If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(Ars Technica)   US patent chief to software patent critics: stop hitting me   (arstechnica.com) divider line 167
    More: Unlikely, Trademark Office, midgets, business method patents, u.s. patent, AIA  
•       •       •

4400 clicks; posted to Geek » on 21 Nov 2012 at 11:03 AM (1 year ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



167 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

Archived thread

First | « | 1 | 2 | 3 | 4 | » | Last | Show all
 
2012-11-21 01:21:51 PM

roc6783: Theaetetus: tgambitg: Theaetetus: A software lock isn't the same as a physical lock. Ignore the legalese for a second - does your physical slide lock have a touch screen? Does it detect a touch at a first location, and does it move an image in response to motion of the touch? No. The result may look the same and the general concept of "unlocking something" may be the same, but the way of getting there is different. Patents are about implementations, not general concepts, and here, the implementation is different.

There is your example of something that exists, then was patented as "on a computer". Take away the "on a computer part" and it is fundamentally the same as the physical slide lock. As you said, the implementation is "on a computer", which is the key to the patent eligibility.

//I would like my month of TF medium-rare, please.


Sorry, no. See the strike-through version above. In order to get to the "fundamentally the same" part, you have to strip out 95% of the claim. Not just "on a computer".

So, I'll expect that TF signup in the next few minutes, eh?
 
2012-11-21 01:24:45 PM

Theaetetus: A software lock isn't the same as a physical lock. Ignore the legalese for a second - does your physical slide lock have a touch screen? Does it detect a touch at a first location, and does it move an image in response to motion of the touch? No. The result may look the same and the general concept of "unlocking something" may be the same, but the way of getting there is different. Patents are about implementations, not general concepts, and here, the implementation is different.


imo the folks ragging on you have a point. all that text above completely misses the point which is the exploding software patent problem. issuing new patents for everything that now happens in software/digital that once was physical leads to the slide lock patent nonsense and patent trolls.
 
2012-11-21 01:26:10 PM

Theaetetus: Patents are about implementations, not general concepts, and here, the implementation is different.


As I understand it, one of Apple's major victories over Samsung includes the 'slide to unlock' patent. Apple takes a solid bar with a slide motion across the screen from left to right to unlock, Samsung (Android) has a round circle where you drag from the center to a point on the outer ring. Same general concept, different implementation, and the court determined it was infringing.

Part of the problem people are having with these patents, and I think the reason so many of them are actually being invalidated on court review is that so many of them these days are so broad they encompass a general concept, and don't distinguish between specific implementations.
 
2012-11-21 01:26:23 PM
FTFA: "In a system like ours in which innovation is happening faster than people can keep up, it cannot be said that the patent system is broken," he said.

That's the very definition of broken.
 
2012-11-21 01:28:48 PM

Kinek: Theaetetus: Kinek: Theaetetus: Kinek: Yes, airplanes were invented, patented, and didn't do fark-shiat in the US for at least a decade because the Wright brothers proceeded to get all litigious on everybody. That's a great example of how the patent system encourages innovation.

Actually, it is... Curtiss and others did a whole bunch of innovating on their own (because they already had access to the roll control methods in the Wrights' patents), resulting in a bunch of opposing lawsuits that eventually got turned into a patent pool, the Manufacturer's Aircraft Association. As a result, the industry really took off from the simple powered and unpowered gliders to the amazing flying machines of WWII, in just a decade.

So you're going to ignore the fact that it took near the lifetime of the patent, WWI, and the US government taking legal action for the Wright brothers to shut the fark up to get the industry to take off? So, the 12 year period where fark all happened is a positive?

You seem to be ignoring the fact that those 12 years were full of tons of new innovations by Curtiss and others. Do you really think the entire patent pool was one patent by the Wright brothers?

Except for the obvious fact that there were no goddamned planes working. The wright brothers, with a few patents, managed to stall the entire industry. Go patents.


Except for the obvious fact that once the pool was established, the industry leaped ahead suddenly due to the wide number of innovations that had been made in the previous 12 years. The patents actually encouraged that innovation, because otherwise, the other manufacturers wouldn't have had anything to add to the pool and they would have lost out.
How do you think we went from the 30 mph Wright Flyer with a maximum ceiling of 30ft to, for example, the Avro 504, with a top speed of 90 mph and a ceiling of 16,000 ft in just 10 years?
 
2012-11-21 01:29:00 PM

Curious: Theaetetus: A software lock isn't the same as a physical lock. Ignore the legalese for a second - does your physical slide lock have a touch screen? Does it detect a touch at a first location, and does it move an image in response to motion of the touch? No. The result may look the same and the general concept of "unlocking something" may be the same, but the way of getting there is different. Patents are about implementations, not general concepts, and here, the implementation is different.

imo the folks ragging on you have a point. all that text above completely misses the point which is the exploding software patent problem. issuing new patents for everything that now happens in software/digital that once was physical leads to the slide lock patent nonsense and patent trolls.


Don't forget Amazon's "One Click."
 
2012-11-21 01:29:39 PM

Kinek: I've noticed this before, that patents, even in their claims, are very....vague. Yes, they're specific in some point, but actual implementation is hardly ever mentioned. Which is understandable when you think about the motivations behind the patents now, they want them to be as broad as possible while still being patentable. That way they can cover as much territory as possible.


This. That's what I was trying to say. When you seek to earn a patent, you don't just cover what your product does, you are trying to patent the CONCEPT of your invention that makes it unique, hoping it'll both better protect your investment, and also allow for it's use in areas that aren't necessarily obvious today to you.

If you read my two patents I've earned over my career with various companies, you'd unlikely be able to recognize the actual functions they're protecting in the product due to how generic the lawyers made the wording. One of them was actually brought forth in litigation against a competitor for a completely different product that didn't actually use the circuit I had originally patented, but the implementation it used was apparently similar enough to the CONCEPT of the patent to make it valid.

/not necessarily saying you shouldn't be able to patent concepts...just saying you definitely can (and everyone certainly tries to)
 
2012-11-21 01:32:24 PM

tgambitg: He noted that during a time of growing litigation in the smartphone industry, "innovation continues at an absolutely breakneck pace. In a system like ours in which innovation is happening faster than people can keep up, it cannot be said that the patent system is broken," he said.

That right there is why it's broken you jackass....


You're so correct that I inadvertently repeated your sentiment, so I "Smart"-ed your post in a display of solidarity. Sorry about the repeat. I'd "Dumb" my own post, if that were possible.
 
2012-11-21 01:33:34 PM

UNAUTHORIZED FINGER: tgambitg: He noted that during a time of growing litigation in the smartphone industry, "innovation continues at an absolutely breakneck pace. In a system like ours in which innovation is happening faster than people can keep up, it cannot be said that the patent system is broken," he said.

That right there is why it's broken you jackass....

You're so correct that I inadvertently repeated your sentiment, so I "Smart"-ed your post in a display of solidarity. Sorry about the repeat. I'd "Dumb" my own post, if that were possible.


No worries. Thank you for the smart, in solidarity, I will do the same with yours.
 
2012-11-21 01:34:15 PM
Give the AIA a chance to work. Give it a chance to even get started.

I think that's fair. Seems like most people in the thread are ignoring this.
 
2012-11-21 01:34:55 PM

Marine1: Then fix it, dickweed.


You didn't read the article did you? They are trying to change it.
 
2012-11-21 01:38:39 PM

Curious: Theaetetus: A software lock isn't the same as a physical lock. Ignore the legalese for a second - does your physical slide lock have a touch screen? Does it detect a touch at a first location, and does it move an image in response to motion of the touch? No. The result may look the same and the general concept of "unlocking something" may be the same, but the way of getting there is different. Patents are about implementations, not general concepts, and here, the implementation is different.

imo the folks ragging on you have a point. all that text above completely misses the point which is the exploding software patent problem.


So, rather than actually focusing on the differences between the patent and the alleged prior art, we should instead... focus on the fact that lots of patents get issued, so therefore we should abolish all patents?
Lots of people are in jail - rather than having trials, let's just deport everyone who gets arrested?
 
2012-11-21 01:40:05 PM

Driedsponge: Theaetetus: Patents are about implementations, not general concepts, and here, the implementation is different.

As I understand it, one of Apple's major victories over Samsung includes the 'slide to unlock' patent. Apple takes a solid bar with a slide motion across the screen from left to right to unlock, Samsung (Android) has a round circle where you drag from the center to a point on the outer ring. Same general concept, different implementation, and the court determined it was infringing.

Part of the problem people are having with these patents, and I think the reason so many of them are actually being invalidated on court review is that so many of them these days are so broad they encompass a general concept, and don't distinguish between specific implementations.


Well lets take a look at the claims shall we?

1. A method of unlocking a hand-held electronic device (Yes), the device including a touch-sensitive display (Yes, the method comprising:
-detecting a contact with the touch-sensitive display at a first predefined location corresponding to an unlock image (Yes, center for android);
-continuously moving the unlock image (The circle) on the touch-sensitive display in accordance with movement of the contact (Your finger) while continuous contact with the touch screen is maintained (Finger never leaves), wherein the unlock image is a graphical, interactive user-interface object with which a user interacts in order to unlock the device (The circle); and
unlocking the hand-held electronic device if the moving the unlock image on the touch-sensitive display results in movement of the unlock image from the first predefined location (The center) to a predefined unlock region on the touch-sensitive display (Corner point).

From what you said looks like it infringes to me. There are plenty of ways to get around this, make the unlock image move discretely, for example, or tapping your finger in discrete locations to unlock (like the num key unlock codes) would also get around this.

But you know, that would require innovation rather than just copying.
 
2012-11-21 01:40:23 PM

Kinek: It's almost like disclosure isn't even the point anymore. You can't build a working program from most of these claims. It's Almost as if patents are being used as some sort of anti-competitive tool. But nobody would ever do that.


This is the point I think Theaetetus is missing about patents. You should be able to replicate exactly what is there from the patents. Without computer code showing the implementation, the patents are too broad, and once you put in the code, you are showing the mathematical means of getting there, and math can't be patented. Software is also an abstract idea, and under patent law cannot be patented. It took a court case to make them patentable, and I truly believe that is erroneous...
 
2012-11-21 01:44:31 PM

Driedsponge: Theaetetus: Patents are about implementations, not general concepts, and here, the implementation is different.

As I understand it, one of Apple's major victories over Samsung includes the 'slide to unlock' patent. Apple takes a solid bar with a slide motion across the screen from left to right to unlock, Samsung (Android) has a round circle where you drag from the center to a point on the outer ring. Same general concept, different implementation, and the court determined it was infringing.


Except that you're comparing Apple's iOS to Samsung's Android implementation... not Apple's patent to Samsung's implementation. The patent doesn't require a solid bar or a motion from left to right.

Incidentally, the same mistaken comparison was the issue in the UK Apple-Samsung design patent fight - Apple was comparing the Galaxy Tab to the iPad, but not the Galaxy Tab to the patent, which wasn't quite the same.

Part of the problem people are having with these patents, and I think the reason so many of them are actually being invalidated on court review is that so many of them these days are so broad they encompass a general concept, and don't distinguish between specific implementations.

Actually, over 80% of patents are upheld by courts on review, and of those decisions that are appealed, 95% are upheld by the Federal Circuit. Kappos talks about it in the speech this article is about.
 
2012-11-21 01:46:23 PM
tgambitg
You really are clueless to what everyone is trying to say, aren't you? All that stuff I took out was the 'on the electronic device' stuff. And all of that is software based. Copyright the code that does it, but it should never have been eligible for a patent. It literally copies a physical slide lock in digital form. You're getting hung up on the legalese that tells you 'oh this isn't a physical lock, see all this stuff? it means a software lock so it can't be the same as the physical lock.'


Thinking about it..
software development or even big parts of computer science and math are about finding abstract representations and solutions for specific problems.
That's where heavy lifting is done in those fields.
Just think of graph theory.
A new, more efficient algorithm for finding some sort of special path (think Euler tour or traveling salesman) could be an invention.
Whether you apply this algorithm to facebook's friendship graph or in a navigation system instead of the bridges of Königsberg means fark all for innovation in that field.

Looking at the slide-to-unlock example, it's no wonder this mindset collides with one that considers the "limitations" as what makes the idea patent-worthy.
This works for physical innovations where you have to do actual work to specialize the solution instead of simplifying or generalizing the concept behind it.
If someone makes a machine that sorts coins by size and then someone comes along and builds one that sorts them by their actual value - that sounds like an innovation.
If you come to a CS person and say "I've written a program that can sort letters instead of numbers" - that's just cute compared to one sorting generic objects by every property you like by passing a comparator function as parameter.
 
2012-11-21 01:46:56 PM

Theaetetus: roc6783: Theaetetus: tgambitg: Theaetetus:

***snip***
Sorry, no. See the strike-through version above. In order to get to the "fundamentally the same" part, you have to strip out 95% of the claim. Not just "on a computer".

So, I'll expect that TF signup in the next few minutes, eh?


Your challenge: No, there aren't. Bet you can't find a single one. I'll bet a month of TF that you can't find an independent claim in a patent that says "1. A method for doing [known process], comprising: [doing known process] ON COMPUTERS!"

Theaetetus:
tgambitg:

Yeah, it doesn't work that way. Here's the claim, and with what you just removed crossed out:
1. A method of unlocking a hand-held electronic device, the device including a touch-sensitive display, the method comprising:
detecting a contact with the touch-sensitive display at a first predefined location corresponding to an unlock image;
continuously moving the unlock image on the touch-sensitive display in accordance with movement of the contact while continuous contact with the touch screen is maintained, wherein the unlock image is a graphical, interactive user-interface object with which a user interacts in order to unlock the device; and
unlocking the hand-held electronic device if the moving the unlock image on the touch-sensitive display results in movement of the unlock image from the first predefined location to a predefined unlock region on the touch-sensitive display.

Or:
1. A method of unlocking a device, the method comprising:
a contact at a first predefined location;
continuously moving the contact; and
unlocking the device.


Here's the full text and the leftover text of your strike-through version.

Strike-through version = Physical latch, to which you have already agreed
Rest of text = on a computer hand-held device, which you have said makes this implementation patent eligible because it is a novel and nonobvious.


I guess you are correct, it says hand-held device rather than computer.
 
2012-11-21 01:49:12 PM

tgambitg: Kinek: It's almost like disclosure isn't even the point anymore. You can't build a working program from most of these claims. It's Almost as if patents are being used as some sort of anti-competitive tool. But nobody would ever do that.

This is the point I think Theaetetus is missing about patents. You should be able to replicate exactly what is there from the patents. Without computer code showing the implementation, the patents are too broad


Patents have to have sufficient disclosure to enable one of ordinary skill in the art to make the invention... Patents have flow charts. I can write a program from a flow chart. I don't need to copy-paste someone's code. Why should specific code be required?

and once you put in the code, you are showing the mathematical means of getting there, and math can't be patented.

Not so - there are plenty of patents that do contain code (even though it's not required). They're not patenting math.

Software is also an abstract idea, and under patent law cannot be patented. It took a court case to make them patentable, and I truly believe that is erroneous...

Except that software isn't necessarily an "abstract idea".
 
2012-11-21 01:56:06 PM

roc6783: Your challenge: No, there aren't. Bet you can't find a single one. I'll bet a month of TF that you can't find an independent claim in a patent that says "1. A method for doing [known process], comprising: [doing known process] ON COMPUTERS!"


Yep.

A method of unlocking a hand-held electronic device, the device including a touch-sensitive display, the method comprising:
detecting a contact with the touch-sensitive display at a first predefined location corresponding to an unlock image;
continuously moving the unlock image on the touch-sensitive display in accordance with movement of the contact while continuous contact with the touch screen is maintained, wherein the unlock image is a graphical, interactive user-interface object with which a user interacts in order to unlock the device; and
unlocking the hand-held electronic device if the moving the unlock image on the touch-sensitive display results in movement of the unlock image from the first predefined location to a predefined unlock region on the touch-sensitive display.

Or:
1. A method of unlocking a device, the method comprising:
a contact at a first predefined location;
continuously moving the contact; and
unlocking the device.

Here's the full text and the leftover text of your strike-through version.

Strike-through version = Physical latch, to which you have already agreed
Rest of text = on a computer hand-held device, which you have said makes this implementation patent eligible because it is a novel and nonobvious. "a hand-held electronic device, the device including a touch-sensitive display... detecting [a contact] with the touch-sensitive display... corresponding to an unlock image... [moving] the unlock image on the touch-sensitive display in accordance with movement of [the contact] while continuous contact with the touch screen is maintained, wherein the unlock image is a graphical, interactive user-interface object with which a user interacts in order to unlock the device; and
[unlocking] the hand-held electronic device if the moving the unlock image on the touch-sensitive display results in movement of the unlock image from the first predefined location to a predefined unlock region on the touch-sensitive display."


No, there's the leftover text. Notice how there's a lot of words there? A lot more than just "on a hand-held device"? In fact, I count several dozen additional words, and they're all different than just "on a hand-held device". They talk about determinations and responsive actions, additional actions to be taken during the determination, etc.

Now, here's what you're claiming you found:
"1. A method of unlocking a hand-held device, the method comprising:
a contact at a first predefined location;
continuously moving the contact; and
unlocking the hand-held device."

Not the same at all. And you can't find a patent with that claim, as I said.

So, I'll expect that TF signup in the next few minutes, eh?
 
2012-11-21 01:56:58 PM

Theaetetus: focus on the fact that lots of patents get issued, so therefore we should abolish all patents?


no but it's obvious to me now that you are a troll. or just like straw man arguments. in either case

bye bye
 
2012-11-21 01:58:54 PM

tgambitg: Kinek: It's almost like disclosure isn't even the point anymore. You can't build a working program from most of these claims. It's Almost as if patents are being used as some sort of anti-competitive tool. But nobody would ever do that.

This is the point I think Theaetetus is missing about patents. You should be able to replicate exactly what is there from the patents. Without computer code showing the implementation, the patents are too broad, and once you put in the code, you are showing the mathematical means of getting there, and math can't be patented. Software is also an abstract idea, and under patent law cannot be patented. It took a court case to make them patentable, and I truly believe that is erroneous...


Well, it's one of the big reasons that I think his theory of Patents as disclosure keeping us from the apocalyptic doom of Trade Secrets is full of crap. Because the patents today AREN'T DISCLOSURE. At least they're not being used that way. I'd honestly rather take trade secrets over this kind of crap. At least with Trade secrets you can't just sue your competitor. Really it's the question, is this promoting the progress? Let me write a formula for when patents should succeed.

(Aggregate cost of duplicative research)*(Time until Trade secrets leak)>(Man hours)*(Cost of man hours)+(Cost of patent litigation and/or competitive stifling)-(FRAND Licensing fees)

Under this model, it is cheaper to deal with the monopoly than to research the same thing over and over. Eventually even trade secrets leak. Maybe due to some shadowrunners.
That's my ground. If we abolish the Patent system and revert to trade secrets, we'll get Shadowrun.
 
2012-11-21 02:01:31 PM

Theaetetus: tgambitg: Kinek: It's almost like disclosure isn't even the point anymore. You can't build a working program from most of these claims. It's Almost as if patents are being used as some sort of anti-competitive tool. But nobody would ever do that.

This is the point I think Theaetetus is missing about patents. You should be able to replicate exactly what is there from the patents. Without computer code showing the implementation, the patents are too broad

Patents have to have sufficient disclosure to enable one of ordinary skill in the art to make the invention... Patents have flow charts. I can write a program from a flow chart. I don't need to copy-paste someone's code. Why should specific code be required?

and once you put in the code, you are showing the mathematical means of getting there, and math can't be patented.

Not so - there are plenty of patents that do contain code (even though it's not required). They're not patenting math.

Software is also an abstract idea, and under patent law cannot be patented. It took a court case to make them patentable, and I truly believe that is erroneous...

Except that software isn't necessarily an "abstract idea".


Alright. Patent from last thursday. The Time Machine patent. I didn't see a flow chart on that. Or anything beyond some nebulous claims. Look at that patent, and build it for me.
 
2012-11-21 02:01:46 PM

Curious: Theaetetus: focus on the fact that lots of patents get issued, so therefore we should abolish all patents?

no but it's obvious to me now that you are a troll. or just like straw man arguments. in either case


If you want to talk about the "exploding patent" problem, I'm happy to do so. But to claim that I'm trolling, or making strawman arguments, or missing the point by not talking about it, when we're talking about something different - namely validity of Apple's slide-to-unlock patent over a physical sliding latch - is just silly. I mean, I noticed you haven't addressed the problem of water shortages in Africa - so, therefore, you're a troll? No, we just aren't talking about that at the moment.

bye bye

4.bp.blogspot.com
Happy Thanksgiving.
 
2012-11-21 02:06:55 PM

Theaetetus: Not the same at all. And you can't find a patent with that claim, as I said.


Now you're just arguing semantics.
 
2012-11-21 02:07:40 PM
Theaetetus, does arguing on Fark count towards your billable hours for your firm or something? I would've long since given up, if I ever had the inclination to argue about patents on the internet in the first place.
 
2012-11-21 02:08:08 PM

Kinek: Theaetetus: tgambitg: Kinek: It's almost like disclosure isn't even the point anymore. You can't build a working program from most of these claims. It's Almost as if patents are being used as some sort of anti-competitive tool. But nobody would ever do that.

This is the point I think Theaetetus is missing about patents. You should be able to replicate exactly what is there from the patents. Without computer code showing the implementation, the patents are too broad

Patents have to have sufficient disclosure to enable one of ordinary skill in the art to make the invention... Patents have flow charts. I can write a program from a flow chart. I don't need to copy-paste someone's code. Why should specific code be required?

and once you put in the code, you are showing the mathematical means of getting there, and math can't be patented.

Not so - there are plenty of patents that do contain code (even though it's not required). They're not patenting math.

Software is also an abstract idea, and under patent law cannot be patented. It took a court case to make them patentable, and I truly believe that is erroneous...

Except that software isn't necessarily an "abstract idea".

Alright. Patent from last thursday. The Time Machine patent. I didn't see a flow chart on that. Or anything beyond some nebulous claims. Look at that patent, and build it for me.


Look again? Fig. 5 looks like a pretty nice looking flow chart to me.
 
2012-11-21 02:11:44 PM

Theaetetus: roc6783: ***snip***

"a hand-held electronic device, the device including a touch-sensitive display... detecting [a contact] with the touch-sensitive display... corresponding to an unlock image... [moving] the unlock image on the touch-sensitive display in accordance with movement of [the contact] while continuous contact with the touch screen is maintained, wherein the unlock image is a graphical, interactive user-interface object with which a user interacts in order to unlock the device; and
[unlocking] the hand-held electronic device if the moving the unlock image on the touch-sensitive display results in movement of the unlock image from the first predefined location to a predefined unlock region on the touch-sensitive display."



The fact that Apple's version of the phone lock imitates a physical slide lock is simply a matter of digitizing an existing concept, or in other words "1. A method for doing [known process], comprising: [doing known process] ON COMPUTERS a hand-held device!"

You disagree because you really believe that the leftover text isn't a description of how to use the digital version of the physical latch on the hand-held device. It's fine, and you clearly have more weight behind your opinion due to the fact that the patent was granted, but I still disagree with you.
 
2012-11-21 02:11:54 PM

UNAUTHORIZED FINGER: FTFA: "In a system like ours in which innovation is happening faster than people can keep up, it cannot be said that the patent system is broken," he said.

That's the very definition of broken.


If not 'broken', then at least out-dated or insufficient. "This suit of armor can't stop a rifle round, but it's perfectly effective against arrows," let's outfit the Marines with them.

roc6783: I guess you are correct, it says hand-held device rather than computer.


That latch stuck to a piece of wood IS a "hand-held device," given sufficiently small wood.
 
2012-11-21 02:16:20 PM

ProfessorOhki: UNAUTHORIZED FINGER: ***snip***

roc6783: I guess you are correct, it says hand-held device rather than computer.

That latch stuck to a piece of wood IS a "hand-held device," given sufficiently small wood.


Why do you always have to go there? And tell your mom to stop calling me.
 
2012-11-21 02:18:26 PM

Kinek: Alright. Patent from last thursday. The Time Machine patent. I didn't see a flow chart on that. Or anything beyond some nebulous claims.


Maybe you should look again, then.

Look at that patent, and build it for me.

I think a skilled programmer could easily build it from that specification and those figures.

But that aside:

Kinek: Well, it's one of the big reasons that I think his theory of Patents as disclosure keeping us from the apocalyptic doom of Trade Secrets is full of crap. Because the patents today AREN'T DISCLOSURE. At least they're not being used that way.


You seem to be under the misconception that patents are the only disclosures. They aren't. The patent actually protects all disclosures of the invention. If you have a trade secret, you can't publish a white paper, a thesis, a review, a journal article, a brochure, a functional specification, a pseudocode library, your source code, your schematics, etc., without destroying the secret. If you have a patent, you can do any of those. So, yeah, patents aren't being used as the sole, exclusive disclosure of something, because they don't have to be.

I'd honestly rather take trade secrets over this kind of crap. At least with Trade secrets you can't just sue your competitor.

Yes, you can, and they're outrageously expensive lawsuits.
 
2012-11-21 02:20:37 PM

Theaetetus: No, there aren't. Bet you can't find a single one. I'll bet a month of TF that you can't find an independent claim in a patent that says "1. A method for doing [known process], comprising: [doing known process] ON COMPUTERS!"


Are abstracts good enough, or are you insisting on claims?

A record button that facilitates audiovisual input into a computer system without requiring manual interaction (direct manipulation interaction) with software. The record button may be grouped with transport controls, a standalone button, or grouped with other controls.
 
2012-11-21 02:21:29 PM

Curious: Theaetetus: A software lock isn't the same as a physical lock. Ignore the legalese for a second - does your physical slide lock have a touch screen? Does it detect a touch at a first location, and does it move an image in response to motion of the touch? No. The result may look the same and the general concept of "unlocking something" may be the same, but the way of getting there is different. Patents are about implementations, not general concepts, and here, the implementation is different.

imo the folks ragging on you have a point. all that text above completely misses the point which is the exploding software patent problem. issuing new patents for everything that now happens in software/digital that once was physical leads to the slide lock patent nonsense and patent trolls.


I'm a 26 year veteran of software engineering/programming and I approve this message.

Back in the DOS, early Windows days I implemented, *invented* I guess, a slide-to-unlock (same as slide-to-lock, IMO) that was used in some industrial machines I programmed. One version was on a DOS system with a display and mouse - you used the mouse or scratch-pad to move the cursor thru an envelope (no, not the paper kind). It did NOT have a graphic icon other than the cursor that dragged along with it, but that was a conscious omission since we didn't want the machine operators to even know there was such a function. A variation was implemented in a headless, mouse-less system where a sequence across a matrix of machine control buttons did the unlocking.

You can argue all day that what I did 15-20 years ago is not at all the same and I could spend the next day refuting your claims by use of words, diagrams and flowcharts showing how all the motion sensing elements and all the visual indicators are basically the same, and that all the code to implement any part of it is basically the same, it's only ON A HANDHELD that is any different.
 
2012-11-21 02:26:21 PM

Theaetetus: Kinek: Alright. Patent from last thursday. The Time Machine patent. I didn't see a flow chart on that. Or anything beyond some nebulous claims.

Maybe you should look again, then.

Look at that patent, and build it for me.

I think a skilled programmer could easily build it from that specification and those figures.


Yes, we could. You know why? Because it's a fairly common pattern that existed all over the place before Time Machine.
1) Perform operation operation
2) Validate operation
3) If data became invalid during operation, correct

The moment the problem is framed as "back up files," that entire flowchart is obvious. That's the thing about patents like this: Yes, I could design a similar system from that patent. Would it be the same system with the same performance, same protocols, same data storage, same implementation? No, it would be something completely different. The only thing it would have in common is a process flow that's common in solid state memory, log processing, enterprise backup, etc. etc.
 
2012-11-21 02:27:42 PM

ProfessorOhki: 1) Perform operation operation


Apparently, I really like operations.
 
2012-11-21 02:38:49 PM

ProfessorOhki: Theaetetus: No, there aren't. Bet you can't find a single one. I'll bet a month of TF that you can't find an independent claim in a patent that says "1. A method for doing [known process], comprising: [doing known process] ON COMPUTERS!"

Are abstracts good enough, or are you insisting on claims?

A record button that facilitates audiovisual input into a computer system without requiring manual interaction (direct manipulation interaction) with software. The record button may be grouped with transport controls, a standalone button, or grouped with other controls.


As abstracts are about as useless as titles in defining the scope of a patent no, it doesn't.

Abstracts are vague summaries that do nothing to describe the meat of the invention. An abstract could say "this application relates to a computer" and it won't mean they are patenting all computers.
 
2012-11-21 02:40:46 PM

Theaetetus: torusXL: Patents were an early constitutional amendment.

[cough]

Shouldn't someone stop to think that a concept from the 1780s is maaaaaaybe out of date?

The patent act was revised last year.


Uhh you put [cough] over the part that changes the whole meaning of what I said.

I said we need a whole new patent system, not piddly little "revisions".

Maybe go back and read the words that I typed? Or is that too hard?
 
2012-11-21 02:41:17 PM

Theaetetus: Driedsponge: Theaetetus: Patents are about implementations, not general concepts, and here, the implementation is different.

As I understand it, one of Apple's major victories over Samsung includes the 'slide to unlock' patent. Apple takes a solid bar with a slide motion across the screen from left to right to unlock, Samsung (Android) has a round circle where you drag from the center to a point on the outer ring. Same general concept, different implementation, and the court determined it was infringing.

Except that you're comparing Apple's iOS to Samsung's Android implementation... not Apple's patent to Samsung's implementation. The patent doesn't require a solid bar or a motion from left to right.

Incidentally, the same mistaken comparison was the issue in the UK Apple-Samsung design patent fight - Apple was comparing the Galaxy Tab to the iPad, but not the Galaxy Tab to the patent, which wasn't quite the same.

Part of the problem people are having with these patents, and I think the reason so many of them are actually being invalidated on court review is that so many of them these days are so broad they encompass a general concept, and don't distinguish between specific implementations.

Actually, over 80% of patents are upheld by courts on review, and of those decisions that are appealed, 95% are upheld by the Federal Circuit. Kappos talks about it in the speech this article is about.


I agree that in the terms of the patent claim, Samsung did infringe. My actual point was that Apple's patent is encroaching on encompassing a general concept as opposed to an exact implementation.

And the courts overturning 20% of patents on review is an astronomically high number. Period. Using 80% as an argument for your case just proves to me the patent approval system has some major flaws. Even 5% on something that is supposed to go through an extensive internal review is bordering on excessive.
 
2012-11-21 02:42:25 PM
Hahaha wait, Theaetetus is a patent attorney according to his profile.

Let him be, guys. He's blind.
 
2012-11-21 02:46:50 PM

torusXL: Theaetetus: torusXL: Patents were an early constitutional amendment.

[cough]

Shouldn't someone stop to think that a concept from the 1780s is maaaaaaybe out of date?

The patent act was revised last year.

Uhh you put [cough] over the part that changes the whole meaning of what I said.

I said we need a whole new patent system, not piddly little "revisions".

Maybe go back and read the words that I typed? Or is that too hard?


Okay, I'll bite.

What's the new patent system going to look like?
 
2012-11-21 02:49:23 PM
Someone should file a patent on the how a human being can complain about how people are unfairly critizing the process.

I mean if we are allowing the patents of how to turn a book page and the shape of a rectangle on plastic. Granting a patent on complaining should be second nature.

Then we sue him for infringement.
 
2012-11-21 02:52:45 PM
In my mind I see software patents as only covering general ideas, and not the implementation. From what I've seen they only cover the input/output, not the procedures. It's like patenting "a device that traps a mouse that has interacted with it" and not the system of trigger, springs, and levers.
And now my cat is a patent violation.
 
2012-11-21 02:53:43 PM

The Voice of Doom: tgambitg
You really are clueless to what everyone is trying to say, aren't you? All that stuff I took out was the 'on the electronic device' stuff. And all of that is software based. Copyright the code that does it, but it should never have been eligible for a patent. It literally copies a physical slide lock in digital form. You're getting hung up on the legalese that tells you 'oh this isn't a physical lock, see all this stuff? it means a software lock so it can't be the same as the physical lock.'

Thinking about it..
software development or even big parts of computer science and math are about finding abstract representations and solutions for specific problems.
That's where heavy lifting is done in those fields.
Just think of graph theory.
A new, more efficient algorithm for finding some sort of special path (think Euler tour or traveling salesman) could be an invention.
Whether you apply this algorithm to facebook's friendship graph or in a navigation system instead of the bridges of Königsberg means fark all for innovation in that field.

Looking at the slide-to-unlock example, it's no wonder this mindset collides with one that considers the "limitations" as what makes the idea patent-worthy.
This works for physical innovations where you have to do actual work to specialize the solution instead of simplifying or generalizing the concept behind it.
If someone makes a machine that sorts coins by size and then someone comes along and builds one that sorts them by their actual value - that sounds like an innovation.
If you come to a CS person and say "I've written a program that can sort letters instead of numbers" - that's just cute compared to one sorting generic objects by every property you like by passing a comparator function as parameter.


Computer science and, in general, virtual representations of things are just not fit for a patent system. We're entering into a bigger Information Age than will be expected by the dinosaurs running the slow beuracracy. Everything will become information, everything will be copyable. Everything will be abstractable and everything will have some generalized theorem hovering silently behind it, ready for some patent troll to seize and thereby take control of whole market segments.

Earthly organisms are information processing machines. Genetic code is activated and deactivated in networks of interactions to produce proteins. If the patent system sits as it is today, it won't be long until someone patents the Genetic Code Protein Synthesizer Device and patents all life on Earth.

How about 50 years from now (not long ag all) when there might be technology for total realism virtual reality worlds? That's could be owned by a patent troll. What will you do if you're born inside one, but have no idea that actually your life and existence is owned by an investment board for VR Corporation?

It sounds silly, but these kinds of insanities are what's at stake when allowing patents on ethereal "information".
 
2012-11-21 02:56:21 PM

Driedsponge: Theaetetus: Driedsponge: Theaetetus: Patents are about implementations, not general concepts, and here, the implementation is different.

As I understand it, one of Apple's major victories over Samsung includes the 'slide to unlock' patent. Apple takes a solid bar with a slide motion across the screen from left to right to unlock, Samsung (Android) has a round circle where you drag from the center to a point on the outer ring. Same general concept, different implementation, and the court determined it was infringing.

Except that you're comparing Apple's iOS to Samsung's Android implementation... not Apple's patent to Samsung's implementation. The patent doesn't require a solid bar or a motion from left to right.

Incidentally, the same mistaken comparison was the issue in the UK Apple-Samsung design patent fight - Apple was comparing the Galaxy Tab to the iPad, but not the Galaxy Tab to the patent, which wasn't quite the same.

Part of the problem people are having with these patents, and I think the reason so many of them are actually being invalidated on court review is that so many of them these days are so broad they encompass a general concept, and don't distinguish between specific implementations.

Actually, over 80% of patents are upheld by courts on review, and of those decisions that are appealed, 95% are upheld by the Federal Circuit. Kappos talks about it in the speech this article is about.

I agree that in the terms of the patent claim, Samsung did infringe. My actual point was that Apple's patent is encroaching on encompassing a general concept as opposed to an exact implementation.

And the courts overturning 20% of patents on review is an astronomically high number. Period. Using 80% as an argument for your case just proves to me the patent approval system has some major flaws. Even 5% on something that is supposed to go through an extensive internal review is bordering on excessive.


Apple's patent is specific. It requires continuous contact with the screen, a continual UI tracing the actions of your finger, and a predefined start and end location. All of these things are easy to manipulate your device around to not infringe (like making the UI discrete, like a dotted line that follows your finger every .5s for instance).

And I'd argue 5% is perfectly acceptable as that means the amount of incorrect patents is far less than that (95% of cases that are BROUGHT TO THE COURTS are upheld, not all of the patents in existance).
 
2012-11-21 02:59:57 PM

RatOmeter: Curious: Theaetetus:

You can argue all day that what I did 15-20 years ago is not at all the same ...


Meant to add that I didn't describe my old work as proof of prior art or demonstration of my awesomeness. It's that I agree that making a little bit of code on a computer do what an old mechanical lock did before does not instantly qualify as a new, non-obvious, patentable invention.
 
2012-11-21 03:15:40 PM

Shazam999: torusXL: Theaetetus: torusXL: Patents were an early constitutional amendment.

[cough]

Shouldn't someone stop to think that a concept from the 1780s is maaaaaaybe out of date?

The patent act was revised last year.

Uhh you put [cough] over the part that changes the whole meaning of what I said.

I said we need a whole new patent system, not piddly little "revisions".

Maybe go back and read the words that I typed? Or is that too hard?

Okay, I'll bite.

What's the new patent system going to look like?


Use your imagination, it's what needs to be figured out. Did I ever say I have a new system ready to go in my back pocket? What's with this pure pessimism you've got going?

Here's an idea I came up with in 5 minutes (you might be able to do the same if you squeeze your ass cheeks together really hard).

First, some background. Algorithms can be considered equivalent if written in a Turing Complete language. "Turing languages" are actually just instantiations of the same underlying concept of a specific type of virtual machinery. Patents are fundamentally wrong for computer software because the underlying mechanism creates equivalences between algorithms that may not even appear the same at first glance. In addition, a programming language means nothing until it has been transformed into machine usable code.

What is needed is a different classification all together for virtual products, which would leave software out of the patent system (where it does not belong). For example, it could be stipulated by law that: To gain a software patent, the source code is required to be printed in the patent. Software would still be patentable, but it would be much more risky from a business perspective to do so.

Therefore, most companies would prefer to classify their software as a trade secret, just like Coke and Pepsi with their recipes. Then the company can't go out and troll the world when someone inevitably creates an equivalent algorithm. Just like if I use my own time and effort to create a soda recipe that tastes just like Coke, Coke can't do anything about it since I used legitimate effort and did not steal their trade secret recipe.

The overall message here is that software is only patentable when examining the actual source code and understanding the target hardware platform. Exactly like regular industry trade secrets. By letting patent trolls in the software industry like Apple run these crazy lawsuits, we are slowly legalizing a fundamentally wrong intellectual property process.

As far as Apple v. Samsung goes, well, the problem seems to be that no one on the jury had any specific knowledge about how to compare the tablets. Most likely, each company used different hardware formulations for things such as the touch screen (which in my opinion is the main patentable part of a tablet computer). If the jury could review the actual hardware implementation of the screens, it would have been trivial to say "Samsung and Apple would have had to use their own time and effort to develop software on top of these differing touchscreen hardware platforms. There is most obviously no infringement". With just a few minutes lazily searching the internet, I can see that ipads have LCD screens and Galaxies use an AMOLED display. Seriously, fark Apple. But most importantly when it comes to software, fark the patent system.
 
2012-11-21 03:19:20 PM

hammer85: Apple's patent is specific. It requires continuous contact with the screen, a continual UI tracing the actions of your finger, and a predefined start and end location. All of these things are easy to manipulate your device around to not infringe (like making the UI discrete, like a dotted line that follows your finger every .5s for instance).


There are only so many motions you can make on a touchscreen with a finger. There's tap, and slide. Patenting the use of one of those motions to unlock a screen is overly broad. Not to mention the fact that this entire process can be described internally to the system as a mathematical algorithm (a function describing the motion of an input variable in space utilizing starting and ending initial conditions).

And I'd argue 5% is perfectly acceptable as that means the amount of incorrect patents is far less than that (95% of cases that are BROUGHT TO THE COURTS are upheld, not all of the patents in existance).

That's just about the most ridiculous assumption in the thread so far. 5% of all patents that are brought under further scrutiny outside of the patent office are invalidated. That does not automatically imply the actual number of granted, invalid patents is
 
2012-11-21 03:23:31 PM

Theaetetus: Kinek: Alright. Patent from last thursday. The Time Machine patent. I didn't see a flow chart on that. Or anything beyond some nebulous claims.

Maybe you should look again, then.

Look at that patent, and build it for me.

I think a skilled programmer could easily build it from that specification and those figures.

But that aside:
Kinek: Well, it's one of the big reasons that I think his theory of Patents as disclosure keeping us from the apocalyptic doom of Trade Secrets is full of crap. Because the patents today AREN'T DISCLOSURE. At least they're not being used that way.

You seem to be under the misconception that patents are the only disclosures. They aren't. The patent actually protects all disclosures of the invention. If you have a trade secret, you can't publish a white paper, a thesis, a review, a journal article, a brochure, a functional specification, a pseudocode library, your source code, your schematics, etc., without destroying the secret. If you have a patent, you can do any of those. So, yeah, patents aren't being used as the sole, exclusive disclosure of something, because they don't have to be.

I'd honestly rather take trade secrets over this kind of crap. At least with Trade secrets you can't just sue your competitor.

Yes, you can, and they're outrageously expensive lawsuits.


Wait. So the billion dollar patent lawsuit -isn't- outrageously expensive?

Farking hell. Soon we'll be levying trillion dollar judgements over 'A device to facilitate the flow of power through a system.'
 
2012-11-21 03:25:03 PM

tgambitg: Theaetetus: Not the same at all. And you can't find a patent with that claim, as I said.

Now you're just arguing semantics.


Not at all. I'm saying that the claim "[known invention] on a computer" doesn't exist. You're apparently agreeing with me, but saying that we can ignore the majority of "[known invention] on a computer [with the computer doing lots of new and nonobvious things]" and that any complaints about that are just semantics.
 
2012-11-21 03:25:53 PM

ProfessorOhki: Theaetetus: No, there aren't. Bet you can't find a single one. I'll bet a month of TF that you can't find an independent claim in a patent that says "1. A method for doing [known process], comprising: [doing known process] ON COMPUTERS!"

Are abstracts good enough, or are you insisting on claims?

A record button that facilitates audiovisual input into a computer system without requiring manual interaction (direct manipulation interaction) with software. The record button may be grouped with transport controls, a standalone button, or grouped with other controls.


Claims... Abstracts have no legal weight. They also represent the application as filed, which may not represent the application as issued and patented.
 
2012-11-21 03:27:46 PM

torusXL: Shazam999: torusXL: Theaetetus: torusXL: Patents were an early constitutional amendment.

[cough]

Shouldn't someone stop to think that a concept from the 1780s is maaaaaaybe out of date?

The patent act was revised last year.

Uhh you put [cough] over the part that changes the whole meaning of what I said.

I said we need a whole new patent system, not piddly little "revisions".

Maybe go back and read the words that I typed? Or is that too hard?

Okay, I'll bite.

What's the new patent system going to look like?

Use your imagination, it's what needs to be figured out. Did I ever say I have a new system ready to go in my back pocket? What's with this pure pessimism you've got going?

Here's an idea I came up with in 5 minutes (you might be able to do the same if you squeeze your ass cheeks together really hard).

First, some background. Algorithms can be considered equivalent if written in a Turing Complete language. "Turing languages" are actually just instantiations of the same underlying concept of a specific type of virtual machinery. Patents are fundamentally wrong for computer software because the underlying mechanism creates equivalences between algorithms that may not even appear the same at first glance. In addition, a programming language means nothing until it has been transformed into machine usable code.

What is needed is a different classification all together for virtual products, which would leave software out of the patent system (where it does not belong). For example, it could be stipulated by law that: To gain a software patent, the source code is required to be printed in the patent. Software would still be patentable, but it would be much more risky from a business perspective to do so.

Therefore, most companies would prefer to classify their software as a trade secret, just like Coke and Pepsi with their recipes. Then the company can't go out and troll the world when someone inevitably creates an equivalent algorithm. Just like if I use my ...


Dude, I'm not the one that wants a whole new patent system. You are. So provide something tangible.

I'm not sure why you keep bringing up trade secrets as a protection mechanism. UI elements especially could never be protected that way, since you can't really hide the way they work.
 
Displayed 50 of 167 comments

First | « | 1 | 2 | 3 | 4 | » | Last | Show all

View Voting Results: Smartest and Funniest


This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »






Report