LGPL VP6 decoder released

[dragongodz]

i also fail to see how having a lawyer email someone is in any way "kind". especially when the email has "you know we can sue you if you dont comply" type remarks in it.

[rjamorim]

CEC: Like you said, we all believe that the code was originally written by On2 either way. And I'm feeling sorry for On2 too, but if Zeitoun Padli created libvp62 by decompiling that .jar, it could be leagal (in many countries incl. the US), and if so, libvp62 does not belong to On2.

http://en.wikipedia.org/wiki/Reverse_engineering

In other words, simply put, On2 owns that binary ("machine") and sorce code ("step by step recipe: how to make the machine"), but they do not own the algorithm ("idea") behind it (in some freakin' countries where software patents are legal, this gets yet another condition: "unless it is patented").

It is quite possible that Zeitoun Padli actually did decompile that jar. If so, libvp62 is legal. We have a written permission to use libvp62 under LGPL anyway, so, unless otherwise proven, it is assumed to be LGPL'ed. On2's lawyer keeps saying "No, it's On2's. No it's On2's." but that doesn't change anything. What if, for instance, I said "CEC, what you just posted is my property. You are infringing my copyright. Delete your post right away, or I have to, reluctantly, sue you"...? You would be like "Huh?!" no? You would think for a while and then say "Show me a proof." That's normal. If I could let you delete your post by just saying "It's mine, not yours." that would be terrible. In that sense that lawyer is doing a really poor job, but it can't be helped, because On2's devs probebly don't believe him enough to show him the source code. And that lawyer doesn't know how to use diff anyway.

Someone would have to give him a nice gui like WinMerge and explain how to use it. It's just a waste of time to involve non-geek lawyers here. On2Tech could just PM rjamorim, and everything will be settled and that'd cost 0$ and just a few minutes.

The disturbing fact is, now Zeitoun Padli can say "I decompiled that jar to make libvp62" even tho actually he or she didn't. It is still possible too that the code was simply leaked or stolen. The situation being like above, probably On2 couldn't prove what they first insisted ("the code was solen") even if they did their best. Maybe they don't know the fact themselves either. Maybe the code was not stolen after all, but just was RE'ed.

So my point is, it is only natural for rjamorim to say "Show me a proof" but in this case because of the existence of that jar, rjamorim is actually asking what is technically impossible. Some kind of compromise is needed.

That's very interesting. Thanks for the long explanation.

Now let's just wait and see if On2 themselves ever get in touch with me...

@rjamorim

http://multimedia.cx/eggs/index.php?p=12

That's interesting. Thanks.

Regards;

Roberto.

[rjamorim]

i also fail to see how having a lawyer email someone is in any way "kind". especially when the email has "you know we can sue you if you dont comply" type remarks in it.

These guys expect whoever receives these e-mails to be scared shitless of C&D letters and promptly comply. From an efficiency point of view, it's great, since most people have very little clue about law and their rights. But from a PR point of view, it's a disaster...

[Doom9]

they have customers that are paying for that decoderAnd those will have to continue paying license fees - just like a commercial entity can't take the XviD decoder and stop paying MPEG-4 licensing fees. With MPEG-4 we actually have the 50k excemption but that's not applicable for On2.. so anybody using a VP62 decoder in a commercial environment is likely to be eligible for license fees. Hence, they lose nothing.

[Liisachan]

@Doom9
Ok, then... is this thing patented? That's the question.
If no, one can just decompile that .jar that is lawfully accessible, to ask someone in "Cleanroom" to re-make VP62dec.cpp, which is 100% leagal AND ethical.
http://en.wikipedia.org/wiki/Clean_room_design
http://en.wikipedia.org/wiki/Cleanroom_%28Software_engineering%29

[rjamorim]

@Doom9
Ok, then... is this thing patented? That's the question.
If no, one can just decompile that .jar that is lawfully accessible, to ask someone in "Cleanroom" to re-make VP62dec.cpp, which is 100% leagal AND ethical.

Yes, but even if it wasn't patented, On2 could force people already licensing the decoder to keep licensing it even with open source alternatives.

All the companies I mentioned in my e-mail (Adobe, Autodesk, AOL) license not only the decoder, but also the encoder. So, they could say "You want our encoder? Then you have to license the decoder too. We don't license the encoder alone".

[foxyshadis]

On2 might have to drag them through a long court case first though - see the Burst.com court cases. Of course for the encoder licensees it's different, until an encoder pops up anyway.

[On2Tech]

This is how I see the situation. First let me apologize in advance for the length of this note but I have a lot that I want to say.

The views expressed by LATEXXX, GodofaGap are pretty much in line with my understanding of copyright law. Specifically, that binaries ARE protected by copyright law in the same way that source code or the text of a novel are protected.

It is also my understanding, that just as translation of a novel from say English to French or indeed English to French and then back to English, does not circumvent the original author’s copyright, so compilation of source code and de-compilation, (both forms of translation) do not circumvent copyright. If this were not true then all copyright protection would be meaningless.

The argument that taking a stolen copy of the code and doing a few global search and replaces on variable names is illegal but it is somehow fine to do a trivial de-compilation, or for that matter take code written in one high level language and convert it to another, and then distribute it as “your own work”, seems to me ridiculous and both morally and intellectually reprehensible.

There follow some more common sense examples that I feel illustrate the fact that copyright does not so easily just ”disappear”.

Consider a company that took some open source software and in a jurisdiction not covered by the GPL, compiled it and then decompiled it. If that company took the decompiled code, changed the license, claimed copyright for themselves and distributed it, saying that they were no longer in any way encumbered by the original license or copyright, then the open source community would quite rightly be outraged.

Yet another pertinent example. If you were to take a camera into a movie theatre and film it this does not circumvent the studio’s copyright. Similarly if you were to take a movie encoded in MPEG-2, decode it and then recode into another format such as VP7 or H.264, this would not circumvent the original copyright. You would not then be able to claim that the re-encoded movie was your “own original work” and that you were entitled to redistribute it in any way you see fit without let or hindrance! Changing the format or way in which something is encoded, or translating something, simply DOES NOT make the original copyright disappear. As long as it can be shown to be clearly derivative or substantially similar, the original copyright will remain in force.

I think that there is also a fallacy in the arguments that are being presented on the board re. the significance of the lack of an EULA in the JAVA code. An act of translation itself (including compilation, de-compilation or conversion from one high level language to another), may not necessarily be an offense, depending on the precise details of any license and the jurisdiction, but this does not in ANY WAY imply a right to distribute the results. It is true that a translated work may itself be copyrighted, but that does not undermine a case for infringement of the original copyright and people on Doom9 have already commented on striking similarities between the decompiled Java and the LIBVP62 C++ code!

Putting the legal issues aside for a moment, (as disagreements regarding the interpretation of the law cannot unfortunately be resolved without actually taking this to court), I and others here at On2 put in a lot of work on VP6 and when I look at the LIBVP62 code it is absolutely clear to me that it is NOT an “original work”. Somebody has just taken our code (by one means or another) dressed it up a bit, and is now claiming it as their own work! Some on the Doom9 board seem to think that defending and promoting this behaviour is somehow taking the moral high ground! I don’t mind saying that this annoys me big time.

A further interesting point was raised by VLADA regarding and LATEXXX regarding the situation for binaries where there is no EULA. As I see it, EULA’s cut both ways. They may impose specific limitations or constraints but they also grant rights that would not exist by default (e.g. because of copyright legislation). Without an EULA, any distribution of, copying of (including creation of any derivative work) and potentially even USE of a binary, may constitute an infringement of copyright.

I think that Rjamorim is right therefore, when he observes that his recent tactic of distributing the VP6 .JAR file together with tools and instructions on how to decompile it, was extremely unwise. There is absolutely no dispute regarding the authorship of the JAR file and it is clearly not the case that Rarewares has been granted permission by On2 to distribute it. Therefore he has unequivocally infringed our copyright.

Rjamorim previously asked if he was breaking any forum rules. Just as we are not an inherently litigious company and have always been reluctant to resort to the courts (expensive and damaging to everyone concerned), I have no desire to see anyone “struck” or to stifle legitimate debate. However, I would refer you forum rule 6 which in my opinion is pretty unambiguous.

“6) No warez, cracks, serials or illegally obtained copyrighted content! Links to content of a questionable nature, asking for, offering, or asking for help/helping to process such content in any way or form is not tolerated.”

Even if he refuses to accept our assertions that the LIBVP62 code is not legal (or even that it is “QUESTIONABLE”), there can be no doubt that the unauthorized redistribution of one of our binaries did infringe. Therefore, yes, he has in my view broken Doom9 forum rules.

As Rjamorim has shown, simply decompiling other peoples binaries be they JAR files or other compiled binaries is not an inherently clever thing that should be applauded as an academic achievement by the open source community. There are tools out there that do this at the click of a button. Quite frankly a trained monkey could do it. Creating those tools might be pretty clever as is the creation of good compilers, but using them is not. Neither is converting from JAVA to C++ particularly difficult or clever, because the languages are very similar in syntax and structure.

I accept that the JAVA binary is out there and cannot be put back into the bag so to speak (as are our other binaries) but I believe that both it and any work that can clearly be shown to be derivative, are still protected by copyright law, and I personally do not think it is unreasonable of On2 to assert its rights if it feels there is a threat to its commercial interests. I just cannot understand, why Rjamorim is continuing in this highly confrontational vein. Unless of course he has some ulterior motive, and is determined to back us into a corner where we have no choice but to take legal action.

“If they come bitching, I'll follow master Anakata's steps and tell them to go sodomize themselves with retractable batons ”

Is this kind of thing really helpful or necessary? Doesn’t the forum also have rules against bad mouthing people? Perhaps he thinks it is legitimate because it is directed at our lawyers rather than at another member of the forum, but I think that is a pretty lame excuse given the courteous if firm tone of their letters.

Many on the forum, have asked why we have not supplied proof. We have, what I believe is a compelling case that copyright has been infringed and also that licensing terms may well have been breached. In addition we have patents pending which when granted will almost certainly be infringed. Surely it is obvious that the place for presenting proof per sé is a courtroom (if it comes to that), and that any public disclosure of our evidence in advance might act to forewarn and forearm any adversary and hence prejudice our position.

In any case, even if we convinced Rjamorim, what is to stop someone else putting the stuff up and demanding proof, then another and another?

In regard to the question of loss / damage that this “illegal” code may cause us: I obviously cannot give details of our business plans in a public forum. However, the decoder source provides information needed to develop an encoder (and indeed this was the stated plan when it was first released). This could obviously impact our sales of encoder products and SDKs. Secondly, though a Flash decoder binary is available for certain platforms, we consider that there are very real and significant ongoing opportunities for revenue in respect of ports to other devices, both in terms of NRE and through source code licenses etc.

Others have asked why On2Tech did not chime in earlier or more often. I must confess that this has been a source of great frustration to me. I am not usually backward in coming forward so to speak, but please understand my position, namely that because of the possibility of litigation at some point in the future, it would not be appropriate of me to sound off on the forum without any reference to our lawyers and members of the management team. FoxyShadis actually put it quite well if slightly condescendingly :rolleyes:

“He's an engineer, he's probably on rather unsteady footing just talking about it. Statements on here could be used in discovery or court, like email, at a later date. (I'm not sure what the exact legal grounds of forum postings are, but they have to be a lot firmer than heresay.) If you want to rant to someone, call up On2's legal, or directors, or someone else higher up in the food chain.”

I think most, if not all of the members of the forum would now accept that the ONLY credible explanations for how this code came into being are either that it was based on stolen code or on decompiled code (perhaps the Java as this is probably the easiest to decompile). I would argue that both are illegal, so proving which of these two actually took place is irrelevant. Even if you wish to try and deny the legal argument, surely none of you can seriously give credit to this individual for creating an ORIGINAL WORK.

I sincerely hope that we can put this to bed once and for all as it is a huge distraction to me and to other members of the engineering team.

That’s it from me ….

THE END


P.S. IMHO, de-compilation is to genuine reverse engineering what paint by numbers is to a Picasso.

P.P.S. © On2Tech !

[Sirber]

That’s it from me ….

THE ENDNo need to hang yourself ;)

[On2Tech]

No need to hang yourself ;)

No need to do it myself certainly... no doubt there will be a queue forming shortly.

[foxyshadis]

Sorry, I didn't mean to be condescending, whatever your position is. I just know that many companies have policies where employees are allowed to discuss certain topics and not others (Financials, touchy legal matters, and upcoming product specifics in particular), at least without consultation or deferring to others, and it seemed that people were missing that.

[On2Tech]

Sorry, I didn't mean to be condescending, whatever your position is. I just know that many companies have policies where employees are allowed to discuss certain topics and not others (Financials, touchy legal matters, and upcoming product specifics in particular), at least without consultation or deferring to others, and it seemed that people were missing that.

I realize no offence was intended and none was taken. In any case the gist of what you said then and above is quite right.

Thanks for your concern.

[soresu]

Originally posted by On2Tech

P.S. IMHO, de-compilation is to genuine reverse engineering what paint by numbers is to a Picasso.

Surely aside from a binary/hex view of the decoders output, de-compilation is the surest first step to reverse-engineering?
Sorry if this sounds stupid... im an artist, not an engineer! :D

[On2Tech]

Surely aside from a binary/hex view of the decoders output, de-compilation is the surest first step to reverse-engineering?
Sorry if this sounds stupid... im an artist, not an engineer! :D

It was intended as a “slightly” flippant comment :o

The serious point I was making though is that to work out something by black box analysis can be very difficult (in the case of trying to work out something like the VP6 bitstream from analysis of content, nigh on impossible). On the other hand to simply run a de-compiler tool and copy what someone else has done does not strike me as clever at all.

But what would I know. I am a strictly FORWARD engineer!

[GodofaGap]

Perhaps for EU people this (http://www.jenkins-ip.com/serv/serv_6.htm#a9) is an interesting read. Amongst other things, it says that according to the EU copyright directive decompilation is not a valid way of reverse engineering (or at least analysis, but what is the difference then?) and that the reverse engineering must be indispensable. I do not know if the RE of VP6 code was indispensable.

But of course, different countries, different laws.

[dragongodz]

Many on the forum, have asked why we have not supplied proof. We have, what I believe is a compelling case that copyright has been infringed and also that licensing terms may well have been breached.
you have ? exactly where, when and to whom have you given any sort of proof to substantiate what you are saying ? without something concrete everything, including discussion in this thread, is suposition at best.

Surely it is obvious that the place for presenting proof per sé is a courtroom (if it comes to that), and that any public disclosure of our evidence in advance might act to forewarn and forearm any adversary and hence prejudice our position.
that quite frankley is a copout. its already been said before that it doesnt have to be a public disclosure. privatly showing SOME proof to a few indepandant and respected third parties, who could then back up your claims, would work aswell.

In any case, even if we convinced Rjamorim, what is to stop someone else putting the stuff up and demanding proof, then another and another?
nothing. however if you provided the proof to select individuals ,as i said above, you could get the support so that if anybody did try and post the sources in or at respectable forums or sites would be told not to etc. now i am guessing that isnt enough for you. consider however that even huge companies like MS cant stop dodgy versions of their software from appearing on the net and what exactly do you expect ? do you really think having lawyers harrasing people who do and have complied with legalities when shown it was real is going to get you better results ? is that really going to stop the code being somewhere on the net ?

Some on the Doom9 board seem to think that defending and promoting this behaviour is somehow taking the moral high ground! I don’t mind saying that this annoys me big time.
but have you considered its On2's behaviour that has actually annoyed people and not taking some moral high ground as you say ?

[Liisachan]

compilation of source code and de-compilation, (both forms of translation) do not circumvent copyright.

Wrong. Especially if de-compiled thing is re-implemented in "clean room". I'm not talking about my personal opinion, just saying the fact already established in key cases.

"Under United States law, reverse engineering a patented item can be infringement; however, if the artifact or process is protected by trade secrets instead of by a patent, then reverse engineering the artifact or process is lawful as long as the artifact or process is obtained legitimately." ... "This process is sometimes termed Reverse Code Engineering or RCE 3. As an example, decompilation of binaries for the Java platform" ...

- Reverse engineering (http://en.wikipedia.org/wiki/Reverse_engineering): From Wikipedia, the free encyclopedia

"Clean room design is useful as a defense against copyright and trade secret infringement because it relies on independent invention. However, because independent invention is not a defense against patents, clean room designs typically cannot be used to circumvent patent restrictions."

- Clean room design (http://en.wikipedia.org/wiki/Clean_room_design) From Wikipedia, the free encyclopedia

That's why you should obfuscate the java. If what you are naively beliving was true, you wouldn't need obfuscater, but if you don't like that "the decoder source provides information needed to develop an encoder" and you "accept that the JAVA binary is out there" then you do have to obfuscate the code. (This part is not a legal duty. I'm just talking about my personal view.)

As Rjamorim has shown, simply decompiling other peoples binaries be they JAR files or other compiled binaries is not an inherently clever thing that should be applauded as an academic achievement by the open source community. He didn't stop that because of such a reason; simply he doesn't have any right to re-distribute that jar. I can see excellent engineers are sometimes not that good at laws, fancying over unrelated metaphors, just like lawers are bad at programming. Whether or not a monky could do the same, is unrelated. You are basically saying "It is too easy; therefore it should be illegal," which is not correct.

I feel great sympathy for you, On2Tech, even though your discussion is incorrect (Ask your lawer if you don't believe me). You are a nice person. And the reason you couldn't make a comment soon is understandable too.

Rjamorim, now, how about being a bit more kind to On2Tech?

[rjamorim]

Rjamorim, now, how about being a bit more kind to On2Tech?

Elaborate, please

[Liisachan]

Already told you, "they could ask you to help them" blah blah. ..."Some kind of compromise is needed" blah blah.

On2Tech said "I sincerely hope that we can put this to bed once and for all as it is a huge distraction to me and to other members of the engineering team." which must be true.

Personally I am more interested in supporting new non-MPEG non-MS codecs, or in spending time for something more productive, than in wasting more time here. From what was posted by On2Tech, I assume they can't give you any solid evidence even if they want to, as they don't know the truth either. Nothing productive is coming out of demanding something impossible. If "they didn't show me the proof" is the only reason for you not to do what is asked, that is.

As a side note, that I'm feeling sorry for On2 doesn't mean RE is illegal. RE is generally legal, no matter how one feels. (On2Tech has a wrong idea that RE should be something very difficult.) But that's not related here.

[maven]

It is also my understanding, that just as translation of a novel from say English to French or indeed English to French and then back to English, does not circumvent the original author’s copyright, so compilation of source code and de-compilation, (both forms of translation) do not circumvent copyright. If this were not true then all copyright protection would be meaningless.

Merely as an observation, but what if someone read the novel, and then wrote his own novel with a very similar story, maybe even in a different language, but without any verbatim copying? There was a very recent case of this where Dan Brown was accused of this (for "The DaVinci Code" IIRC).

[GodofaGap]

that quite frankley is a copout. its already been said before that it doesnt have to be a public disclosure. privatly showing SOME proof to a few indepandant and respected third parties, who could then back up your claims, would work aswell.
It's not a copout, and what you suggest does not work at all. Why should On2 even trust these people? On2 has nothing to prove to them. Court is the only place where On2 will have to substantiate their claims.

nothing. however if you provided the proof to select individuals ,as i said above, you could get the support so that if anybody did try and post the sources in or at respectable forums or sites would be told not to etc.
Don't you realize how meaningless this sounds? You expect On2 to trust on other's people good-will, while they already feel it has been violated. These people you suggest can't stop anyone from re-publishing this code. They just can't. So why should On2 even try to do what you say?

consider however that even huge companies like MS cant stop dodgy versions of their software from appearing on the net and what exactly do you expect ?
Well "MS and other huge companies" are in quite a different position than On2, this comparison doesn't hold at all. But actually what you are saying is:

"We can decompile your code, we can violate your copyright, we can violate your EULA's, and guess what On2? There's not a SINGLE thing you can do about it. Haha!"


but have you considered its On2's behaviour that has actually annoyed people and not taking some moral high ground as you say ?
Don't anyone here even dare speak of "moral high ground".

[dragongodz]

It's not a copout, and what you suggest does not work at all. Why should On2 even trust these people?
hmm these same people On2 came to to help with beta testing ? your right of course, we are all just scum who you really have to keep your eye on.(sarcasm for those that missed it)

Court is the only place where On2 will have to substantiate their claims.
if a person or company decides to say they dont really want to sue people then they should expect to provide some proof of their claims at the same time. otherwise it amounts to no more than threats and FUD.

These people you suggest can't stop anyone from re-publishing this code. They just can't. So why should On2 even try to do what you say?

because it could stop it from being publicised on respectable forums(like this) or mailing lists(such as ffmpeg) etc. if On2 is not interested in it not being spread through these avenues without resorting to court then why are they posting on them ? why are they advertising their latest codec ? why have they asked for testers in the past ?

what you are saying is:

"We can decompile your code, we can violate your copyright, we can violate your EULA's, and guess what On2? There's not a SINGLE thing you can do about it. Haha!"
may i suggest you dont ever try and put words in my mouth. i said exactly what i meant. you obviously seem to think this threatening people with lawyers is going to stop it from being out there somewhere. its not. all it is doing is pissing off people on respectable sites.

Don't anyone here even dare speak of "moral high ground".
jeez what exactly is your problem ? On2Tech was the one who said some peoples actions are based on a supposed moral high ground. i asked him if he has considered actions could be based on other. i really dont care what you think about it, your posts speak volumes.

[Doom9]

Without an EULA, any distribution of, copying of (including creation of any derivative work) and potentially even USE of a binary, may constitute an infringement of copyright.That simply cannot be true. I've been both buyer and creator of individual software and we always set the license restrictions in the contract that was signed prior to the start of development. So unless otherwise mentioned, customer X using application A of mine, can keep on using the application indefinitely and for an unlimited amount of users, as long as the contract is not violated. You don't need an EULA to grant permission to use an application.
And many EULA prohibitions, especially when it comes to reverse engineering, or void in many countries (and I fully support that.. it's not right if a company can impose arbitrary restrictions.. and copyright law covers unauthorized use of an application just fine).

If I can sum up your argument, basically it's that libvp62 was created by decompiling the jar file and translating the code from java to c++. Unless my mind fails me, I see nothing in copyright law to directly cover this. The decompilation shouldn't infringe upon anything, but what comes thereafter can be problematic. Say, after looking at the decompiled sources, somebody writes a spec of the format, and the same or another party writes a decoder based on that specification. That constitutes clean room reverse engineering. However, if you copy parts of that code, then it's copyright infringement (just as is using parts of an open source work in a commercial application might violate the open source license, at least if that license is the GPL and the sources of the commercial application are not made available upon request).

So it comes down to how libvp62 was created. If the spec approach was taken, then there's no problem. If the code was read in one language and translated to another, then likely that's in violation of copyright law. If the code was read in one language, and translated not basically line for line, but as maven described "in your own words", then that can hardly by copyright infringement. Just as the Da Vinci Code case was turned down, I'm quite confident that there are many commercial programs out there that were inspired by open source software .. but inspiration isn't forbidden (even though the software lobby tries to patent mere ideas and workflows so that in the end one cannot write software without unwillingly and unknowingly make use of certain patents).

[GodofaGap]

hmm these same people On2 came to to help with beta testing ? your right of course, we are all just scum who you really have to keep your eye on.(sarcasm for those that missed it)
I am sure you remember the Ateme beta-test debacle. No, not all are scum but not all are saints either. And on the internet it is very hard to tell who is which.

if a person or company decides to say they don't really want to sue people then they should expect to provide some proof of their claims at the same time.
Sure, but not under these conditions, not anymore. I don't know where rjamorim got the truemotion.jar file from, but if it originates from On2 (I wonder from where else), it would appear he is violating their copyright.


because it could stop it from being publicised on respectable forums(like this) or mailing lists(such as ffmpeg) etc. if On2 is not interested in it not being spread through these avenues without resorting to court then why are they posting on them ? why are they advertising their latest codec ? why have they asked for testers in the past ?
Well, that appeared all to have happened before they were allegedly 'stolen' (in whatever way you want interpret that) from, didn't it? I haven't seen any respect in any direction from then on.


may i suggest you dont ever try and put words in my mouth. i said exactly what i meant.
May I suggest you to rephrase your words then? You clearly make the suggestion that it is futile for On2 to pursue this issue, and that they should accept the spreading of this code as a fact of life. Just like other companies (e.g. MS and other huge ones) did.

you obviously seem to think this threatening people with lawyers is going to stop it from being out there somewhere. its not. all it is doing is pissing off people on respectable sites.
And you obviously seem to think that it is quite incomprehensible that On2 seeks for legal council in this matter. That it is quite incomprehensible that On2 doesn't want to make mistakes with disclosing source (in public or private) that can possibly hurt their business even more in the future. On2 is not trying to win a popularity quiz here. Now you may not like this, but you really need to do a reality check and understand that On2 simply cannot rely on 'the community' to solve this issue. It just doesn't have the authority to do that.

I have read the letters posted by rjamorim: and at one point this (http://forum.doom9.org/showpost.php?p=816575&postcount=134) was written.
Now I do not know if you will voluntarily remove the VP6.2 code from your website, but I hope you will immediately. To the extent there is any way to prove our issues without jeopardizing On2's proprietary position, we are willing to do so, but first the code must be removed now.
Why wasn't this opportunity taken? Why was the chance to find out the real intentions of On2 omitted? Did everyone just miss it?

jeez what exactly is your problem ? On2Tech was the one who said some peoples actions are based on a supposed moral high ground. i asked him if he has considered actions could be based on other. i really dont care what you think about it, your posts speak volumes.
I will tell you what I think about it anyway. As you long as you don't get any further than being childish and say "Well, they started first!", you have no right to talk about moral high ground, respect or any other words of these nature. You just don't.

[On2Tech]

LIISACHAN: I am really pleased that you draw attention to these definitions because far from contradicting my position I think they re-enforce it!

Regarding “Clean Room”: The first paragraph talks about the use of clean room reverse engineering as a strategy that may circumvent copyright, BUT the second paragraph defines it. I will not quote too much directly for fear of copyright infringement :D and you can check for yourself, but the key thing is that you MUST be able to demonstrate that the implementation was not contaminated by ANY knowledge of the proprietary techniques used.

How can someone claim this if they have simply decompiled code and copied it or done a translation into another language. This precisely is NOT a clean room design.

It is also clear to me that just claiming that something is a clean room implementation is not a get out of jail free card for copyright infringers. You have to be able to back up the claim.

Even their definition of reverse engineering has an interesting phrase namely "without actually copying ANYTHING from the original". But, some on the boards have themselves observed that there are chucks of the LibVp62 C++ code that are almost identical to the decompiled Java (variable names aside)... In fact, in many situations porting decompiled Java to C++ would require little more than cut and paste.

I do not think that there is even a vaguely credible case that this code is a genuine "clean room" implementation, created without any reference to or copying from our code (be that physically stolen code or code that was obtained by de-compilation of one of our binary objects). Is anyone even suggesting that this is the case?

Regarding checking with a lawyer: I would be the first to admit that I am not a lawyer and that my common sense understanding of the law might well be no better than the average lawyers understanding of software. For this reason I did check with them before posting (I would probably have got into deep water if I had not done so). Basically, they agreed that if you can show that something is directly derivative or substantially similar and hence obviously NOT a clean room implementation, then copyright will apply. They also agreed that my examples were basically sound.

There are a number of other interesting things that I have read in Wickepedia today having first followed LIISACHAN ‘s link, though I doubt this can be considered to be a definitive guide to the law. In talking about derivative works it makes clear that where a work is based on or contains elements of a previous work, it may be deemed a “derivative work” or “new version” and be entitled to its own copyright. However, this is ONLY true if the original copyright owner gives permission… otherwise it is simply considered to be a COPY. Translations and even DRAMATIZATIONS (where clearly the words will not stay the same and there is some re-telling / interpretation of a story) are quoted as examples.

To address MAVEN’s point. Dan Brown wrote a story based on some ideas contained in what was claimed to be a work of non-fiction. One expert I heard talking on the subject said that if the original had been a work of fiction and they had not claimed it as an academic work, the ruling might have gone the other way.

Why… because just re-telling something that is identifiably the same basic story in your own words is not enough to get around copyright (see above re. dramatizations). By the way, for a video decoder, just how useful would “similar story” be? One bit wrong and all hell breaks loose.

@ Dragongodz.

I think a superfluous stray comma may have misled you here. It should read:

“We have what I believe is a compelling case that copyright has been infringed and also that licensing terms may well have been breached.”

According to my (and our IP lawyers) interpretation of the law, the community has already furnished itself with evidence that the LibVP62 code at the very least infringes copyright, namely its similarity with the decompiled JAVA.

Some of you may disagree with this interpretation, but I do not see how we can be expected to resolve that without resort to the courts, which NONE of us want and some would still dispute. We will just have to agree to differ. I am not threatened by people disagreeing with me, nor am I inclined to easily take offence, (though I have a sneaking suspicion that “VideoMixer” may have intended his last post to be slightly offensive ;) ).

It is very easy to say with hindsight that we should have handled things differently, but we have not been in this position before and were in uncertain waters. We did not and do not want to litigate unnecessarily, but we DO want to protect our assets (and ultimately our jobs :( ).

For the record, I have a huge amount of respect for the open source community and the good works that they do and our actions absolutely 100% are NOT and never have been intended to be an attack on that community.

On2Tech

[dragongodz]

No, not all are scum but not all are saints either. And on the internet it is very hard to tell who is which.
you may not be able to be sure with a lot but there are some pretty promanint people who it would be hard to call discreditable. take doom9(the person) as just 1 example. he has been sent early betas of software for years and never leaked anything. is it hard to tell he would be trustworthy enough ?

Sure, but not under these conditions, not anymore.
why ? On2's claims have not really changed, ok shifted about a bit to try and cover RE but thats all.

I don't know where rjamorim got the truemotion.jar file from, but if it originates from On2 (I wonder from where else), it would appear he is violating their copyright.
if you read his posts you will see he already removed the .jar on his own because on thinking about it he thought it would be of real questionable legality.

Well, that appeared all to have happened before they were allegedly 'stolen' (in whatever way you want interpret that) from, didn't it? I haven't seen any respect in any direction from then on.
the respect started to dry up when On2 decided to use lawyers and provide no proof. until then nobody had any real problems with them. if they had come to this forum, for example, and simply stated that they believed the source code was stolen and would provide some limited proof to a couple of trustworthy people to back that up this would never have got this far here. so you can not complain about On2 not getting a lot of respect back when its they who treated people with no respect first.

You clearly make the suggestion that it is futile for On2 to pursue this issue
please reread. i do not say it is futile. i infact say it can be stopped from spreading around reputable sites etc.

and that they should accept the spreading of this code as a fact of life. Just like other companies (e.g. MS and other huge ones) did.
accept that there will still probably be places on the internet where the source code is ? yes, thats a reality. however limiting those places by not having big and reputable sites have it is a different thing that is possible.

Now you may not like this, but you really need to do a reality check and understand that On2 simply cannot rely on 'the community' to solve this issue. It just doesn't have the authority to do that.
hmm i think you need to do a reality check aswell. of course the community can not stop it everywere, i already said that. it certainly CAN help to reduce it though. alienating it though is not going to help anyones cause.

Why wasn't this opportunity taken? Why was the chance to find out the real intentions of On2 omitted? Did everyone just miss it?

actually rjamorim did remove the code at 1 point. however if you keep reading all the emails the lawyer has also said theres basically no way anyone is going to see the proof/source outside of court. so who missed the opportunity again ?

As you long as you don't get any further than being childish and say "Well, they started first!", you have no right to talk about moral high ground, respect or any other words of these nature. You just don't.
HAHAHA now thats rich. i am curious though as to where you think you have ANY right to tell me what i can or can not talk about. "You just don't." thats got to be one of the weakest reasons for anything. thanks for the laugh though. :)

[Liisachan]

@On2Tech Your discussion is wrong: "clean room" is just a 120% sufficient condition, not a necessary condition at all. But let's stop wasting our time... all you want is to let Rareware stop hosting that src code, right? If so, I already PMed rjamorim about that like 24 hours ago, and he/she was like "Yeah, I guess..." It's not like I'm his/her friend, but I figured that would be the fastest if you were not going to PM him/her yourself.

Like you said, Rarewares can't show any decisive evidence that it was RE'ed. Whereas On2 can't show any decisive evidence that it was not RE'ed. So I guess this will get nowhere without some kind of compromise from both sides. You said you wouldn't mind binary files being around (.class files in that .jar, that .jar itself, or anything). Or maybe you rather meant "it can't be helped" but anyway, it's a sort of compromise you made. So I guess now is when Rarewares should be a bit kinder to you.

[GodofaGap]

if you read his posts you will see he already removed the .jar on his own because on thinking about it he thought it would be of real questionable legality.
http://www.rarewares.org/files/others/VP6_src_legal.zip

This link shouldn't be working then?


actually rjamorim did remove the code at 1 point. however if you keep reading all the emails the lawyer has also said theres basically no way anyone is going to see the proof/source outside of court. so who missed the opportunity again ?
You are just being funny now right? He removed the source code and replaced it with the jar file, and you would consider that complying with the request? Erm... ok.

relevant post (http://forum.doom9.org/showpost.php?p=817652&postcount=160)

HAHAHA now thats rich. i am curious though as to where you think you have ANY right to tell me what i can or can not talk about. "You just don't." thats got to be one of the weakest reasons for anything. thanks for the laugh though. :)
Yeah, whatever.

[On2Tech]

DragonGodz

I recognize that you were previously involved in beta testing for us and respect you for that, so I am truly sorry that you now appear to hold On2 in such unreserved contempt.

Perhaps if we were able to turn back the clocks we would have handled things differently from the start. But that unfortunately is not an option.

If we were now to provide you or anyone else with an original code extract, for example, which shows just how strikingly similar the LibVp62 code is; how would that help? People will just point out that the same can be said of the decompiled Java. I have presented an argument that it does not matter. That it clearly was not a clean room implementation and matching up code either to the original or decompiled code is more than enough to demonstrate infringement of copyright. The latter has already been done by some members of this forum so where is the need for further proof?

If you will not accept this argument, then I can’t see that there is much more I can say. If you think that our lawyers would be willing to release, or allow me to release, any other more specific evidence outside of a courtroom, then I think that this is naive.

I certainly would not suggest that you have acted out of malice in this matter. Also, if you feel that On2 have treated you or others on this forum with disrespect then I would like to reassure you that this was not intended and offer my apologies.

But I also want you to know that I think we have been judged a bit harshly and that there has at times been a lack of balance. It seems that we are being assumed by many to be in the wrong unless we can prove otherwise. Our code has been ripped off and suddenly we are the villains! To say that we are in the wrong simply because our lawyers got involved is I think a bit disingenuous. Rather it would have been surprising for a publicly traded company if they had not.


On2Tech

[On2Tech]

@On2Tech Your discussion is wrong: "clean room" is just a 120% sufficient condition, not a necessary condition at all. But let's stop wasting our time... all you want is to let Rareware stop hosting that src code, right? If so, I already PMed rjamorim about that like 24 hours ago, and he/she was like "Yeah, I guess..." It's not like I'm his/her friend, but I figured that would be the fastest if you were not going to PM him/her yourself.

Like you said, Rarewares can't show any decisive evidence that it was RE'ed. Whereas On2 can't show any decisive evidence that it was not RE'ed. So I guess this will get nowhere without some kind of compromise from both sides. You said you wouldn't mind binary files being around (.class files in that .jar, that .jar itself, or anything). Or maybe you rather meant "it can't be helped" but anyway, it's a sort of compromise you made. So I guess now is when Rarewares should be a bit kinder to you.

I certainly don't want to argue the point with you and I appreciate that you are trying help us find some middle ground... for which I am grateful.

[Sharktooth]

Well, i already proposed a solution for both sides that would bring some advantages to both on2 and the OSS community.
Releasing the decoder source will make *nix and mac users able to playback vp6.x contents and will help the spreading of the vp6.x format.
That is a GREAT advantage for users and for YOU (on2).
Yes, you can guess the bitstream and specs from the decoder sources... but again it will be a winning situation for on2 since the vp6 format will be spread even more...

[Sirber]

would be spred even more with a CLI encoder instead of crap VFW.

[dragongodz]

Yeah, whatever.
you bring absolutly nothing useful to this thread and are just wasting my time. so consider yourself ignored by me.

that you now appear to hold On2 in such unreserved contempt.
not at all. if you look back at my posts you will see i have said i have lost respect because of the course of action taken when it could have been handled differently. so no its not contempt at all and there has been no malice from me either.

If we were now to provide you or anyone else with an original code extract, for example, which shows just how strikingly similar the LibVp62 code is; how would that help? People will just point out that the same can be said of the decompiled Java.
yes NOW they could. however i mentioned that you could do a limited showing of some proof, even just a little, before that arguement even started.

The latter has already been done by some members of this forum so where is the need for further proof?
and this was as you say done by other members here and not by On2. its unfortunate that its been left to others instead of On2 clearing this up without the need for this mess.

If you think that our lawyers would be willing to release, or allow me to release,
sorry but thats plain wrong. the lawyers are hired by you(On2) and can only advise you. they can not tell you what you can and can not release.

But I also want you to know that I think we have been judged a bit harshly and that there has at times been a lack of balance.
maybe but on the same hand you must also admit that neither you nor those emails from your lawyer has really been the most help to clear this up.

It seems that we are being assumed by many to be in the wrong unless we can prove otherwise. Our code has been ripped off and suddenly we are the villains!
you know thats not true. it was On2 that did the accusing and in a rather unfreindly way. as i said before, that sort of action is more likely to annoy and get peoples backs up.

To say that we are in the wrong simply because our lawyers got involved is I think a bit disingenuous.
no, to treat a community that you have had long time contact with by using lawyers, especially when it could have been cleared up here and other places without them, is what was criticised, by me atleast. if you read back you will see at no point do i say you are lying but at no point do i say you are telling the truth. the simple fact is you gave nothing to go by and sent lawyers to talk for you. are you really surprised some people didnt like that ?

[GodofaGap]

you bring absolutly nothing useful to this thread and are just wasting my time. so consider yourself ignored by me.
You haven't had any contribution to this thread besides mentioning "FUD" a few dozen times, and some fairy-tale about how On2 should have done this all without lawyers. You do not live in the real world. There are too many problems with your approach, yet you fail to see that. So resorting to personal attacks is the best you can do?

[Sirber]

Guys guys! This is a forum, this is internet, it is virtual ;)

[GodofaGap]

Yes, you are right. I apologize for my behaviour.

My views are so different from some of the others particiapting in this thread that I doubt any further discussion would have brought us closer anyway. I will refrain from posting anymore in this thread therefor. In reality, it is On2's turn to make a move now, whatever it might be.

Sorry, again.

[On2Tech]

not at all. if you look back at my posts you will see i have said i have lost respect because of the course of action taken when it could have been handled differently. so no its not contempt at all and there has been no malice from me either.


I have already accepted that you intend no malice and also that with the benefit of hindsight we might have handled things on this board differently.


sorry but thats plain wrong. the lawyers are hired by you(On2) and can only advise you. they can not tell you what you can and can not release.

Sadly I think that you misjudge my (On2Tech's) influence if you think that I am in a position to overrule the advice of our lawyers of many years standing.


you know thats not true. it was On2 that did the accusing and in a rather unfreindly way. as i said before, that sort of action is more likely to annoy and get peoples backs up.

no, to treat a community that you have had long time contact with by using lawyers, especially when it could have been cleared up here and other places without them, is what was criticised, by me atleast. if you read back you will see at no point do i say you are lying but at no point do i say you are telling the truth. the simple fact is you gave nothing to go by and sent lawyers to talk for you. are you really surprised some people didnt like that ?

I am sorry but I must have missed something here. I was not aware that our lawyers ever had any direct contact with this community (Doom9, have you received any emails or letters)? I know that they sent letters to all sites who published the code saying that it infringes our rights (which I am certain that it does) and asking them to take it down. Surely you would not expect me to handle that? None of these were particularly aggressive or threatening by legal standards. In all cases apart from one, these sites responded promptly. The most recent site to post the code just happens to be run by a member of this forum. He received emails in the same way as they others, though perhaps by this time our lawyers were getting more impatient. Even so I think their correspondence with Rjamorim has been broadly respectful. The fact that he published their letters and his replies to the forum should not be confused with us in some way setting our lawyers on the community.

Perhaps, I should have said more and said it earlier. Perhaps if I had done so, and had presented our case to you all on day one (actually I was travelling at the time so there would always have been at least a couple of days delay), things would have come out differently. But as I have explained, I am not always free to say what I want when I want, and we cannot now turn back the clock. So please, lets reflect on lessons learned and move on.

Now I would like to forget about this for a while and try and enjoy the weekend.

Respectfully

On2Tech

[dragongodz]

Sadly I think that you misjudge my (On2Tech's) influence if you think that I am in a position to overrule the advice of our lawyers of many years standing.
i clearly said On2 meaning the company and not yourself. so no i did not mean you had the power to do as you wished but rather the company that hired and pays the fees of the lawyers.

I was not aware that our lawyers ever had any direct contact with this community
to be clear i mean contact with this community(i dont just mean this forum) through its members and site ops(who can also be members) of course.

Now I would like to forget about this for a while and try and enjoy the weekend.

agreed. my last post and this one were more inline with trying to clear up some misconceptions you seem to have had or do still have. such as saying i know held On2 or yourself in utter contempt.
so have a good weekend. :)

[avih]

FWIW, I think the issue is rather simple.
Let's put the legal stuff aside here for a sec.

What we've got here is a code which was written by On2, and a copy of it which got circulating, something that On2 isn't interested in.

Whether or not it's legal to distribute it in some form or another is not really the important part IMHO. If one respects On2 then the code should be removed. If there's no respect for On2 then one should be ready to figh or hide or face possible consequences (fully legitimate decision BTW IMO).

Since we don't approve laws that seem unreasonable from our perspective (DCMA et al), we shouldn't look only at the legal perspective as it's different from place to place, and will eventually be decided upon the amount of money spent on lawyers.

Forget the law. They request that it gets removed. Do we respect that or not?

avih

ps.
I'd give On2 a big thumb up if they eventually (rather sooner than later though) decide to release the decoder as OSS, possibly with multiple licenses, such that they can still license it to 3rd parties which sell products, and allow free usage on OSS apps. Not as simple as it might sound, but there are precedences.

peace.

[smok3]

yeah, they should release the code under some oss licence, thats pretty much the only clean thing they can do right now (the 'stolen' code is allready in cirulation, so why allow the possibility that some half-broken decoder will come into works from that?), and yes, since they are the only code owners, there is no problem in multilicensing either (oss + commercial license is not a problem in that case afaik).

or should i say, learn from the big boys (since your such a tiny/small company i mean):
http://www.apple.com/macosx/bootcamp/

[Doom9]

Doom9, have you received any emails or lettersNo I have not. But then again I have not been hosting the material in question either and there's no working link to a libvp62 in this thread as far as my search goes. I have had a look at libvp62 and the results of jad but first of all it's a video codec so there's lots of maths to give me a headache (I've had more than my share at college and I'm extremely glad not having to worry about matrices, integrals, differentials, probability calculations and the likes on a daily basis anymore.. I never quite liked that stuff), and it's not like you can make any reasonable judgement without at least trying to understand how the code works.. and I just don't have the time and motivation to do so.

Since the cat's out of the bag either way, what's the harm in posting how libvp62 resembles On2's java code? Say you go to court, at the very latest at this point that information is going to leak anyway.. just as Groklaw is reporting everything on the SCO cases.

[rjamorim]

You guys are all nuts, but in a nice sort of way :D

Well, I decided to compromise. Not because On2 ever contacted me in reasonable terms (they didn't, all they did was to brandish the DMCA as some soft of Mighty Sword of Destruction +5), but because I discussed this a lot with Liisachan and we came to conclusions that seem to be the best to all parties. Liisachan is smart and reasonable. On2 are <censored> that lost whatever modicum of respect I had for them due to that old VP3 thing.

So, the compromise is: I'll stop hosting the sources but I'll keep hosting the directshow filter and demo decoder. The directshow filter is much more useful than sources for the vast majority of users out there anyway. And I'm considering linking to other binaries using libVP62 (I know of one being worked on as we speak) at RareWares as well.

Should be enough for everyone. If it isn't... tough luck. I have more valuable stuff to worry about anyway.

Best regards (to the good guys ;) )

Roberto.

Edit: these changes will only appear on RW tomorrow (sunday)

[Liisachan]

@On2Tech
I refrained from posting anything here to make sure you can have a good weekend, but since I've been named...
What I actually did was sending 2 PMs to rjamorim, costing a few minutes and like 10 minutes respectively. Rjamorim is giving me too much credit.

Here's 2 things I wanted to say to you:
(1) You said "I accept that the JAVA binary is out there and cannot be put back into the bag so to speak" which is true too. But accepting that .jar being around, is practically giving anyone the same (or the similar) thing than libvp62. Any good programmers (not me) can decompile it and re-implement it easily. So, how about considering what Sharktooth, smok3 and others suggested? If this goes on, there will be buggy VP62 decoders out there (you said that code contained a bug which was fixed later). Buggy decoders could unfairly damage the reputation of VP62. This already happened in the relationship between LQ Tremor decoder in ffdshow and Ogg Vorbis. I'm sure many users who use ffdshow's def Vorbis decoder without a 2nd thought assume that Vorbis is not so good. Which is really frustrating, because it's not Vorbis' fault...

(2) I'd like to support anyone who might create an alternative for AVC/VC-1. Many said they started disliking On2 because of this mess, but I won't. I once got unmotivated to support On2, but greatly changed my mind again after reading what you posted in person. I believe I'm not the only one feeling that way. This was an unfortunate accident after all.
On the other hand, On2 is responsible too, for not obfuscating the Java code (if that is the origin), or (if it's stolen by a hacker) for not keeping your "intellectual property" secure if it's that precious to you.

VP6 can't be an alternative but VP7 can. I hope On2 will be concentrating on VP7 (or maybe VP8?).

We too got a legal letter from lawyers hired by Microsoft when their Windows 2000 code was leaked and we quoted a few lines in our (technical and harmless) news article where one of us tried to judge if it was authentic or fake by testing an unknown feature of TaskMgr which could be read from the src. We complied with their request but felt really strange, because it was their fault not ours that the code was leaked, yet we were treated as if we were criminals. It was like, they let out a big fart, and then got angry saying "You unlawfully listened to it. It's our intellectual property and you have no rights to blah blah"
Um, anyway, maybe that traumatic experience was the reason I overreacted in a weird way here in this thread.

[falcon2000eg]

The question really should be what the benefits of distributing libvp62 under lgpl are, is it useful to OSS community .I don't think so .another reason is the moral issue making a codec is not like making word processing program, there is a lot of math and algorithms especially if it is non-mpeg .On2 made there own standards with their original ideas and they deserve some respect for their work (let the law things aside for while) why harm the company just because we can I cannot see the point nor understand why some one do that .
Sorry for my bad English that’s why I do not post a lot here but I follow most of the topics and this topic and the discussion in it is very rich.

[temporance]

rjamorim:

An idea for rarewares: how about a setting wiki page devoted to describing the vp6 format. People who have learnt from the libvp62 code could write a full description of the format. I would be willing to help out in a little way.

And why stop with a vp6 format wiki? The OS world at large would benefit from wiki pages describing "closed" formats.

The beauty of this idea is that, as far as I know, it would not be infringing anything as it could be written without copying anything -- it would be an original work, created by multiple authors who themselves own the copyright. A bit like wikipedia.

Open source developers could quite ligitimately use the wiki as a specification to create clean-room implementatinos of formats like vp6.

[Manao]

On2 made there own standards with their original ideas and they deserve some respect for their workAll codecs use the same principles. The hardest part of making a video codec is to avoid the overpatented mine field that concerns video algorithms.is it useful to OSS communityAsk the linux users.why harm the companyHow can that harm the company ? I still don't see why allowing the linux people to watch VP6 videos would harm On2. I don't think somebody in the OSS community would do a VP6 encoder when there are more efficient standards out there. Since the VP6 format is patented, no firm can take the libvp62 code, and sell it without paying licenses fee to On2.

In the end, imho, it can only promote the format, not harm it.

[Kopernikus]

rjamorim:

An idea for rarewares: how about a setting wiki page devoted to describing the vp6 format. People who have learnt from the libvp62 code could write a full description of the format. I would be willing to help out in a little way.

And why stop with a vp6 format wiki? The OS world at large would benefit from wiki pages describing "closed" formats.

The beauty of this idea is that, as far as I know, it would not be infringing anything as it could be written without copying anything -- it would be an original work, created by multiple authors who themselves own the copyright. A bit like wikipedia.

Open source developers could quite ligitimately use the wiki as a specification to create clean-room implementatinos of formats like vp6.


Have a look here:

http://wiki.multimedia.cx/index.php?title=Main_Page

[temporance]

Have a look here:

http://wiki.multimedia.cx/index.php?title=Main_Page

Perfect. Take a look at http://wiki.multimedia.cx/index.php?title=On2_VP6

Contributions more than welcome, if only prettification.

[frex]

sorry but any working links to the source code available now?

[Sirber]

not sure it's legal to repost it, but I'm sure many has them.