|
Authored by: jesse on Wednesday, January 02 2013 @ 03:33 PM EST |
In the old days, computers had front panel lights.
One of the standard functions of the null task was to rotate the lights as a
status indicator of normal operation. If the system was busy, the lights would
pause, or jump around a bit (going fast/or slow).
The challenge was to replace the pattern and make it go backwards...
Or bounce between the left most light and the right most light.
I did this on a PDP-10 around 1974/1975. Of course the operators objected -
things didn't "look right". No system penetration was required - just
a program to tell the system what pattern to display, running in user mode with
no privileges whatever.[ Reply to This | Parent | # ]
|
|
Authored by: pem on Wednesday, January 02 2013 @ 04:42 PM EST |
I'm not a mathematician, but then I don't think you are either, so
I'd like you to tell me what application of 1+1=2 makes that into applied
maths?
Any application. If it's used in an application, it's
applied, no? :-)
You talk about the doctrine of equivalents
however the problem I think you have is that you do not extend it as far as you
should be doing. Consider that we both have similarly built computers running
the same OS. My computer running my software (a ballistics simulation of firing
a gun) is equivalent to your computer running your software (a simulation to
create an optimal drill bit design) irrespective of the fact that
computer+software produce different outputs. Our computers run the same set of
instructions (that is CISC/RISC/whatever). They are actually doing the same
things down at the minutest level - manipulating bits.
What you
have described is a machine that is a simulator, with two different simulations
loaded. The simulations are completely different, and the software being
simulated might or might not, in some other context, be part of a patentable
invention. Following Diamond, an invention would probably require something
more -- for example, the ability to connect the computer to a gun. The drill
bit generator is an interesting case -- would a patentable invention be created
by hooking up a printer so you can get drawings?
But outside simulations, in
many applications, the computer itself is of much less complexity than the
software. Some view it as a computer with a sprinkling of software on top, when
in reality, it is often a huge complex piece of software, which requires a
certain amount of hardware (including, btw, a computer) to make it do something
useful. But the hardware can be swapped out quite easily in many cases. I
think in your example, assuming you wrote in a high level language, you could
probably swap either of your applications from an X86 to an ARM with much
greater ease than morphing one of the applications into the
other.
We could then look at another pair of computers, one is a
WDR-1-Bit-Computer, the other is some ARM jobby, lets say a Samsung Android
phone since we love them. 2 really different machines. Both are hooked up,
whichever way you want, to a simple LED board which has 8 x 1 LED bulbs. I
personally wouldn't be able to do it (at the moment) but I'm pretty sure that
some clever programmer would easily be able to create the bounce effect on an
"image" of 3 bulbs lit which then "move" along the board to the end then, with
the right timing, show 2 bulbs, 1 bulb then bounce back to show 2 bulbs then 3
bulbs. They are special purpose machines that do the same thing so doesn't that
mean that the doctrine of equivalents would tell you that these are the same
machine? Are they?
The two machines you describe would probably
be the same, but would probably not infringe on Apple's patent. The doctrine of
equivalents has to read on the claims. The claims will probably state something
about a two dimensional image. Bingo -- your 8x1 device would not
infringe.
The doctrine of equivalents would also tell you that 2
similar cafetiere's (ie French press), one with coffee the other with tea, are
the same even though they produce different drinks when you put hot water in
then (after a few minutes) pour out the resulting liquid. Or should that be 2
different cafetiere's, both with coffee, are the same because they both produce
coffee?
Again, it would be highly fact-specific, partly depending
on whether the claim was narrow
ed in the patent office during examination.
So, for example, if there was a
preexisting French Press for coffee, and there was a slight modification (what?
larger holes? I don't know) that made it suitable for tea, perhaps you could
still get a patent on the tea one. But that patent would not cover a
smaller-holed version for chicory, because the examiner would have narrowed the
claim to not cover the prior art coffee one that also had smaller holes.
On the
other hand, if the French Press for coffee was brand new and patented, and then
I subsequently patented one with larger holes for tea, then (depending on how
the claims were worded), the French Press for tea might be subject to both
patents -- the new one directly for the larger holes, and the old one, for the
idea of the plunger, disc with holes, etc., under the doctrine of equivalents.
Ownership of a patent doesn't actually allow you to make anything; it allows you
to keep others from making stuff. So the manufacturer of the tea French Press
might have to pay off two patent owners.
Is the 3d printer a
different 3d printer because it produces something
different?
Obviously, the physical printer is not different. And
in many cases, the entire system (hardware + software) will not be different,
just the data load for the specific object to be printed. But there may be
software-only improvements (for speed, for automatically inserting internal
structures, whatever). It is interesting that several companies sell pure
software upgrades as different physical products. For example, you can buy an
oscilloscope that runs at one speed, or one that is faster for more money. But
guess what? It's just a different software load. So, it's not physically
different. But is it a different machine? That's one for the courts, but if
the decision in Diamond is followed faithfully, the answer could possibly be yes
-- but Diamond was about process, so more about that later. I'll give you
another real-world example from personal experience -- I used to work for a
company that made embedded modem integrated circuits. A customer could buy a
1200 baud modem, a 2400 baud modem, a V.92 modem, etc. You get the picture.
Different patents were involved in each of these (more patents on the higher
speed ones). The chips were physically the same die, but when packaging them,
bond wires could be used to attach some of the internal pads to ground or not.
So the physical modem chip, sent to the customer, was exactly the same for any
of the modems, but had its personality altered by how it was packaged. Similar
to the oscilloscope example I gave, the chip was capable of functionality that
was sometimes not available to the customer. We had to pay more royalties on
the higher speed version because it practiced more patents. The software to
practice those patents existed in all the chips, but was never executed except
on the higher speed chips.
But bear in mind that the decision in Diamond was
about the process. I will readily admit that sometimes I write the wrong thing,
so I will probably say "different machine" in several contexts where really, it
could be considered the same machine, but will be running a different
process. If I'm confusing, sometimes it's because I'm confused, but if you
step back and think about it, the process-based approach makes a lot more sense
for software. You don't have to decide whether or not it's a different machine,
because that's not the focus of the patent at all.
If the only
thing that distinguishes one 3d printer which produces a sphere from a 3d
printer that produces a cube is pure maths and pure maths is abstract then where
is your doctrine of equivalents now?
I don't know. Ask the
supreme court. There may be answers soon. But I don't think a patent on the 3d
printer would be about whether it produces spheres or cubes in any case. It
might be about how fast it can print, or how internal structures are
automatically inserted into the produced object. Another issue is that the data
could be pre-processed by whatever special algorithm on a general purpose
computer, and then fed to the printer later. But people have tried for years
(even before software) to get around patents by separating out the steps or
components that practice a patent. They usually lose, although with method
patents it would appear that a patent holder might have to go after the end user, rather than a supplier. In any case, if the
supreme court makes a definitive ruling that "on a computer" is not sufficient
to make something patentable, it will be interesting to see how that
jurisprudence is squared up with the prior case law that says "if A+B infringes,
and you do A followed by B then you also infringe" in the case where A by itself
is ruled to be non-statutory subject matter. OTOH, in Diamond, they effectively
said that non-statutory subject matter could be a component of a patented
process, so they might not have any trouble ruling that way.
You
may say that my points are silly but I don't care. I'm
right.
Your questions aren't silly. Did you make any definitive
statements?
(This isn't necessarily correct in all instances. I'm
just parodying your answers to certain points that PolR mentioned which appear
to be quite petulant.
If I was petulant, it was because I was
tired and, frankly, the "book as patentable object, but I'm going to go back and
forth discussing the book vs the machine that makes the book" was esoteric and
not at all enlightening (to me), and that made irritable.
I can't
pretend to understand everything that PolR has posted previously but everything
he has said that I understand I do agree with. In those previous posts PolR did
provide logical explanations and links which supported what he
stated.
Not that well IMO. For example, if I understand his
position correctly, he has effectively stated that an implementation in an FPGA
of circuit X should not be patentable, even if circuit X were otherwise
patentable, while simultaneously appearing to give lip service to Diamond. This
makes no sense. Even if the algorithm/logic used is non-statutory matter, if
creating a circuit embodying it makes a patentable invention, and if Diamond
says that creating a process embodying non-statutory matter can be patentable,
then certainly the same circuit (the FPGA acts exactly the same as the
patentable non-FPGA version) should be patentable, no?
You provide
no links to evidence for your assertions in this discussion and I can't remember
what previous points you may have raised in other discussions on this subject so
you do start from a worse position with your point of view.)
I
have not participated in these discussions in any meaningful way since about 3
years ago, because it can be quite tiring, because people (yes, like PolR)
present really esoteric arguments as if they are common sense, when they aren't,
and when common sense (and, AFAIK, the Doctrine of Equivalents, which is itself
originally based on common sense) says that whether I wire up a few gates
together by hand, or program a CPLD to get those same gates, I have built the
same thing.
I think PolR argues software=maths etc because those
that the things that he knows about and can argue about with expertise based on
fact and logic.
I suppose, but a lot of what he writes really
doesn't make sense to me. It might look good on the surface, but doesn't stand
up to scrutiny. The patented book is one such example. The problem is, in some
cases, it takes a lot of energy to get past the reasonable sounding surface, and
it takes a _lot_ of energy to argue persuasively enough to bring others along
with you. In the meanwhile, sometimes you suffer abuse. Having said that,
there has been practically zero abuse in the comments on this particular
article, and for that, everybody is to be commended.
Arguing for
the things that you want the focus to be on are things which are subjective and,
in some cases, are already being done poorly by the USPTO.
I
think a lot of them can be done objectively. As I have written several times in
the past, if I were in charge, I would mandate a period of a few months where
the problem description of a patent was published (but not the claims, drawings,
or descriptions), and during that time, any submitted "solutions" to the
described problem would be out-of-bounds for the patent. If someone can come up
with a solution that quickly, either it was obvious, or they had prior
art.
Another thing I would change is that there should be a rigorous
classification for field-of-use. If your patent isn't correctly classified as
to what it covers, too bad. This would make prior art searches a lot simpler.
[ Reply to This | Parent | # ]
|
- I doubt it - Authored by: Anonymous on Thursday, January 03 2013 @ 08:33 AM EST
- I doubt it - Authored by: pem on Thursday, January 03 2013 @ 12:28 PM EST
- I doubt it - Authored by: Anonymous on Friday, January 04 2013 @ 09:07 AM EST
- I doubt it - Authored by: Anonymous on Friday, January 04 2013 @ 03:59 PM EST
- I doubt it - Authored by: bprice on Saturday, January 05 2013 @ 07:52 AM EST
- No, I'm not - Authored by: Anonymous on Saturday, January 05 2013 @ 09:51 AM EST
|
|
|
|