User: Password:
|
|
Subscribe / Log in / New account

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 15, 2012 22:24 UTC (Thu) by SecretEuroPatentAgentMan (guest, #66656)
In reply to: Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired) by pboddie
Parent article: Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

> You have to understand the nature and relatively frequent occurrence of independent discovery to understand the point.

I think I have seen such a statement before. And I believe I asked just how frequent this occurrence is. Cold facts and figures would be appreciated.


(Log in to post comments)

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 15, 2012 22:37 UTC (Thu) by dlang (subscriber, #313) [Link]

In the software field, independent discovery is the overwelming reason for patent disputes.

It's practically unheard of for anyone to go read patents to discover how to do something in software.

It's more common to hear a description of the results that someone else is achieving and create the code to achieve these results.

remember that patents are supposed to be about a specific way to do something, not about all possible ways to achieve the results.

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 19, 2012 23:03 UTC (Mon) by SecretEuroPatentAgentMan (guest, #66656) [Link]

> In the software field, independent discovery is the overwelming reason for patent disputes.

Care to provide some actual figures and references? Since wilful infringement in the US can carry triple damages there are all incentives to claim independent discovery. It would seem plausible that if this truly was the case there should be an substantial statistics for prior use rights. And that is not exactly overwhelming.

> It's more common to hear a description of the results that someone else is achieving and create the code to achieve these results.

That is no guarantee that the result you achieve is through the same algorithms or methods. Data compression, to take an example, can be achieved in many very different ways. So I cannot see that this makes independent discovery more probable.

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 20, 2012 0:57 UTC (Tue) by dlang (subscriber, #313) [Link]

>> In the software field, independent discovery is the overwelming reason for patent disputes.

> Care to provide some actual figures and references?

the simple fact that people don't have access to the source code that's implementing things, and therefor have no way of knowing exactly how the patent was actually implemented is one very strong thing.

The second simple fact is that the common practice in the software industry is to not read patents, so this means that anything that really 'infringes' had to be independent invention.

> It would seem plausible that if this truly was the case there should be an substantial statistics for prior use rights.

I don't know what country you are from, but there is no such thing as 'prior use rights' in the US patent law.

You may be able to claim that your use constitutes 'prior art', but only if you can prove that you implemented the invention more than two years prior to the person with the patent (and now that we are 'first to file', rather than 'first to invent', even that may not be valid)

> So I cannot see that this makes independent discovery more probable.

The problem is that when patents are being evaluated by the court in terms of infringement, they tend to read the claims in broad terms (Like Apple claiming that a click violates it's 'slide to unlock' patent because 'a click is a zero length slide'. In that case they happened to run into a Judge with a clue, but usually that sort of argument wins)

And the sheer cost of defending against a patent infringement suit can be crippling, even if there is zero real merit to the case. And on top of that, there's always the possibility of getting the same sort of 'justice' that Samsung got in it's case. As a result, the vast majority of software patent lawsuits get settled without going to court.

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 20, 2012 22:06 UTC (Tue) by SecretEuroPatentAgentMan (guest, #66656) [Link]

> the simple fact that people don't have access to the source code that's implementing things, and therefor have no way of knowing exactly how the patent was actually implemented is one very strong thing.

Source code is not required to work a patent but sufficient information for a person skilled in the art is a requirements for a patent.

> The second simple fact is that the common practice in the software industry is to not read patents, so this means that anything that really 'infringes' had to be independent invention.

That is a rather sweeping statement. Does your experience include cell phone technologies?

> I don't know what country you are from, but there is no such thing as 'prior use rights' in the US patent law.

Laws like this exists in many countries and under different names. In Germany this is known as "vorbenutzungsrecht" and German patent law has influenced much of European and some Asian patent laws. At the EPO this is implemented in Art 70 (4) EPC
http://www.epo.org/law-practice/legal-texts/html/epc/2010...

It appears that the US has something similar called "Prior user rights":
http://www.uspto.gov/aia_implementation/20120113-pur_repo...
35 U.S.C. 273:
http://www.uspto.gov/aia_implementation/20110916-pub-l112...

> (and now that we are 'first to file', rather than 'first to invent', even that may not be valid)

I was told the US moved to "first inventor to file" which is different from "first to file".

> The problem is that when patents are being evaluated by the court in terms of infringement, they tend to read the claims in broad terms

Are you referring to Marksman hearing? And how do you find it tends to be broad?

> And the sheer cost of defending against a patent infringement suit can be crippling

You can get insurance for such cases.

> As a result, the vast majority of software patent lawsuits get settled without going to court.

Out of court settlement is preferred to keep cost down and to make sure the settlement is not made public. I was not aware that bizarre judges was a factor in this.

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 20, 2012 22:28 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link]

> Source code is not required to work a patent but sufficient information for a person skilled in the art is a requirements for a patent.

Bullshit. A common way to patent something is to patent a basic obvious functionality that everyone would be _forced_ to replicate and which is NOT ENOUGH to produce actual working implementation.

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 22, 2012 21:42 UTC (Thu) by SecretEuroPatentAgentMan (guest, #66656) [Link]

>> Source code is not required to work a patent but sufficient information for a person skilled in the art is a requirements for a patent.

> Bullshit.

Are you objecting to source code not being required or to the requirement of sufficiency?

> A common way to patent something is to patent a basic obvious functionality that everyone would be _forced_ to replicate and which is NOT ENOUGH to produce actual working implementation.

How about a good example for this debate?

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 22, 2012 22:39 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link]

> Are you objecting to source code not being required or to the requirement of sufficiency?
I'm objecting to the fact that in reality software patents often don't have anything close to "sufficiency".

> How about a good example for this debate?
Practically any codec patent. Or Microsoft's flash storage patent.

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 24, 2012 11:58 UTC (Sat) by SecretEuroPatentAgentMan (guest, #66656) [Link]

>> Are you objecting to source code not being required or to the requirement of sufficiency?
> I'm objecting to the fact that in reality software patents often don't have anything close to "sufficiency".

That would be grounds for revocation.

>> How about a good example for this debate?
> Practically any codec patent. Or Microsoft's flash storage patent.

I did a clearance and neither my present employer nor I have ever worked with or for Microsoft. So let us take a hard look. Can you provide a patent number, preferably for a recently granted EP patent?

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 24, 2012 15:53 UTC (Sat) by Cyberax (✭ supporter ✭, #52523) [Link]

>> I'm objecting to the fact that in reality software patents often don't have anything close to "sufficiency".
>That would be grounds for revocation.
Yeah, sure. In the ideal world. While in reality it often takes many years to get a revocation.

> I did a clearance and neither my present employer nor I have ever worked with or for Microsoft. So let us take a hard look. Can you provide a patent number, preferably for a recently granted EP patent?
US5634050 - they've used this patent against TomTom, as far as I remember. However, this patent is not nearly sufficient to produce a working flash-based storage. There's no mention of load leveling, write amplification mitigation, etc.

However, should you create a real flash storage - you're likely to infringe this patent.

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 24, 2012 20:26 UTC (Sat) by SecretEuroPatentAgentMan (guest, #66656) [Link]

> US5634050 - they've used this patent against TomTom, as far as I remember. However, this patent is not nearly sufficient to produce a working flash-based storage. There's no mention of load leveling, write amplification mitigation, etc.

OK, I had a look.

Strictly speaking this does not state a flash-based storage, rather "A block-erasable, write-once, multiple-read memory device, the memory device having one or more blocks". This has several effects:
- it would cover much more than just flash-based devices
- it avoids issues relating to load levelling etc.

Also, strictly speaking, a flash-based system would work without wear-levelling, it would not just work that well or for very long. Independent claims are drafted with the fewest features possible that enables the invention to work, even though it might not be optimal.

Features that make the invention work better or well are typically found in dependent claims. You see a little about this in claim 10:

10. The method of claim 8 wherein each block has an erase count that indicates the number of times the block has been erased and wherein the step of selecting a block in the memory further comprises the step of selecting the block in the memory based on the erase count.

Most patent offices have requirements to unity, that the claims should be directed to a single invention. Supposedly this is to make the claims easy to read, in reality I suspect it is to earn more fees. This means that claims directed to wear levelling would be in a different application and claims relating to write amplifications would be in yet another application.

> However, should you create a real flash storage - you're likely to infringe this patent.

Considering priority according to parallel EP0557736 is 29.01.1992 the patent will have lapsed most places. The US has however different rules though that also have changed over time.

A quick review showed no third party observations, oppositions or re-examinations. It would appear that at the time competitors did not consider the patent a problem.

If instead you can find something fresher you could try a third party observation. In a few cases there are no fees and you do not require a professional representative.

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 20, 2012 22:49 UTC (Tue) by dlang (subscriber, #313) [Link]

> That is a rather sweeping statement. Does your experience include cell phone technologies?

Yes, it includes software used on cell phones.

Note that I am talking about SOFTWARE related patents here. Patents on the hardware (including the chips) are a different ball game.

> I was told the US moved to "first inventor to file" which is different from "first to file".

in the case of multiple independent inventors, not really.

>> And the sheer cost of defending against a patent infringement suit can be crippling

> You can get insurance for such cases.

not really, to get the insurance, the insurance company needs to understand what risk they are taking on. That would require an extensive (and very expensive) patent search.

Show me one case where someone has been protected by this 'patent insurance'

Let’s Limit the Effect of Software Patents, Since We Can’t Eliminate Them (Wired)

Posted Nov 22, 2012 22:00 UTC (Thu) by SecretEuroPatentAgentMan (guest, #66656) [Link]

[about reading patents]
>> That is a rather sweeping statement. Does your experience include cell phone technologies?

> Yes, it includes software used on cell phones.

Well, I have only had a few clients in cell phone technologies so I am not making wide claims here. These companies take a dim view on "IPR bigamy" (other companies don't have such limitations).

> Note that I am talking about SOFTWARE related patents here. Patents on the hardware (including the chips) are a different ball game.

I was referring to software (and firmware, some differentiate between software and firmware).

>> You can get insurance for such cases.

> not really, to get the insurance, the insurance company needs to understand what risk they are taking on. That would require an extensive (and very expensive) patent search.

A Google search for "IPR insurance" gives more than 1 million hits. And the insurance company will make a risk assessment.

> Show me one case where someone has been protected by this 'patent insurance'

One of my clients with such an insurance has been able to take on several large opponents in infringement cases so I know these work. Since settlements are out of court and secret and since I prefer not to use this forum to advertise my services (and risk engrumpening the Editor) I hope you understand my reluctance to be specific here.


Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds