Introducing myself and my project

[Pete Bleackley]

Hello.

I'm Pete Bleackley. I'm a Research Engineer, working for the BBC. My job is
to implement the Dirac Open Source video compression system in VHDL. I've
just completed modules for arithmetic coding and decoding, which can be
found at

http://www.opencores.org/projects.cgi/web/dirac/overview

This code is available under the MPL, GPL and LGPL.

I'd be interested to hear from any other Open Source hardware hackers who
might be interested in what I'm doing.

Pete

 

[burn.sir]

Hello.

I'm Pete Bleackley. I'm a Research Engineer, working for the BBC. My job is
to implement the Dirac Open Source video compression system in VHDL. I've
just completed modules for arithmetic coding and decoding, which can be
found at

http://www.opencores.org/projects.cgi/web/dirac/overview

This code is available under the MPL, GPL and LGPL.

I'd be interested to hear from any other Open Source hardware hackers who
might be interested in what I'm doing.

Pete

kudos to BBC for this work.

I understand that the code is under xxGL, but is the codec itself under
any patents?

burns

 

[Pete Bleackley]

kudos to BBC for this work.

Thank you.

I understand that the code is under xxGL, but is the codec itself under
any patents?

The BBC has some purely defensive patents on some parts of the algorithm,
mainly because that seems to be the only effective way to stop anybody else
from patenting the techniques. Where possible we've used techniques that
have a long history in academic literature that we can cite as prior art
against hostile patent claims. With regard to our own patents, one of the
provisions of the MPL is that it requires all contributors to licence
applicable patents free of charge, irrevocably. This was one of our reasons
for licensing under MPL, since we knew that this was an issue we'd have to
deal with.

Thank you for asking - it's a frequently asked question, and an important
one for me to answer.

Pete

 

[rickman]

The BBC has some purely defensive patents on some parts of the algorithm,
mainly because that seems to be the only effective way to stop anybody else
from patenting the techniques. Where possible we've used techniques that
have a long history in academic literature that we can cite as prior art
against hostile patent claims. With regard to our own patents, one of the
provisions of the MPL is that it requires all contributors to licence
applicable patents free of charge, irrevocably. This was one of our reasons
for licensing under MPL, since we knew that this was an issue we'd have to
deal with.

Thank you for asking - it's a frequently asked question, and an important
one for me to answer.

Maybe I don't understand patents. I thought a patent had to be an
original work and not in the public domain. If you provided public
info on a patent more than a year before you apply your patent can be
denied. So I guess someone would have a year after you made it public
to steal the patent... oh well!