Slashdot Mirror
Do Patents Stop Companies From Creating 'Perfect' Products?
Chris M writes "In a recent CNET article, the mobile phone editor writes about what he thinks would make a perfect phone. Unfortunately, as someone in the comments section points out, much of the technology that is used in this concept phone belongs to separate companies. 'I'm sorry to be the Devil's Advocate here, but most of those feautres are patented to separate companies. It would require almost all the major manufacturers [working together] to do this, which is highly unlikely.' Do you think patents are stopping companies from creating 'perfect' devices, or are there other factors at work?"
Except that in the modern world, features frequently are patented. Or the method patented is so broad that it covers all possible implementations.
I still have more fans than freaks. WTF is wrong with you people?
I agree with the first part. His perfect phone certainly looks very different from my perfect phone.
Currently, I own the Samsung i607 (blackjack) which earns about an 8/10 from me, which no other phone ever did. It's very thin, light, durable, and has an easy-to-care-for matte finish. It has a full QWERTY keyboard, and a very nice screen, fast processor, 3G, etc... If it had a 4-day battery life instead of 1.5-2 day, and a standard bluetooth stack that would let me sync and tether with ease from my Ubuntu laptop, it would match my dream phone... Not so hard, really...
Company A patents technology X, but has no interest in making a product that has technology X plus feature Y. Company B would like to make a product with technology X and feature Y, but is stumped by the patent. Result: the world never gets an X+Y product.
This is not just theoretical. I work in a field knee deep in patents and I see this sort of nonsense all the time.
Engineering is the art of compromise.
CDMA.
Try building a phone that works with the CDMA cellular network, and doesn't violate Qualcomm's patent. It's not going to happen. They've patented too much stuff that's too fundamental to making an interoperable device.
If you piss them off, or if they decide for some reason not to license their patent(s) to you (e.g., you want to make a multi-network phone and their other customers -- the telcos -- don't like the idea), you're S.O.L. as far as most of the U.S. cellphone market is concerned.
Video compression is the same way. Try to build an MPEG-4 encoder that doesn't violate the MPEGLA's patents; it's not going to happen. Sure, you could build some completely unrelated video encoder, but that's of extremely limited utility in a world where standards matter.
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
That's not how patents work. If you change one thing and patent the new "invention", it's a new patent, completely separate from the original one. You have to reference the original, but there are no fees to be paid.
True. But if you try to create any instances of your patent, then you will have to pay fees to the original patent-holder (or get sued.) See, patents are like class definitions, but to create an instance of that class, you have to pay. You may extend someone else's definition, and then they will need your permission to change it, but you still need to create an instance of the base class with yours.
Your ad here. Ask me how!
But his conclusions are entirely wrong. Patents give you the right to keep others from using your invention, but they don't give you the right to use it. Thus, when you take someone else's patent and improve it, you can patent the improvement, but you still can't make the device without the permission of the original patent holder. Likewise, the original patent holder can't make use of your improvement without your permission.
You are reading a copy of my copyrighted post.
You mentioned FUD. A big problem is patent FUD. What matters is what people think. Chilling effects. Many people think patents might be violated even when not. Or they feel they're not in violation but don't want to get in long expensive court battles even if victory is certain. Or they know they will be in violation of patents that should never have been granted, and calculate that trying to get them invalidated isn't worth the expense or that going ahead and hoping they aren't sued is too risky. These problems are made worse by the government's attitude of "when in doubt, grant the patents and let the courts sort 'em out." The patent office rakes in more fees, the lawyers get richer, the justice system hires more people to handle the load, and the rest of us foot the bill for this very expensive system.
As for examples, how about Apple's "Look and Feel" patents on the MacIntosh GUI? Microsoft was not allowed to use a trashcan icon. When they opted for a "recycle bin" they were nonetheless challenged again by Apple, weren't they? How about the hundreds of patents on the hundreds of tiny variations on arithmetic coding? While LZW and the GIF image format weren't too hard to work around, arithmetic coding was another matter. Only takes one of those hundreds of patent holders deciding to sue to make arithmetic coding not worth the trouble, not when Huffman is almost as good. The result is, no one will touch arithmetic coding even though the basic algorithm is free of patent protection. Instead, everyone uses Huffman coding. The primary difference between the long vanished "bzip" and "bzip2" is in bzip2, bzip's arithmetic coding was replaced with Huffman. People would not even use bzip because of that.
Note that we're talking about the users who by rights should have nothing to fear from whatever alleged patent violations authors may have committed. Instead, the possibility that a court might pull the plug, as nearly happened to the Blackberry, is enough to scare off users. Then there's the stunt SCO pulled, trying to shake down Linux users for $699 each in licensing fees, lest they convince a court that they aren't blowing a bunch of smoke and do manage to get an insanely draconian and unenforceable injunction ordering everyone to stop using Linux. That anyone actually paid SCO is sad. That patent trolls might have such leverage is ridiculous. Dell customers don't have to worry should Dell be found in violation of some patent. Dell's PCs won't suddenly stop working. But somehow Linux users do have to worry, with MS claiming Linux is in violation? WTF?
Intellectual Property is a monopolistic, selfish, and defective concept. It is "tyranny over the mind of man"
This is a well known phenomenon, referred to as the Tragedy of the Anticommons. Yochai Benkler describes how multiple patent holders delayed the development of radio until the U.S. government intervened:
For an interesting essay on the topic, read Boldrin & Levine: Against Intellectual Monopoly. Actually, just read the first couple of pages, describing the effects of James Watt's patent had on the development of steam engines in the late 1700s. It seems that his rabid enforcement of his patent pretty much blocked all further progress in the technology until his patent ran out in 1800. He didn't even profit from it himself until after his patent ran out, when he switched from patent enforcement to steam-engine manufacturing. And his own development was blocked by others' patents on other parts of the mechanism.
;-)
There's plenty of historical evidence that things like patents and copyrights are primarily a barrier to progress, unless the government steps in a second time and decrees some sort of mandatory licensing. (But this doesn't have much effect on the doctrinaire arguments that we read here so often.
Those who do study history are doomed to stand helplessly by while everyone else repeats it.