[Vorbis] open source vs closed source compression

[Eric Dantan Rzewnicki]

On Wed, Mar 07, 2007 at 11:29:09AM +0000, Ivo Emanuel Gonçalves wrote:
> On 3/7/07, Ulrich Windl <ulrich.windl at rz.uni-regensburg.de> wrote:
> >Ignoring whether the claim is valid or not, an open-source compression 
> >does in no
> >way guarantee that software patents are not involved. It's the software 
> >patents,
> >not the "binary only". However many companies think if they publish 
> >binary-only,
> >noone can see where their code came from, not to talk about intellectual 
> >rights,
> >etc.
> >The other thing is that most users of MP3 don't care about owning a 
> >license for
> >using it, and some people even think that when using LAME (a binary using 
> >LAME)
> >does not require them to have an MP3 license. I also think that most 
> >vendors of
> >MP3 players in the far east do not care much about paying license fees for 
> >MP3.
> What Ulrich said is quite true.  In the big scheme of things, nobody
> cares about patents and free software, and sometimes not even if the
> quality is good.  While marketing Vorbis, we have to understand that
> what makes it a good format is not its patent status, but its high
> fidelity and multi-use qualities.
> 
> In the end of the day, only lawyers care about patents; the average
> idio--person cares only about marketing hype.
> 
> >Long Live open source.
> 
> You mean free software.  Be sure to keep that in mind.  Open source is
> a term used by Eric S. Raymond, another hack like Steve Gibson, and it
> means a totally different thing.  Ever heard of that free as in beer
> and speech?  You can't claim that about "open source".  That's why
> using the word free software is always better.

As I understand it, BSD licensing is not strictly Free Software. In a
simplistic view the difference between OSS and Free Software boils down
to "may" vs. "must" with regard to the rights you pass on when you
redistribute modified code. GPL says you must pass on the rights. BSD
says you may, but don't have to as long as you credit the original
authors, which is why it is used for xiph codecs. 

As I understand it RMS agrees with this for the purpose of encouraging
adoption of xiph codecs. Hardware developers may be under NDAs with the
suppliers of their various chips. If they need to modify code they get
from xiph.org to make it work with such hardware they will not be able
to release the modifications due to the NDA.

I welcome any corrections or additions to this summary as I'm sure I
don't have it completely correct. I think it is an important point that
must be understood in promoting this stuff and I want to make sure I
understand it fully myself.

-Eric Rz.