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 66616 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Vorbis development, status & patent issues

Reply #150 –
Quote
So all Xiph could say is "we think we avoid patents A,B,... because this and this"

Which is off course better that the current situation.

And the point is, if they have even thought about being "patent free", then they must already have looked at this, and taken technical decisions based on ensuring that they do avoid Patents A, B,...

So we're not asking them to do any more work, or disclose something they've paid someone else to do. Just explain their technical decisions.


The cynical observer would wonder why it's not in the source code! If you're trying to comment the code, it's not unreasonable that you might say "here, we do x, y, then z (not the more obvious a, b, then c because this would infringe US pat no. x,xxx,xxx)" - otherwise you risk someone "tweaking" the code at a later stage, to use the better(?) and more obvious(?) technique, which also happens to be patented!

They must already have internal documentation for this.

Surely?


If not, they have no case for using the words "Patent Free". If "Patent Free" just means "we haven't applied for a patent for any of this stuff", then I've got a great idea for a patent free video disc format!

If they have worked all these issues through internally, then maybe they should make their reasoning public. Unless they aren't confident of their reasoning?


Cheers,
David.

Vorbis development, status & patent issues

Reply #151 –
Quote
And the point is, if they have even thought about being "patent free", then they must already have looked at this, and taken technical decisions based on ensuring that they do avoid Patents A, B,...

So we're not asking them to do any more work, or disclose something they've paid someone else to do. Just explain their technical decisions.


It could start with Xiph giving us a technical explanation or their view on a certain patent. Then someone comes in and expresses a different view on this patent and how it applies to vorbis. Then Xiph will say they had lawyers looking at this but that the detailed infromation cannot be disclosed (e.g. since they paid for it).

This would not clear up the patent situation, instead we'd then have even more reason to believe they have something to hide, since there exist different opinions on their explanation.

Quote
The cynical observer would wonder why it's not in the source code! If you're trying to comment the code, it's not unreasonable that you might say "here, we do x, y, then z (not the more obvious a, b, then c because this would infringe US pat no. x,xxx,xxx)" - otherwise you risk someone "tweaking" the code at a later stage, to use the better(?) and more obvious(?) technique, which also happens to be patented!


Although IMO this would not solve the patent problems, I still like this approach very much. The example you give is perfect.

Quote
They must already have internal documentation for this.

Surely?

If not, they have no case for using the words "Patent Free". If "Patent Free" just means "we haven't applied for a patent for any of this stuff", then I've got a great idea for a patent free video disc format!

If they have worked all these issues through internally, then maybe they should make their reasoning public. Unless they aren't confident of their reasoning?


There is still value in claiming that they did not patent any of the underlying ideas. By publishing these ideas (together with code) no other party can patent it, thereby making it 'patent-free'.

Unfortunately, the other part of 'patent-free' involves making sure that other exsting patents do not apply to the code, and that's where the problem is.

Vorbis development, status & patent issues

Reply #152 –
Quote
You can have something that is completely 100% patent free. To say otherwise is nonsense!

You are right, of course; something can be patent free. But from what I have seen, only the court system can legally determine if a technology infrings on a patent that is still in effect.

I suppose that Xiph could go to court, seeking a declaration on non-infrigment, but barring that they are speaking their (perhaps correct) opinion when they claim that Ogg Vorbis is patent free.

Vorbis development, status & patent issues

Reply #153 –
Quote
I suppose that Xiph could go to court, seeking a declaration on non-infrigment, but barring that they are speaking their (perhaps correct) opinion when they claim that Ogg Vorbis is patent free.

Nobody is expecting Xiph to go to court to prove something. All that is wanted is at least some kind of clarifications and a bit more detailed technical explanations how they in their opinion managed to circumvent patents, so that their advertizement of Vorbis being patent free codec could be considered valid in their opinion.

Currently there's absolutely nothing concrete, except some unofficial comments on some irc-channel that insist not infringing...
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #154 –
I've had a little chat with Monty about this, and to summarize:

1) He is certain that Vorbis is not infringing on any of the patents mentioned.

2) He clarified that in a claim like:

Quote
1. In a method of transmitting a signal including the steps of dividing the signal into successive, overlapping blocks by means of windows, converting the partial signals contained in the blocks into a spectrum by transformation utilizing window functions within the respective said windows such that the resultant of the window functions in the respective overlapping regions of successive blocks equals 1, subsequently, coding and transmitting the resulting spectra, receiving and decoding the transmitted coded spectra, converting the decoded spectra back into partial signals by transformation, and finally, joining the blocks containing the partial signals in an overlapping manner; the improvement comprising selecting the length of the respective window functions as a function of signal amplitude changes.


The only part that is relevant is after the 'the improvement comprising...', and only if the former part applies.

3) Xiph doesn't elaborate on the patent issues because:

a) They were told by their laywers not to do so.
b) They want to avoid situations like this:

FhG guy: 'Vorbis infringes patent X!'
Public: 'OMG OMG OMG'
Vorbis guy: 'Let me spend a few hours reading through the claims of X to explain why you're wrong'
....
Vorbis guy: 'No we dont. We do xyzzy and the patent only talks about yzzy'.
AAC guy: 'But Vorbis infringens patent Y!'
Public 'OMG OMG OMG'
Vorbis guy: 'Let me spend a few hours (again) reading through the claims of Y to explain why you're wrong'
...
Vorbis guy: 'No we dont. We don't do xzxz at all but zxxz instead, and the patent only applies if you also do zyzy, whereas vorbis does zzyy.'
FhG guy: 'Vorbis infringes patent Z!'
Public: 'OMG OMG OMG'
Vorbis guy: 'AAAAAAAAAAARGH CANT GET ANY WORK DONE'

and rather have:

FhG guy: 'Vorbis infringes patent X!'
Public: 'OMG OMG OMG'
Vorbis guy: 'No.'
Public: 'Oh it was FUD'
AAC guy: 'Vorbis infringes patent Y!'
Public: 'OMG OMG OMG'
Vorbis guy: 'No.'
Public: 'Oh, it was just more FUD'
FhG guy: 'Vorbis infringes Y!'
Vorbis guy: 'Wrong again. You are stupid. Just sue us next time'
Public: 'OMG OMG OMG'

Basically, elaborating on the patent issues would not stop FUD from being used against Vorbis, might be used against it to confuse matters even more, and might make it harder for Xiph to effectively defend should it ever come to a real lawsuit.

Note, this is not an official Xiph position, it's my interepretation of the chat with Monty.

Vorbis development, status & patent issues

Reply #155 –
The problem is that already in this thread some people went to the specifics, and anybody can see that things are unclear and there's nothing "FUD" in that...
If discussion like this thread is marked FUD by Xiph, well then I don't know what to say, except that I disagree and can only wonder why Xiph doesn't want to discuss about this at HA. After all, there are no FhG guys posting here, as far as I can see, so it wouldn't be too bad situation for Xiph..
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #156 –
You did not read my post carefully enough, notably the very first sentence. According to Xiph, the relevant claims in the patents mentioned here do no apply to Vorbis. They don't want to elaborate each and every time someone _thinks_ something infringes upon Vorbis, because it is very easy to come up with something that sounds somewhat like what Vorbis might possible be doing but it takes a lot of time to actually throughly investigate each claim, let alone explain why it is not relevant.

Monty can and did point out why it didn't apply, but he is not interested in doing a throughout research each and every time someone thinks something might possibly be a problem for Vorbis. They don't have the manpower for that, and there's no point in doing it anyway.

The point about FUD is that it is much easier to spread the suspicion that Vorbis is infringing rather than to actually throughly check it out.

Vorbis development, status & patent issues

Reply #157 –
Quote
You did not read my post carefully enough, notably the very first sentence. According to Xiph, the relevant claims in the patents mentioned here do no apply to Vorbis.

Which boils down to whether one trusts their "patent-free"-claim and research on the subject or not.

Vorbis development, status & patent issues

Reply #158 –
Quote
You did not read my post carefully enough, notably the very first sentence. According to Xiph, the relevant claims in the patents mentioned here do no apply to Vorbis. They don't want to elaborate each and every time someone _thinks_ something infringes upon Vorbis, because it is very easy to come up with something that sounds somewhat like what Vorbis might possible be doing but it takes a lot of time to actually throughly investigate each claim, let alone explain why it is not relevant.

This has been covered in this thread already..

Quote
Monty can and did point out why it didn't apply, but he is not interested in doing a throughout research each and every time someone thinks something might possibly be a problem for Vorbis. They don't have the manpower for that, and there's no point in doing it anyway.
This is the first time, that I know a patent has been inspected more specifically publicly. So imo this "each and everytime" doesn't apply here.. There isn't even first time yet...
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #159 –
Quote
This is the first time, that I know a patent has been inspected more specifically. So imo this "each and everytime" doesn't apply here.. There isn't even first time yet...

You don't know everything JohnV

We went through this at least once before with menno.

Vorbis development, status & patent issues

Reply #160 –
Quote
Quote
You did not read my post carefully enough, notably the very first sentence. According to Xiph, the relevant claims in the patents mentioned here do no apply to Vorbis.

Which boils down to whether one trusts their "patent-free"-claim and research on the subject or not.

This is very much true.

Edit: Unless you pay your own lawyers.

Vorbis development, status & patent issues

Reply #161 –
Quote
Quote
This is the first time, that I know a patent has been inspected more specifically. So imo this "each and everytime" doesn't apply here.. There isn't even first time yet...

You don't know everything JohnV

We went through this at least once before with menno.

Obviously you replied before you saw my editing..
"This is the first time, that I know a patent has been inspected more specifically publicly."
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #162 –
Quote
So imo this "each and everytime" doesn't apply here..

Why not? It's not Xiph's job to do FhG's homework, obviously

Vorbis development, status & patent issues

Reply #163 –
Quote
Quote
So imo this "each and everytime" doesn't apply here..

Why not? It's not Xiph's job to do FhG's homework, obviously 

We are discussing about this HA thread here. This is HA thread, and HA is covering technical and scientific discussion about lossy audio codecs. There's very specific public discussion in this thread (especially in the last 2 thread pages) regarding an unclear patent and Vorbis. This hasn't happened here or anywhere else publicly before as far as I know. I don't know what FhG even has to do anymore with this HA discussion. FhG is not even here discussing...

Even more, Xiph people including you and Carsten Haese have expressed that they'd like to have an official explanation to the specific issue covered in this HA thread.. But reading the unofficial Monty's reply (via you), I fear that it starts to be more and more unlikely.
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #164 –
Quote
Quote
So imo this "each and everytime" doesn't apply here..

Why not? It's not Xiph's job to do FhG's homework, obviously

I think this is really sidestepping the larger issue here.  Fhg was never really asking for any clarification from the Vorbis guys, they simply made a statement about a possibility of infringement.  That possibility was brought up in this thread and basically exposed a problem with the current situation, namely that there is no clarification of "patent free" for 3rd parties, and that there is no way for 3rd parties to verify the reliability of this claim.  Further, it's questionable as to how extensive Xiph's own knowledge of their non-infringement is given this scenario, at least as far as convincing others goes.

Whether or not Xiph thinks any of this matters or whether or not they should waste their time with the situation, the fact still remains that the "patent free" claim is, at least as far as I understand, basically the prime selling point of Vorbis.  If this claim isn't treated in a more rigorous and methodical manner (as far as the public is concerned), whether Xiph has the resources to do this or not, it would definitely seem to create a conflict of interest.

Vorbis development, status & patent issues

Reply #165 –
Quote
The problem is that already in this thread some people went to the specifics, and anybody can see that things are unclear and there's nothing "FUD" in that...
If discussion like this thread is marked FUD by Xiph, well then I don't know what to say, except that I disagree and can only wonder why Xiph doesn't want to discuss about this at HA. After all, there are no FhG guys posting here, as far as I can see, so it wouldn't be too bad situation for Xiph..

OK, both c_haese and I have wasted way too much time on this thread. Now, you'd like Monty to spend his time on HA instead of Vorbis? Seriously, if we need to really spend time trying to convince someone of "patent non-infringement", it will be a judge, not you.

Vorbis development, status & patent issues

Reply #166 –
Quote
Quote
The problem is that already in this thread some people went to the specifics, and anybody can see that things are unclear and there's nothing "FUD" in that...
If discussion like this thread is marked FUD by Xiph, well then I don't know what to say, except that I disagree and can only wonder why Xiph doesn't want to discuss about this at HA. After all, there are no FhG guys posting here, as far as I can see, so it wouldn't be too bad situation for Xiph..

OK, both c_haese and I have wasted way too much time on this thread. Now, you'd like Monty to spend his time on HA instead of Vorbis? Seriously, if we need to really spend time trying to convince someone of "patent non-infringement", it will be a judge, not you.

You're missing the point here.  It's not about HA, it's about the clarity of the patent situation.  It's not about spending time on HA, it's about spending time clarifying the patent situation (which everyone seems to take for granted), or pointing to information already gathered so it is clarified to others.

If none of this is going to happen, or if it's simply "not important", then we're back to square one.  Vorbis is "patent free", only so far as you trust unofficial statements made by the developers.  For any further clarity, each individual themselves would have to put forth significant information to do an independent patent search.  Not a very ideal situation it would seem...

It seems that everyone keeps going back to the "impossibility" of being absolute sure of the validity of the "patent free" claim, but I don't even think that's necessary.  It would definitely go a long way towards improving the situation if there were perhaps a document with detailed information on how Vorbis manages to avoid some of the more prominent patents that Ivan was talking about.  "All or nothing" aren't the only two options here -- something at all would be better than what we have now.

Oh, and just in case you're wondering why HA might be making such a big deal out of this itself, the reason is because Vorbis is often touted as an alternative if you want free licensing and assurance not having to deal with patents.  The problem is that this assurance seems rather false or shallow now.  As far as HA is concerned, I would like to see this matter clarified if people are going to continue asserting this "fact" on the boards here.

Vorbis development, status & patent issues

Reply #167 –
Quote
Whether or not Xiph thinks any of this matters or whether or not they should waste their time with the situation, the fact still remains that the "patent free" claim is, at least as far as I understand, basically the prime selling point of Vorbis.  If this claim isn't treated in a more rigorous and methodical manner (as far as the public is concerned), whether Xiph has the resources to do this or not, it would definitely seem to create a conflict of interest.

Could be.
But I get the impression that Xiph.org is using their time and effort in a strictly prioritized sense, only divulging patent issues where they might deem it a practical issue/problem (convincing hardware-manufactures for support and issues where it  hinders exposure of the format) and leaving the end-users to do the discussion regarding FUD (or not) on their own. In a "time is money" cost-beneficial analysis kinda way.

Call me naive, but I do believe that they had it covered before they decided on using such a strong term as "patent-free" in the meaning "not infringing".

If Ivan and Petracci discovers something tangible things might alter, but until then it is just FUD. The discussion is sound, but we cannot expect Xiph to correct everything.

Vorbis development, status & patent issues

Reply #168 –
Quote
Vorbis is "patent free", only so far as you trust unofficial statements made by the developers. 


Yes.

Quote
For any further clarity, each individual themselves would have to put forth significant information to do an independent patent search.  Not a very ideal situation it would seem...


Yes.

Quote
It would definitely go a long way towards improving the situation if there were perhaps a document with detailed information on how Vorbis manages to avoid some of the more prominent patents that Ivan was talking about.  "All or nothing" aren't the only two options here -- something at all would be better than what we have now.


As explained above, Xiph thinks this is very much contrary to their interests and would only worsen the situation for them.

Quote
Oh, and just in case you're wondering why HA might be making such a big deal out of this itself, the reason is because Vorbis is often touted as an alternative if you want free licensing and assurance not having to deal with patents. The problem is that this assurance seems rather false or shallow now. As far as HA is concerned, I would like to see this matter clarified if people are going to continue asserting this "fact" on the boards here.


It seems to be basically a matter of faith.

Vorbis development, status & patent issues

Reply #169 –
Quote
OK, both c_haese and I have wasted way too much time on this thread. Now, you'd like Monty to spend his time on HA instead of Vorbis? Seriously, if we need to really spend time trying to convince someone of "patent non-infringement", it will be a judge, not you.

You'll never arrive at the need to convince a judge if you can't convince me.

I have (game) developers asking me whether they should use Vorbis and whether using it won't get them sued.

The answer I usually give is 'Unreal uses it, they'd be a bigger target than you are.'

Convincing? Don't think so.

Vorbis development, status & patent issues

Reply #170 –
Quote
As explained above, Xiph thinks this is very much contrary to their interests and would only worsen the situation for them.

Quote

Oh, and just in case you're wondering why HA might be making such a big deal out of this itself, the reason is because Vorbis is often touted as an alternative if you want free licensing and assurance not having to deal with patents. The problem is that this assurance seems rather false or shallow now. As far as HA is concerned, I would like to see this matter clarified if people are going to continue asserting this "fact" on the boards here.


It seems to be basically a matter of faith.

Right.. anybody checked if they registered the "Church of Xiph" name yet?

Just joking, but it's weird that Xiph, a non-profit organization, developing an open source, free software, completely refuses from open discussion on a pretty much neutral environment, where every party has a chance to express their opinion.

Well, it's certainly their right, if that is what they want...
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #171 –
Quote
We are discussing about this HA thread here. This is HA thread, and HA is covering technical and scientific discussion about lossy audio codecs. There's very specific public discussion in this thread (especially in the last 2 thread pages) regarding an unclear patent and Vorbis. This hasn't happened here or anywhere else publicly before as far as I know. I don't know what FhG even has to do anymore with this HA discussion. FhG is not even here discussing...

The issue is that the best judges of this are the people that wrote the codec and the people that invented the methods in question. Neither of us are lawyers, neither of us don't know anything about the technical nitty-gritty, most of us don't even have detailed knowledge of how a patent claim works.

We can go on for 10 more pages about whether Vorbis infringes or not, I don't think anything useful will come of it as long as the above doesn't apply.

Apparently Xiph's lawyers told them to STFU, and they're STFU. I doubt they will change this unless there's pressing reasons not to, and I haven't seen any so far.

Vorbis development, status & patent issues

Reply #172 –
Quote
We can go on for 10 more pages about whether Vorbis infringes or not, I don't think anything useful will come of it as long as the above doesn't apply.

So you don't think an official explanation from Xiph is any useful or clarifies the situation, or explains their position on this matter.. ok..
Juha Laaksonheimo

Vorbis development, status & patent issues

Reply #173 –
Time for a touch of reality here. The major number of contributors to this thread are in search of a statement, with supporting evidence, from Xiph as to how Ogg Vorbis does not infringe certain patents.

I'm sorry, but this is all back-to-front. The situation regarding Patent Law is that where infringement is claimed by a Patent Holder, the burden of proof rests with the Patent Holder to prove infringement, it does not rest with the defendant to prove in what way they do not infringe.

Xiph has taken expert legal opinion that leaves them in the position of confidence that they are not infringing any current patents. Whether Fraunhoffer, or anyone else, has asked a vague question regarding possible infringement is irrelevant to the situation. Unless and until they, or anyone else, files a suit alleging infringement there will be no direct comment from Xiph, and nor should there be. Would it be prudent for Xiph to make detailed statements regarding these matters? No, of course it wouldn't. Patent Law, almost probably more than any other, is a minefield and the last thing Xiph should be doing is providing ammunition to anyone who may consider instigating an action. I'm afraid it is very naive to expect otherwise.

So, in short, any statements other than those already made are very unlikely to be made. Indeed, it would be extremely foolish for Xiph to make any detailed statement.

This is not what people want to hear, but, as I said, it's time for a reality check.

Vorbis development, status & patent issues

Reply #174 –
Thank you for a good summary of the situation.

The problem is that many people want to know better why exactly Xiph thinks they are not infringing. But it is not in Xiph's (current) interest to elaborate on that.