IETF patent policy (Was: [hybi] Process!)

On Tue, 2 Feb 2010, "Martin J. Dürst" wrote:
> 
> Well, I am not a lawyer, and the simple summary here may ignore all kinds of
> details, but essentially a lot of communities (standards organizations as well
> as open source communities) and the people and companies involved therein
> have, over the years, made painful experiences with the following scenario:
> 
> Person A: "wouldn't it be great if you spec did X."
> 
> Editor, WG: "not a bad idea, we'll take that"
> 
> (some years later)
> 
> Person A: "Hello everybody, I have a patent on X, please start paying."

Can you cite an example of that?

The only cases I've heard of are:

WG Member A: "wouldn't it be great if you spec did X."
Editor: "not a bad idea, we'll take that"
(some years later)
Person B, unrelated to anyone in the WG: "Hello everybody, I have a patent 
on X, please start paying."


> Patent disclosure obligations are designed to avoid the above scenario. 
> People participating in the IETF (and the organizations supporting them) 
> appreciate the fact that the IETF has patent disclosure obligations.

To be honest, I do not find the IETF IPR obligations to be especially 
compelling. First the concept of "reasonable and non-discriminatory" terms 
is almost meaningless in the face of free-license software (where the only 
acceptable patent license is a transferable or global RF one). Second, the 
IETF rules are about individuals and are only useful to the extent that 
individuals are aware of relevant patents. Given that it can take patent 
lawyers months of work to determine if a spec does or doesn't infringe on 
a typical company's patent portfolio [1], it seems naively optimistic to 
think that an engineer is going to know anything about which patents are 
relevant. Thirdly, there appears to be no way to enforce the IETF policy, 
so under the IETF scheme, nothing stops the scenario you describe above 
from happening, even intentionally.

[1] e.g. it took Microsoft something like half a year to establish whether 
they had any relevant patents on the HTML5 effort.

The WHATWG has no patent policy at all, but in practice that doesn't seem 
to be much worse than the IETF policy. One could argue that it's better, 
because it doesn't lead to people having a false sense of security!

The W3C's patent policy is a real patent policy -- taking part in the 
working group automatically forces you and your employer to RF-license any 
patents you own but don't disclose. This still doesn't stop the real 
problem (patent trolls coming out of nowhere later, the scenario I 
describe above), but it does stop participating competitors using patents 
against each other, which the IETF's policy doesn't.


> [Copyright is much easier to deal with, because it essentially would go 
> like this:
> 
> Person B: "Here is some proposal Y for your spec."
> 
> Editor, WG: "Great text, we'll just paste it in."
> 
> (later)
> 
> Person B: "Why the hell are you taking my text? It's me who wrote it, 
> it's my copyright, and I don't like it sitting there after that other 
> paragrap."
> 
> Editor: "Sorry, didn't know. Too bad, have to rewrite these few 
> paragraphs so that they say the same thing with different words." (way 
> cheaper than changing implementations to get around a patent or paying 
> royalities)]

Technically, you're still liable for the copyright violations during the 
period where the copyright was being violated, as I understand it.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'