3 ms·
Let's be honest here. Some of the blame goes back to the FFMPEG folks. They choose the LGPL which is certainly a less known/understood license and they tack on
by jsz0 17y ago
Let's be honest here. Some of the blame goes back to the FFMPEG folks. They choose the LGPL which is certainly a less known/understood license and they tack on a bunch of extra requirements. I'm looking at their license policy page and I don't see any obvious link to a non-English version. If you're not a license expert and you happen to lookup LGPL on the wiki for some guidance you get:
The main difference between the GPL and the LGPL is that the latter can be linked to (in the case of a library, 'used by') a non-(L)GPLed program, and regardless of whether it is free software or proprietary software.[1] This non-(L)GPLed program can then be distributed under any chosen terms if it is not a derivative work. If it is a derivative work, then the terms must allow "modification for the customer's own use and reverse engineering for debugging such modifications."
I'm not arguing that the FFMPEG folks don't deserve respect and adherence to their license terms. Just seems like they aren't really helping themselves here either. If you make a situation confusing you cannot be angry when people get confused. My personal feeling is if you want to do "free software" you should remember the whole free part. The GPL and it's variants are too overtly political for my tastes. The BSD license seems a bit more intellectually honest.
- Locke1689 17y agoActually, they're being a little hypocritical too. They are very protective of their license, but I know for a fact that a number of developers work in the United States where large portions of the codec library violate US patent law. Can't have it both ways guys...
- noonespecial 17y agoActually, you can and should. Its the difference between patents and copyrights. The (L)GPL pertains to the copyright on the actual code and the laws are (in comparison) fairly universal and well understood. Patents on the other hand pertain to a method of doing something (separate from actual code) and are far from certain and vary widely between jurisdictions. The outcomes are decided in local courts based on a mix of precedent and who can spend the most on lawyers. In short, copyrights and patents have nothing whatsoever to do with each other.
- Locke1689 17y agoWell, I wouldn't say they don't have anything to do with each other. They are both forms of intellectual property law. While I would agree that patents are often subject to different regulations and burdens of proof, your assumption that all patent cases are flimsy is without basis. First, you mistake software patents for many traditional patents. For example, patents on drugs are routinely upheld in US courts. You are correct that software patents can be nebulous. However, in the case of codec patents, like that of H.264, the new MPEG video standard, SCOTUS and patent cases have routinely upheld the validity of similar software patents. For an overview of relevant statutes, I would point to the USPTO documents, which outline the standards that most software patents are viewed under (by the USPTO and many practicing US patent lawyers)[1]. While In re Bilski[2], recently granted a writ, may present a new view on some of the more spurious (and business-oriented) software patents, it seems that SCOTUS is more than willing to uphold the patents in practical implementation cases[3][4][5]. If this were a discussion based purely on ideology then upholding copyright but ignoring patent rights may be defensible, however to suggest that they will bring legal action to defend their copyright but to ignore or refuse that they should be held accountable in patent litigation is a tad hypocritical. [1] http://www.uspto.gov/web/offices/pac/mpep/documents/2100.htm http://www.uspto.gov/web/offices/pac/mpep/documents/2100.htm [2] http://www.cafc.uscourts.gov/opinions/07-1130.pdf http://www.cafc.uscourts.gov/opinions/07-1130.pdf [3] http://www.oyez.org/cases/1980-1989/1980/1980_79_1112/ http://www.oyez.org/cases/1980-1989/1980/1980_79_1112/ [4] http://digital-law-online.info/cases/32PQ2D1031.htm http://digital-law-online.info/cases/32PQ2D1031.htm [5] http://digital-law-online.info/cases/31PQ2D1545.htm http://digital-law-online.info/cases/31PQ2D1545.htm
- noonespecial 17y agoYou can infringe on a patent without any knowledge of said patent (by now, in software, I'd say this is the norm). Its a bit harder to do that when it comes to copyright! Plus ffmpeg is an international project. The SCOTUS (and its rulings on patents), while important, is still SCOTUS. That was all I was getting at.
- DarkShikari 17y agoPatents do not apply to the developers, they apply to the users of the software; this is generally the model that open source works under in general (see, for example, LAME). In many jurisdictions, users can and do pay MPEG-LA licensing fees when using ffmpeg; no "infringement" takes place. I have worked for companies that have used ffmpeg and x264 in their systems; this is extremely standard procedure. Also, most of the developers are based in Europe, where software patents are much more dubious than in the US.
- jai 17y agoNo not really, most of the devs are european...
- astrange 17y ago> They choose the LGPL which is certainly a less known/understood license and they tack on a bunch of extra requirements. If it had extra requirements, it wouldn't be using the LGPL (and I don't think it would even be compatible). Those are just restatements of parts of the license; you can see which sections just by reading it, as it's quite short.