would you really want a colonoscopy without the jelly?
Open source and software patents are generally regarded as two things that go together like peanut butter and a punch to the face*. Open-source advocates talk of vague, badly written US patents lurking in the system, waiting only to pounce on unsuspecting devs and condemn them to 1,001 years litigation. Open sourcers actively …
would you really want a colonoscopy without the jelly?
An organisation the size of IBM is bound to include both good and bad. The lawyer who wrote the above evidently knows the theory but not the practice of patents in our time, and is spouting the former in blissful ignorance of the latter.
"without patent protection, the incentives to innovate in the field of software are significantly reduced"
If that were true, there would have been significantly reduced innovation in software before 1994 and most innovations afterwards would be accompanied by a few patents. Riiight.
When I write code, it is (c)jake as soon as I publish it, anywhere.
How do patents help me in this process?
Methinks that IBM Marketing has the wrong end of the stick. Again.
...if you read the brief backwards, actually, IBM are against software patents
Certainly much better than peanut butter and a punch in the face. Seriously, you don't ever want to try an *un*lubricated colonoscopy.
(Although there is some debate over whether oil-based lubricants work better than the standard water-soluble jelly.)
Did IBM just hire someone from Ubersoft marketing or something?
Thanks for the heads up, AC. Valuable Info now commited to System Memory. :-) Is it Friday, already?
must apply to retro-engineered proprietary work too.
So IBM are saying OK Microsoft, we forgive you the OS/2 / Win 95 thing, OK?
water based lubricants seem to require more frequent reapplication in sexual liaisons ;) oil based is however more difficult to clean afterwards. i prefer oil based.
For that to work ALL software would need to be open source, otherwise IBM would simply patent open source versions of others closed source software, cutting the ground from beneath their products.
IBM clearly don't back up their own argument, because nearly all their software is closed source and protected by trade secrets. Apparently trade secrets work extremely well!
As for this: "After all, IBM is each year awarded more US patents than anyone else around - all the while being one of the industry's biggest backers of Linux."
When was the last time you bought any piece of software made by IBM? They don't make PCs any more (sold to Lenova), they don't make hard disks, boxed retail software, etc. .... so this image of IBM as some mighty developer, making new software that powers the world, it's a fantasy.
My view of IBM is of a failing hardward/software vendor turned into a services company/patent troll. Their friendship to Linux is because they lost to Microsoft, it's not altruistic. Their obsession with Patents is because their patent troll approach lets them leach of more inventive companies.
also from IBM
War = Peace
Companies like IBM and Oracle just use open source for their own agenda. Are they open sourcing their truly valuable software properties like DB2 or Oracle's database system? Of course not. Oracle makes a fortune off the fact that MySQL can't compete with it for heavy duty transaction processing. But they do want to harness the polticial mass movement to use to their advantage and against their enemies like Microsoft.
A copyright protects your *expression* of an invention. A patent protects the invention itself.
Here, the invention would be the algorithm, and the expression is the code that implements the algorithm. If the algorithm is patented, you need a license from the patent owner in order to create an implementation. Your copyright prevents others from legally duplicating your code without your permission.
Not saying I favor software patents. I don't. Also, IANAL.
"A patent protects the invention itself."
Yes. I know.
"Here, the invention would be the algorithm,"
I disagree. "Algorithm" is just a fancy way of saying "Math(s)" ... which existed long before humans learned to count stuff ... or indeed, before humans existed.
"and the expression is the code that implements the algorithm."
The exact method of implementing that bit of math(s) in any given computer language is what is copyrighted as soon as it is published.
"If the algorithm is patented, you need a license from the patent owner in order to create an implementation."
Except there is a Universe full of prior art, and thus it's not patentable. Yes, I know, the current US patent system says otherwise. I disagree, and fully expect this to be changed before my life is over ... We'll see :-)
"Your copyright prevents others from legally duplicating your code without your permission."
Correct. And in my opinion, that should be all that is necessary.
Firstly, don't forget that the *primary* protection of the openness of open source is copyright law: open source licences use copyright law as their main tool.
Secondly, IF a patent holder is a Good Guy, the patent can be licenced in such a way that it can be freely used in open source software. IBM has done this, for a whole bunch of patents.
But those patents are not freely licenced for proprietary software written by IBM's competitors.
If the patent holder is a Bad Guy, of course, the patent can be used to bugger up the open source software. So far as I know, IBM has never done this.
Disclosure: I used to work for IBM.
IBM actually develop micro-structure hardware & the patents on these forms would be much of IBM's future-proofing & fair enough if another company's development has the first tier already designed for them. The building of these micro-structures, lower striates & the similar are expensive dev. That IBM give away their software equates more to their lack of saleability than anything else so build up a clientèle base by giving it away then eventually charge for it.
That idea is fine but they'll never make any money out of it. IBM also buy up as many patents as possible to future-proof against M$ or Sun Micro which has hurt them in the past; ie M$ succeeding with Windows & IBM missing its own boat. It's not so much about screwing the open-sourcers but intervening when someone makes a huge pay check off IBM's coattails. They don't care about a million or so which isn't even the last day's anything to them but big windfalls that are using their invention are not to be ignored. Can you blame them here & wouldn't you like the payoffs?
I doubt that OSS is doing anything but helping IBM's future-proofing with free dev but we won't be making big money any time soon with IBM assisting us. They do offer some protection from M$ & from past behaviours no-one in their right mind would trust M$. You could probably paint IBM with a similar brush except that so far they've not seemed to be interested in hurting the OSS movement unlike M$.
The important bit of this paragraph is "...on a patentee's terms...". In other words, IBM's take seems to be that Patents are just a tool - it's how you use them that creates the Evil.
If I patent an invention but I make it available to everyone on the terms that you cannot charge money for any item which uses my patent, does that make me Evil simply because I have a patent? Or am I in fact following in the spirit of FOSS by ensuring that my invention (and its descendants) can be distributed freely?
Knee-jerk reactions like the Author's is what is making the majority of FOSS advocates look like idiots: IBM is correct on at least one point - Patents (like knives and guns) are nothing but a tool. The rest depends on how you use said tool.