> ...come to think of it, I do vaguely recall somebody attempting to claim a 
> patent on patent trolling...

The fun part comes when you are designing something that will be used in 
products sold in many different countries (eg a car or phone), then you have 
to make sure you check the patent situation from every possible country.

Post a reply to this message

scott <sco### [at] scottcom> wrote:
> It's not just software

  My point is that with software that kind of job is ridiculous.

  In the European patenting system the principle is that ideas and methods
cannot be patented, only real devices. I think it was like that in the US
as well, a century ago, but it gradually changed.

  (Of course companies have their ways to get around the European patenting
system by playing with words, but at least the principle is there. A granted
software patent is much less likely to hold up in court in Europe than it is
in the US.)

-- 
                                                          - Warp

Post a reply to this message

Warp wrote:
>   Now we only have about 10 million other software patents to worry about.

Yep. But as I said, I think the problem is not so much that they're 
"software" as that they get argued to mean things entirely outside the scope 
of what they're talking about.

> I think some software companies (at least in the US) have dedicated people
> whose only purpose is to research whether the company is breaking some
> software patents. And it's a full-time job.

Are you kidding? There are entire companies who do nothing but buy patents 
and then go sue people over them. They're called "patent trolls". :-)

-- 
   Darren New, San Diego CA, USA (PST)
   Insanity is a small city on the western
   border of the State of Mind.

Post a reply to this message

Darren New wrote:
> No. Really. It hasn't.  You can use that formula to draw pretty 
> pictures, for example, and they have no patent on that.

Or, to put it another way...

JPEG, MPEG, and MP3 are all patented. That doesn't stop you from using DFT 
for other purposes.

-- 
   Darren New, San Diego CA, USA (PST)
   Insanity is a small city on the western
   border of the State of Mind.

Post a reply to this message

Warp wrote:
>   In the European patenting system the principle is that ideas and methods
> cannot be patented, only real devices. I think it was like that in the US
> as well, a century ago, but it gradually changed.

Not particularly. The patents cover computers doing certain things. 
"Software patent" isn't really a good name for it as such, since all the 
"software patents" are actually patenting computers doing particular 
processes. Just like manufacturing patents cover lathes doing particular 
processes, and drug patents cover chemists doing particular processes.

>   (Of course companies have their ways to get around the European patenting
> system by playing with words, but at least the principle is there. A granted
> software patent is much less likely to hold up in court in Europe than it is
> in the US.)

Yeah. From my experience, I'd say just reading the patent narrowly, instead 
of letting people argue "well, obviously anything you could do by speaking 
on a telephone you could also do with a computer bouncing a laser off the 
space shuttle."

-- 
   Darren New, San Diego CA, USA (PST)
   Insanity is a small city on the western
   border of the State of Mind.

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> Warp wrote:
> >   In the European patenting system the principle is that ideas and methods
> > cannot be patented, only real devices. I think it was like that in the US
> > as well, a century ago, but it gradually changed.

> Not particularly. The patents cover computers doing certain things. 

  That's my point: Here ideas and ways of doing things (algorithms) cannot
be patented. Only real, physical devices (with working prototypes) can be
patented (at least in principle). You don't patent what the device *does*,
you patent the design of the device itself.

  In the US you can get patents for completely ridiculous things. For
example, check patent number 4022227, which is basically a patent for
combing your hair in a certain way. Or patent number 6368227: Swinging
a swing in a certain way.

  Or the infamous patent number 5443036: Excercising a cat using a laser
pointer. Note that it does not patent the laser pointer device. It patents
*using* it in a certain way, which is just ridiculous. I'd say limiting
how people can behave in the privacy of their own home is a violation of
basic human rights and constitutional law.

  If you want the *device* itself patented, check patents 6505576, 6557495,
6651591 and 6701872 (all of which patent basically the same thing, even
though that should be impossible).

-- 
                                                          - Warp

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> Are you kidding? There are entire companies who do nothing but buy patents 
> and then go sue people over them. They're called "patent trolls". :-)

  Let's hope the patent trolling patent is granted. That should result in
interesting lawsuits.

-- 
                                                          - Warp

Post a reply to this message

Darren New wrote:
> Darren New wrote:
>> No. Really. It hasn't.  You can use that formula to draw pretty 
>> pictures, for example, and they have no patent on that.
> 
> Or, to put it another way...
> 
> JPEG, MPEG, and MP3 are all patented. That doesn't stop you from using 
> DFT for other purposes.

And (Hi Andrew!) CDMA is patented for cell phones, but maybe not for other 
processes. You could probably use CDMA to deliver cable TV without running 
into the same patents you would using it for cell phones.

Now, another problem with "software" patents are that they're hard to 
research. Since they get read so widely, it's difficult to do a patent 
search and figure out if it's already invented.  If you're working on a 
machine to make aspirin, it's a safe bet that the patents that cover it 
contain the word "aspirin", "analgesic", etc. If you're working on an 
authentication system, most anything might be interpreted as authenticating 
the user.

For example, there was one argument wherein the patent said "the user's 
computer", and the defense had a system wherein the user could use any 
computer (and indeed was intended to prove it was the right user at any 
arbitrary computer), and the patent holder argued that it was "the user's 
computer" due to the fact that it was the computer he was using, even tho 
the patent talked about specialized hardware dongles attached to "the user's 
computer", and the patent holder wanted "specialized dongle" to be read as 
"any password-like mechanism, including passwords".  They also argued that 
the SSL nonce acted as the dongle. That's the sort of crap that causes 
trouble. Maybe there's just more stupid patent trolls in software, or maybe 
this crap goes on all the time and I just don't see it since I do software.

Then again, I heard of a very expensive patent case where basically 
backoff/retry was patented for a particular yet wide-spread purpose, and the 
patent was upheld.  Which seems rather wrong to me, as it really was the 
obvious way to do it.  I think they need a test like taking a random 
professional and asking "If you had to solve problem X, how would you solve 
it?" And if the person gives the solution in the patent, throw it out.

"If you needed to talk on a communication channel, and you wanted to reduce 
interference during collisions, what would you do?"  "Well, try, and if you 
get a collision, wait a little while and try again."  Bzzzt. Throw out the 
patent.

-- 
   Darren New, San Diego CA, USA (PST)
   Insanity is a small city on the western
   border of the State of Mind.

Post a reply to this message

Warp wrote:
> Darren New <dne### [at] sanrrcom> wrote:
>> Warp wrote:
>>>   In the European patenting system the principle is that ideas and methods
>>> cannot be patented, only real devices. I think it was like that in the US
>>> as well, a century ago, but it gradually changed.
> 
>> Not particularly. The patents cover computers doing certain things. 
> 
>   That's my point: Here ideas and ways of doing things (algorithms) cannot
> be patented. Only real, physical devices (with working prototypes) can be
> patented (at least in principle).

Certainly.  I understand what you're saying. I'm saying that people get 
around it by patenting "a computing device that computes XYZ using process 
PDQ."  That's the "design of the device."

It *might* be possible to word thing in a way that makes it impossible to 
patent something that can be run on a COTS computer, perhaps, but I don't 
see any good way of doing that offhand.

I've also seen patents where they give a mechanical device that does what 
you want to patent without even any electricity, and then add "by the way, 
it might be possible to build the same machine using computer parts, and the 
patent should cover that too."  I don't know how you'd avoid such a patent 
without also making useless patents on things like a new kind of drug that 
gets cooked up in a certain way. You cook it in a round pot, I cook it in a 
square pot, so I'm not violating your patent?

 > You don't patent what the device *does*,
> you patent the design of the device itself.

So you have no patents on drugs, or on processes for making drugs, or 
methods for curing rubber or hardening steel? No patents on constructing 
radial tires? On more efficient ways to manufacture paper? Or anti-lock 
brakes, or collision preparation systems, which are all computerized and 
wouldn't work but for the computer components?

Could you have patented an ink-jet printer in such a way that nobody else 
could make an ink-jet printer? Surely the shape of the case or the color of 
the ink can't be a determining factor. Can you patent putting a chip in a 
cartridge to prevent its refilling?

I'm just saying I think it's a grey area, with a broad range of things that 
can be interpreted widely that really shouldn't be, and if you try to 
legislate that down, you're going to have trouble with things most people 
think should make sense to patent.

>   In the US you can get patents for completely ridiculous things. 

Agreed. For one thing, the patent examiners are apparently greatly 
overworked. Who wants to have both a law degree and an engineering degree, 
and work for a government salary?

>   If you want the *device* itself patented, check patents 6505576, 6557495,
> 6651591 and 6701872 (all of which patent basically the same thing, even
> though that should be impossible).

The other problem with a lot of patents is they're not "enabling". This 
means they don't tell you how to build the actual device. Look at, for 
example, the patent on the Segway scooter. It basically says "We patent any 
device with two wheels on one axel where the rider's center of weight is 
above the wheels." There's no discussion about *how* you make it work. 
There's nothing in the patent that you couldn't learn by watching someone 
ride a Segway for five seconds. There's no "this is the secret we're 
revealing in return for the monopoly on manufacturing this" in the patent.

Another one I saw was a mechanism for making phone calls. The whole thing 
was three black boxes: A computer, a device to hook the computer to the 
phone, and the phone. It read like a requirements spec, rather than a 
finished device. *That* is a bad patent and it *should* be thrown out. I.e., 
in theory that patent is already invalid, not just "the law should be 
changed to make it invalid."

Unfortunately, the "loser pays until he wins" crapshoot combined with the 
"when in doubt, grant the patent and let the courts fight it out", tend to 
be the real problem.

-- 
   Darren New, San Diego CA, USA (PST)
   Insanity is a small city on the western
   border of the State of Mind.

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> The other problem with a lot of patents is they're not "enabling". This 
> means they don't tell you how to build the actual device.

  How about patents which require fictional laws of physics in order to
work? http://www.freepatentsonline.com/6960975.html

  This kind of patent would never be granted in Europe (at least in theory)
because of the requirement that a *working* prototype of the device must
exist.

-- 
                                                          - Warp

Post a reply to this message