bouncer.
James Gleick's "Patently Absurd" is a decent post-one-click overview of the topic.
[RMS has] taken a basic word, "free" and redefined it.
Without context, "free" has about as much objective meaning as "good."
You'll find a simple description of the context in which RMS uses the word in Categories of Free and Non-Free Software.
Compared to Gosling's parroted pejoratives, like "viral infection," this is a much more useful basis for discussion.
What is it about testers and their negative reactions to developers writing tests?...
What are your criteria for considering a test valid or valuable?...
I would expect testers to gleefully accept developer unit tests.
Hard-earned experience shows that it's risky to depend on developers to test their own code.
It's kind of like playing a chess game with yourself.
That said, I disagree with MosesJones's claim that unit testing is just hacking.
It's an excellent part of a balanced breakfast.
Ironically, if testers do their job too well, developers begin to treat them as a safety net, like the next step after compiling and linking.
I spend too much of my time in this mode, so that there is little time or incentive to take it to the next level; in effect, I write the unit tests that the developer should have written.
The value of a particular test depends on the maturity of both the testing and development process.
If I'm hurting for coverage, I may well incorporate a developer's unit test vebatim, especially if experience shows the feature under test to be fragile.
Ideally, however, an automated regression test has some stricter criteria than "it's already written":
Clarity.
A test should pass or fail.
"Inspect output for anomalous values" is okay for a developer's tool box, but doesn't cut it for an automated regression test.
Granularity.
Related to clarity is an appropriate level of granularity.
I can't necessarily allocate five hours of a nightly run to a detailed set of tests for one API function.
On the other hand, it's difficult to diagnose a failure in an omnibus test that fails if any of a hundred different assertions fails, even it runs in a fraction of the time.
Reliability.
I have a limited amount of time to write tests, run them, and analyse their results.
A test with too many spurious failures wastes my time during analysis.
These failures may occur because of unstated dependencies (e.g., must be run in a certain order), race conditions, etc.
Generality.
If the feature under test is applicable on several platforms and will be relevant to future versions of the product, then I want a general, portable test.
This portability may be best achieved by working within a framework that may not be appropriate for use in a unit test.
Testing involves many of the same complexities as development of the software itself, with a less tangible deliverable and a (usually much) smaller budget.
Probably the most important new feature of OpenGL 2.0 was going to be the GLSL high level shader language.
GLSL is clearly OpenGL 2.0's killer feature.
Check out the GLSL section of 3Dlab's site for some resources.
The slide show from Randi Rost's book tour are on-line as a 9 MB PDF.
Before too long, you should be able to download RenderMonkey, a cool tool for playing with GLSL and D3D shaders.
Unfortunately, NVIDIA's GLSL implementation has a ways to go.
You won't even get the necessary ARB extensions unless you set a debug registry flag (google "ShaderObjects").
After that, good luck.
I'm anxious to see how the industry receives GLSL.
Will we get some decent drivers before Doom 4?
So what you're saying is: we have a failure to communicate.
Much of The Mythical Man Month deals with the difficulty of communication within a programming project.
Too many cooks spoil the broth, so why not make the smallest possible team as productive as possible?
Bridging the gap between the producer and the consumer is the discipline of systems analysis.
Regardless of how productive or gelled the team, is it building the right product?
In the limit, I guess the ideal software is developed by an uber genius for his own use.
Moore's Law is one reason why software still stinks....
If people design software with the assumption that it will be totally obsolete and replaced in 18 months, they create software that is so badly designed that it must be replaced in 18 months.
Software is obsoleted by a relative lack of features, not by a doubling of processor speed.
Moore's law increases programmer productivity by making it feasible to use higher level languages and tools.
Software really would stink if computers never got faster, because time spent optimizing would take away from time spent fixing bugs and adding features.
Slashdot Mirror
There are those who would see this as a vindication of the current system. What's the harm in issuing bad patents when they are inevitably invalidated? I do not share this veiw. Palm has obviously been damaged by Xerox's patent aggression: the cost to license Jot, the R&D and marketing to incorporate it into the product, and the consumer confusion, to say nothing of the effort to actually invalidate the patent.
Until its incentive structure changes, the patent office will continue to hand out patents like leis at a luau. A recent Slashdot article notes that Microsoft is applying for ten a day. That's called working the system (or, as those who have already worked the system say, catching up). I'm willing to believe that a few tens of thousands of the world's highest paid engineers can come up with a few novel ideas, but I also believe that Edison was too liberal in his assessment that genius is only 99% perspiration.
Grafitti 2 is based on Jot. If the license is fully paid up, I don't see Palm going back to the original Grafitti.
No kidding! My wife called me in Moscow twice, for the princely sum of $500. (I think it was a little over $4 per minute). I hadn't expensed anything else on the trip, so the company paid for the calls without objection.
I later found out that AT&T had an international plan that would have lowered the cost to about $5 per month plus twenty-five cents per minute.
Would you put Gamebyro (nee NetImmerse) in the same category? I was interested to see that Mad Doc is using it for Empire Earth 2, even as Stainless Steel Studios is introducing an engine licensing program of its own.
This aging architecture has maintained an incredible price/performance ratio. At this price level, the only thing that compares is the G5. A comparable UltraSPARC, Itanium, POWER, or PA-RISC system will cost much more.
As for registers, AMD64 doubled the number of general-purpose registers, which are already subject to register renaming.
I haven't looked at GNOME 2.6, yet, but I've been troubled by the "have it our way" philosophy of the likes of Havoc Pennington and Marco Pesenti. There was a discussion on the Epiphany list about abolishing the font dialog, because the user may not understand "serif" and "monospace."
Petreley's point is that you ignore the power user at your own peril. The absence of configuration options (and don't kid yourself that the debate is about just one option) serves to polarize the user base into the "daft simpletons" and the gurus. There's a healthy middle class of users who have no patience for crap like GConf, when it comes to a simple preference like not littering the desktop with innumerable Nautilus windows.
I used to think the same, but I read an interview in which an id executive claimed it was a surprisingly small portion of their revenue. I don't have the link handy, but Google found a quote from Todd Hollenshead: "engine licensing accounts for 20% of id Software's revenue." 20% is a big piece of the pie, but it seems the games are still the moneymakers.
Kodak only acquired the patents in 1997. Wang didn't know or didn't care about the alleged infringement.
"A patentee has the exclusive right to make, use, or sell the invention ...
Anyone who, without permission, makes, uses or sells the patented invention is a direct infringer of the patent" [Intellectual Property in a Nutshell].
We could consider numerous "what-if" scenarios. Most of the good ones would require smarter, more visionary executives than Caldera or SCO ever employed. Part of Red Hat's continued success is due to strategic investments in research and development, especially the kernel improvements by Alan Cox, Stephen Tweedie, and Ingo Molnar. SuSE has made similar investments, garnering the attention of IBM and Novell.
I don't think OpenServer and UnixWare were good enough, especially under the care of such a small company, to compete with Linux or BSD. System V was a significant, though largely symbolic, asset. Caldera could have parlayed it into success in the Linux market. (Who knows: UnitedLinux might have grown to 1.1 or 2.0.) Instead, it was eaten from the inside out by old-SCO opportunists.
I'm looking for a light notebook with long battery life. I thought about going without a CD-ROM drive, and the ThinkPad X40 looks good in this category. However, I'd like to watch movies and play the occasional copy-protected game, so I'm waffling between an iBook and the ThinkPad T41. The iBook advertises an improbable six hours of battery life. The T41 can swap the CD-ROM drive for a battery.
Of course, these are all above your $1,000 limit.
Home builders clamored for the K6-2+ and K6-III+ processors for the same reason, but they were difficult to obtain. I happily ran a Celeron 850 (~20 W, IIRC) without a CPU fan.
Interestingly, Motorola markets the G4 (7457 and 7447A) used in the iBook, Powerbook, and iMac as an embedded processor. Now, if only Apple would market a headless iMac, like the old cube ...
[While I don't love the term GNU/Linux, I'll use it here to distinguish the O/S from the kernel.]
If you paste the (proprietary) OS X GUI onto GNU/Linux, you haven't improved GNU/Linux; you've forked OS X. This isn't semantics or zealotry: if your app. runs on OSX/Linux, then you aren't really gaining GNU/Linux users. A significant reason to use GNU/Linux is that it's free (beer and speech), so a free solution like GNUstep is probably your best bet for an easy port.
That said, I'm not completely sold on the idea of a cross-platform GUI. If you care about the user experience, then it's important to be native, meaning that in addition to using native widgets, you observe the conventions. For now, the easiest way to do this is to actually use the target O/S's framework. Recognize this early and architect your app. to make it less painful.
With friends like these ...
It's a new field with no masters, but I I hope you're wrong. Consider the Kuleshov effect, described in Hamlet on the Holodeck, as follows:
I seem to recall one of the Mapplethorpe photos playing a similar trick with the words "fond" and "fondle." In sufficiently skilled hands, I think this can be a powerful technique to prune the exponential branches a thorougly non-linear story would require.Amen. Story telling is fine, insofar as it goes, but the real promise of the medium is interactivity. The most critically acclaimed games have told decent stories (System Shock 2, Half-Life, Max Payne), but were almost completely linear.
I keep hearing that the game industry doesn't nead any more idea people; if that's true, then why does each year bring more of the same? Video games will take an important step when they match Choose Your Own Adventure and Clue for interactivity. If you're a game designer, please read the following books: Janet Murray's Hamlet on the Holodeck, Chris Crawford's Chris Crawford on Game Design, and Andrew Rollings's Game Architecture and Design.
I know everyone does it, but it's always a jolt to see a product advertised with a stock photograph I've already seen. The picture of the woman displaying a "thumbs up" sign is used by a local copy shop (without the slingshot).
Yes, the analogy is bad, as are most comparisons of "intellectual property" to real property. After ten years of open and notorious use, including an ISO standard, a speculator buys a patent on which JPEG possibly infringes. Consider that some states allow real property to be stripped by adverse possession in that amount of time.
It's more like an out-of-state lawyer bought your neighbor's lot after discovering that your house overlaps it by a foot.
The defendants will claim noninfringement, implied license, estoppel by laches, and that the patent is invalid. At worst, they should be barred from continuing to infringe the patent. Some of them may be able to live six months without it, depending on how much stock is already in the retail channel.
As for the details, the most important reform is a change in the incentive structure. It seems like it's easier to get a patent than a municipal parking permit, because the office subsists on application fees. If there were a reapplication fee for every rejected application, the office would change overnight from a Walmart greeter to a Viper Club bouncer.
James Gleick's "Patently Absurd" is a decent post-one-click overview of the topic.
Without context, "free" has about as much objective meaning as "good." You'll find a simple description of the context in which RMS uses the word in Categories of Free and Non-Free Software. Compared to Gosling's parroted pejoratives, like "viral infection," this is a much more useful basis for discussion.
This announcement is the first time I've seen a list of the board members. By whom was he elected?
Hard-earned experience shows that it's risky to depend on developers to test their own code. It's kind of like playing a chess game with yourself. That said, I disagree with MosesJones's claim that unit testing is just hacking. It's an excellent part of a balanced breakfast. Ironically, if testers do their job too well, developers begin to treat them as a safety net, like the next step after compiling and linking. I spend too much of my time in this mode, so that there is little time or incentive to take it to the next level; in effect, I write the unit tests that the developer should have written.
The value of a particular test depends on the maturity of both the testing and development process. If I'm hurting for coverage, I may well incorporate a developer's unit test vebatim, especially if experience shows the feature under test to be fragile. Ideally, however, an automated regression test has some stricter criteria than "it's already written":
Testing involves many of the same complexities as development of the software itself, with a less tangible deliverable and a (usually much) smaller budget.GLSL is clearly OpenGL 2.0's killer feature. Check out the GLSL section of 3Dlab's site for some resources. The slide show from Randi Rost's book tour are on-line as a 9 MB PDF. Before too long, you should be able to download RenderMonkey, a cool tool for playing with GLSL and D3D shaders.
Unfortunately, NVIDIA's GLSL implementation has a ways to go. You won't even get the necessary ARB extensions unless you set a debug registry flag (google "ShaderObjects"). After that, good luck.
I'm anxious to see how the industry receives GLSL. Will we get some decent drivers before Doom 4?
Bridging the gap between the producer and the consumer is the discipline of systems analysis. Regardless of how productive or gelled the team, is it building the right product?
In the limit, I guess the ideal software is developed by an uber genius for his own use.
Software is obsoleted by a relative lack of features, not by a doubling of processor speed. Moore's law increases programmer productivity by making it feasible to use higher level languages and tools. Software really would stink if computers never got faster, because time spent optimizing would take away from time spent fixing bugs and adding features.