Working Toward a Patent-Agnostic Open Source License

Glyn Moody writes "Are there ever circumstances when software patents that require payment might be permitted by an open source license? That's the question posed by a new license that is being submitted to the Open Source Initiative (OSI) for review. The MPEG Working Group wants to release a reference implementation of the new MPEG eXtensible Middleware (MXM) standard as open source, but it also wants to be able to sell patent licenses. If it can't, it might not make the implementation open source; but if it does, it might undermine the fight against software patent proliferation."

Software patents.

[Z00L00K]

It's just a way of trying to make software patents more valid.

I would say that any patent that lacks hardware (chemical compound or physical device) wouldn't be valid.

What does it mean?

[BradleyUffner]

Does this mean that the source code would be freely available, but that you couldn't use it without paying them? I skimmed the artical and linked pages, but can't figure out what this would actually mean.

Make the reader open, the writer licensed

[Anonymous Coward]

Everybody should be able to use, say, H.264 and AAC if your software only does playback.

To record/create files, you would need a license.

Re:Software patents.

[kansas1051]

I would say that any patent that lacks hardware (chemical compound or physical device) wouldn't be valid.

Few patents target software per se. Most "software patents" actually claim (cover) computing hardware that implements some allegedly novel/non-obvious functionality. In any event, why allow someone to patent an application-specific integrated circuit that performs new function X but not a FGPA configured via a HDL that performs new function X?

This Doesn't Make a Whole Lot of Sense

[eldavojohn]

What is the motive behind this new license? To cherry pick a few of the ideals of Open Source Software (OSS)?

It sounds like, from the license, that they want the openness of many eyes reviewing and improving the code with derivative work while at the same time licensing that idea to other companies. Which, frankly, I cannot comprehend as any company would just opt for the open source community code to integrate into their product than pay the patent holder to roll their own. Or are they planning on charging you for the "open source" version like normal software? If so, how is that any difference from a commercial license modified so that you receive the code to review with the product?

I mean, I'm happy for them to do whatever they feel like ... I don't mind more licenses and I think the MPL was a step in the right direction but not perfect. Either way, observers can be sure of one thing, there are at least some aspects of open source that appeal very much to a lot of people. It will be interesting to see what results from this endeavor.

Re:Software patents.

[Samschnooks]

In any event, why allow someone to patent an application-specific integrated circuit that performs new function X but not a FGPA configured via a HDL that performs new function X?

Then a mystery writer could patent a method of committing a crime and the method of solving the crime. Royalties from the mafia and from the police! Woohoo!

Re:MIT/BSD licenses

[Wesley Felter]

According to the thread, many people in OSI believe that MIT/BSD licenses do (implicitly) grant patent rights. This was a surprise to me.

Re:Software patents.

[h4rr4r]

Then why is does their patent on a system comprising a specially-programmed computer to perform the steps of x, and/or an apparatus comprising a processor, memory element, and means for performing x, prevent me from makeing my own?

It would be like patenting the claw hammer and suggesting that you now own the patent on all devices that can pound nails into walls.

Re:MIT/BSD licenses

[funkatron]

It's pretty hard to read

Redistribution and use in source and binary forms, with or without modification, are permitted

or

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so

as being anything other than statements giving permission to use the software. Even if the software is patented wouldn't distributing it with a license saying that people can use it mean that you were giving permission to use it and were therefore allowing the use of your patent.

Re:Software patents.

[Schraegstrichpunkt]

The problem isn't really patents per se, but patent infringement lawsuits and the resulting court orders. It would make more sense to continue to issue patents as usual, but to amend patent law so that it's unenforceable under certain circumstances deemed problematic (e.g. mere software). Similarly, you could have different durations of enforceability depending on the type of infringement.

Patent law can be useful, but it needs to be far more carefully controlled than it is right now.

Re:Software patents.

[Zordak]

Well, I guess to the extent that murdering somebody is considered "transforming matter" under Bilski, you could patent the murder process. But I don't know that there's a good way to claim solving a crime under Bilski, unless you're claiming some specific technique like DNA analysis, which I'm sure was patented at some point.