Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Vorbis development, status & patent issues (Read 65930 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Vorbis development, status & patent issues

Reply #75 –
>Wouldn't Xiph holding a patent on Vorbis technology put them in a stronger position lawsuit-wise to defend themselves?

Only if they patented some technology that the company sueing vorbis infringes upon, then you get a patent swap - happens all the time, someone tries to sue IBM, they will say lets see what we can sue you on...

Quote
So you're saying MPC is patent free?


What MPC has going for it is it is small fry with no real identifiable company behind it.

Vorbis development, status & patent issues

Reply #76 –
edit: removed double quote

Quote
But the problem, as has been repeated many times in this thread, is how can we be sure of this patent search?  Not in whether or not it was actually thorough enough, but more fundamentally, that it even exists in any meaningful sense.


I imagine that you have asked FhG, MS and Real for prof that their codec didn't include any other 3rd party license, right?

Quote
So far, nobody will disclose any information about it other than to say that it does indeed exist.  How is this any more useful than simply saying nothing at all, especially when there is no real reputation here to trust or any promises of compensation, etc. ?


Can we just forget about 100% bulletproof proofs? There ain't no such thing, either for codecs like Speex and Vorbis or for 3rd party claims in MP3/WMA/Real. MS just got sued over plugins in IE. Look at the GIF/JPEG/RAMBUS fiascos, ...

Quote
Considering the following, is it actually reasonable for us to simply accept Vorbis as being patent free?

1.  A patent search which you won't give us any details of.
2.  A lack of indemnification.
3.  A lack of conditions which would at least imply the correctness of the search (who exactly did the search, what did they search for, etc.. even if we can't have the results).
4.  Statements from others (Fhg) likely to be knowledgeable on the matter that indeed there may be patent infringement.


I guess it all depends on the meaning you give to patent-free. The only one that makes sense is to say that it means: "we didn't patent it and we did our best to make sure there aren't other patents".

As for 4), does it mean that if I claim I own a patent on MP3, you'll stop using it?

Now regarding patent searches, eventhough it's a good thing, it doesn't guarantee as much as you'd think (this once again applies to both free and proprietary codecs). First, there are probably patents on digital computers and stupidities like that. Of course, there's prior art, but only a court can reverse the patent. That alone would mean that all software is patented. There are lots of stupid patents like that that would have to go before a judge to be (most likely) turned down. That's why, in the end, only a judge can say patent-free or not. Even then, there's submarine patents. Some have perfected the art of holding patents "under water" for *decades*, so in theory, even MP3 is not clear from those.

So I hope you understand more clearly what we mean when we say "patent-free".

Vorbis development, status & patent issues

Reply #77 –
Quote
Quote
I don't have/know the actual patent number, but the description of the patent is pretty clear, so it shoudn't be too hard to verify.

US5357594 (Not from FhG, but one of the Ahead AAC developers was questioning this specifically)

I had a look at that patent and while IANAL, it looks pretty phony to me. What they're claiming is basically a patent on overlap-and-add and transform codecs in general. However, both were invented long before 1993. Their claims are so broad, some would cover anything from MP3 to JPEG, MPEG and the rest. It's one of those "a digital device that can programmed to perform different tasks" patents. The only specific claims are about how to create the window and they don't really fit with the way the Vorbis window is made.

Vorbis development, status & patent issues

Reply #78 –
Quote
Quote
Quote
I don't have/know the actual patent number, but the description of the patent is pretty clear, so it shoudn't be too hard to verify.

US5357594 (Not from FhG, but one of the Ahead AAC developers was questioning this specifically)

I had a look at that patent and while IANAL, it looks pretty phony to me. What they're claiming is basically a patent on overlap-and-add and transform codecs in general. However, both were invented long before 1993. Their claims are so broad, some would cover anything from MP3 to JPEG, MPEG and the rest. It's one of those "a digital device that can programmed to perform different tasks" patents. The only specific claims are about how to create the window and they don't really fit with the way the Vorbis window is made.

Umm, so did you check US5357594 or US5214742? The interesting/correct patent meant was mainly the US5214742.
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #79 –
Quote
Can we just forget about 100% bulletproof proofs? There ain't no such thing, either for codecs like Speex and Vorbis or for 3rd party claims in MP3/WMA/Real. MS just got sued over plugins in IE. Look at the GIF/JPEG/RAMBUS fiascos, ...

I guess it all depends on the meaning you give to patent-free. The only one that makes sense is to say that it means: "we didn't patent it and we did our best to make sure there aren't other patents".

As for 4), does it mean that if I claim I own a patent on MP3, you'll stop using it?

Now regarding patent searches, eventhough it's a good thing, it doesn't guarantee as much as you'd think (this once again applies to both free and proprietary codecs). First, there are probably patents on digital computers and stupidities like that. Of course, there's prior art, but only a court can reverse the patent. That alone would mean that all software is patented. There are lots of stupid patents like that that would have to go before a judge to be (most likely) turned down. That's why, in the end, only a judge can say patent-free or not. Even then, there's submarine patents. Some have perfected the art of holding patents "under water" for *decades*, so in theory, even MP3 is not clear from those.

So I hope you understand more clearly what we mean when we say "patent-free".

even as a dissapointed ex vorbis user, i must totally agree with jmvalin ... no one can ever claim that a program is patents free ... this is because the very nature of the patent process ... i mean, look at the linux case, and many, many others ... (and no, hp indemnifying users does not counts) ... it is obvius that the patents system is overhelmed and very ineficient ... what matters to me most, is when, if ever, vorbis would be a viable coded ...

PS1: garf, why are you so harsh on vorbis, being a "vorbis developer"?

PS2: i'm still re-ripping my collection of CDs (used to be vorbis) to "mp3 aps" to use on my SlimX

Vorbis development, status & patent issues

Reply #80 –
Quote
PS1: garf, why are you so harsh on vorbis, being a "vorbis developer"?

Delusion?

Vorbis development, status & patent issues

Reply #81 –
Quote
Umm, so did you check US5357594 or US5214742? The interesting/correct patent meant was mainly the US5214742.

Just looked at that one. I think Vorbis is probably OK because of the "as a function of signal amplitude changes" at the end of claim 1. I think Vorbis chooses the length of the window better than that. Otherwise, it would be easy to change the encoder to choose the window differently without much problem.

Vorbis development, status & patent issues

Reply #82 –
Quote
I imagine that you have asked FhG, MS and Real for prof that their codec didn't include any other 3rd party license, right?

No, and that's wholly irrelevant to my point.  I'm not aiming my discussion here at the patent situation regarding Vorbis, vs the patent situation regarding other codecs per se.  Rather I'm trying to examine the statement: "Vorbis is patent-free," in so far as what it means, the reasoning behind it, whether or not it is true or false, and how we can  test it's condition.

Whether or not MS or Real or Fhg use patented technology doesn't relate to that.

Xiph is saying that Vorbis is patent-free, yet there appear to be some flaws in how exactly that conclusion was reached (peering in from the outside at least, which in this context is all that matters), and there appears to be a lack of evidence which would allow a third party to verify this.

Quote
Quote
So far, nobody will disclose any information about it other than to say that it does indeed exist.  How is this any more useful than simply saying nothing at all, especially when there is no real reputation here to trust or any promises of compensation, etc. ?


Can we just forget about 100% bulletproof proofs? There ain't no such thing, either for codecs like Speex and Vorbis or for 3rd party claims in MP3/WMA/Real. MS just got sued over plugins in IE. Look at the GIF/JPEG/RAMBUS fiascos, ...


Sure.  We can just use "reasonable likelyhood" instead, if you'd prefer that.

As I tried to examine earlier though, there appears to be no way for one to reason that it is likely that Vorbis is patent free other than:

1.  Xiph says so (and simply that).
2.  AOL used Vorbis in Winamp (but did not say anything about patents).
3.  Xiph has not been sued yet.

Considering the significance of the statement that is being made (that Vorbis is "patent free"), and what this means to most companies (significant monetary value or rather, their livelihood), the premises being used to support this conclusion are somewhat flimsy, especially if a few other factors are considered:

1.  Xiph refuses to disclose patent search information.
2.  Xiph refuses to give indemnification.
3.  There would be extremely little for a company suing them to gain even if they were infringing.

Quote
Quote
Considering the following, is it actually reasonable for us to simply accept Vorbis as being patent free?

1.  A patent search which you won't give us any details of.
2.  A lack of indemnification.
3.  A lack of conditions which would at least imply the correctness of the search (who exactly did the search, what did they search for, etc.. even if we can't have the results).
4.  Statements from others (Fhg) likely to be knowledgeable on the matter that indeed there may be patent infringement.


I guess it all depends on the meaning you give to patent-free. The only one that makes sense is to say that it means: "we didn't patent it and we did our best to make sure there aren't other patents".


Then why do you say: "Vorbis is 'patent-free'", vs. "Vorbis likely uses unpatented technology."

I'll take a guess and assume because the former is more significant than the latter, and that most companies would not be willing to base their production around a technology with such a lack of assurance in security.

However you look at it though, by asserting that Vorbis is patent-free without actually being sure is bending the truth.  How important this is could be arguable, but I'm willing to bet that most of the people who have taken this statement for granted are not even aware of the relative uncertainty of it.

Furthermore, it seems absurd to make assumptions about the superiority of Vorbis over other alternatives based on something which you don't even know to be a definite attribute of Vorbis in the first place, nor which you can reasonably demonstrate (since we don't like absolutes here).

Quote
As for 4), does it mean that if I claim I own a patent on MP3, you'll stop using it?


No.  First of all, I wasn't relating in that statement any sort of judgement as to what course of action would be appropriate.  The statement was only made to be used as a foundation for reasoning as to whether or not it is likely that Vorbis uses patented technology or not.

As for the significance of #4, Fhg has been shown to work within the realm of patents in a known and given manner, and the status of the patents on their technology is openly disclosed, even if it may be incomplete or not entirely accurate.

Given this, it would be reasonable to assume that their statements on the status of something being patented would, at the very least and if nothing else, hold more weight than the pure lack of evidence whatsoever (that the patent search information has not been disclosed).

Quote
Now regarding patent searches, eventhough it's a good thing, it doesn't guarantee as much as you'd think (this once again applies to both free and proprietary codecs). First, there are probably patents on digital computers and stupidities like that. Of course, there's prior art, but only a court can reverse the patent. That alone would mean that all software is patented. There are lots of stupid patents like that that would have to go before a judge to be (most likely) turned down. That's why, in the end, only a judge can say patent-free or not. Even then, there's submarine patents. Some have perfected the art of holding patents "under water" for *decades*, so in theory, even MP3 is not clear from those.


I see a few more problems here:

If you dismiss or designify the importance of patent searches and the disclosure of this information, then how is being "patent free" (a state in which all patents are known, or disclosed, not to apply) important any longer?

This selling point no longer becomes very important because you admit that it is rather arbitrary, yet the problem is that this is the sole reason many people use to Vorbis in the first place.

Secondly, if you use the premise that only a Judge or court of law can determine whether something is "patent free" or not, then how can you rightly expect people to simply accept that something is patent free on word alone?

It may be the case that a patent search would not be legally significant in absolute sense, but at the very least it is closer to the legal system than nothing at all (and especially if we don't like absolutes, this is definitely an improvement).

Quote
So I hope you understand more clearly what we mean when we say "patent-free".


I do, I think, though I don't see how it makes sense or how the conclusions that are being drawn from that "matter of fact" make sense either.

Vorbis development, status & patent issues

Reply #83 –
Quote
Quote
I imagine that you have asked FhG, MS and Real for prof that their codec didn't include any other 3rd party license, right?

No, and that's wholly irrelevant to my point.  I'm not aiming my discussion here at the patent situation regarding Vorbis, vs the patent situation regarding other codecs per se.  Rather I'm trying to examine the statement: "Vorbis is patent-free," in so far as what it means, the reasoning behind it, whether or not it is true or false, and how we can  test it's condition.

Whether or not MS or Real or Fhg use patented technology doesn't relate to that.


What I meant is that if you were to buy boxed software from Real, there would likely be something along the lines of "this product is protected under patents #so and so". Do you see anything saying "We don't think there are other patents but we don't think so"? This is just sort of implied by the fact that the patent situation is such that you never really know for sure. Now, if you take Vorbis (or Speex), the logical adaptation of "this product is protected under patents #so and so" becomes "this product is not protected by patent" or "this is patent-free". That exactly what is meant by patent-free.

Quote
However you look at it though, by asserting that Vorbis is patent-free without actually being sure is bending the truth.  How important this is could be arguable, but I'm willing to bet that most of the people who have taken this statement for granted are not even aware of the relative uncertainty of it.


Your definition of patent-free seems to imply 100% certainty. There is no such thing when it comes to patents. That's why we use patent-free to mean something which actually makes sense.

Quote
Furthermore, it seems absurd to make assumptions about the superiority of Vorbis over other alternatives based on something which you don't even know to be a definite attribute of Vorbis in the first place, nor which you can reasonably demonstrate (since we don't like absolutes here).


You need to cross the lake and there are two boats. The first one says "hole-free hull" and by looking at it, you see no hole, but of course you can never be sure. The second boat says "there are 25 holes in the hull" and you can find a detailed description of each one. Which boat will you take.

Quote
If you dismiss or designify the importance of patent searches and the disclosure of this information, then how is being "patent free" (a state in which all patents are known, or disclosed, not to apply) important any longer?


I do believe they are useful if you can afford them. However, by your definition of patent-free, it wouldn't help in any way because it doesn't provide a 100% guarantee.

Vorbis development, status & patent issues

Reply #84 –
Quote
So I hope you understand more clearly what we mean when we say "patent-free".

If I understand you correctly, the only thing that can prove Ogg Vorbis to be patent free is a court of law. Ironically, this won't prove anything if Xiph wins; it's one patent down, and many to go. There are so many patents world wide, that for all practical concerns, they're infinite. So claiming that Ogg Vorbis is patent free is in fact an impossible claim.

Perhaps what you mean is that it is unpatented* and free**?

* You have not seen any evidence to the contrary, and have not patented it yourself.
** In every possible sense of the word. FSF or salesman...

Vorbis development, status & patent issues

Reply #85 –
Quote
The answer is simple: Yes. An accused person is innocent until proven guilty. A technology is patent free until proven to be infringing in a court of law. Since Vorbis has not been proven to be infringing on any patents (in fact, it has as yet not even been credibly accused of doing so), it is justified to say it's patent free.


I agree this is legally true (IANAL) and the whole point of this discussion is that Xiph claims its codecs to be patent free and other codecs (Musepack) don't. If someone claims that Xiph technologies have any patents then they have to prove it, not just say it. If they prove it, Xiph will probably die, I don't know (doesn't matter for the discussion, I think). If this doesn't happen, then Xiph can make "patent-free" claims.


btw: Didn't Buschmann say that Musepack infringed (or could be infringing) some patents? At least the early encoders?

cya

Vorbis development, status & patent issues

Reply #86 –
Quote
btw: Didn't Buschmann say that Musepack infringed (or could be infringing) some patents? At least the early encoders?

He guessed it was infriging MP2 patents owned by Philips. But he's a programmer, not a lawyer, so you can't expect that he has done a patent search on Musepack algorithms.

I think he mostly said that to warn people. "I'm not sure, but there MIGHT be patents owned by Philips, so be careful"

Vorbis development, status & patent issues

Reply #87 –
Quote
Quote
So I hope you understand more clearly what we mean when we say "patent-free".

If I understand you correctly, the only thing that can prove Ogg Vorbis to be patent free is a court of law. Ironically, this won't prove anything if Xiph wins; it's one patent down, and many to go. There are so many patents world wide, that for all practical concerns, they're infinite. So claiming that Ogg Vorbis is patent free is in fact an impossible claim.

What I meant is that only a court can say it's not patent-free. Of course, if one finds a patent that is obviously infringed, it might also qualify. Otherwise, patent-free and unpatented mean roughly the same thing to me, no?

Vorbis development, status & patent issues

Reply #88 –
Quote from: jmvalin,Sep 24 2003, 03:48 PM
Quote
Quote from: jmvalin,Sep 24 2003, 02:37 PM
I imagine that you have asked FhG, MS and Real for prof that their codec didn't include any other 3rd party license, right?

No, and that's wholly irrelevant to my point.  I'm not aiming my discussion here at the patent situation regarding Vorbis, vs the patent situation regarding other codecs per se.  Rather I'm trying to examine the statement: "Vorbis is patent-free," in so far as what it means, the reasoning behind it, whether or not it is true or false, and how we can  test it's condition.

Whether or not MS or Real or Fhg use patented technology doesn't relate to that.


Do you see anything saying "We don't think there are other patents but we don't think so"? This is just sort of implied by the fact that the patent situation is such that you never really know for sure.

I assume you meant to say:

"There may be other patents but we don't think so."

If that's what you meant to say, then, no, I don't think you would usually see that.  However, the fact that patent information is disclosed at all, is more significant than not disclosing this information.

At the very least it provides some place to start in attempting to make a judgement as to whether or not it is secure to use the technology, in contrast to a vacuous uncertainty and an empty assurance.

Quote
Now, if you take Vorbis (or Speex), the logical adaptation of "this product is protected under patents #so and so" becomes "this product is not protected by patent" or "this is patent-free". That exactly what is meant by patent-free.


Right, I understand what "patent free" means.  The problem is that you have demonstrated that "patent free" is the actual condition that exists.  And as for the absolute or not thing, you haven't even demonstrated the reasonable likelyhood that it is "patent free" (whether this is an absolute condition in this sense is not as important as the likelyhood that it could be of this condition).  See the premises listed in my previous posts about this.

Quote
Quote
However you look at it though, by asserting that Vorbis is patent-free without actually being sure is bending the truth.  How important this is could be arguable, but I'm willing to bet that most of the people who have taken this statement for granted are not even aware of the relative uncertainty of it.


Your definition of patent-free seems to imply 100% certainty. There is no such thing when it comes to patents. That's why we use patent-free to mean something which actually makes sense.


I allowed flexibility in the definition.  "Patent free" may not be an absolute thing, but whether it is or not, to say that something is "patent free," one would still need some sort of reasonable evidence to draw that conclusion, at least if we are going to presume that this non-aboluste condition is significant in any sense (the looser we allow the definition to be, the less significant being of the condition becomes, and this goes right back again to causing problems for Vorbis, if Vorbis is important by virtue of being "patent free").  Again, see my previous posts as for why this doesn't seem to be so cut and dry.

Quote
Quote
Furthermore, it seems absurd to make assumptions about the superiority of Vorbis over other alternatives based on something which you don't even know to be a definite attribute of Vorbis in the first place, nor which you can reasonably demonstrate (since we don't like absolutes here).


You need to cross the lake and there are two boats. The first one says "hole-free hull" and by looking at it, you see no hole, but of course you can never be sure. The second boat says "there are 25 holes in the hull" and you can find a detailed description of each one. Which boat will you take.


This analogy is flawed because you assume the data you are looking for is negative towards your purpose.  Crossing a lake in a boat with holes in it does not make sense obviously because it already gives something negative to your goal.  However, if holes are equated to patents, the reverse is true in the case of companies licensing technologies.  Patent certainty (or the likelihood if you prever) is definitely positive and not negative to the goal of security.  The more that you know your technology is related to something else, the less chance that you will find a suprise.  In this sense, instead of the data being negative, it is positive.  "The more the better".  The more patents that you are aware of as existing (or the more that you are aware, with reason, that no existing patents apply, if you prefer), and which are licensed, the less chance that a patent will be found which will cause you harm.

If you want to use a similar analogy, you could think of the holes as patents, and licensing them as patching the holes.  The more holes you are aware of and that have been patched, when given the entirety of the boat, the less likely that you will find more, especially since there is only a finite space which the boat can occupy (or in this case, a finite base of technology which is used which could be patented).

Quote
Quote
If you dismiss or designify the importance of patent searches and the disclosure of this information, then how is being "patent free" (a state in which all patents are known, or disclosed, not to apply) important any longer?


I do believe they are useful if you can afford them.


This makes no sense.  First you devalue patent searches saying that decisions about patents by the legal system are mostly arbitrary.  Then you come back and say that they are important.

Quote
However, by your definition of patent-free, it wouldn't help in any way because it doesn't provide a 100% guarantee.


My definition doesn't have to be absolute, the arguments work just as well even if it not.  You still need some measure of significance, and you don't seem to have that it would seem, at least not externally or in that way that you seem to be claiming.

Vorbis development, status & patent issues

Reply #89 –
Quote
You need to cross the lake and there are two boats. The first one says "hole-free hull" and by looking at it, you see no hole, but of course you can never be sure. The second boat says "there are 25 holes in the hull" and you can find a detailed description of each one. Which boat will you take.

Let me try to drag your analogy. Actually, it works very well for defending patented algorithms, and not the other way around.

If I were to choose, I would take the boat with the well documented holes. Because then, I can fix them all (or "pay the license fees") before setting sail.

If I take the boat that claims to be without any holes, I might very well find a hole (or "be sued over patent licensing") in the middle of my lake crossing. And then I'm screwed, my boat will sink (or "my company will be dragged to court").

Vorbis development, status & patent issues

Reply #90 –
Quote
If that's what you meant to say, then, no, I don't think you would usually see that.  However, the fact that patent information is disclosed at all, is more significant than not disclosing this information.

At the very least it provides some place to start in attempting to make a judgement as to whether or not it is secure to use the technology, in contrast to a vacuous uncertainty and an empty assurance.


Actually, what these statements say is "we know that this product is covered by these patents" (usually because the patents are owned by the company). We say exactly the same except that the "these patents" is the null set.

Quote
Right, I understand what "patent free" means.  The problem is that you have demonstrated that "patent free" is the actual condition that exists.  And as for the absolute or not thing, you haven't even demonstrated the reasonable likelyhood that it is "patent free" (whether this is an absolute condition in this sense is not as important as the likelyhood that it could be of this condition).  See the premises listed in my previous posts about this.


So you're the only one that gets to define what the "reasonable likelyhood" threshold is, right?

Quote
If you want to use a similar analogy, you could think of the holes as patents, and licensing them as patching the holes.  The more holes you are aware of and that have been patched, when given the entirety of the boat, the less likely that you will find more, especially since there is only a finite space which the boat can occupy (or in this case, a finite base of technology which is used which could be patented).


You mean that a boat with 10 patches and no apparent hole is safer than a boat with 0 patch and no apparent patch?

Quote
Quote
I do believe they are useful if you can afford them.


This makes no sense.  First you devalue patent searches saying that decisions about patents by the legal system are mostly arbitrary.  Then you come back and say that they are important.


I devalued patent searches as a way to get 100% certainty. I sure would like to get one for Speex "just to be sure", eventhough I'm pretty confident I evaded the patents (by using older technology in my case).

Vorbis development, status & patent issues

Reply #91 –
Quote
You mean that a boat with 10 patches and no apparent hole is safer than a boat with 0 patch and no apparent patch?

You just changed your analogy. Formerly, you didn't say absolutely nothing about patches. You just said one boat had well documented holes, and the other seemed to have none, but you coudln't get any guarantee.

BTW, patches make even less sense comparing to the whole patent issue.

Vorbis development, status & patent issues

Reply #92 –
The discussiong goes boring with all these analogies.. Why not just stick to the actual issue instead of arguing about analogies.. 

This is pretty interesting what jmvalin said:
Quote
Quote
Umm, so did you check US5357594 or US5214742? The interesting/correct patent meant was mainly the US5214742.

Just looked at that one. I think Vorbis is probably OK because of the "as a function of signal amplitude changes" at the end of claim 1. I think Vorbis chooses the length of the window better than that. Otherwise, it would be easy to change the encoder to choose the window differently without much problem.

Hopefully some knowledgeable people can elaborate this further..
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #93 –
Quote
Quote
You mean that a boat with 10 patches and no apparent hole is safer than a boat with 0 patch and no apparent patch?

You just changed your analogy. Formerly, you didn't say absolutely nothing about patches. You just said one boat had well documented holes, and the other seemed to have none, but you coudln't get any guarantee.

BTW, patches make even less sense comparing to the whole patent issue.

I was answering to Dibrom's modified analogy. Anyway, the analogy was there to illustrate my point, but since it's not an "formal" argument, it's not worth depating. So case closed.

Vorbis development, status & patent issues

Reply #94 –
Quote
Quote
If that's what you meant to say, then, no, I don't think you would usually see that.  However, the fact that patent information is disclosed at all, is more significant than not disclosing this information.

At the very least it provides some place to start in attempting to make a judgement as to whether or not it is secure to use the technology, in contrast to a vacuous uncertainty and an empty assurance.


Actually, what these statements say is "we know that this product is covered by these patents" (usually because the patents are owned by the company). We say exactly the same except that the "these patents" is the null set.

You're explaining "patent free" again when there is no need to.  As I said in the last post, I understand what you are meaning, however, what is important is how you came to this conclusion.

Where is the information for which you based your conclusion that Vorbis is "patent free" upon?  Can we see this? It seems that the answer is "no" judging the quotes from Jack earlier.  If we can't see this, why not?

Again, I don't need an explaination of what "patent free" is, rather what we're looking for is how do we know Vorbis is "patent free," especially in light of the circumstances.

Quote
Quote
Right, I understand what "patent free" means.  The problem is that you have demonstrated that "patent free" is the actual condition that exists.  And as for the absolute or not thing, you haven't even demonstrated the reasonable likelyhood that it is "patent free" (whether this is an absolute condition in this sense is not as important as the likelyhood that it could be of this condition).  See the premises listed in my previous posts about this.


So you're the only one that gets to define what the "reasonable likelyhood" threshold is, right?


No.  This is certainly open to interpretation.  However, if we are going to be using an existing system (patent system), it would make sense to adopt it's common conventions.

We need a patent search or some sort of technical information as to why Vorbis is "patent free."  Anyone can then draw a conclusion as to how likely it is that this is significantly so or not, but we at least need some starting point.

Whether or not the Xiph guys have this, most common people on the outside, especially those who would be lead to believe that Vorbis is "patent free," most certainly do not.

Quote
Quote
If you want to use a similar analogy, you could think of the holes as patents, and licensing them as patching the holes.  The more holes you are aware of and that have been patched, when given the entirety of the boat, the less likely that you will find more, especially since there is only a finite space which the boat can occupy (or in this case, a finite base of technology which is used which could be patented).


You mean that a boat with 10 patches and no apparent hole is safer than a boat with 0 patch and no apparent patch?


You misunderstand.  It would be safer to choose a boat with 10 patches and no apparent hole, for which you can verify and see for yourself, and which has a life jacket (indemnification) than a boat which someone has told you about but which will not show you a condition report for (expert analysis of the possibility of patents; the source code being available alone doesn't cut it since most people would be unable to draw conclusions about it, and most companies wouldn't bother to check it), which someone else has said may be severely damaged (Fhg), and further for which you are given no guarentees of.

And even if you wanted to consider a company which is interested in a technology with is subject to a lack of patents, the same could be held; the holes are the possibility of a patentable technology: if you have them and they are patched, then it is technology which you are aware of which could patented, which either 1) is patented and you have licensed it, or 2) is not patented and would not be a risk (such as the supposed case of Vorbis); if you have them and they are not patched, then that represents unmitigated risk.

Vorbis development, status & patent issues

Reply #95 –
Quote
The discussiong goes boring with all these analogies.. Why not just stick to the actual issue instead of arguing about analogies.. 

I would tend to agree, though it seems that "patent free" keeps needing constant redefinition, and that we can't actually get to whether Vorbis is "patent free" or not (regardless of definition), even though it's been asked for countless times in this thread.

Vorbis development, status & patent issues

Reply #96 –
This is how I see this discussion:

1. - Monty came and said "Hey, I made a patent free codec"
2. - A patent search is claimed to have been done, but nobody gets any proof of it.
3. - Neverthless, people started spreading the word, even though they didn't veryify the facts first.
4. - Some years later, HA members bring the issue "Hey, what were they saying about no patents? Is it really true? Did they offer any documents attempting to prove it back then?"
5. - Vorbis advocates get pissed because "patent free" had mostly became vox populi, something you take for granted and ask no questions about. So, "what's up with these people, questioning after all these years"?

Vorbis development, status & patent issues

Reply #97 –
Quote
You're explaining "patent free" again when there is no need to.  As I said in the last post, I understand what you are meaning, however, what is important is how you came to this conclusion.

Where is the information for which you based your conclusion that Vorbis is "patent free" upon?  Can we see this? It seems that the answer is "no" judging the quotes from Jack earlier.  If we can't see this, why not?

Again, I don't need an explaination of what "patent free" is, rather what we're looking for is how do we know Vorbis is "patent free," especially in light of the circumstances.


OK, I'll talk about Speex here because that's what I know best (of course), but the same would apply for Vorbis. When writing Speex I made sure I avoided the known patents, I used older techniques and I searched the USPTO database. I can't prove that I did that, you'll have to take my word for it. I still think this has much more value than a patent search (though a search could add to the "certainty") and I consider this enough to say patent-free. I'm not sure what other proofs you want.

Seriously, it took years for the JPEG patent to surface and even now, it's not clear if the patent is really valid. What more do you want?? The patent system is a big lottery, regardless of whether you're dealing with free or proprietary software. I'm sorry, but right now, there isn't much more we can do.

Quote

Vorbis development, status & patent issues

Reply #98 –
Quote
This is how I see this discussion:

1. - Monty came and said "Hey, I made a patent free codec"

Hehheh.. Well I think that's a bit too much simplified. After all Monty has definitely used very considerable amount of time and effort (as a sole employee of Xiph) to build the codec and all things around it, so gotta give lots of credit for doing that..

But as said, it would be good for everybody if the patent issues were cleared somewhat better than what has been done so far...
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #99 –
Quote
PS1: garf, why are you so harsh on vorbis, being a "vorbis developer"?

Well, considering how much time Garf spent on tuning in the Vorbis platform, its perfectly understandable that he's very concerned about being a developer "on thin ice".  Certainty is a big plus for developers and whether all the work they poured in was worth it. 

If Xiph.org can be crystal clear on patenting issues and it's enough to reassure everyone that Vorbis isn't infringing patents and that it won't be sued in the future, then it would alleviate a lot of this concern and doubt so people can continue going about contributing to it.  It is a lot  like doctors who suddenly find their medical indemnity insurance has disappeared.  Would they continue to work with the comfort and confidence knowing that they might get sued in the future?