Author Topic: Advise the President  (Read 4919 times)

0 Members and 4 Guests are viewing this topic.

Offline aldo_14

  • Gunnery Control
  • 213
That doesn't matter to me.  From what you've stated, the PO holds all the aces it needs to push through funding requests.  I don't believe an organisation should be exempted from moderating itself - at the moment the PO system will not improve, because it suits the major companies; the system allows the types of patents like IS-NOT to be approved, does so in an amount of time that handicaps small business (all that time their patent is available for reverse engineering by companies with sufficient R&D budgets), and allows these nonsense patents to be created for the sole purpose of being a big stick to intimidate smaller competitors without the financial wherewithal to legally challenge them.

I actually worked on implementing a newly-applied software patent last yer  (07/04 to 04/05 to be precise).  I'm under an effective NDA not to reveal any details about that.  But I can search the US patent system, and find that patent - and specifically the text that details exactly how that design was implemented (and this is still a pending application).  On the UK side of things - not a sniff of those details.  But that's massively open to abuse right there, because anyone can grab that, and begin an implementation.

 

Offline Osiri

  • 24
Okay you are misinterpreting what I  have said.  I never said the USPTO holds all the aces.  Congress and the President do.  The invalid patent are issued to small companies too.  The time to reverse engineer does not matter because you can keep someone from using your patent with the current delay.  Right now there is a balance struck where your invention is protected if you secure a valid patent.  

Plus there is no sense in even saying reverse engineering when it comes to patents because when you file for a patent you disclose every relevant detail of the invention.  Reverse engineering only applies to trade secrets.  

As to your last comment about the patent office search... THAT IS EXACTLY WHAT PATENTS ARE FOR :D.  The point of the US Patent Practice is to make a deal with inventors.  You disclose to the public every stinking detail of your invention and we will grant you the EXCLUSIVE RIGHT to use it.

What I was saying about the delay would only be if there was a delay of 5 - 10 years before publication or something like that.  If the delay got so bad that there was a 5 year delay in the publishing after filing there I believe the doctrine of Laches would apply and if not then I'm sure some other doctrine would.  I mean if you don't know your knocking off someone elses invention then your pretty safe.

EDIT: the point of the publishing of every detail is basically a early warning and protection concept.
« Last Edit: October 16, 2005, 05:48:47 pm by 3173 »
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.

 

Offline Kamikaze

  • A Complacent Wind
  • 29
    • http://www.nodewar.com
Preferably pure software patents will be dropped entirely, regardless of if they have prior art or not. The USPTO used to reject all pure software patents, but recently has succumbed to the stupidity. If you can't patent mathematical truths, why can you patent algorithms?
Science alone of all the subjects contains within itself the lesson of the danger of belief in the infallibility of the greatest teachers in the preceding generation . . .Learn from science that you must doubt the experts. As a matter of fact, I can also define science another way: Science is the belief in the ignorance of experts. - Richard Feynman

 

Offline aldo_14

  • Gunnery Control
  • 213
If the patent office can hold up the granting of patents for proper prudence, and the avoidance of delay is the key issue here, then of course the PO has control.  Are you saying that Congress would sit back and do bugger all if the PO decided it needed to double it's examination time for due prudence?

Reverse engineering makes perfect sense, in 2 manners.  Firstly, foreign markets.  Secondly (and more importantly), you can incorporate existing knowledge in order to create a derivative that does not infringe the patent.  

I'd point out that the details (I'm referring to) published are effectively technical details; they're the exact same blueprint I used to write the (well, bits of) code.  They tie in a very tight implementation to the concept, but the concept itself would, IMO, be easily bastardizable to be a just-enough-different version.  And that in turn would mean the tech advantage from being first to market would be wiped out.

 

Offline Osiri

  • 24
Because every thing in the world is a mathmatical algorithm.  Lawyers have a rule that someone should not be able to use someone elses brilliance to do thier own job.(this is the work product doctrine to all you law types)  Physics is mathmatical truths.  Chemists work with mathmatical formulas.  Computer hardware works with mathmatical truths.  They are all simply the practical use of it.  No formula like E=MC2 should not be patentable.  But if it were brand new, a bubble sort should be able to be patented.  Just because you can express it as mathmatics does not mean it is not an invention.

Further, I am sorry but if you are going to take away my rights to my innovative software what the hell is the point of me developing it.  What you are asking for is to hand the software industry to the big corps.  

Say a little company makes a new innovation.  Well, if they cannot have patent protection the big companies steal..errrr integrate that innovation into thier own software and bam.  The big company got someone to do free research for them.  Any other questions.
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.

 

Offline TrashMan

  • T-tower Avenger. srsly.
  • 213
  • God-Emperor of your kind!
    • FLAMES OF WAR
Quote
Originally posted by Osiri

He is like Bush 2.0 except made by Microsoft.  He is gauranteed to have bugs.  They need almost daily updates on him to keep him from pushing the little red button.  Last week Cheaney was cheering him on as the Secret Service was desparately trying to keep him from pushing it while they updated his software.  We are not safe folks.


LOL! good one!

And those bugs are incorectable.. the code is the mother of all spaghetti!

Of course, while I hate Bush, I don't belive EVERYTHING he does is wrong..
Nobody dies as a virgin - the life ****s us all!

You're a wrongularity from which no right can escape!

 

Offline Osiri

  • 24
AHHHHHHH It just lost a 2 page response.
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.

 

Offline Kamikaze

  • A Complacent Wind
  • 29
    • http://www.nodewar.com
Quote
Originally posted by Osiri
But if it were brand new, a bubble sort should be able to be patented.  Just because you can express it as mathmatics does not mean it is not an invention.


That's the scariest thing I've read today. You're telling me a simple mathematical algorithm that can be applied in thousands of different applications should be patented? Mathematical algorithms aren't "invented", they're discovered. There is a optimal solution to many programming problems after all (in terms of efficiency). Physical inventions are complex and don't restrict innovation much because it usually isn't possible to have a perfect machine.

Meh. I suppose Newton should've patented the derivative when he "invented it" so Leibniz couldn't develop it any more...

Quote

Further, I am sorry but if you are going to take away my rights to my innovative software what the hell is the point of me developing it.  What you are asking for is to hand the software industry to the big corps.  


Uhh, no. Most of the pure software patents are made by the big corps. In fact, most software patents are not even made by software companies. What you want to do is create a situation where virtually no software can be developed because every bit of code is patented.

Software patents just stifle development. You won't be able to make your "innovative software" because all the algorithms you'd want to use are patented. Furthermore, good applications aren't based on the algorithms they use. They're based on the quality of the whole thing. 3DSMax's 3ds file format doesn't necessarily make the software better than, say, Softimage or Lightwave. The GIF format's compression algorithm doesn't make it better than PNG or JPEG.

EDIT: To give some numerical backup for my argument, here's some info from a paper I wrote:

Quote
Firms in the software industry received about 7% of software patents in the second half of the 1990s (Bessen 6). 75% of software patents were granted to manufacturers, not software developers (6).


The citation is for "Bessen J. & Hunt M. R. (2004). The Software Patent Experiment."
« Last Edit: October 16, 2005, 06:38:48 pm by 179 »
Science alone of all the subjects contains within itself the lesson of the danger of belief in the infallibility of the greatest teachers in the preceding generation . . .Learn from science that you must doubt the experts. As a matter of fact, I can also define science another way: Science is the belief in the ignorance of experts. - Richard Feynman

 

Offline aldo_14

  • Gunnery Control
  • 213
There is a difference between innovative software and basic algorithms.  If you hold back concepts at a base level, you hold back technological progress.   What if someone trademarks the concept of sending messages in binary? bye-bye internet.

Or the good old IS-NOT case?  Well, bye-bye every single program ever that uses a !=.

Or amazons patent on business systems?  Enforce it, and say goodbye to every single on-line shop (pretty much; IIRC it covers use of a credit card for online transactions or somesuch).

That's the problem of nonsense patents; patents on prior art that are granted without proper examination, and financially uncontestable for many small businesses.  And crippling for open-source software, which is what their real purpose is for.

Ok, fair enough, that's not patents per se.  If you actually noticed, my arguement was partially about publishing patents when we have that sort of flimsy system.

Now...  I don't agree with algorithmic patents, simple things like bubble sort, at a low level.  Largely as it restricts technological progress.  Demand royalties for something like bubble sort and you kill 90% of the work that would go on to improve that.   Or at the very least, waive it for non-financial use like academic or OS.  An algorithm is, after all, nothing but a formula (Specifically; An explicit step-by-step procedure for producing a solution to a given problem).

Patent an implementation using that, sure, like maybe a distributed bubble sorting topology for ordered message sending.  But otherwise it's no different from patenting the equals operator.  (incidentally, UK patent law forbids patenting methods for performing mental or computing acts; so a bubble sort would not be patentable)

  To apply your analogy to the real world - we'd have one car manufacturer (patent the use of internal combustion, or some key component part), etc, because basic principles would be patented.  Now that may be nice for the car manufacturer or whatnot, but for the rest of us, it's !!!!.

So perhaps I'm hurting small companies a bit, but in my mind the benefit to academic and private work is invaluable in actually progressing the field.

And my whole (other) point was... publish a patent details, and a big company can take that concept, reverse engineer it, incorporate it and make it just different enough not to be sued.  Assuming the patent holder can identify the use of it, of course, in which case it is invisible.

EDIT; the European Patent Convention of 1973 bars Software from being patented, for the same reason as scientific theories, mathematical methods, medical discoveries, etc are barred.  The regarding of software as purely a product is, I think, a very poor attitude to have of what is fundamentally a science.  If we begin patenting software, why not physics theories?  So that anyone wanting to study the Theory of Relativity has to pay £1,000 for each month of work?
« Last Edit: October 16, 2005, 06:36:21 pm by 181 »

 

Offline Osiri

  • 24
QUOTE ALDO
___________________________________----
If the patent office can hold up the granting of patents for proper prudence, and the avoidance of delay is the key issue here, then of course the PO has control. Are you saying that Congress would sit back and do bugger all if the PO decided it needed to double it's examination time for due prudence?
_________________________________________

Aldo, Congress does whatever the hell those 100+400 fat lazy !!!!ing !!!!!!!s want to do.  They are too busy worrying that some 14 year old rape victim might get an abortion in her third week of pregnacy or that two gay guys might be plugging each other somewhere to worry about the inconsequential things like patents and a balanced budget.  

There is also the problem of statutorially mandated response and examination time limits(IOW deadlines).  A applicant has a maximum of 6 months statutorially to reply to the USPTO(fine there but thats paying a 3 month extension fee too most of the time). But the USPTO has its own time limits set out in the USC, the CFR, and the MPEP.  Reexamination has the shortest time period of all which is a good thing too. Some of these time periods can be as short as a month.  
EDIT: WHAT I AM SAYING IS THEY CAN'T DOUBLE EXAMINATION TIME FOR PROPER PRUDENCE BECAUSE OF THE STATUTORY DEADLINES.


QUOTE ALDO
___________________________________----
Reverse engineering makes perfect sense, in 2 manners. Firstly, foreign markets. Secondly (and more importantly), you can incorporate existing knowledge in order to create a derivative that does not infringe the patent.

I'd point out that the details (I'm referring to) published are effectively technical details; they're the exact same blueprint I used to write the (well, bits of) code. They tie in a very tight implementation to the concept, but the concept itself would, IMO, be easily bastardizable to be a just-enough-different version. And that in turn would mean the tech advantage from being first to market would be wiped out.

____________

Aldo, you make an excellent point.  However, you are missing the fact that Patents don't work that way.  First, patents exist for two reasons.  Most importantly, to disclose the FULL OPERATION of any new invention.  One of the requirements of a patent is an enabling disclosure that must enable one skilled in the art to make use etc the invention.  Second and less important, reward the inventor for the disclosure by giving him the exclusive monopoly for 10-20 years.  

As far as simple derivation, patents unlike trade secrets cannot be ripped off in that manner.  Actually if you can figure out a trade secret by reverse engineering its yours.  Back to patents.  To derive a new patentable invention from an old one you must do more than make some simple improvement.  You must have, as my patent professor would say, that BOLT OF LIGHTNING TYPE INNOVATION.  In other words, it must be a new invention in and of itself.  I could not simply take your code and reverse and modify a couple of things and call it mine.  We don't award things that anyone could do.  

If I have an invention and you make a patentable extension of it you are likely to have to pay me a license fee just to use your own patent.  

As far as foreign markets... that is what the WTO PCT is for.  



THERE I THINK I GOT IT ALL IN THERE from where it lost it all last time
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.

 

Offline vyper

  • 210
  • The Sexy Scotsman
Use the quote button or I'll beat you with a shivan plushy...
"But you live, you learn.  Unless you die.  Then you're ****ed." - aldo14

 

Offline Osiri

  • 24
LOL a shivan PLUSHY


Aldo, if you invented this brilliant new type of sorting algorithm that was different from anything ever done before, would you want a big company to be able to use it freely whenever they wanted.  Essentially you get nothing for your work.  

The end result.  I WOULD NEVER DISCLOSE MY NEW ALGORITHM TO THE WORLD AT ALL.  I WOULD KEEP IT AS A TRADE SECRET AND SAY F**K ALL YOU.  YOU ARE NOW HURTING THE PROGRESS OF SCIENCE.

FURTHER TO SAY SCIENCE IS UNPATENTABLE IS WRONG.
The US Constitution give patent to inventors in the useful arts.  I would say science is a useful art.  No a formula without meaning is not useful.  But an algorithm has meaning.  It does something.  A sorting algorithm on a computer is a useful process(Processes are patentable in the US).

As far as your car engine analogy.  The inventor of the internal combustion engine could have patented it if he had done it in the 1800s.  And rightfully so.  If I invented the perfect way to utilize a hydrogen powered car engine I would want the rights to that invention.  

Open source is still infringement.  And rightfully so again.  PATENT IS A MONOPOLY.  NO ONE ELSE gets to use it.

A algorithmic graphics renderer is still an algorithm.  The question is what level of basic algorithm are you wanting to protect.  Must it be 10 basic algorithms working together.  How bout 15,  50.  Where does it quit being scientific fact and become an innovative idea.  These are serious question not to be answered quickly.  

Most if not all of you understand O(N) notation.  Operation counts right(well I am changing it a bit but anyway).

Let just say the best IS-NOT operator on the market takes 3 cycles to operate.(again not reality but whatever.)  ;)  Now, if I made a IS-NOT operator that did the operation in 2 cycles that would be a hell of a lot faster overall for a computer program.  I could not keep the world from using the 3 cycle IS-NOT nor could I keep them from using a 1 or 2 cycle is not that was different from my own.  What I could keep the world from using is my 2 cycle IS-NOT routine.

That is likely what happened but I don't know for sure.  

Patents restrict and help technological progress.  It gives the world the state of the art.  We can then experiment to move beyond that state of the art but the person who advanced it gets to keep that level of the state of the art for 10-20 years because he got us there.  Does that make sense?
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.

 

Offline Osiri

  • 24
Quote
Originally posted by Kamikaze


That's the scariest thing I've read today. You're telling me a simple mathematical algorithm that can be applied in thousands of different applications should be patented? Mathematical algorithms aren't "invented", they're discovered. There is a optimal solution to many programming problems after all (in terms of efficiency). Physical inventions are complex and don't restrict innovation much because it usually isn't possible to have a perfect machine.

Meh. I suppose Newton should've patented the derivative when he "invented it" so Leibniz couldn't develop it any more...



Uhh, no. Most of the pure software patents are made by the big corps. In fact, most software patents are not even made by software companies. What you want to do is create a situation where virtually no software can be developed because every bit of code is patented.

Software patents just stifle development. You won't be able to make your "innovative software" because all the algorithms you'd want to use are patented. Furthermore, good applications aren't based on the algorithms they use. They're based on the quality of the whole thing. 3DSMax's 3ds file format doesn't necessarily make the software better than, say, Softimage or Lightwave. The GIF format's compression algorithm doesn't make it better than PNG or JPEG.

EDIT: To give some numerical backup for my argument, here's some info from a paper I wrote:



The citation is for "Bessen J. & Hunt M. R. (2004). The Software Patent Experiment."




I was not really saying something as simple as bubble sort should be patentable ever.  It was completely obvious as a process before computers were invented.  

Hash sorts on the other hand are a bit more tricky.  Developing the perfect hash sort is a bit of an art.  There is no mathmatical perfection with one.  

But if mine works better than yours whose to say I should be able to keep it.  

What you are missing is in terms of scope.  Most of these patents are toilet paper because of the scope of the claims.  

They may look vicious but they have no bite.  If they can only claim a bubble sort in the most narrow of circumstances they may as well have the copyright to thier software.  The patent will run out sooner and the copyright will actually protect more.

(WHICH IS REALLY PATHETIC SINCE COPYRIGHT OFFERS THE LEAST PROTECTION OF ALL IN MOST CIRCUMSTANCES)
« Last Edit: October 16, 2005, 07:14:38 pm by 3173 »
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.

 

Offline Osiri

  • 24
As a final edit that I wanted separate from my prior responses...

Everyone must realize ther is a basic factor many of you are missing.  That factor is one of scope of protection.

To simplify let me say it like this.  I can patent a combustion engine.  

Can I keep everyone else from using combustion engines if I get that patent?

No.  

Saying I am going to patent a combustion engine is alot like saying I am going to patent a sort algorithm.  

You know I don't mean all combustion engines.  What I mean is one specific engine or improvement on all engines.  For example, Dodge might want to patent the new improved Hemi but that would not prevent all Hemi engines from being produced.  Just ones using the improvement.  

In software.  I cannot patent the basic idea of a hash sort or a bubble sort.  But if I figured out a way to make a new sort using both hash and bubble sort I could patent it.  

Saying that by allowing software patents we will allow all algorithms to be used up is like saying that by allowing books to be copyrighted all stories will be used up.  Its a bit of a THE SKY IS FALLING type attitude.  

The simple fact is there is no way to use up all ways to do something in software.  NOT EVEN CLOSE.  YOU CAN'T USE UP THE BASIC TOOLS OF COMPUTER SCIENCE.  THEY ARE ALL PRIOR ART.  IF YOU ARE GOING TO MAKE AN INNOVATION IT WILL USE THE BASIC TOOLS NOT SOMEONE ELSES EXACT IMPLEMENTATION OF THE SOFTWARE.  If you do that you are infringing on copyright anyhow and that lasts 76 years after the writer dies.  If you used his actual code or even a very close derivation it is copyright infringment.  But if you modify substantially his code to make some sort of innovative difference you can get a copyright and a patent of your own.  

With just the routines currently available and hence unpatentable everything can be done.  Turing machines you know.  Just because someone is patenting a particular way of doing it is not going to foreclose all ways of doing it.

SO CHILL.

In Intellectual Property law there is a fragile balance.  You must not stiffle progress.  BUT AT THE SAME TIME YOU MUST REWARD THOSE WHO ARE MAKING THE PROGRESS. (IOW these poor saps making this !!!! up gotta eat you know).

Oh and as to your statistics.  Most smaller companies are going to put out valid patent applications.  Those huge companies playing it by the shotgun approach are likely to get them invalidated.

Anyhow, I'm bored this is a lot like taking work home with you.  Actually that is almost exactly like what this is.
« Last Edit: October 16, 2005, 08:17:37 pm by 3173 »
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.

  

Offline aldo_14

  • Gunnery Control
  • 213
Quote

Osiri wrote on 10-17-2005 12:53 AM:
Aldo I will try to publish this on the board except this first line
but man you need to calm down.

Patent practice is not as bad as all you think it is.  Trust me I am doing in depth study on it and I talk with actual patent practitioners alot.

I would truly like to see this IS-NOT patent.

If you invented this brilliant new type of sorting algorithm that was different from anything ever done before, would you want a big company to be able to use it freely whenever they wanted.  Essentially you get nothing for your work.

As far as your car engine analogy.  The inventor of the internal combustion engine could have patented it if he had done it in the 1800s.  And rightfully so.  If I invented the perfect way to utilize a hydrogen powered car engine I would want the rights to that invention.  

Open source is still infringement.  And rightfully so again.  PATENT IS A MONOPOLY.  NO ONE ELSE gets to use it.

A algorithmic graphics renderer is still an algorithm.  The question is what level of basic algorithm are you wanting to protect.  Must it be 10 basic algorithms working together.  How bout 15,  50.  Where does it quit being scientific fact and become an innovative idea.  These are serious question not to be answered quickly.  

Most if not all of you understand O(N) notation.  Operation counts right(well I am changing it a bit but anyway).

Let just say the best IS-NOT operator on the market takes 3 cycles to operate.(again not reality but whatever.)  ;)  Now, if I made a IS-NOT operator that did the operation in 2 cycles that would be a hell of a lot faster overall for a computer program.  I could not keep the world from using the 3 cycle IS-NOT nor could I keep them from using a 1 or 2 cycle is not that was different from my own.  What I could keep the world from using is my 2 cycle IS-NOT routine.

That is likely what happened but I don't know for sure.  

Patents restrict and help technological progress.  It gives the world the state of the art.  We can then experiment to move beyond that state of the art but the person who advanced it gets to keep that level of the state of the art for 10-20 years because he got us there.  Does that make sense?


Would I make the bubble sort, or equivalent, routine free for wide use?

Absolutely, positively, unequivocally yes.

You're viewing it from the point of view of business; defend investment, prevent advancement (i.e. stop companies eroding an advantage), and soforth.  I'm viewing from the point of science, which is what it really is; would you restrict the advancement of the theory of relativity to one person, the single patentholder?

By your standards, 90% of the work that I did for my degree should havce been already under patent.  Bye-bye Computer Science departments, then.

Can't you see that holding the state of the art for 10-20 years represents a complete halt in progress?

You've pointed out the problems in your previous code - if all the current tools of computer software were patented at creation, we wouldn't be able to type this.  We'd probably be sitting at CGA screens that cost £1000 each, if atall.  An algorithm is exactly one such tool; a series of steps, to solve a problem.  If you include that as patentable, you must add the likes of the Observer pattern.

Also, you pointed out large companies use a scattergun approach - that's because the validation has failed so many times in evaluating nonsense patents that it is a no risk operation for them.

This is the ISNOT patent; http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.html&r=1&f=G&l=50&s1=%2220040230959%22.PGNR.&OS=DN/20040230959&RS=DN/20040230959

Abstract;  A system, method and computer-readable medium support the use of a single operator that allows a comparison of two variables to determine if the two variables point to the same location in memory.

That abstract alone should point out this is not just prior art, it's an attempt to patent mathematics.

 

Offline Osiri

  • 24
Quote
Originally posted by aldo_14


Would I make the bubble sort, or equivalent, routine free for wide use?

Absolutely, positively, unequivocally yes.

You're viewing it from the point of view of business; defend investment, prevent advancement (i.e. stop companies eroding an advantage), and soforth.  I'm viewing from the point of science, which is what it really is; would you restrict the advancement of the theory of relativity to one person, the single patentholder?

By your standards, 90% of the work that I did for my degree should havce been already under patent.  Bye-bye Computer Science departments, then.

Can't you see that holding the state of the art for 10-20 years represents a complete halt in progress?

You've pointed out the problems in your previous code - if all the current tools of computer software were patented at creation, we wouldn't be able to type this.  We'd probably be sitting at CGA screens that cost £1000 each, if atall.  An algorithm is exactly one such tool; a series of steps, to solve a problem.  If you include that as patentable, you must add the likes of the Observer pattern.

Also, you pointed out large companies use a scattergun approach - that's because the validation has failed so many times in evaluating nonsense patents that it is a no risk operation for them.

This is the ISNOT patent; http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.html&r=1&f=G&l=50&s1=%2220040230959%22.PGNR.&OS=DN/20040230959&RS=DN/20040230959

Abstract;  A system, method and computer-readable medium support the use of a single operator that allows a comparison of two variables to determine if the two variables point to the same location in memory.

That abstract alone should point out this is not just prior art, it's an attempt to patent mathematics.





Okay, first I am glad you are such a humanitarian.  
:rolleyes:
I am more interested in putting food on the table and giving a home to my family. My wife deserves more than that.

I would like to see your response if you had spent 6 years creating a new method and then someone else took it and made millions from it without giving you a dime.  It just like anything else.  If I invent it, I get the temporary rights.


Further, so I suppose if you invent a new computer you will want to dedicate that to the public as well.  Its the same damn thing.  

Again, I will say, none of the crap that you did for your computer science degree is patentable.  Nor is mine.  That is because the tools(algorithm are tools of our trade) have been in service for a long time.  All of the basic operators are unpatentable because they are obvious anyhow.  For example, if I were examining an EQUALS operator I would be laughng sbecause that is simple logic to begin with.

By innovative I am refering to higher level programs.  For example, a new graphics engine that takes a new approach.  That would be patentable. Maybe the language translator, I did but not the graph object I used behind it.  GET IT.  THE WHOLE THING NOT THE COMPONENT PARTS.

I KEEP SAYING ITS A QUESTION OF SCOPE.


As to the IS-NOT application, IT IS AN APPLICATION.  It has not been allowed.  For all we know this is someone's idea of a joke.  It got published under the statuory 18 month publication.

If it gets allowed at the level of the abstract, it would never be enfoceable.

WHAT IN COMPUTER SCIENCE IS PATENTABLE.

BIG THINGS.  GRAPHICS RENDERERS. MAJOR ROUTINES.  

NOT THE STUFF I HAVE BEEN WORKING WITH FOR YEARS.

Okay and don't you see that the state of the art is not the state of the art about 10 minutes after it is the state of the art.  Doom engines already outdated.(I haven't really looked at the engines so don't try to give me a reality check I don't really care)  So the owner might have a patent on it.  In two years no one will care because it will be old news.

Computer Science is a science and an engineering discipline.

Mathmatics is the PURE SCIENCE.
« Last Edit: October 17, 2005, 08:30:32 am by 3173 »
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.

 

Offline aldo_14

  • Gunnery Control
  • 213
Quote
Originally posted by Osiri

Okay, first I am glad you are such a humanitarian.  
:rolleyes:
I am more interested in putting food on the table and giving a home to my family. My wife deserves more than that.

I would like to see your response if you had spent 6 years creating a new method and then someone else took it and made millions from it without giving you a dime.  It just like anything else.  If I invent it, I get the temporary rights.
 


The same as what I just posted.  It's called a 'principle'.  

If I was doing research, it would not be for the money.  Or won't be for the money, dependent on where the next year takes me I suppose (so it might not be an 'if' anyways).

Oh, and the Doom engine isn't really a good relation to base-level algorithms or technology; it's an application.  If it was possible to, say, trademark the concept of 'rending an abstract world description to a 2d representation', then we wouldn't have seen any change in state of the art (incidentally, that's - accidentally - very similar to the patent Oracle was denied for SGML to an.other markup).

You're comparing apples to oranges here.  An algorithm is not an application any more than the principles of combustion are an engine.

(I'd note that a algorithm patent based on concept is too wide, and one based on specific implementation is too narrow, to be of value with such a low level concept)

It's interesting for me to note that reuse is, and has been for many years, the holy grail for the software industry.  Call it hierarchical subdivision into libraries, component code, extensive frameworks, whatever; you seem to be against that concept as advancing it would require free use.  Like with design patterns.  Someone patents the Observer pattern..... well, they'd be rich and the rest be ****ed.

Word processing is an application.  A Mobile Agent system is an application (well, a framework for one).  A game is an application.  An expert system is an application.

Bubble sort, Dijkstras algorithm, recursive traversal, alpha-beta pruning.... all these things are not applications.  They are algorithms.  Steps to solve a problem case.  We don't patent the scientific, mathetical theories when we develop, say, a new engine, why should we do the same for software?  The implication that computer science is not science is to me, complete and utter bollocks.  It is, and you shouldn't be allowed to control the very basics any more than you should be allowed to patent the concept of using ones and zeroes to store and convey information.

(All caps is considered bad form, BTW.  Like shouting.  Plus it kind of gets on my tits in an empty vessels make most noise type manner and leads me to making addendums like this.  which surely annoys everyone.  Small caps with italic/bold emphasis is much easier on the eye IMO)

 

Offline Osiri

  • 24
Aldo what we have here is a failure to communicate.  Or at least a failure for you to listen.  Please read the entire message carefully. I am doing the same for you.

First let me congratulate you on starving yourself.  However, I am sure that you will get paid to do your research so you will not.  Someone has to pay for that research.  They do that by making money on previously completed research.  Else you would starve for the betterment of man.  A manufacturer is not going to pay for research that can it will not get to keep.  

You may be a saint but corporations are chartered to make profit.(I swear this is a requirement in thier charter.)

Realistically you may not do it for the money.  However, even university professors get a cut (50% at my university).  They are doing research for the money. They need it to do more research.

Algorithms like dijkstras, recursive traversal and the like are unpatentable.  There is no doubting that.  

A new algorithm is not so unpatentable.  I don't have an example because I am not researching new algorithms.  Let's say a new improved version of recursive traversal was concieved. (this is not going to happen)  It has never been thought up in logic, theory, or practice.  Say it works better than the version it is based on.
It MIGHT be patentable.

Many of the other things you cite are not patentable because they are simply implementations of things that have been in use for many many years(just packaged or whatever).  To patent a software algorithm takes more.

You need alot more than you have stuck in your head.

A lot more.

You must have a new idea.

A patentable algorithm is not like the principle of combustion.  An algorithm that basic is never to be patentable.  Neither would the idea work for patent in the engine.

However, a new engine design(utility patent) would be highly patentable if it was distinguishable over the prior art.

A basic algorithm newly concieved might be patented.  

I cannot patent recursive traversal.  However, I can restrict the use of my improvement.  That is how the law works.  I am no more making this up than standing on my head.

Look, you are not going to lose your ability to write code.  That cannot happen under the patent laws.  

The pressure for advancement is there.

If someone else has the exclusive use of a algorithm(again not something previously known), and you do not, what are you going to do.  

You are going to either use the old way to do it or find a better way than the other guys way.

That is life.

It is not like you could never improve on someone elses patented idea either.  

Patents are finite for a reason.  

I don't want to say it like this but open your eyes.  You may be as I said a saint but the rest of the world is not.

In a perfect world you would be right.  All advancement would be for the betterment of man.  All of it.  No one would need money because there would be no need for it because we would all work without reward just for the sheer joy of working.  

In the real world corporations control 90% or so of the worlds money.  They must follow thier charters that say they are formed for the business of making a profit for thier investors.  

A corporation cannot fund research that will be unprofitable to them.  If they will not gain a benefit from the research they are not allowed to use it.

As for the Doom Engine example.  I said it was not for reality and I didn't really care.  It was an example.  I figured you have crowned yourself a patent expert you would know what I was really refering to.  

What I meant was... the basic principles of something as simple as bubble sort is unpatentable.  The use of 50 different algorithms in a new and novel combination is going to be patentable.

Say my renderer with a nonobvious code variation does something yours cannot do.  Nonobvious mean others have not thought about it and others would not see the improvement without research as well.  

I can keep my renderer.  I cannot prevent you from improving on it in a nonobvious way.  Hence the improvement of computer science.


As far as your argument that it will hamper computer science.  No.  You are saying it will hamper the development of applications of the idea.  This is true to a small extent.

By your example what you want is when someone invents a hydrogen engine he will not be able to patent it because that would prevent other car manufacturers from using that engine to make thier cars better.  

That is utter and complete Bull ****.  No one would respect that argument.  

No patent will keep you from writing any code you want.  A patent may prevent you from using the particular method that someone else thought up first.

Again, that is life. Get over it. That is the law and it will most likely stay the law.

You can petition your congressman to change it.  Only the legislature can change the law.  Vote for liberal legislatures.

Aldo what you seem to be supporting is open source.   I admire that.  I too like open source.  However, open source does not have the right to keep the rest of the world from keeping our source code to us.  It is our intellectual property.  Whether it is by patent or copy right or trade secret, it is owned by the maker, inventor, conceiever.  

One final thing for you to consider is that without patents, many advances will be trade secrets.  Since you cannot see what method I am using to make the program faster(and in a big application it would take years to figure it out), I can hide my advancement for as long as I want to.  Reverse engineering it would be difficult.  Trade secrets are protectable for as long as you keep them secret and no one else implements the same thing.

Further even if someone else finds the same invention as my trade secret, they too will keep it a trade secret.  In the end there will be a sharp divide among software developers.  There will no longer be a free exchange of these advancements.  There is already not a free use, would it be better to get them to hide it to so that no one can continue working where the patent owner left off.  

Do you understand this last part?

EDIT: I am sorry this was a bit too far.  I will try to get time to edit to be a little less sarcastic when I am out of class.  Again sorry, I was not trying to come off as rude as it seems.
« Last Edit: October 17, 2005, 11:23:44 am by 3173 »
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.

 

Offline aldo_14

  • Gunnery Control
  • 213
[q]As far as your argument that it will hamper computer science. No. You are saying it will hamper the development of applications of the idea.

By your example what you want is when someone invents a hydrogen engine he will not be able to patent it because that would prevent other car manufacturers from using that engine to make thier cars better.

That is utter and complete Bull ****. No one would respect that argument. [/q]

You miss the point.  If you are allowed to patent such basic building blocks, you bring an inherent limit upon the potential applications that can be developed by anyone in future.

I notice you're trying to read me in as being arrogant ('crowned your self a patent expert') whilst switching between strawman and strawman (' I said it was not for reality and I didn't really care') example, without addressing the philosphical point I was making.  That software is a part of scientific endevour, the same as medicinal, mathematical, physical, etc reasearch.  And algorithm is effectively a formula; if you restrict such a base level idea, then you compromise the ability to develop it further.

 At uni, I did work on bubble, quick, etc sorting algorithms as a 2nd year class.  If that level of algorithm was patentable, and hence restrictable, that may not have been legal for me.  And hence even the academic system is compromised.

These algorithms are as basic and - for lack of a better term- sacrosanct to commerical restriction as the theory of combustion is to an engine.  That's from a computer science perspective.

As for misreading my example; my point was very simple, and you completely misread it.  I never suggested that specific implementations should be unpatentable; but that the theoretical basis,the scientific theories, should not be restricted.

I can guess from your signature you have a vested interest in making money off ideas.  As a scientist first and professional second, my vested interest is in releasing and building upon them.

[q]Again, that is life. Get over it. That is the law and it will most likely stay the law.[/q]

Actually, in Europe that's the opposite of patent law in this context, which classes software in the same vein as medical research, and mathematical and scientific theory; i.e. cannot be patented.

 

Offline Osiri

  • 24
Not the basic building blocks
Got any patentable ideas?  Got $20K laying around.  I will need every penny to help you.