[Vorbis] patent issues with Vorbis

[Borphee]

As near as I can tell, this is basically just a re-hash of similar conversations that have happened many times over the years....  Nothing has changed and no new information is available...


From: "Jack Moffitt" <jack at xiph.org>


>> Using vorbis requires faith.  Faith is okay for individuals, but not really for most companies.
> 
> Using any software requires faith.  There is no way to "prove" you don't

True.  True.  I even said that patents weren't clear in general.

At that point, it depends a lot on how much you trust the company you are doing business with.  And whether you feel that company will support you if legal issues do arise later.

> infringe patents other than perhaps seeking a preliniary ruling.
> Problems can be avoided by careful planning and searching (both were
> done here at Xiph)

So people say...  However, Xiph has been rather closed about what steps were actually taken.

Basically, Xiph just says "It's patent free.  Trust us!"

If an  organization isn't willing to discuss openly and publicly why they feel some code or algorithm is patent free and why they feel their product is safe, or at least doesn't have a reputation for dealing with those situations if they later occur, then why should you trust them?

As I said... individuals (and non-profit groups, etc.) can do things (use vorbis etc.) on faith, but that companies can't.

> but patent claims can come from anywhere at any
> time.  Even microsoft has this problem now and then.  No such claims

That is true.

The difference between Microsoft and Xiph though, is that you can trust that Microsoft makes an effort to prevent patented stuff from getting into their products.  And when patent issues do occur, they take responsibility and try to protect their users.  (Hard to belive I'm saying good things about Microsoft...[shudder])

Is Xiph going to do the same thing?

Xiph has a reputation (justified or not) for being very closed about why they believe Vorbis is patent free.  Both the efforts they made during development and coding to avoid patents, and the research they did to discover if patents might aply to the code that had been written and algorithms that had finally been used.

When licensing mp3 or AAC or whatever, you do get a certain amount of protection.  The companies / organizations doing the licensing have made an effort to deal with all the patent issues and properly license each of them.  If something does happen and you end up in court, you can honestly say that because it was a reputable organization, you truely believed they had covered all the patents etc. and that the licensing fee you were paying them covered that, and that the contract you signed with them makes the fault / responsibility their problem, and not your.  And so on.

With Vorbis, there is no guarantee.  There are no details public about what steps were even taken that cause Xiph to say that Vorbis is patent free.  Everything has to be taken on faith.  You are on your own.

And Xiph is not a major company that has earned trust.  It's a small non-profit org that's refusing to discuss the details of certain subjects and refusing to stand behind any company that might get into trouble for using Vorbis.

Not every company is willing to take that risk.  Especially when it comes to encoders.

An invidual can take that risk.  But a lot of companies are going to get nervous about shipping products based on that kind of attitude.


> No such claims have ever been made against us.

That may be true.

But that could just as easily mean they don't consider Xiph and Vorbis worth the effort.

Vorbis is pretty small and limited and Xiph is non-profit.

Things could become very different if a company actually starts selling products based on it and money becomes involved.

In other words, individuals can get away with things that a company can't.  The same is true of small, limited non-profit organizations.


And don't forget... MPC was positively known to have at least one patent infringment (or so they say on HA.org) and nobody tried suing the developers or have it removed from the web, etc.

And LAME mp3 binaries are distributed through-out the world, in spite of patent infriment.  Sure, LAME devs only do source, but a lot of other places do indeed do binaires.  And they don't get bothered, even in countries where they should get bothered.

And people readily distribute mpeg-4 video (xvid, etc.) binaries without getting bothered.

So saying that you haven't been bothered yet really isn't a strong endorsement, considering Xiph's size and vorbis' market size.

You can bet things would be very different if (for example), Microsoft purchased Vorbis and made it the default audio format.  Companies would be showing up weekly with patent issues, and at least some of those would probably be valid.

 
>> (One link you refer to mentioned AOL considering adding Vobis encoding to Winamp.  I don't think that ever occured, did it...  So apparently they backed away too.  So, I would suspect (with no proof) that their team of lawyers either found something or decided it was just too shaky to mess with.  It does have decoding ability, of course.  But that's a simpler situation than encoding.)
> 

> A better theory would be that the licenseholders of the other codecs
> refused good licensing deals if they supported Vorbis.  It's hard for me

I would hardly call that "a better theory" considering they had already licensed those codecs.  Before they even considered adding Vorbis support.

In fact, AOL even had blanket licenses already for some, so they didn't have to ask.  In which case, vorbis wouldn't have even been mentioned.

No, I don't think anybody is truely going to believe that the other codec companies got together and conspired to prevent AOL from adding vorbis encoding.  At that time (and even now), vorbis is simply too minor for them to care.  With mp3 being the standard, and AAC being the next standard (in spite of Microsoft's best efforts), Vorbis is not going to gain a large following, even if Winamp had included vorbis encoding.

And it's a little unreasonable for a company to be willing to ship a product with Vorbis support simply because you say that the reason AOL didn't include a Vorbis encoder in Winamp was because the other codec licenseholders conspired against it.

> to believe that the lawyers were upset about patent issues.  All the

Who said anything about them being "upset".

I said that it might be possible that they either found some issue and they recommended against vorbis, or they found the situation to be murky enough that they recommended against it because it wasn't worth the effort to get waist deep in patent muck to try and decide if certain patents did infringe.  (ie: that they found enough possibilies and probabilities for it to not be worth the effort for such a minor codec.)

If AOL had included a vorbis encoder in Winamp, they would have been large enough to become a target, so basic due dilligence would have to be done before they would have allowed vorbis encoders to ship with it.  If Xiph had done that (and made it public), then AOL or such could point to that and say they believed it was safe because of that search.

But they had to do their own....  And Winamp still doesn't ship with Vorbis encoding.


> major game publishers ship Vorbis.  Several large music companies ship
> Vorbis.  Microsoft's own game companies ship Vorbis.  Several pieces of
> consumer electronics support Vorbis, and some even support FLAC.

Most of that is decoding, isn't it...

As I said... the situation is slightly easier for decoders than encoders.

There are a lot of potential patents that apply to an encoder that doesn't apply to a decoder.

And even still, some companies are indeed willing to take it on faith.  Just because some companies are willing to ship decoders doesn't mean a whole lot.  It doesn't clarify the patent situation in the slightest.
 
> You could ask the same "is it safe to use" question about any piece of
> code you incorporate or write yourself.  Software patents don't just
> affect Vorbis.  At least address the issue head on.

I'm pretty sure the guy didn't ask about general patent theory or whether such and such patent might infringe upon mp3 or AAC.  (And I did say that with patents: either it clearly infringes or you wait to be sued.)

He had already done enough research into Vorbis patents to show that he was already quite familiar with the entire concept.

He asked about Vorbis.  And I did address that head on....  It's a big unknown and you have to take it on faith because Xiph refuses to discuss why they believe it to be patent free, and that if something does happen, you are on your own because Xiph isn't going to stand behind you and support you.


If Xiph wishes to make an *official* public statement to the contrary, I'll certainly appologize.

But considering people have been asking for several years for Xiph to make an official public statment (beyond "it's patent free... Trust us."), somehow I doubt that's going to happen.



As I said at the top of this message... This discussion is simply the same discussion that has occured numerous times.  There is no new information in it.  There is no new argument for either side.