Elan Microelectronics Turns to U.S. International Trade Commission in Multi-Touch Patent Dispute With Apple

[cmaier]

1) you need to pay attention to the claims

Please what have I missed?


2) the only scientific technique we have to determine whether something would have been obvious to someone having ordinary skill in the art in the past is to compare the supposed invention to the state of the art at the time. If the differences are insubstantial, it's obvious.

And hypothesizing putting a single sensor in an xy array of multiple sensors is just that.

The state of the art at the time can be determined by looking at all the publications and products in the world at that time. In court, this is the analysis that took place. It seems a pretty good system to me - I'm not sure how we can do better.

Automatically exclude things that are obvious extensions of existing technology

All you've done is repeatedly declare that it's obvious, and state your incredulity that it could be found not so. What facts do you have to support your conclusion?[/b]

Common sense - if something exists suggesting putting that something in an xy array to make it do what it does multiple times is obvious, its been done before with non-capacitive sensors such as keyboards, that any and all sensors would reasonably have such future application is not inventive, particularly not in each individual case of type of sensor.

If it was so obvious, it would have been done! But it wasn't.
We have no indication that it was 'done' even with this patent - this is not about having actually produced a new type of technology, i.e. a multiple sensor grid, it is about what you would do with one if you had it. And what's being done is, again, intuitively obvious and should never be patentable.

There is no requirement that it be "done" to be patented. Only that the disclosure be sufficient such that a person having ordinary skill in the art would understand the invention and would be enabled to practice the invention.

You want to "automatically exclude things that are obvious extensions of existing technology" but that's a circular definition. An extension of existing technology is obvious when? The test is it's obvious if the extension is insubstantial. The test used by the law has the advantage of not being circular, unlike your "it's obvious when the difference is obvious" test.

The claims do not suggest "putting that something in an xy array to make it do what it does multiple times" despite you repeatedly saying so. They are means+function claims, which mean they are limited to the specific circuits and techniques taught in the rest of the patent specification. Elan does NOT have a patent on putting capacitive sensors in a grid, which is what you keep saying.

 

[BobVB]

There is no requirement that it be "done" to be patented. Only that the disclosure be sufficient such that a person having ordinary skill in the art would understand the invention and would be enabled to practice the invention.

The test used by the law has the advantage of not being circular, unlike your "it's obvious when the difference is obvious" test.

And the disadvantage of licensing patents on things that are totally derivative.

The claims do not suggest "putting that something in an xy array to make it do what it does multiple times" despite you repeatedly saying so.

Yes it does, what follows is what's intuitively obvious if you have an input grid of that nature that is composed of smaller units than the proposed entry device (a finger).

They are means+function claims, which mean they are limited to the specific circuits and techniques taught in the rest of the patent specification.

And those circuits and means would be the intuitively obvious way to deal with such an input matrix. Scan by line, find highs, lows, calculate discrete points - again, how else would anyone do it presuming the existence of such a grid input?

Elan does NOT have a patent on putting capacitive sensors in a grid, which is what you keep saying.
No they have a patent on using such a grid in the totally obvious way it would be used, like a laser pointer and a cat.

Put 5 intelligent people in a room telling them the information you have and what you want to do and they would develop the same basic system before the end of the day. A 20 year patent on such a basic concept is inane.

 

[cmaier]

There is no requirement that it be "done" to be patented. Only that the disclosure be sufficient such that a person having ordinary skill in the art would understand the invention and would be enabled to practice the invention.

The test used by the law has the advantage of not being circular, unlike your "it's obvious when the difference is obvious" test.

And the disadvantage of licensing patents on things that are totally derivative.

The claims do not suggest "putting that something in an xy array to make it do what it does multiple times" despite you repeatedly saying so.

Yes it does, what follows is what's intuitively obvious if you have an input grid of that nature that is composed of smaller units than the proposed entry device (a finger).

They are means+function claims, which mean they are limited to the specific circuits and techniques taught in the rest of the patent specification.

And those circuits and means would be the intuitively obvious way to deal with such an input matrix. Scan by line, find highs, lows, calculate discrete points - again, how else would anyone do it presuming the existence of such a grid input?

Elan does NOT have a patent on putting capacitive sensors in a grid, which is what you keep saying.
No they have a patent on using such a grid in the totally obvious way it would be used, like a laser pointer and a cat.

Put 5 intelligent people in a room telling them the information you have and what you want to do and they would develop the same basic system before the end of the day. A 20 year patent on such a basic concept is inane.

It's not derivative if there's a substantial difference from what has come before, which is what a competent court of law has found.

The rest of your post is more conclusions without facts. You ignore that the claims require all the specific components described in the full text of the patent specification and deride it as "putting capacitive sensors in a grid," which is not what the claims cover. You ignore 25 pages of material that describe the "means" claimed in the claims.

It's so intuitively obvious that it takes 16 columns of text and a dozen figures with complex flow charts to explain it.

And the fact that you deride the question marks in the figures and claim that it means some sort of uncertainty on the part of the inventors tells me you haven't the slightest idea what you are talking about from a technical perspective, and don't know how to read flow charts.

 

[BobVB]

It's so intuitively obvious that it takes 16 columns of text and a dozen figures with complex flow charts to explain it.
As could your process of cooking breakfast - also a non-patentable 'means'. That you would think that such voluminous documentation isn't possible even for intuitive processes makes me think you really don't know how easy it is to make complex decision diagrams for even the simplest of tasks.

Want the simple far less than 16 pages way. - if you have any analog input matrix where the input divide is bigger than the discrete input device(s), determining how many different discrete input devices are currently being detected by the matrix will involve finding the center of each analog input, finding discrete areas of non activation between these centers, and enumerating them. This will be done by scanning for digital values from the analog input device, putting these digital values in a xy matrix and calculating positive value region centers, enumerating them and checking to see if there is a range of base value circuits between all centers.

And the fact that you deride the question marks in the figures and claim that it means some sort of uncertainty on the part of the inventors tells me you haven't the slightest idea what you are talking about from a technical perspective, and don't know how to read flow charts.

Funny tells me the same thing about you. Either you are dividing by 4 or you aren't. You can't say we are going to take n readings and divide by n and consider it a significant part of a patent - you can't patent taking an average.

 

[cmaier]

It's so intuitively obvious that it takes 16 columns of text and a dozen figures with complex flow charts to explain it.
As could your process of cooking breakfast - also a non-patentable 'means'. That you would think that such voluminous documentation isn't possible even for intuitive processes makes me think you really don't know how easy it is to make complex decision diagrams for even the simplest of tasks.

Want the simple far less than 16 pages way. - if you have any analog input matrix where the input divide is bigger than the discrete input device(s), determining how many different discrete input devices are currently being detected by the matrix will involve finding the center of each analog input, finding discrete areas of non activation between these centers, and enumerating them. This will be done by scanning for digital values from the analog input device, putting these digital values in a xy matrix and calculating positive value region centers, enumerating them and checking to see if there is a range of base value circuits between all centers.

And the fact that you deride the question marks in the figures and claim that it means some sort of uncertainty on the part of the inventors tells me you haven't the slightest idea what you are talking about from a technical perspective, and don't know how to read flow charts.

Funny tells me the same thing about you. Either you are dividing by 4 or you aren't. You can't say we are going to take n readings and divide by n and consider it a significant part of a patent - you can't patent taking an average.

What on earth are you talking about? Element 305? The question mark does not represent "maybe we are dividing by 4." The box is a test box, and the test is: "is (Xpeak1-XValley) greater than (Xpeak1)/4 AND is (xpeak-xvalley)>(xpeak2)/4) AND is ... ?" There is a question mark because the box asks a question - is a set of values within certain criteria? If yes, do one thing. If no, do another thing.

I don't mind arguing the merits or lack thereof of the current patent system with someone who knows what they are talking about, but you are deriding a patent disclosure that you don't even begin to understand.

 

[BobVB]

What on earth are you talking about? Element 305? The question mark does not represent "maybe we are dividing by 4." The box is a test box, and the test is: "is (Xpeak1-XValley) greater than (Xpeak1)/4 AND is (xpeak-xvalley)>(xpeak2)/4) AND is ... ?" There is a question mark because the box asks a question - is a set of values within certain criteria? If yes, do one thing. If no, do another thing.

As opposed to the box immediately above that has similar decisions and doesn't use question marks? Please, so now you are saying the application is inconsistent.

Even if you are correct, you have previously stated that these are 'precise' documents so if apple divides by 3 or by 5 or by 4.5 then it doesn't infringe, right?

I don't mind arguing the merits or lack thereof of the current patent system with someone who knows what they are talking about, but you are deriding a patent disclosure that you don't even begin to understand.
Again, I would say the same about you... it is only in such specifics as the 4, or the match of those particular boxes that there is any 'precision' about this patent and all are things that could be done differently with the same functional result. Apple may not use a 25% threshold to detect a 'click' as you are saying 305 demands. Again, that page is merely saying if the two fingers are pressed harder than 1/4 of the base reading and both are larger than the click threshold an xbutton will be considered 'down' otherwise up. I don't care how many pages of such elementary stuff you print up none of it should be patentable.

 

[cmaier]

As opposed to the box immediately above that has similar decisions and doesn't use question marks? Please, so now you are saying the application is inconsistent.

Even if you are correct, you have previously stated that these are 'precise' documents so if apple divides by 3 or by 5 or by 4.5 then it doesn't infringe, right?

I don't mind arguing the merits or lack thereof of the current patent system with someone who knows what they are talking about, but you are deriding a patent disclosure that you don't even begin to understand.
Again, I would say the same about you... it is only in such specifics as the 4, or the match of those particular boxes that there is any 'precision' about this patent and all are things that could be done differently with the same functional result. Apple may not use a 25% threshold to detect a 'click' as you are saying 305 demands. Again, that page is merely saying if the two fingers are pressed harder than 1/4 of the base reading and both are larger than the click threshold an xbutton will be considered 'down' otherwise up. I don't care how many pages of such elementary stuff you print up none of it should be patentable.

You might want to invest in a keyboard with punctuation characters.

Can you at least admit that you are wrong, and that the ?'s refer to "is this value greater than that value?" and stop trying to save face by pointing out that ?'s aren't used everywhere? If so, perhaps we can continue a rational conversation. If not, then it's pointless, since it would indicate that your opinions are uninfluenced by reality.

 

[BobVB]

You might want to invest in a keyboard with punctuation characters.

Can you at least admit that you are wrong, and that the ?'s refer to "is this value greater than that value?" and stop trying to save face by pointing out that ?'s aren't used everywhere? If so, perhaps we can continue a rational conversation. If not, then it's pointless, since it would indicate that your opinions are uninfluenced by reality.

Sure, either way I win - either they were trying to be nonspecific which invalidates your 'precision' claim, or they used a specific figure of 4 when if someone else uses anything else invalidates their 'precise' patent.

Looking through I see the application is inconsistent so you are probably right, rather than just one ? at the end they put one after every clause.

Still when are you going to admit that what that entire page is describing is basic to the point of not being patentable? That all they are describing is the way anyone would do so from an analog input matrix to decide if there were two discrete contacts and that these contacts were above what ever threshold the device owner would care to set? Fill a matrix, find the significant contact points, evaluate their relative proximity. Again, that's simply derivative of the technology, the thing you say they AREN'T patenting.

 

[Liquorpuki]

Still when are you going to admit that what that entire page is describing is basic to the point of not being patentable? That all they are describing is the way anyone would do so from an analog input matrix to decide if there were two discrete contacts and that these contacts were above what ever threshold the device owner would care to set? Fill a matrix, find the significant contact points, evaluate their relative proximity. Again, that's simply derivative of the technology, the thing you say they AREN'T patenting.

So what your saying is all this technology is obvious based on the creation of a single capactive sensor?

 

[cmaier]

Still when are you going to admit that what that entire page is describing is basic to the point of not being patentable? That all they are describing is the way anyone would do so from an analog input matrix to decide if there were two discrete contacts and that these contacts were above what ever threshold the device owner would care to set? Fill a matrix, find the significant contact points, evaluate their relative proximity. Again, that's simply derivative of the technology, the thing you say they AREN'T patenting.

There's no "it's too basic" test. There are two tests: the thing was done before (it's not new - it was "anticipated" by the prior art), or it would have been obvious to do it.

Even if you look just at the flowchart on sheet 17, it is doing far more than what you claim. A specific geometric relationship between the maxima and minima (expressed mathematically) is required.

The damned patent expressly disclaims as prior art "touch screens and capacitive pads which employ a matrix of row and column electrodes and detect, for example, either the transcapacitance between row and column electrodes or the ..." It also expressly disclaims interpolation (Column 1:32-40). By law, the patent does not, therefore, cover these things. And yet that's exactly what you claim the patent covers.

As the patent points out (and as was proven in court), everyone had only ever done this with one finger at a time. No one had done it with two fingers at a time, because either: 1) they didn't know how. 2) it didn't occur to them to try. In either case, they are entitled to a patent. The fact that it is obvious to you, 14 years later, to use two fingers, is meaningless. The fact that once you decide to use two fingers there may be only one way to do it is meaningless.

And I dispute that your characterization is "all that they are describing." They talk for example, about analog multiplexors. You don't even know what an analog multiplexor is, so your assertion that the use of one is somehow so fundamental as to be unpatentable is meaningless. The use of an analog multiplexor followed by an A-D convertor is also not a foregone conclusion. Why not use A-D convertors to sample the conductors and then multiplex the digital signal instead? The method by which thresholds are used is sufficiently arbitrary to not be a foregone conclusion. Why store things in a RAM before processing? That doesn't seem necessary to me (as someone with a Ph.D. in electrical engineering).

You can argue there should be "it's not already done, but it's 'too basic'" test, but I'd be curious how you can state that test in a way that it can actually be applied.

 

[BobVB]

So what your saying is all this technology is obvious based on the creation of a single capactive sensor?

No keep up - he says the sensor array is NOT the patent. Or at least he did when I pointed out that creating was was derivative. He now says it is how they are dealing with the data that is patentable when they aren't really doing anything but stating the obvious way you would deal with data from such a sensor array.

I agree someone who designed a particular sensor array should be able to patent how to make that particular chip, but anyone who wants to design a better one should be free to do the same. But if what it spits out is just an xy matrix of digital values, unless you are doing something far beyond matrix math with simple real world adjustments for tech variances, how you massage that data into useable form shouldn't be patentable unless there is some 'secret sauce' level of effort involved. Just saying I'm taking one a difference value, comparing it to the max value divided by 4 (why 4?) and also requiring the value be above an arbitrary threshold level doesn't cut it.

 

[cmaier]

No keep up - he says the sensor array is NOT the patent. Or at least he did when I pointed out that creating was was derivative. He now says it is how they are dealing with the data that is patentable when they aren't really doing anything but stating the obvious way you would deal with data from such a sensor array.

I agree someone who designed a particular sensor array should be able to patent how to make that particular chip, but anyone who wants to design a better one should be free to do the same. But if what it spits out is just an xy matrix of digital values, unless you are doing something far beyond matrix math with simple real world adjustments for tech variances, how you massage that data into useable form shouldn't be patentable unless there is some 'secret sauce' level of effort involved. Just saying I'm taking one a difference value, comparing it to the max value divided by 4 (why 4?) and also requiring the value be above an arbitrary threshold level doesn't cut it.

It's a noise filter, and Shannon-Nyquist tells you 2 is a lower bound.

 

[BobVB]

There's no "it's too basic" test. There are two tests: the thing was done before (it's not new - it was "anticipated" by the prior art), or it would have been obvious to do it.

Even if you look just at the flowchart on sheet 17, it is doing far more than what you claim. A specific geometric relationship between the maxima and minima (expressed mathematically) is required.

you can't seriously be saying that - there is no such 'geometric relationship' defined on that page and there isn't enough on the page to be 'far more' of anything. Remove all the grade school stuff you and you are down to 'is the difference between the max signal and the intervening minimal signal greater than 1/4 of the max signal for both max signals AND are both signals greater than the defined threshold - if so there are 2 fingers. if not its counted as one finger.

Again, common sense - Apple uses 1/5 the value, or 1/4.5 the value and they aren't infringing on this precise patent.

The damned patent expressly disclaims as prior art "touch screens and capacitive pads which employ a matrix of row and column electrodes and detect, for example, either the transcapacitance between row and column electrodes or the ..." It also expressly disclaims interpolation (Column 1:32-40). By law, the patent does not, therefore, cover these things. And yet that's exactly what you claim the patent covers.

You said it wasn't about the sensor. Again, their setup is about scanning for xy data on a sensor, something you couldn't do with existing sensors.
I have admitted that you an individual sensor would of course be patentable - how you deal with that xy information out of any such sensor would not or have to be so specifically specified as to make it easily avoidable.

And in any case that means, if you are correct, that the Apple patent actually does not infringe on this one as one of their patent's claims is:

2. The touch panel as recited in claim 1 wherein the transparent sensing medium includes a pixilated array of transparent capacitance sensing nodes.

3. The touch panel as recited in claim 1 wherein the transparent capacitive sensing medium comprises:

a transparent electrode layer, the electrode layer including a plurality of electrically isolated electrodes and electrode traces formed from a transparent conductive material, each of the electrodes being placed at different locations in the plane of the touch panel, each of the electrodes having an individual trace for operatively coupling to capacitive monitoring circuitry

In other words an xy grid of electrodes the very thing you say they did not patent.

I guess we will see.

 

[rolphi]

And this is why we have judges

The recent exchange between cmaier and BobVB is an illustration of one of the roles of the judge in such a case.

Instead of having to deal with arguments that do not follow, innapropriate use of legally defined terms, misguided appeals to common sense, claiming facts not in evidence, circular conclusions, claiming victory by fiat, and personal sniping, you can roll your eyes slightly and say to the judge "Really?" and have him/her put the other debater in place. And isn't that nicer for everyone?

Okay, maybe you have to say "Objection" but wouldn't it be more dramatic if snarkiness worked too?

 

[BobVB]

Okay, maybe you have to say "Objection" but wouldn't it be more dramatic if snarkiness worked too?

Fortunately I'm immune to his snarkiness and goes to show if you keep them talking often they prove themselves wrong. If the patent is as specific as he claims there is no infringement on modern touchscreen interfaces that virtually all use which are grids of discrete electrodes which he thinks are specifically excluded. Maybe Synergy used the old tech covered by their patent and that's why they lost but seems that tech has been left behind by almost all smartphone touchscreen producers.

 

[cmaier]

Fortunately I'm immune to his snarkiness and goes to show if you keep them talking often they prove themselves wrong. If the patent is as specific as he claims there is no infringement on modern touchscreen interfaces that virtually all use which are grids of discrete electrodes which he thinks are specifically excluded. Maybe Synergy used the old tech covered by their patent and that's why they lost but seems that tech has been left behind by almost all smartphone touchscreen producers.

1) I think he is implying you are the one engaging in the behavior he was talking about, not me.

2) You didn't prove anything. Your arguments did not make any logical sense, and were based on a lack of understanding of either engineering or the law of patents, two areas with which I am intimately familiar. (For example - I said the patents don't cover X. You took that to mean that anything that includes X cannot infringe. Of course that's not how patents work. Just because the patents don't cover X, does not mean they don't cover X+Y. In fact, that was my argument. The patents don't MERELY cover scanning a grid [as you maintained repeatedly]. They require additional steps/components. You then pointed at dependent claims and indicated something about how if Apple doesn't do the dependent claims it can't infringe the patent, which is also incorrect. )

I see no point in continuing in arguing with someone who:

1) is not an engineer
2) is not a patent lawyer
3) who has no basic understanding of either engineering or patent law.

 

[MorphingDragon]

Why is it the Law systems all over the world make me want to face-palm harder?

 

[cmaier]

Why is it the Law systems all over the world make me want to face-palm harder?

I agree. I think we should adopt New Zealand's system of submitting all disputes to a panel of 12 koala bears, hobbits, and kangaroos.

 

[BobVB]

1) I think he is implying you are the one engaging in the behavior he was talking about, not me.

1) I think he is implying you are the one engaging in the behavior he was talking about, not me.
Unlikely since the snarky comments were initiated by you.

2) You didn't prove anything. Your arguments did not make any logical sense, and were based on a lack of understanding of either engineering or the law of patents, two areas with which I am intimately familiar.

One since you are the one who on one hand says certain sensors are excluded and then immediately imply that statement is not true I doubt mine is the logic problem, two, since my thesis is the laws of patents are flawed that is irrelevant.

(For example - I said the patents don't cover X. You took that to mean that anything that includes X cannot infringe. Of course that's not how patents work. Just because the patents don't cover X, does not mean they don't cover X+Y.

Did you mean, X AND Y because if it doesn't cover X it can't cover X+Y (as obviously X isn't there). If you are saying that some how the trivial math it uses was somehow issued a patent independent of how or the quality of the data collected by the sensor then we are back to my contention that the system is flawed.

In fact, that was my argument. The patents don't MERELY cover scanning a grid [as you maintained repeatedly]. They require additional steps/components.

And you specifically stated that certain sensor types were NOT covered by the patent and the new 'array of sensor' pads that are standard are specifically of one of those precluded types. So which is it, is the patent specific to a particular type of sensor or not? You waffle on this so much you can't really blame me for not being able to keep track on what you are saying.

You then pointed at dependent claims and indicated something about how if Apple doesn't do the dependent claims it can't infringe the patent, which is also incorrect.

yes, first you say that the claims are integral to the patent, and now you seem to be saying the sensor type is just a dependent claim that doesn't really exclude the sensors you 1 message ago said it did. Do you see why I think the system is broken?

I see no point in continuing in arguing with someone who:

1) is not an engineer
2) is not a patent lawyer
3) who has no basic understanding of either engineering or patent law.

and thinks that patent law is seriously flawed.

If your previous statement is correct about excluded sensors is correct then the patent is about the large analog single sensors that early trackpads and the early Palm Pilot's used and I doubt much of anyone uses any more. And if you examine the representations they seem to be talking about getting a single X value profile along its axis and a single y value profile along it which would provide a pseudo xy coordinate matrix where the x value was the same through the column range and the y value the same for all rows. Developing a math system that wasn't fooled by the inherent inaccuracy the single values in rows and columns would be slightly more than trivial and for that particular sensor type would be patentable - they had basically figured out a way to make a single sensor give out a matrix of data that could be used to reliable guess about multitouch points even though it wasn't a true xy value grid.

But no one needs that anymore. They have arrays of sensors that give them true point data sets in an digital coordinate system. There is no need to figure out a way to get good results from an approximated data set - its just simple geometric math of the kind you did in high school. If you have a true point value for the xy domain of data the math used in figuring out how many fingers are on the array of sensors is so trivial and obvious that it shouldn't be patentable. If this patent did so regardless of the kind of sensor used (in spite of you saying the patent specifically excluded just the type of senors now in use) in getting that xy data matrix then I maintain the patent system is seriously broken and not living up to its constitutional mandate in this regard. Letting someone patent basic matrix math from a true xy data grid is not advancing the technological arts for the citizens.

 

[MorphingDragon]

I agree. I think we should adopt New Zealand's system of submitting all disputes to a panel of 12 koala bears, hobbits, and kangaroos.

I think you have the wrong country.

 

[cmaier]

1) I think he is implying you are the one engaging in the behavior he was talking about, not me.

1) I think he is implying you are the one engaging in the behavior he was talking about, not me.
Unlikely since the snarky comments were initiated by you.

2) You didn't prove anything. Your arguments did not make any logical sense, and were based on a lack of understanding of either engineering or the law of patents, two areas with which I am intimately familiar.

One since you are the one who on one hand says certain sensors are excluded and then immediately imply that statement is not true I doubt mine is the logic problem, two, since my thesis is the laws of patents are flawed that is irrelevant.

(For example - I said the patents don't cover X. You took that to mean that anything that includes X cannot infringe. Of course that's not how patents work. Just because the patents don't cover X, does not mean they don't cover X+Y.

Did you mean, X AND Y because if it doesn't cover X it can't cover X+Y (as obviously X isn't there). If you are saying that some how the trivial math it uses was somehow issued a patent independent of how or the quality of the data collected by the sensor then we are back to my contention that the system is flawed.

In fact, that was my argument. The patents don't MERELY cover scanning a grid [as you maintained repeatedly]. They require additional steps/components.

And you specifically stated that certain sensor types were NOT covered by the patent and the new 'array of sensor' pads that are standard are specifically of one of those precluded types. So which is it, is the patent specific to a particular type of sensor or not? You waffle on this so much you can't really blame me for not being able to keep track on what you are saying.

You then pointed at dependent claims and indicated something about how if Apple doesn't do the dependent claims it can't infringe the patent, which is also incorrect.

yes, first you say that the claims are integral to the patent, and now you seem to be saying the sensor type is just a dependent claim that doesn't really exclude the sensors you 1 message ago said it did. Do you see why I think the system is broken?

I see no point in continuing in arguing with someone who:

1) is not an engineer
2) is not a patent lawyer
3) who has no basic understanding of either engineering or patent law.

and thinks that patent law is seriously flawed.

If your previous statement is correct about excluded sensors is correct then the patent is about the large analog single sensors that early trackpads and the early Palm Pilot's used and I doubt much of anyone uses any more. And if you examine the representations they seem to be talking about getting a single X value profile along its axis and a single y value profile along it which would provide a pseudo xy coordinate matrix where the x value was the same through the column range and the y value the same for all rows. Developing a math system that wasn't fooled by the inherent inaccuracy the single values in rows and columns would be slightly more than trivial and for that particular sensor type would be patentable - they had basically figured out a way to make a single sensor give out a matrix of data that could be used to reliable guess about multitouch points even though it wasn't a true xy value grid.

But no one needs that anymore. They have arrays of sensors that give them true point data sets in an digital coordinate system. There is no need to figure out a way to get good results from an approximated data set - its just simple geometric math of the kind you did in high school. If you have a true point value for the xy domain of data the math used in figuring out how many fingers are on the array of sensors is so trivial and obvious that it shouldn't be patentable. If this patent did so regardless of the kind of sensor used (in spite of you saying the patent specifically excluded just the type of senors now in use) in getting that xy data matrix then I maintain the patent system is seriously broken and not living up to its constitutional mandate in this regard. Letting someone patent basic matrix math from a true xy data grid is not advancing the technological arts for the citizens.

Since you seem to enjoy run-on sentences, I am not about to try untangling whatever it is you are trying to say. A couple of things do jump out at me:

1) they are not patenting "basic matrix math."
2) they are patenting a specific way of processing analog values produced by a matrix of analog capacitive transducers into a set of digital values while massaging the data so as to eliminate noise and false readings and improving the signal/noise ratio problems that come from using a finger rather than a pinpoint to generate the capacitance.
3) X+Y vs. X AND Y. I am not sure what point you are trying to make, so let me be clear. The patent does not provide Elan with a monopoly on X, because Elan admits X was in the prior art. It provides Elan with a monopoly on X combined with Y, Y being additional elements that were either not in the prior art (according to a Federal District Court Judge) or were in other areas of prior art but would not have been obvious to combine with capacitive sensors at the time the patent was filed (according to a Federal District Court Judge).
4) "But no one needs that anymore." What does that have to do anything? All that matters is the state of the art at the time the invention was made. Anything that happened after that point (which is presumptively one year prior to the filing date of the patent) is irrelevant.
5) I assume rolphi was talking about you since he included things like "inappropriate use of legal terms" (and he knows I am a patent attorney) and "circular arguments" (which I accused you of).

 

[cmaier]

I think you have the wrong country.

I know. As I've said on here previously, I love when kiwis get annoyed when people think new zealand is australia.

 

[BobVB]

Since you seem to enjoy run-on sentences, I am not about to try untangling whatever it is you are trying to say.

Are you still living under the illusion that you aren't the snarky one?

1) they are not patenting "basic matrix math."
See below.

2) they are patenting a specific way of processing analog values produced by a matrix of analog capacitive transducers into a set of digital values
Which you realize directly conflicts with your quote from 2 of your messages ago, with their diagrams from their patent, and with the technology that is currently available as opposed to what they were using at the time. The Apple patent and current technology produces digital data in the screen, its transmitted by a two sided digital data bus, not an analog source.

That is the problem - you are presenting this as if their particular solution to solving a problem using what is now antiquated technology applies to today's technology. Figure 2 confirms this - they are presenting an array with x and y conductors, current technology is an array of discrete data points. They are getting x and y conductivity profiles, modern technology is getting an specific value for each and every transducer. They are not the same situations, they do not require the same 'matrix math' to get usable results.

Again, read page 20 of the patent, the description of figure two is NOT a dependent claim but specifically states - 'Fig 2 shows in block diagram form the electronics of the present invention. There are NO X Y conductors in the iPhone, iPad or any current multitouch technology, each pixel has its own digital value, and its position is determined by physical placement, not its connection by a 2 conductor axis grid. How you would do 'the matrix math' on such a grid is different than how you would do it with modern technology.

Again, I fully support that they could patent a particular way to enable multitouch on a particular type of sensor. That shouldn't

massaging the data so as to eliminate noise and false readings and improving the signal/noise ratio problems that come from using a finger rather than a pinpoint to generate the capacitance.
Which is basic matrix math.


3) X+Y vs. X AND Y. I am not sure what point you are trying to make, so let me be clear. The patent does not provide Elan with a monopoly on X, because Elan admits X was in the prior art. It provides Elan with a monopoly on X combined with Y, Y being additional elements that were either not in the prior art (according to a Federal District Court Judge) or were in other areas of prior art but would not have been obvious to combine with capacitive sensors at the time the patent was filed (according to a Federal District Court Judge).

And the Y in this case is technology they neither mention, example or described in their patent claim... Do you know what specific technology Synergy was using at the time?

4) "But no one needs that anymore." What does that have to do anything? All that matters is the state of the art at the time the invention was made. Anything that happened after that point (which is presumptively one year prior to the filing date of the patent) is irrelevant.
And they invented how to do multitouch with a device that uses single value axis profiles filling a matrix with pseudo data (single x value for all columns, single y value for all rows )which has nothing to do with a matrix filled with digital individual transducer values. The "present invention" diagramed and mentioned in the patent is as relevant as wooden wheels on a car.

Again, look at page 21 of their patent; they go into great detail how the touchpad operates and that is NOT how current touchpads operate. The specifically state "one conductor for each row and each column" that would produce only the 1 dimensional x and y profiles seen in their diagrams.

5) I assume rolphi was talking about you since he included things like "inappropriate use of legal terms" (and he knows I am a patent attorney) and "circular arguments" (which I accused you of).

Well since I have not used any legal terms at all, and the waffling on the sensor and snarkiness are all yours I assumed the opposite.

Seriously, if their sensor filled the array with identical data as today's tech I could see your point but their very specific description of the 'present invention' doesn't do that - the data matrix of a modern screen is actual values, their invention is based on two linear X Y projection plots collected by arrays of x and y conductors projected into a 2D space. The math necessary for handling their 'pseudo matrix' would have to avoid false detection due to 'shadowing' of the projections, a matrix filled with real data would not have that hurdle and would be trivial to create for any matrix of real values of any sensor system.

Their patent is about a solution to a problem no one has any long.

 

[MorphingDragon]

I know. As I've said on here previously, I love when kiwis get annoyed when people think new zealand is australia.

I'm not annoyed, but I do think there are people stupid enough here to actually believe that.

 

[cmaier]

You seem to be confusing two separate issues. In this post your argument appears to be that the Elan patent is not infringed by Apple. Apple is all digital, or whatever it is you are trying to say.

Ok, fine. I don know what it is Apple actually does, and infringement is an issue that will be determined freshly by this court and on which the previous case has no bearing


But that is a completely different issue than whether or not the patent is VALID. The patent is valid if it was new and unobvious when it wa invented. It's scope is fixed at that time. Some things infringe and some don't. All I have said, repeatedly, is that it is likely valid, since synaptics tied very hard to invalidate it and failed. You had been arguing that it was invalid as it was "intuitively obvious" ("obvious" being the legal term you improperly and circularly define).

I'm trying to understand what your current point is, but it seems to merely be that Apple does not infringe. They may or may not. Many many posts ago someone said they did not, and I disagreed with the reasoning (he misunderstood terms in the patent and misunderstood how one determines infringement). But as we have no evidence as to the actual circuits used in apple devices, there is no way to determine whether there is infringement. But, again, that is a completely different issue than invalidity, which is what you had been arguing.

A court will first determine validity and enforeceability (separate issues) and then determine infringement (or vice versa- whatever is easier). Each question has its own set of rules and its own tests.

Since you seem to enjoy run-on sentences, I am not about to try untangling whatever it is you are trying to say.

Are you still living under the illusion that you aren't the snarky one?

1) they are not patenting "basic matrix math."
See below.

2) they are patenting a specific way of processing analog values produced by a matrix of analog capacitive transducers into a set of digital values
Which you realize directly conflicts with your quote from 2 of your messages ago, with their diagrams from their patent, and with the technology that is currently available as opposed to what they were using at the time. The Apple patent and current technology produces digital data in the screen, its transmitted by a two sided digital data bus, not an analog source.

That is the problem - you are presenting this as if their particular solution to solving a problem using what is now antiquated technology applies to today's technology. Figure 2 confirms this - they are presenting an array with x and y conductors, current technology is an array of discrete data points. They are getting x and y conductivity profiles, modern technology is getting an specific value for each and every transducer. They are not the same situations, they do not require the same 'matrix math' to get usable results.

Again, read page 20 of the patent, the description of figure two is NOT a dependent claim but specifically states - 'Fig 2 shows in block diagram form the electronics of the present invention. There are NO X Y conductors in the iPhone, iPad or any current multitouch technology, each pixel has its own digital value, and its position is determined by physical placement, not its connection by a 2 conductor axis grid. How you would do 'the matrix math' on such a grid is different than how you would do it with modern technology.

Again, I fully support that they could patent a particular way to enable multitouch on a particular type of sensor. That shouldn't

massaging the data so as to eliminate noise and false readings and improving the signal/noise ratio problems that come from using a finger rather than a pinpoint to generate the capacitance.
Which is basic matrix math.


3) X+Y vs. X AND Y. I am not sure what point you are trying to make, so let me be clear. The patent does not provide Elan with a monopoly on X, because Elan admits X was in the prior art. It provides Elan with a monopoly on X combined with Y, Y being additional elements that were either not in the prior art (according to a Federal District Court Judge) or were in other areas of prior art but would not have been obvious to combine with capacitive sensors at the time the patent was filed (according to a Federal District Court Judge).

And the Y in this case is technology they neither mention, example or described in their patent claim... Do you know what specific technology Synergy was using at the time?

4) "But no one needs that anymore." What does that have to do anything? All that matters is the state of the art at the time the invention was made. Anything that happened after that point (which is presumptively one year prior to the filing date of the patent) is irrelevant.
And they invented how to do multitouch with a device that uses single value axis profiles filling a matrix with pseudo data (single x value for all columns, single y value for all rows )which has nothing to do with a matrix filled with digital individual transducer values. The "present invention" diagramed and mentioned in the patent is as relevant as wooden wheels on a car.

Again, look at page 21 of their patent; they go into great detail how the touchpad operates and that is NOT how current touchpads operate. The specifically state "one conductor for each row and each column" that would produce only the 1 dimensional x and y profiles seen in their diagrams.

5) I assume rolphi was talking about you since he included things like "inappropriate use of legal terms" (and he knows I am a patent attorney) and "circular arguments" (which I accused you of).

Well since I have not used any legal terms at all, and the waffling on the sensor and snarkiness are all yours I assumed the opposite.

Seriously, if their sensor filled the array with identical data as today's tech I could see your point but their very specific description of the 'present invention' doesn't do that - the data matrix of a modern screen is actual values, their invention is based on two linear X Y projection plots collected by arrays of x and y conductors projected into a 2D space. The math necessary for handling their 'pseudo matrix' would have to avoid false detection due to 'shadowing' of the projections, a matrix filled with real data would not have that hurdle and would be trivial to create for any matrix of real values of any sensor system.

Their patent is about a solution to a problem no one has any long.