do you actually have a single concrete example of why software patents are sooooooooo bad
That's a really tough question. Picking a single example is not easy since there are so many good ones. Maybe you can help. Start with the LPF's examples to get you the "best of" through 2002. Then read up on the whole Blackberry/email mess. After that maybe we'll have to roll dice or something to pick a single one.
If IBM could have patented software from the get-go, I'm sure they would dominate the market. However, it would be a much smaller market. See quotes against software patentability.
I'm the only one at work who spools things from a Linux box. I occasionally see other people's jobs freeze and have to ask them to delete so mine can get through, but it's never gone the other way around. Sorry you had the opposite experience. I've been printing for about 5 years.
My wife caught CNN imitating Fox's distortion of the Libby case yesterday. This is not atypical of what we see on CNN. Unlike Fox, CNN sometimes covers the left point of view, but that does not make them left-leaning. Right-wingers generally accuse CNN of being left-wing to make themselves seem more centrist.
What Diehr said is that a claim does not become unstatutory just because software is a part of it. They affirmed Benson and Flook, in that if the only novel/non-obvious part of a patent is the software, then the invention as a whole cannot be taken as novel/non-obvious. They also pointed out you can't just remove the software and look at what's left. The way the software interacts with other elements may be novel and non-obvious. However, practically none of the "software patents" that are discussed on slashdot fit that description. The kind of invention Diehr opened the way for would include software elements, but wouldn't be what we call a "software patent".
Yes, I'm aware lots of people see it differently. That's because lots of people are wrong. Hardly anyone reads the case carefully. For example the oyez summary incorrectly attributes to Stevens the idea that no software can ever be patentable. He said no such thing, and in fact said he had no difference of law with the majority, only difference of fact (his reading of the claims.) The bitlaw summary says the majority saw the software as the only novel part of the invention. The majority said the opposite. I don't care that lots of people agree that Diehr opened the way for software patents. Lots of people are wrong.
Innovating is not a problem. If you come up with something that is truly novel and not obvious in software, there's hardly any chance it's patented. The problem is coding straightforward solutions to day-to-day problems. That's where you'll trip all over hundreds of patents.
If Congress were to pass a law granting "exclusive rights" to smiling it would be ruled unconstitutional. The powers of Congress are fairly broad in that a lot can fall under the umbrella of promoting general welfare. But ridiculous laws that forbid smiling are overstepping constitutional authority. It's similarly ridiculous to say you cannot put an idea to use that you came up with yourself just because someone else filed a patent on it. However, this is authorized by the constitution in the context of promoting progress. A few bad patents can be expected, but if the system as a whole impedes progress rather than promoting it, then it's unconstitutional.
In copyright terms a "license" is what you need if you're going to infringe the exclusive rights of the copyright holder. This case helps define what "essential" means in US statute. If your adaptation is an essential step in running a program that you lawfully own a copy of, then you are not infringing any exclusive rights and don't need a "license agreement" to permit you to make such adaptations.
EULAs are no different from other potential contracts. To be valid contracts there has to be "consideration" on both sides. E.g. you give money in exchange for rights. If you've already given your money for a copy and you already have the rights, there's no consideration and no contract. But some courts have held that a prominent notice on the outside of the package serves to tie an agreement to a copy you buy. For serious questions about whether you can ignore an EULA retain a lawyer.
If you buy their product, wouldn't it make sense you have to follow their rules?
No, that makes absolutely no sense. When you buy a product it's yours.
The copyright holder has certain exclusive rights, and those rights are not infringed when you need to adapt a copy you own to use it. Copying CDs to audio tape to play in the car has always been considered fair use, for example, even when record companies could sell you both CD and tape versions.
It has to do with a 1980 US law that explicitly mentions computer programs. Clearly it was meant to eliminate the "every time you run a program you make a copy" argument that people still try to foist on everyone today. Apparently they didn't anticipate the "every time you open a file you make a copy" argument at that time.
For running a small web site, Apache is probably too sophisticated a server, and Linux/BSD/etc are too sophisticated as operating systems. However, if you already have a sophisticated OS, server and database, there's no reason not to use them.
Hopefully from now on all worms and viruses will be named according to the OS they affect. I'm tired of hearing Windows worms/viruses referred to as if they were affecting all computers everywhere.
You ignored "subset" in my post. PostgreSQL hierarchical tables are not part of that subset of SQL and I never use them.
You restrict yourself to "simple" EJBs. Yes, these can be made in 2 minutes. Doing real-world work such as in my example cannot be done in 2 minutes with EJBs, but can be done in 2 minutes if you know SQL.
I've been doing SQL since 1997, and every time I've used GROUP BY I've used exactly the non-aggregate expressions in the SELECT clause. Can you give an example where it would be something else? I too would like GROUP BY DEFAULT or somesuch.
By "real database" I mean nothing more than that subset of SQL that is portable between PostgreSQL, DB2 and Oracle, and easily adaptible to Transact-SQL (MSFT, Sybase). I gave a simple and concrete example of something that takes two minutes to implement in SQL and 2 days to implement using EJBs. You retorted with speculation about my experience. Answer my example first.
Changing database vendors would involve straightforward porting, not rewriting. But even if you did change database vendors five times and did a rewrite each time, it would still make more sense to redo the 2 minutes worth of work six times rather than spend two days and be done with it.
I hope you're right that a lot of techies realize SQL skills are necessary. Got a pointer to any info about Amazon's use of EJB? I'll let go of my lovely queries when it's no longer true that they can be coded quicker and run faster than any of these "abstraction layers" designed to cram the relational model into a limited object-oriented model.
Thanks for the pointer. It confirms what had previously been my sneaking suspicion. Referring to a database as "data persistence" shows ignorance of at least the reporting capabilities of a real database, and likely other things as well. These "object persistence" abstractions will make a 2-day project out of a report that would take 2 minutes to write in SQL. E.g., you wanted to see what magazines people in zip code 90210 subscribe to in order of popularity.
No wasted space and every column is relevant in my example. EVERY time I want EVERY column, no exceptions.
Then your example is contrived. In the real world somebody would ask, "How many people checked item 100?" or "How many people who checked item 33 also checked item 99?"
I'd just as soon give the DB a time interval to analyze and decide by itself what indexes to create based on the SELECT/INSERT/UPDATE/DELETE operations that happen in that interval.
It's amazing to me how right-wingers on slashdot shoot themselves in the foot by citing actual documents rather than just making things up like Bill O'Reilly. Bill O'Reilly knows what he's doing.
Thank you for pointing out the $200B deficits under Clinton. They were a huge improvement. Look at the 4th paragraph on page 3 about three straight years of deficit reduction for the first time since Harry Truman was President. Thank you for pointing out that this was in 1996, before the dot-com boom. After the dot-com boom the deficits disappeared and we finally had a surplus, only to be squandered by Republican don't-tax-just-spend budgets under GW Bush.
Clinton, while less than perfect, at least had some economic sense. Republicans now attribute his success to luck. From Clinton's 2000 DNC speech: "The Republicans said then they would not be held responsible for the
results of our economic policies. I hope the American people will take
them at their word."
Slashdot Mirror
If IBM could have patented software from the get-go, I'm sure they would dominate the market. However, it would be a much smaller market. See quotes against software patentability.
Um, you may have missed this news, but it's now possible to make prints from digital too.
I'm the only one at work who spools things from a Linux box. I occasionally see other people's jobs freeze and have to ask them to delete so mine can get through, but it's never gone the other way around. Sorry you had the opposite experience. I've been printing for about 5 years.
My wife caught CNN imitating Fox's distortion of the Libby case yesterday. This is not atypical of what we see on CNN. Unlike Fox, CNN sometimes covers the left point of view, but that does not make them left-leaning. Right-wingers generally accuse CNN of being left-wing to make themselves seem more centrist.
What Diehr said is that a claim does not become unstatutory just because software is a part of it. They affirmed Benson and Flook, in that if the only novel/non-obvious part of a patent is the software, then the invention as a whole cannot be taken as novel/non-obvious. They also pointed out you can't just remove the software and look at what's left. The way the software interacts with other elements may be novel and non-obvious. However, practically none of the "software patents" that are discussed on slashdot fit that description. The kind of invention Diehr opened the way for would include software elements, but wouldn't be what we call a "software patent".
Yes, I'm aware lots of people see it differently. That's because lots of people are wrong. Hardly anyone reads the case carefully. For example the oyez summary incorrectly attributes to Stevens the idea that no software can ever be patentable. He said no such thing, and in fact said he had no difference of law with the majority, only difference of fact (his reading of the claims.) The bitlaw summary says the majority saw the software as the only novel part of the invention. The majority said the opposite. I don't care that lots of people agree that Diehr opened the way for software patents. Lots of people are wrong.
SCOTUS did not legalize software patents. If Diamond v. Diehr and prior rulings were followed, we would not have software patents today.
Innovating is not a problem. If you come up with something that is truly novel and not obvious in software, there's hardly any chance it's patented. The problem is coding straightforward solutions to day-to-day problems. That's where you'll trip all over hundreds of patents.
If Congress were to pass a law granting "exclusive rights" to smiling it would be ruled unconstitutional. The powers of Congress are fairly broad in that a lot can fall under the umbrella of promoting general welfare. But ridiculous laws that forbid smiling are overstepping constitutional authority. It's similarly ridiculous to say you cannot put an idea to use that you came up with yourself just because someone else filed a patent on it. However, this is authorized by the constitution in the context of promoting progress. A few bad patents can be expected, but if the system as a whole impedes progress rather than promoting it, then it's unconstitutional.
In copyright terms a "license" is what you need if you're going to infringe the exclusive rights of the copyright holder. This case helps define what "essential" means in US statute. If your adaptation is an essential step in running a program that you lawfully own a copy of, then you are not infringing any exclusive rights and don't need a "license agreement" to permit you to make such adaptations.
EULAs are no different from other potential contracts. To be valid contracts there has to be "consideration" on both sides. E.g. you give money in exchange for rights. If you've already given your money for a copy and you already have the rights, there's no consideration and no contract. But some courts have held that a prominent notice on the outside of the package serves to tie an agreement to a copy you buy. For serious questions about whether you can ignore an EULA retain a lawyer.
No, that makes absolutely no sense. When you buy a product it's yours.
The copyright holder has certain exclusive rights, and those rights are not infringed when you need to adapt a copy you own to use it. Copying CDs to audio tape to play in the car has always been considered fair use, for example, even when record companies could sell you both CD and tape versions.
It has to do with a 1980 US law that explicitly mentions computer programs. Clearly it was meant to eliminate the "every time you run a program you make a copy" argument that people still try to foist on everyone today. Apparently they didn't anticipate the "every time you open a file you make a copy" argument at that time.
For running a small web site, Apache is probably too sophisticated a server, and Linux/BSD/etc are too sophisticated as operating systems. However, if you already have a sophisticated OS, server and database, there's no reason not to use them.
Hopefully from now on all worms and viruses will be named according to the OS they affect. I'm tired of hearing Windows worms/viruses referred to as if they were affecting all computers everywhere.
I've been doing SQL since 1997, and every time I've used GROUP BY I've used exactly the non-aggregate expressions in the SELECT clause. Can you give an example where it would be something else? I too would like GROUP BY DEFAULT or somesuch.
By "real database" I mean nothing more than that subset of SQL that is portable between PostgreSQL, DB2 and Oracle, and easily adaptible to Transact-SQL (MSFT, Sybase). I gave a simple and concrete example of something that takes two minutes to implement in SQL and 2 days to implement using EJBs. You retorted with speculation about my experience. Answer my example first.
Changing database vendors would involve straightforward porting, not rewriting. But even if you did change database vendors five times and did a rewrite each time, it would still make more sense to redo the 2 minutes worth of work six times rather than spend two days and be done with it.
I hope you're right that a lot of techies realize SQL skills are necessary. Got a pointer to any info about Amazon's use of EJB? I'll let go of my lovely queries when it's no longer true that they can be coded quicker and run faster than any of these "abstraction layers" designed to cram the relational model into a limited object-oriented model.
Thanks for the pointer. It confirms what had previously been my sneaking suspicion. Referring to a database as "data persistence" shows ignorance of at least the reporting capabilities of a real database, and likely other things as well. These "object persistence" abstractions will make a 2-day project out of a report that would take 2 minutes to write in SQL. E.g., you wanted to see what magazines people in zip code 90210 subscribe to in order of popularity.
I'd just as soon give the DB a time interval to analyze and decide by itself what indexes to create based on the SELECT/INSERT/UPDATE/DELETE operations that happen in that interval.
It's amazing to me how right-wingers on slashdot shoot themselves in the foot by citing actual documents rather than just making things up like Bill O'Reilly. Bill O'Reilly knows what he's doing.
Thank you for pointing out the $200B deficits under Clinton. They were a huge improvement. Look at the 4th paragraph on page 3 about three straight years of deficit reduction for the first time since Harry Truman was President. Thank you for pointing out that this was in 1996, before the dot-com boom. After the dot-com boom the deficits disappeared and we finally had a surplus, only to be squandered by Republican don't-tax-just-spend budgets under GW Bush.
Economic conservatives endorsed Kerry mainly because Bush was and is so incredibly bad. The damage from Bush deficits will be with us for years to come.
Clinton, while less than perfect, at least had some economic sense. Republicans now attribute his success to luck. From Clinton's 2000 DNC speech: "The Republicans said then they would not be held responsible for the results of our economic policies. I hope the American people will take them at their word."