decoration decoration
Stories

GROKLAW
When you want to know more... 		decoration
For layout only
Home
Archives
Site Map
Search
About Groklaw
Awards
Legal Research
Timelines
ApplevSamsung
ApplevSamsung p.2
ArchiveExplorer
Autozone
Bilski
Cases
Cast: Lawyers
Comes v. MS
Contracts/Documents
Courts
DRM
Gordon v MS
GPL
Grokdoc
HTML How To
IPI v RH
IV v. Google
Legal Docs
Lodsys
MS Litigations
MSvB&N
News Picks
Novell v. MS
Novell-MS Deal
ODF/OOXML
OOXML Appeals
OraclevGoogle
Patents
ProjectMonterey
Psystar
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v IBM
SCO v Novell
SCO:Soup2Nuts
SCOsource
Sean Daly
Software Patents
Switch to Linux
Transcripts
Unix Books

Gear

Groklaw Gear

Click here to send an email to the editor of this weblog.

You won't find me on Facebook

Donate

Donate Paypal

No Legal Advice

The information on Groklaw is not intended to constitute legal advice. While Mark is a lawyer and he has asked other lawyers and law students to contribute articles, all of these articles are offered to help educate, not to provide specific legal advice. They are not your lawyers.

Here's Groklaw's comments policy.

What's New

STORIES
No new stories

COMMENTS last 48 hrs
No new comments

Sponsors

Hosting:
hosted by ibiblio

On servers donated to ibiblio by AMD.

Webmaster
Comes may come here..... | 456 comments | Create New Account
Comments belong to whoever posts them. Please notify us of inappropriate comments.
Corrections here please
Authored by: tiger99 on Thursday, November 29 2012 @ 08:01 AM EST
If needed.

[ Reply to This | # ]

Not the official off topic thread. Publicity is good
Authored by: Anonymous on Thursday, November 29 2012 @ 08:01 AM EST
Let Daryl know Groklaw is still around Or mog lol.

[ Reply to This | # ]

Off topic here please
Authored by: tiger99 on Thursday, November 29 2012 @ 08:02 AM EST
We will devise a suitable punishment for anyone who posts on-topic stuff here!

[ Reply to This | # ]

Newspicks here please
Authored by: tiger99 on Thursday, November 29 2012 @ 08:05 AM EST
Comments on the Groklaw Newspicks should go here. Please make a link so the item
is still accessible once it scrolls down off the page.

[ Reply to This | # ]

Comes may come here.....
Authored by: tiger99 on Thursday, November 29 2012 @ 08:06 AM EST
Transcripts of the Comes documents, that is.

[ Reply to This | # ]

Vote for Groklaw thread.
Authored by: tiger99 on Thursday, November 29 2012 @ 08:07 AM EST
See previous article for what this is about, and please do cast your vote!

[ Reply to This | # ]

Step one: define your terms
Authored by: chriseyre2000 on Thursday, November 29 2012 @ 08:11 AM EST
In order to properly make the argument that software is mathematics you need to
start by explaining to lawyers what mathematics is. There is far more to this
subject than simple arithmatic. Mathematics is formal reasoning about abstract
patterns.

This will make arguments that software is mathematics much easier.

[ Reply to This | # ]

What Does "Software Is Mathematics" Mean? Part 2: A Semiotics Approach to the Patent Eligibility of Software by PolR
Authored by: Anonymous on Thursday, November 29 2012 @ 09:06 AM EST
Even a three year old child knows that software is not mathematics. Show her a
iPhone without a game and an iPhone with a game. She will pick the one with a
game because she will have much more fun with the second device than the first.
She will probably even call it "mine."

Although no government in any country on earth agrees with you, it is legitimate
to believe that patents are bad. However, there really is nothing special about
software patents that distinguishes them over other type of patents. There are
regularly patent lawsuits in all technology areas. Litigation is almost always
reserved for the worst patents, because the parties typically settle if the
asserted patent is a strong one.

Your assumption that the lawyers do not understand computers is incorrect. Many
of the patent attorneys involved in patent litigation involving software have
advanced degrees and extensive industry experience in software. What is lacking
on this blog is an understanding of the legal and business worlds.

Contrary to the apparent assertions in this blog, software patents cannot be
infringed by merely performing steps in one's mind. Under current law, any
method claim that can be infringed by performing all of the steps of the claim
in a person's head is invalid.

[ Reply to This | # ]

What Does "Software Is Mathematics" Mean? Part 2: A Semiotics Approach to the Patent Eligibility of Software by PolR
Authored by: Anonymous on Thursday, November 29 2012 @ 09:45 AM EST
I read this (with great difficulty) and this from someone
who obtained a physics degree and has been in the IT field
as a programmer, and system admin for over 20 years.

However, as I read it, I think I achieved my own "epiphany"
regarding software patents. And that is, software patents
don't meet the contract of disseminating the knowledge of
"how to" do something.

Look at virtually any software patent (let's take
bounceback).

It describes "what happens" (a finger drags a UI element to
a border, and it bounces back).

It doesn't tell me as "one reasonably knowledgeable of the
art" how to do it. How do I write the code to make that
happen?

I always thought patents were about the "how" not the
"what".

-Paul

[ Reply to This | # ]

What Does "Software Is Mathematics" Mean? Part 2: A Semiotics Approach to the Patent Eligibility of Software by PolR
Authored by: Anonymous on Thursday, November 29 2012 @ 09:52 AM EST
> What Does "Software Is Mathematics" Mean?

Not much, if anything. It would not be the first area of an otherwise
"unrestricted" area that legislators have carved out and made non-free
(e.g. free-speech/crowded-theater). It is frustrating that so much energy is
wasted on this kind of hair splitting on groklaw, when there are real problems
with software patents.

[ Reply to This | # ]

What Does "Software Is Mathematics" Mean? Part 2: A Semiotics Approach to the Patent Eligibility of Software by PolR
Authored by: Anonymous on Thursday, November 29 2012 @ 11:03 AM EST
Not done reading yet, but i got to this part:
For a programmer, an unattended computer is like a book sitting on a shelf. This book isn't currently read but its meaning is available to whoever opens it. Similarly, no one checks the internal state of a programmed computer once the program is done debugging. But programmers know that if they inspect it they will read the meanings.10

The implication is that the internal activity of a computer is a sign even when it is not actually examined by a live human. This is the correct result. According to semiotics, a device is a sign whenever there is some convention on how to interpret its meaning. The requirement is not that someone actually applies the convention. It is that the convention must be defined. Then if someone comes along and uses the convention, the meaning could be understood.
Couldn't it be reasonably argued that production code has removed the debugging capabilities, therefore even a programmer can't "open the book and read it"? Thus the internal activity of a computer is not a sign because there is no convention on how to interpret its meaning.

Just food for thought. I like the read so far though. Thanks.

~ukjaybrat - IANAL

[ Reply to This | # ]

There's a reference that may help those of us who are programmers
Authored by: dio gratia on Thursday, November 29 2012 @ 12:10 PM EST

A Theory of Computer Semiotics by Peter Bøgh Andersen. My kid sister was kind enough to send me a copy for my last birthday, although she paid less that what Amazon is currently asking.

There's a review of the book at UPenn (PDF). The author has an abstract of the second edition that can be found here. It's part of the Cambridge Series on Human-Computer Interaction, and I believe you can get paid access to an Adobe ebook version through Cambridge Books Online.

[ Reply to This | # ]

I'm hoping this beats Bilski
Authored by: Ian Al on Thursday, November 29 2012 @ 12:17 PM EST
The Supremes said that the Bilski patent was a patent on abstract ideas, but I have always been uncomfortable because the approach they took was to look for evidence of non-abstract ideas and come to their conclusion on that basis.
The concept of hedging, described in claim 1 and reduced to a mathematical formula in claim 4, is an unpatentable abstract idea, just like the algorithms at issue in Benson and Flook . Allowing petitioners to patent risk hedging would pre-empt use of this approach in all fields, and would effectively grant a monopoly over an abstract idea.

Petitioners’ remaining claims are broad examples of how hedging can be used in commodities and energy markets. Flook established that limiting an abstract idea to one field of use or adding token postsolution components did not make the concept patentable.

That is exactly what the remaining claims in petitioners’ application do. These claims attempt to patent the use of the abstract idea of hedging risk in the energy market and then instruct the use of well-known random analysis techniques to help establish some of the inputs into the equation.

Indeed, these claims add even less to the underlying abstract principle than the invention in Flook did, for the Flook invention was at least directed to the narrower domain of signaling dangers in operating a catalytic converter.
There seems to be nothing in that explanation that indicated what sort of hedging method invention would not be an abstract idea. If the process actually made the investments and counted the profits, would that transform it into non-abstract, patentable subject matter?

I have, of recent days, been saying that patenting functions documented in patent claims was patenting abstract ideas. I have had my fingers crossed in case someone thought to challenge that assertion. The abstract ideas used in writing software have a non-abstract conclusion since they end in written (hopefully working) code. However, if Bilski was abstract ideas, it shows that the ideas can still be abstract even when the subject of those ideas, hedging, is not. I will read your latest piece with enthusiasm in the hope that you have provided a scientific and rigorous proof of what is abstract thought and what is not.

I have, of recent days, been saying that patenting functions documented in patent claims was patenting abstract ideas. I have had my fingers crossed in case someone thought to challenge that assertion. The abstract ideas used in writing software have a non-abstract conclusion since they end in written (hopefully working) code. However, if Bilski was abstract ideas, it shows that the ideas can still be abstract even when the subject of those ideas, hedging, is not.

I will read your latest piece with enthusiasm in the hope that you have provided a scientific and rigorous proof of what is abstract thought and what is not.

---
Regards
Ian Al
Software Patents: It's the disclosed functions in the patent, stupid!

[ Reply to This | # ]

What Does "Software Is Mathematics" Mean? Part 2: A Semiotics Approach to the Patent Eligibility of Software by PolR
Authored by: Anonymous on Thursday, November 29 2012 @ 12:27 PM EST
Your clock analogy is not correct.
The convention for reading time may be changed. This makes the clock a sign according to semiotics. For example, suppose we decide to use a decimal time system. In this system there are 20 hours in a day, 10 before noon and 10 after. Each hour is divided in 100 minutes and each minute is divided in 100 seconds. We can easily build a clock that keeps time according to this new system. We just have to paint the numbers 1 to 10 on the old clock in place of the old numbers. Then the position of the handles can be interpreted according to the new convention. The short handle indicates the hour and the long handle indicates the minutes if you mentally multiply the numbers by 10.

Here is the abstract idea question. Did we invent a new clockwork which performs its functions according to the law of physics? Or did we invent a new way to interpret the meaning of an old clock? Of course, in the decimal clock example the clockwork was not changed. The invention amounts to painting new numbers on the clock.

This is not inventing a new clock.

Getting a clock to turn the hands to represent half of a 20 hour / 100 minute day would require change in the gearing of the minute hand. Simply relabeling the face does not change the fact that that the minute hand of a 12/60 clockwork would rotate 12 times per half day, not 10. Having said that, I don't know if a 10/100 clockwork would be considered patentable subject material.

[ Reply to This | # ]

Wrong branch of government
Authored by: Anonymous on Thursday, November 29 2012 @ 12:58 PM EST
This always seemed to me to be a pointless approach. It is Congress you need to
influence, not the Supreme Court.

The Constitution grants extremely broad powers to Congress:

"Congress shall have the Power to promote the Progress of Science and
useful Arts, by securing for limited Times to Authors and Inventors the
exclusive Right to their respective Writings and Discoveries"

There is no fine nuance here regarding what kinds of inventions, etc., are
eligible. That is the matter for legislation. The Supreme Court does not
decide on policy.

If you know anything about the Supreme Court you know that they are very
hesitant to act as a legislator. They defer to the intent of Congress, and the
Executive branch, unless they have acted in contradiction of the Constitution.
That is how our government works.

But the Executive branch, through the USPTO, has repeatedly acted in support of
software patents. This has been constant policy, via both Republican and
Democratic administrations for many years. And Congress has expressed no
willingness or even interest in changing this, even though it is 100% within
their power to do so.

Expecting the Supreme Court to change fundamental policy in this area is rather
naive.

[ Reply to This | # ]

What Does "Software Is Mathematics" Mean? Part 2: A Semiotics Approach to the Patent Eligibility of Software by PolR
Authored by: Anonymous on Thursday, November 29 2012 @ 01:27 PM EST
The Functions of Programmed Computers Depend on Human Understanding

Before I start, I'm unable to create a new account despite my announced intention to do so here as they have been temporarily disabled.

I commend you for such a detailed and elaborate presentation for your view point. I admit that I have not yet finished reading let alone fully understand everything you present. My intention with this posting is not to challenge or undermine your position in relation to software and mathematics, but perhaps more to cause to to rethink some aspect of your position and it's for that reason that I will do my best _not_ to engage in any dialectic with the intention of supporting my differing viewpoint.

That being said, I feel compelled to comment on the above blockquote.

1. "Human Understanding" applies to the humanity in general my comments are made where such an expression is understood to be inclusive of any human beings that have mental capabilities within one or even two standard deviations of the statistical norm of metal capabilities, whatever that may be. As such we certainly can understand it _not_ to refer to individuals who are outliers of that norm, such as mentally deficient individuals, or exceedingly brilliant or studied individuals insofar as the requirements of understanding is not a set of the priory stated norm.

2. "Programmed Computer" is understood to refer to a hardware platform that holds software functionality that may or may not be executing but only present in some part of its memory system.

3. "Functions" is understood to refer to a some functional characteristic present in or on a Programmed Computer whether they be machine specific or caused by machine operations related to its execution some software. No presumption exists that would cause a reasonable individual to believe the function is required or executing, only that the function exists and may be executed.

So both definitions 2 and 3 also contain no presumption that the Programmed Computer is even turned on. In that way, it's capabilities is purely theoretical and certainly no presumption exists that would suggest that the Programmed Computer is performing anything useful that is not purely a result of its tangibility.

With that, I challenge your presumption that the Functions of a Programmed Computer depends on Human Understanding. Consider:

You certainly don't understand the specifics of the caching functionality Google uses, or the transaction characteristics in the filesystem they use. However such a lack of understanding does not prevent that functionality from performing its task(s) within the relevant "Programmed Computers" if and when conditions may cause the the function to be performed.

Certainly the creation of the software component that is part of the "programmed computer" requires the understanding of one or more engineers, however that specific knowledge is not "Human Understanding" since the engineers' specific understanding as it relates to the making of the software component(s) is an outlier compared to normal human mental capacities.

If your intention is to alter my definition of "Human Understanding" to be inclusive of any and all human knowledge and mental capacities, than you may have a valid counterargument to my challenge. But such a homogenization of human mental capacities is a declension to a human beings natural instinct, will and capacity to enhance his/her lives and capabilities by acquiring new skills or assets.

Such a alteration is communistic in nature rather than individualistic and to truly follow such a principle in an unbiased way is to challenge the very foundation of capitalism as well as our cultural and natural values. We may as well be trying to claim that men and woman should not be selective about the characteristics of their friends or spouses, let alone consider any individuals claim to intellectual property whether it be copyright or patents.

[ Reply to This | # ]

My "Aha" Moment
Authored by: OpenSourceFTW on Thursday, November 29 2012 @ 02:01 PM EST
This article made my brain hurt, but in a good way. Thanks PoIR!

I now understand a very important point, namely, about how software does not
transform a machine.

Lets say I run Software X, which calculates the the Dow Jones average. Yes the
logic behind it may be innovative (but not patentable, it's an algorithm), but
does the execution of it significantly change the underlying computer? No, it is
still doing the same low-level operations it has always done (ADD, SUB, NEG,
etc). Nothing is inherently different. Only the meaning we attach to it is
different, and that is not enough to make it patentable.

So how can this be patentable if the machine is the same?

Another question: If someone argues that the exact number and order of these low
level operations is different for this software than for preceding software, and
that this should be patentable, then this is easy to circumvent.

Suppose the patent applicant used a particular Intel processor. What if I simply
use an AMD processor? Won't the process be somewhat different due to the
difference in hardware? Doesn't that mean I made an even newer machine? What if
I run it on my Android device? It will certainly be different there. Therefore,
the whole machine transformation argument is moot even if it was valid, because
simply changing the execution environment would produce a different machine.

The argument attempts to straddle both sides at once by claiming a specific
transformation permits the patenting of all transformations, no matter how
different, as long as they "do about the same thing." It permits the
innovation to be specific but the protection to be broad.

---
I voted for Groklaw (Legal Technology Category) in the 2012 ABA Journal Blawg
100. Did you? http://www.abajournal.com/blawg100. Voting ends Dec 21.

[ Reply to This | # ]

"Information" is abstract
Authored by: Anonymous on Thursday, November 29 2012 @ 02:31 PM EST
The information processing functions of a computer cannot be separated from human understanding.

100% agree.

My view is that all information is abstract, in the same way that numbers are abstract (i.e. they are symbols that can be manipulated mentally, not physical phenomena that can be moved around by machines).

"Information processing" is the manipulation of abstract symbols. Its an abstract activity: a mental, semantic activity. We might assist it by using mechanical machines that manipulate physical phenomena (matter or electrical charges or magnetic fields) but really those machines just push physical stuff around, not "information". The "information" is a semantic assocation between those physical things and human-meaningful abstract concepts (symbols). The semantic association exists only in the minds of the programmers and users of the machine. There is no physical representation or embodiment of it in the machine itself. So even if you want to do an "information processing task" with the help of the machine, the first step (programming the machine) and the last step (interpreting the results) require a semantic mental process to be done by a human, and the calculations done in-between by the machine do not directly manipulate "information" at all.

In summary:
"Information" is an abstract thing that exists only in our minds, and "information processing" is an entirely mental activity! (In other words, its mathematics).

[ Reply to This | # ]

Reductionism
Authored by: danielpf on Thursday, November 29 2012 @ 02:34 PM EST
To equate all software to mathematics is as reductionist as equating all
material objects to atoms.

Between writing high level code and bit flipping there are many levels of data
transformation which progressively change the meaning of what software is,
exactly as between atoms and devices there are many levels of matter
organization allowing to pass between non-patentable natural phenomenon at the
atomic level to inventions at the device level.

The high level human work may be just painting something on screen which is then
translated to a postscript program, and after some more levels to bit flipping
and finally to ink dots; it would be abusive to say that using a computer is
always software creation, so math, yet there is no sharp distinctions between
all the possible ways to create software at different levels, from assembly
language to compiled language to interpreted language to spreadsheet composition
to word processor.

Mathematicians do their best to have a precise and unique definition of the
words they use, while most other humans use language in a fuzzy way: words have
different meanings in different contexts. In between legislators try to
introduce some rigor in the laws but by far do not reach the mathematical rigor.


So in the real world the word software has a different meaning in different
contexts. I pretend that what legislators mean by software covers a different
semantic field than what computer scientists mean.
Not understanding that produces sterile discussions.

Note that I am strongly against software patents, but the software=math argument
appears to me counter-productive.





[ Reply to This | # ]

aside: when i program...
Authored by: sumzero on Thursday, November 29 2012 @ 03:00 PM EST
i am inventing nothing. all i am doing is figuring out the
correct way to request from a general purpose computer
something it can already do in a way that it understands. it's
just translation work.

sum.zero

---
48. The best book on programming for the layman is "alice in wonderland"; but
that's because it's the best book on anything for the layman.

alan j perlis

[ Reply to This | # ]

Ceci n'est pas un pipe
Authored by: Anonymous on Thursday, November 29 2012 @ 03:26 PM EST
I can't believe we went through an entire discussion of semiotics without you using the painting "Ceci n'est pas un pipe" ("This is not a pipe") as an illustration of the difference between the sign carrier (the painting) and the referent (the pipe).

[ Reply to This | # ]

Is there a point to this?
Authored by: Anonymous on Thursday, November 29 2012 @ 03:35 PM EST

This "Software is Mathematics" jabber keeps coming back to Groklaw.

Is there a point to it?

[ Reply to This | # ]

Reverse Logic
Authored by: Anonymous on Thursday, November 29 2012 @ 03:41 PM EST
I note in the footnote about the ENIAC a mention that no voltage is
"usually" interpreted as a logical zero. This is correct, but it
might be worth calling out a little more explicitly: the ones and zeroes are
purely convention. There is reverse logic where no voltage is a logical one and
vice versa. And we're free to choose either way when designing circuits.

So, as you say, the meaning of the voltages exists in the mind of the one using
the computer, not in the signals themselves.

[ Reply to This | # ]

What Does "Software Is Mathematics" Mean? Part 2: A Semiotics Approach to the Patent Eligibility of Software by PolR
Authored by: Anonymous on Thursday, November 29 2012 @ 04:00 PM EST
As a programmer for over 20 years, I know that programming
is math. Back in those days, computer science was often
taught by mathematicians. I would assume that any competent
professor in computing today would understand that
programming is math. So, why doesn't someone go ask the top
computer science professors whether programming is math or
not?

If I wanted to know about laws, I would ask a judge or
lawyer instead of a computer science professor. Yet when
people want to know about programming, we are asking judges
and lawyers. Makes no sense.

[ Reply to This | # ]

Brilliant! Excellent! Now go find someone to read it.
Authored by: Anonymous on Thursday, November 29 2012 @ 04:55 PM EST
This is an excellent onion gum.

It will be of exactly ZERO use to convince Senators, Representatives, Judges,
the media, lawyers, advocates, lobbyists, and decision makers of ANYTHING.

You do such a great job of making simple nuances complicated that I can't
imagine you have a career outside of academia. In the real world there's simply
no opportunity to employ anyone who takes ten words to say what could be said in
three.

Sadly it took great effort to write this, unless such obfuscatory drivel comes
naturally, and this effort will NEVER be used to effect change, improve patents,
remove software from patentability, or do more than put Groklaw readers to
sleep.

Speaking of which I didn't realize it was naptime - till I read part 2. I was
still waking up from part 1.

Anon - my job involves law and patents - clearly yours does not.

[ Reply to This | # ]

Troll index
Authored by: OpenSourceFTW on Thursday, November 29 2012 @ 06:06 PM EST
I believe the effectiveness of an article is inversely related to the number and
vehemence of the trollish comments left.

By that criteria, this article is a success. :)

---
I voted for Groklaw (Legal Technology Category) in the 2012 ABA Journal Blawg
100. Did you? http://www.abajournal.com/blawg100. Voting ends Dec 21.

[ Reply to This | # ]

PJ - can we have a permanent PolR section in the sidebar?
Authored by: Anonymous on Thursday, November 29 2012 @ 06:52 PM EST
I mentioned this on the last article, too. These essays are
enormously powerful, and if we could get our policy makers
to read them, we might have a shot at ending this plague.

I looked through the Patents section just now, and I see
there is an impressive collection of links and articles. It
is possible that PolR's work is in there somewhere. But
these essays are of such importance that I think they
deserve a dedicated section to which decision makers and
other stakeholders can be directed.

Thanks,
David Bruce

[ Reply to This | # ]

Interpretation of a programmed "General Purpose" Quantum computer
Authored by: Anonymous on Thursday, November 29 2012 @ 07:01 PM EST
PoIR, thank you for the thorough analysis in your continuing series. We often
have a tendency to stop right after proving to ourselves that software is a
branch of mathematics, without placing that fact in the context of an actual
patent test. I am enthusiastic about this semiotic approach that you're
describing, as a test based on the exclusion of advancements to interpretants
seems like it might be more attractive to the courts and examiners office
by-and-large than the direct proof of software being a branch of mathematics. We
know that the arguments are the same but the "Software is Mathematics"
mantra seems to be difficult for many, particularly non-programmers, to
reconcile with their own perception of software as collections of interactive
"objects" and "tools".

For the sake of completeness though, I would like to bring up one issue that may
arise from interpreting the computer as a sign when determining infringement.
You write, "The opposite view -- that there is no sign unless someone must
actually watch the sign-vehicle -- leads to absurd results." However this
is precisely how a quantum computer operates. Unlike a bit which always must
exist in one of two states, a qubit does NOT exist in any defined state until a
measurement is taken. This is not merely semantics or a lack of adequate
measuring tools, it is a fundamental principle of quantum mechanics: that qubit
does not exist in any one state until it is measured.

For the time being this is mostly an academic exercise, but we are not far away
from this becoming reality. I wonder if "quantum software" would still
be safe from patents using this test?

[ Reply to This | # ]

New machines and toilet doors
Authored by: soronlin on Friday, November 30 2012 @ 06:14 AM EST
Old Machines

The central processor (CPU) and central memory (CM) [of the CDC Cyber 72] operated in units of 60-bit words. In CDC lingo, the term "byte" referred to 12-bit entities (which coincided with the word size used by the peripheral processors). Characters were six bits, operation codes were six bits, and central memory addresses were 18 bits. Central processor instructions were either 15 bits or 30 bits,...

eight 18-bit address registers (A0 through A7), eight 18-bit index registers (B0 through B7), and eight 60-bit operand registers (X0 through X7). [There were no LOAD or STORE instructions] Seven of the A registers were tied to their corresponding X register. Setting A1 through A5 read that address and fetched it into the corresponding X1 through X5 register. Likewise, setting register A6 or A7 wrote the corresponding X6 or X7 register to central memory at the address written to the A register. A0 was effectively a scratch register....

ones complement...

-- Wikipedia

The Cyber series of computers, in common with many computers of their era, was a totally different beast than computers of today. Whereas we have settled on word lengths that are a power of two, two's complement arithmetic and either a register or a stack-based approach, the Cyber series differed in all of these respects. Programming it in assembler was a fascinating adventure.

All of which means that to implement an invention on it, every single aspect of that implementation must necessarily be different from the implementation on a more modern machine, even down to the representations of the numbers, and the encoding of characters.

Toilet Doors

On a toilet cubicle door you have a lock. This lock has a knob that you have to move through a defined path in order to lock or unlock the door. Except for such convoluted possibilities as RFID tags, there is no getting away from that fact. To lock or unlock the door you have to move something in some defined way.

That is not, in itself, patentable. It is obvious that anyone skilled in the art of lock design could come up with several different ways to implement such a lock:

  1. A simple bolt action bolt.
  2. A rotating knob that uses gear teeth to move a bolt.
  3. A rotating knob that turns an asymetrical plate such that in one position the plate engages the lock plate.
  4. An electrical switch that powers an electromagnet.
  5. A pivoted bar that rests in a bracket on the door frame in the locked position or can be rotated 180 degrees so that it doesn't.
Each of these implementations is a patentable invention, (excepting novelty and obviousness.)

New Machines?

Let us now suppose that I had an implementation of a Cyber 72 on an integrated circuit, rather than a room full of equipment racks. Let us suppose that I gave it a FORTRAN compiler. Let us suppose that I put this chip into a mobile phone and, with malice aforethought, I implemented slide-to-unlock. However, just to be obtuse, I made it rotate to unlock much like invention 2 above. It still infringes the patent, but is dissimilar to the iPhone embodiment.

One imagines that the implementation of Apple's slide-to-unlock was written in Objective-C, or assembler. As such the bones of their implementation is full of classes and objects, lists and collections. FORTRAN has integers, reals and arrays, and that's all. (OK, it's got complex numbers too.) And on the Cyber72, the integers are 60 bits, not 32 or 64, and negative numbers are represented in one's complement rather than two's complement. Any text is encoded in a different scheme that only has 64 characters in it rather than ASCII's 95 or Unicode's 100,000 or so.

There would be no similarity between Apple's software and the Cyber 72 software. They would look totally different. The APIs would be totally different. The implementation of slide-to-unlock would therefore be totally different. If one skilled in the art of analysing software was to compare them, the only similarity would be at the most abstract level: "they both detect a finger moving along a path, and they both tell the phone's OS to unlock." Possibly, but not necessarily, they might use the same algorithm for tracking the finger movement, but at such a high level of abstraction that one might assume that I and Apple read the same academic paper. Algorithms as such are not patentable.

So if programming a computer creates a new machine, then different software must create a different machine. An infringing product must therefore create a similar machine to any embodiment of the invention. We have shown above that, by the rules and law, no patentable aspect of our Cyber72 phone is similar to any aspect of the iPhone in regards to the slide-to-unlock patent. And yet it implements every aspect of the patent claims.

We have a situation where a product fully implements a patent in every particular, and yet it does not infringe it. That is a paradoxical result. In Maths, Logic and any other rigorous field, a paradox indicates that one of your assumptions is incorrect. We have three options to escape it:

  1. Our product does not infringe because the Cyber72 implementation is not the same sort of computer with the same sort of OS and the same sort of software. But that would significantly reduce, possibly to zero, the number of infringements of any software patent. It would also be a minefield to determine.
  2. The law is wrong. But that means that algorithms are patentable subject matter.
  3. The patent is wrong, since it does not seem to describe any patentable subject.
One might think that we could discard the notion that programming a computer makes a new machine, but the argument still stands, albeit in more abstract terms; we have an embodiment that shares no aspect of its constitution with any other embodiment. The only commonality is the abstract algorithm and the abstract concept.

I believe that everyone, even Apple, given only those three alternatives, would choose to invalidate the patent.

It's Mathematics, Stupid

Maths works. Every time. Consistently. It is not like the law, which can be argued different ways on different days. On days when "is" may be ambiguous, "=" still means the same.

Because software is mathematics, the above argument can be applied to every patent. It will invalidate every patent that does not have significant non-software subject matter.

[ Reply to This | # ]

What Does "Software Is Mathematics" Mean? Part 2: A Semiotics Approach to the Patent Eligibility of Software by PolR
Authored by: Anonymous on Friday, November 30 2012 @ 09:56 AM EST

Most of mathematics is equational or relational -- describing what a formal abstraction is in terms of simpler parts or properties, or establishing bounds using inequality relations.

Both mathematics and CS are complicated by divergence. Infinite series can be convergent (1/2 + 1/4 + 1/8 + ... = 1) or divergent which, in mathematics, allows "equations" that are not equal to anything (1 + 1 + 1 + ... = ?).

Most software abstractions are highly divergent. You can't reason equationally about any general purpose program language because for a program to be reducible to an equation requires it to be proven halting which cannot be done in the general case. For example, the program

def f(i): return f(i+1)
has no answer for the same reason as the divergent series above.

That mathematics at large is so intolerant of divergence while software muddles along so well in the absence of proofs of termination indicates that the value of a program lies not in its ability to define what things are by equational thinking or to establish bounds, but to perform a computations that are actualized, so I think software artifacts are substantially different from the artifacts that result from mathematical methods.

[ Reply to This | # ]

While I agree that software is math,
Authored by: albert on Friday, November 30 2012 @ 05:49 PM EST
and PoIR did his/her usual fine analysis, I'm afraid it's a case of pearls
before swine*. First, I'd like to see the courts definition of an abstract idea.
To me, the phrase "abstract idea" is an abstract idea! Just look up
the definitions. The phrase can have many meanings. This isn't good law. Is it
defined anywhere in a court ruling? Neither word shows up in the USPTOs document
MPEP, Appendix L. Perhaps the concept can be defined in relation to software
patents, but I think it will always be subjective. Patents are effective when
the subject is a machine. That was the original objective. You can't apply the
old law to software; it didn't exist back then. The USPTO started patenting math
circa 1976, IRRC. Now, there are at least 10 important ones (see rsa.com).
Congress should have stopped it dead then. If you can't patent math, then those
math patents are invalid.

Point #1, we already have math patents. How does this help the 'software is
math' argument? The software patent problem should not be decided in the courts,
which have been shown to be incapable of doing so. That's not a criticism of the
courts, they're just trying to apply ancient laws to modern technology. We've
seen what happens when the courts try to extend the law (Citizens United).
Sometimes, it's a disaster.

Point #2, 'IP lawyers' think software is 'intellectual property', a term with no
formal definition. Is a patent or copyright 'property'? They are treated as
property, but no, they are _protections_ of useful property, building a monetary
wall around an idea (in copyright, the _specific expression of ideas_, in
software patents, the algorithm(s) expressing the ideas). A patent is much like
a restraining order. Since patents are 'intellectual protection', we can
eliminate trolls, since they have no useful property to protect, only patents
describing the property, and universally, poor descriptions at that..


Point #3, Machines(circuits) vs. Software.
I build a really cool circuit that compresses/decompresses data in a unique way,
and patent it. I can sue anyone who copies the circuit. Now, someone comes along
with a general purpose computer that does the same algorithm in software.
Drat! I should have patented the algorithm! A circuit is a physical object.
Patent infringing is easy to detect, and deal with. What if I use gate arrays?
The programming of a gate array can be unknowable, just like a ROM program in a
CPU. How do I detect infringement? I can't, unless I can subpoena the source
code. A competitor could create a circuit that accomplished identical results,
in an entirely different manner, and escape infringement. I leave it to the
mathematicians to decide whether different algorithms can produce identical
results given the same input data, and, if so, is this a rigorous proof of
infringement.


Point #4, Algorithms vs. Processes
Some folks say that algorithms are processes, and so are patentable. They take
refuge in defining both as 'a sequence of steps that accomplish some end'. In
the old days, a process ending with a product: a chemical element, compound, or
alloy; or a physical item like a bolt, tube, rail, etc. Often, a machine wasn't
necessary. With the advent of software in everything, things got fuzzy. Let's
take the 'swipe patent', since we're sick of hearing about it. It's essentially
a patent on a gesture, or strictly, the interpretation of a gesture. How does
this bogopatent even remotely resemble a process? Does it produce a product? No,
it produces data patterns, theoretically hundreds of thousands of data patterns,
not to mention the interpretation of those patterns. Clearly, this does not
conform to the traditional concept of a product, so 'process' cannot be applied
here.

Point #5, Detection of Software Patent Infringement
How is software patent infringement detected in the marketplace? It can't be
detected, unless the situation involves a patent on an action, or series of
actions, steps, if you will (like the swipe bogopatent), playing a audio or
video file with proprietary encoding, or existing in an environment where
proprietary patents must be used (like the mobile phone), etc.

Conclusions
1. Software patents need to be eliminated. It's the simplest, fairest, and most
economical thing to do. All existing s/w patents should expire in 1 year. There
will be an explosion of innovation and competition, I guarantee.
2. Patenting algorithms needs to be eliminated, and all such patents
invalidated.
3. Process patents should produce a product, in the original sense of the word.
4. Business methods patents should be eliminated. They are even more absurd then
s/w patents (if that's possible).

Changing the law to accommodate s/w patents would be a magnitude 11 disaster.

Footnote

In my years of reading Groklaw (Thanks PJ, Mark, et.al.), I've learned a lot,
and often about things that were of no interest to me, that I now find
fascinating. Outside of this community, I still feel like a voice crying out in
the wilderness, which is frustrating sometimes. So, everyone, keep up the good
work,
I'm pullin' for ya, we're all in this together...

Albert


----------------------------------------------------------
*Caution, resist the urge to respond, look it up instead.

[ Reply to This | # ]

I've skipped over the important bits
Authored by: Ian Al on Saturday, December 01 2012 @ 12:50 PM EST
I realised, early on, that PolR had missed the key aspect of applying the theory to inventions on-a-computer. I can't believe he laid out all the facts so well and missed this major issue!

Here is where he went wrong:
The programmer must define the conventions on the syntactic organization of the bits and the corresponding semantics. Then he must define the operations of arithmetic and logic which will solve the problem. This is all interpretants.
The day some clever clogs invented assembler language programming and soon after when more clever clogs invented higher level languages, the programmers and computer designers were divided into many different camps, each with their own common symbol set. There was the processor circuit designer who shared the same sign language with the programmers who originally coded direct to binary instructions. Assembly language programmers had the compiler to translate the symbology defined as the assembler language into machine code binary.

Now, as PolR points out, one can go up, layer after layer, via a software based universal algorithm to the Dalvik assembler level. Even then, programmers of apps do not, typically, use Dalvik assembler.

They use a higher level language such as Java because there is more in the Dalvik virtual machine than the universal algorithm. There is also the manipulation of objects that have Dalvik symbology and the object based programming language must be translated into Dalvik objects (complex symbols with properties) which are then manipulated using the Dalvik universal algorithm.

There is also a horizontal separation which applies at each layer of programming. It starts at the BIOS and works its way across via the operating system, across APAs and, potentially, across further divides until it is presented as a symbol set of library functions to the programmer. The programmer who writes a 'Hello World' program only has access to the print-to-screen symbols in his semiotics workspace. He knows nothing of OS display APAs and video driver symbology.

The textual symbol, CAT, does not refer to a specific cat, but the interpretant will usually be a small feline suitable to be a pet (or not, if you don't like cats). Pets, feral cats and wild cats would be referents, although this might not extend to the big cats. The sign-vehicle in this case is the text displayed on your computer monitor. Between the sign-vehicle I used when composing this comment and the sign-vehicle you used when reading the sign, the symbol transferred to many other sign vehicles.

In the good old days, one of the sign-vehicles would have been a series of addresses containing eight-bit byte signs of ASCII code in the guts of the computer (I'm so glad they did not patent that world standard!). I doubt I would recognise the sign in the computer I am now using. How it looks in an IP packet is anyone's guess.

The important thing is that I was the author that wrote the sign. Only the interpretant is shared between us and it exists only in our minds. The referent is sat on the mat. The sign (CAT) was passed from sign-vehicle to sign-vehicle between us.

A high-level language programmer does not pass the referent or the interpretant down to the processor or out to the BIOS by the act of interpreting or compilation. In fact, there are only signs in the source code and so there cannot be a referent or an interpretant in the source code. The symbols specific to the program are translated by the programmer into a mathematically valid programming language and the interpretants of the program-specific symbols have to be described in source code comments if they are to be shared with other programmers. Only the interpretants of symbols in the programming language, itself, are shared between programmers who 'know' the language (their minds contain the sign/interpretant relationships of the computer language). The code is always just the sign part of the triad.

Interpretants are always in the mind of the programmer and the programmer will also have to know what the referent is in order to write the program. The referent is what in the real world is represented by the signs in the computer. The signs do not, typically appear to the program user. The programmer intends the result of symbol manipulation to be 'seen' by the user (the symbol manipulation is the sole purpose of the program).

The vital question is which of any of this is patent protectable under patent law and is there a test of a patent to determine if the user or installer of software written by a writer and compiled/interpreted at several software boundaries into machine code and installed on-a-computer can infringe on a valid patent?

Please note that the Supreme Court has opined that no software writing can infringe on a patent. Only the manufacture, importing, selling or using of machines, following processes, executing methods and transforming of materials can infringe on patents.

Lets start with Bilski because computers are so hard. Bilski was not an on-a-computer invention. It could not be used without a computer because of the extensive and complex mathematical analysis and the need for extensive and timely data on energy suppliers current deals and the institutional energy users business models. It made a buck by estimating which energy deals could be 'sold on' to energy users, thus making a middle man margin. The patent did not claim a computer.

The Supreme Court found that patent law did not exclude business methods because parts of the word of the law would have no meaning if that were the case.

The Supreme Court found that their previous landmark cases of Benson, Flook and Diehr showed that Bilski was insufficient to be a business method patent and was, thus, abstract ideas.

A business method is the manipulation of symbols. It does not matter that a symbol in the Bilski business method is of any one energy deal any more than it is important that 'cat' means any one cat. The Supreme Court set the principle in Bilski, Flook and Diehr that there had to be specific and significant post-process production of a referent for a patent to be valid. From Bilski, with all the twiddly bits removed:
Even though petitioners’ application is not categorically outside of §101 under the two broad and atextual approaches the Court rejects today, that does not mean it is a “process” under §101. Petitioners seek to patent both the concept of hedging risk and the application of that concept to energy markets. Rather than adopting categorical rules that might have wide-ranging and unforeseen impacts, the Court resolves this case narrowly on the basis of this Court’s decisions in Benson , Flook , and Diehr , which show that petitioners’ claims are not patentable processes because they are attempts to patent abstract ideas. Indeed, all members of the Court agree that the patent application at issue here falls outside of §101 because it claims an abstract idea.

In Benson , the Court considered whether a patent application for an algorithm to convert binary-coded decimal numerals into pure binary code was a “process” under §101. The Court first explained that “ ‘[a] principle, in the abstract, is a fundamental truth; an original cause; a motive; these cannot be patented, as no one can claim in either of them an exclusive right.’ ” (quoting Le Roy). The Court then held the application at issue was not a “process,” but an unpatentable abstract idea. “It is conceded that one may not patent an idea. But in practical effect that would be the result if the formula for converting . . . numerals to pure binary numerals were patented in this case.” A contrary holding “would wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself.”

In Flook, the Court considered the next logical step after Benson . The applicant there attempted to patent a procedure for monitoring the conditions during the catalytic conversion process in the petrochemical and oil-refining industries. The application’s only innovation was reliance on a mathematical algorithm. Flook held the invention was not a patentable “process.”

The Court conceded the invention at issue, unlike the algorithm in Benson, had been limited so that it could still be freely used outside the petrochemical and oil-refining industries. Nevertheless, Flook rejected “[t]he notion that post-solution activity, no matter how conventional or obvious in itself, can transform an unpatentable principle into a patentable process.” The Court concluded that the process at issue there was “unpatentable under §101, not because it contain[ed] a mathematical algorithm as one component, but because once that algorithm [wa]s assumed to be within the prior art, the application, considered as a whole, contain[ed] no patentable invention.” As the Court later explained, Flook stands for the proposition that the prohibition against patenting abstract ideas “cannot be circumvented by attempting to limit the use of the formula to a particular technological environment” or adding “insignificant postsolution activity.”(Diehr). Finally, in Diehr, the Court established a limitation on the principles articulated in Benson and Flook . The application in Diehr claimed a previously unknown method for “molding raw, uncured synthetic rubber into cured precision products,” using a mathematical formula to complete some of its several steps by way of a computer.

Diehr explained that while an abstract idea, law of nature, or mathematical formula could not be patented, “an application of a law of nature or mathematical formula to a known structure or process may well be deserving of patent protection.”. Diehr emphasized the need to consider the invention as a whole, rather than “dissect[ing] the claims into old and new elements and then … ignor[ing] the presence of the old elements in the analysis.” Finally, the Court concluded that because the claim was not “an attempt to patent a mathematical formula, but rather [was] an industrial process for the molding of rubber products, ” it fell within §101’s patentable subject matter.
Benson translated one symbol into another symbol with the identical interpretant. Flook manipulated one set of symbols and transferred the result to the production process as alarm limit symbols with different interpretants. Diehr used the novel process to produce referents; precision, cured-rubber components. It did not matter that the actual referents, say rubber ducks, were symbolised in the process. It mattered that, once the process was finished, the process produced significant referents as claimed in the patent.

So, why was Bilski abstract ideas? The process provided a symbol set to guide the middle man. It did not result in a new referent or referent set. As the courts should always do, §101 was considered first and the invention was ruled out as unpatentable subject matter because it was abstract ideas. If the court had gone on to consider the other sections, they would have found that the only novel concept was the math algorithms used. Hedging and dealing in the energy market were prior art. The math algorithms were non patentable subject matter and also prior art because the courts treat all math as prior art no matter how novel it may be. From the quotation, above,
The Court concluded that the process at issue there was “unpatentable under §101, not because it contain[ed] a mathematical algorithm as one component, but because once that algorithm [wa]s assumed to be within the prior art, the application, considered as a whole, contain[ed] no patentable invention.
If the execution of a computer program results in symbols being displayed on a screen, printed on paper or stored in a medium, that is not the production of referent.

Let's consider a system comprising a computer and a mixing desk used to record music and duplicate a music CD of the music. The music CD is the referent being produced. I doubt that there is a patentable novel invention in there, but, if there was, it would have to be the whole process of recording the performance and duplicating the CDs and not just the burning of CDs.

Without the music recording process, running the program to burn several music CDs is no different to copying any data file to any CD. It is only the tying-in of the recording of the music file into the invention that provides any potential for referent production. '[T]he prohibition against patenting abstract ideas “cannot be circumvented by attempting to limit the use of the formula to a particular technological environment. or adding “insignificant postsolution activity.”' It does not matter how often the symbols are transferred from one sign vehicle (the file) to another sign vehicle (the CDs) no referent has been involved. Narrowing the invention to just symbols representing music does not change this into patentable subject matter.

Looking at the 'swipe to lock' invention, a swipe of a finger and 'locking' are two abstract ideas. In a touch-screen device, a programmer may put symbols into his program that represent a particular sort of finger swipe across the touch screen panel and narrow the 'locking' symbol to 'preventing some or any physical buttons on the device, or symbols representing the actions of physical buttons, initiating their symbol manipulative actions'.

A finger swipe and a button operation cannot be signs in a computer program. They can only be represented by signs. The interpretants of those two signs will only be known to the programmer, unless he adds program comments to share the interpretants with other programmers. The programmer can only manipulate those and other symbols in the computer.

This particular program can only be modelling the finger swipe and button push referents. Abstract ideas cannot be patented and the modelling of abstract ideas or anything else by manipulating symbols cannot infringe on a valid referent-based patent.

A programmer cannot create referents by manipulating signs from the sign-interpretant-referent triad. Only by tying other specific systems with the smartphone and having the process encompassing the program generate a referent such as a precision rubber product can the overall process have any potential for patent protection. The program in the smartphone is not such a process/method as a stand-alone device. Nor is it a new machine in its own right.

The invention, without limiting the manipulation of the symbols to a particular technological environment, is just abstract ideas. One can neither have a box of finger swipes nor a box of locking. Limiting the abstract ideas to the technological environment of a touch-screen device is only possible if a programmer redefines the symbols to make sense in that technology. The symbols created by the programmer are not the abstract ideas in the patent. Therefore, they cannot infringe on the patent even if the patent were valid, which it cannot be.

It is worth repeating that patents cannot be infringed upon by just 'doing things' even if it is doing things with a smartphone or doing things in programs. Only the manufacture, importing, selling or using of machines, following processes, executing methods and transforming of materials can infringe on patents.

What if the touch-screen device was not a mobile phone? What, in the invention, is to be locked? If the patent had said 'the iPhone has a lot of virtual buttons that do stuff. I want to monopolise the concept of disabling some or all of those virtual buttons when used on a touch-screen phone' then it would have been obvious that the invention was bogus.

Applying the science of computing and the philosophy of semiotics, any invention to do solely with the manipulation or the modification of the manipulation of symbols cannot, by definition, act on referents and is both math (manipulation of symbols in a computer) and abstract ideas (the selection of symbols shared between the inventor and the user) and is non-patentable subject matter on both counts.

The philosophy of semiotics fits the cornerstone patent decisions of the Supreme Court and provides the intellectual backing to the, hitherto, pragmatic decisions of the court ('Indeed, all members of the Court agree that the patent application at issue here falls outside of §101 because it claims an abstract idea').

How very satisfactory!

---
Regards
Ian Al
Software Patents: It's the disclosed functions in the patent, stupid!

[ Reply to This | # ]

What is art?
Authored by: Ian Al on Sunday, December 02 2012 @ 04:28 AM EST
Once there were two camps of artist; the figurative artist and the abstract
artist. Art is the manipulation of paints, pencil or charcoal to produce the art
which may be symbolic of something real or it may be abstract.

That's when it is not sculpture. Or performance art. Or an installation.

The figurative paintings are symbols conforming to the semiotics triad. However,
the symbols are not what they seem. Magrite painted a picture of a pipe entitled
"Ceci n'est pas une pipe" - 'this is not a pipe'. The painting was a
symbol of a pipe. The interpretant is not 'a pipe in the mind of a viewer'.
There is no referent. The painting and the title of the painting are essential
components of the painting. The interpretant of art is in the eye of the
beholder.

It is not possible to patent the method or process that an artist uses, because
it cannot be completely expressed in patent claims. The art is only protectable
by copyright because it is creative expression. It is not a manufactured
machine.

The artist might create an expressive work of art by arranging an unmade bed, or
a tent with the names of previous lovers pinned inside (Tracey Emin). Another
artist may 'paint' by trickling paint in a stream from a paint can or splashing
paint from a brush (Jackson Pollock).

Computer artists have used the computer screen and mouse as canvas, paintbrush
and palette. Other computer artists have programmed the manipulation of shapes
and colours by using computer language programs executed by the computer. Yet
more have provided the facility of tactile surfaces for the viewer to interact
with the art. Again, the bed, the tent, and the computer are not the referent
nor the interpretant. Often the title is an essential part of the art.

As the exercise of the law in the US lower courts stands, what a programmer
makes visible on a computer screen or printer, or audible via the speakers is
patentable as long as the symbols in the sign-vehicle that he manipulates have a
fixed interpretant and a fixed referent.

The vital legal question that the courts fail to put is 'is it art?'. If it is
art then it is non-statutory subject matter.

If it is the mathematical manipulation of symbols or the symbolising of abstract
ideas in a real medium whereby the signs have a specific interpretant and
referent , then, in the eyes of the lower court, it is patentable, statutory
subject matter. This is even though both abstract ideas and mathematical
manipulation of symbols are ruled non-statutory by the Supreme Court.

The lower courts are ruling that if the symbols manipulated in a program have
the same interpretant and referent as the symbols claimed in a patent, then the
patent is infringed even if the programmer had no intention of using symbols
with the same interpretant.

The only interpretants in source code are those formed by using the source code
computer language. The patent system has precluded the programmer from using any
symbols in his program that have the same interpretant set and are manipulated
in the same way as the claims in a patent.

When it comes to evaluating whether the programmer has used the symbols and
symbol manipulation in his program that is protected by the patent, then the
judge or jury decide on the basis, not of the symbols used and manipulated, but
whether it looks like it to the lay person.

---
Regards
Ian Al
Software Patents: It's the disclosed functions in the patent, stupid!

[ Reply to This | # ]

What Does "Software Is Mathematics" Mean? Part 2: A Semiotics Approach to the Patent Eligibility of Software by PolR
Authored by: Anonymous on Sunday, December 02 2012 @ 08:58 AM EST
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
examines the nature of software.

It is reviewed by Slate at:
http://www.slate.com/articles/technology/books/2012/11/computer_programming_10_p
rint_chr_205_5_rnd_1_goto_10_from_mit_press_reviewed.html

From the book:

"Computer programs process and display critical data, facilitate
communication, monitor and report on sensor networks, and shoot down incoming
missiles. But computer code is not merely functional. Code is a peculiar
kind of text, written, maintained, and modified by programmers to make
a machine operate. It is a text nonetheless, with many of the properties of
more familiar documents. Code is not purely abstract and mathematical; it
has significant social, political, and aesthetic dimensions. The way in which
code connects to culture, affecting it and being influenced by it, can be
traced by examining the specifics of programs by reading the code itself
attentively.

Like a diary from the forgotten past, computer code is embedded with
stories of a program’s making, its purpose, its assumptions, and more. Every
symbol within a program can help to illuminate these stories and open
historical and critical lines of inquiry."

Free download from: http://10print.org/ Creative Commons licence

[ Reply to This | # ]

Distinction of Utilitarian / Ethics / etc aspects
Authored by: Anonymous on Wednesday, December 05 2012 @ 09:31 AM EST
You start with a distinction of different aspects of patentability of software.
The article would in my opinion profit from making these distinctions more clear
and where your argument stands.
These distinctions by the way also apply to other forms of "intellectual
property"; obviously, one must make the

1. Ethics / morals. There are different opinions on the ethics of IP
protection.
E.g. Ayn Rand would say that intellectual property is, like property in general,
some sort of sacred "natural right" and society should try to honor IP
as much as possible.
E.g. Stallman would say that "information wants to be free" /
computation and information sharing is a human right and society should try to
honor IP as little as possible.
E.g. the US constitution seems to take the same view as Stallman (but is willing
to take different compromises). This is ideology and not politics.

2. Utilitarian. Which does benefit society more?
There are good arguments for either conclusion on calidity of IP. Luckily, this
is mostly an empirical question and not a moral one- so it is for scientists to
decide, not for judges or politicians.

3. Compromise. After the first two questions are decided, a compromise must be
reached. That's politics.

4. Coherence. Whatever compromise society is going to make should not be too
contradictory- but some contradictions are OK. Compromise again.

5. Existing Law. IANAL, so this is for you.


You seem to argue that "software patents are valid" and
"mathematics is not patentable" are contradictory statements.
Even more, you seem to argue that "there exist valid patents" and
"mathematics is not patentable" are less contradictory than the first
pair of statements and we therefore lose coherence by such a state of affairs.

I see three immediate problems with your argument.

1. You do not argue, why the contradiction is so bad. Sometimes two rules clash;
so what?

2. Your argument does not make it clear, where computers/ computer programs are
different from e.g. a windmill. A physicist might see
"windmill" = "laws of nature" + "abstract maths"
"computer program" = "abstract maths"
Since "laws of nature" are not patentable and the relevant and known
ones fit on a single sheet of paper anyway (and are discovered for more than 30
years), the difference is not obvious.

3. Your coherence argument contains at its core something like "there is no
coherent rule separating abstract ideas and software patents". You use
arguments like "the interpretation of voltages is up to human
interpreters".

This lacks imagination on how complex and still coherent a rule might be. For
example, the interpretation of voltages is a single bit. You do not need to know
the charset used by a text-outputting program in order to interpret it: This is
just a Cesar's chiffre, which is easily crypto-analysed.

Software which is reverse-engeneerable does not need to carry context with it,
it is self-interpreting.

Or, in a different notation: (Compressible) Data does not need context
(conventions for human interpreters) in order to be understood. Good reading:
wikipdia "Kolmogorov Entropy".

[ Reply to This | # ]
Groklaw © Copyright 2003-2013 Pamela Jones.
All trademarks and copyrights on this page are owned by their respective owners.
Comments are owned by the individual posters.

PJ's articles are licensed under a Creative Commons License. ( Details ) 		Site layout based on Woodlands theme by Bryan Bell.
Groklaw logo by John Crowley.
News Picks logo by Ted Thompson.
Powered By GeekLog
Created this page in 0.05 seconds