Re: Patents and public (was: RE: Proposal to Adopt HTML5)

On Apr 13, 2007, at 10:09 AM, Chris Wilson wrote:

>
> Maciej Stachowiak [mailto:mjs@apple.com] wrote:
>> You might expect that, but I have never seen a W3C working group do
>> such a thing, or even try to determine what might be a patentable
>> invention. I personally don't even know how to judge if an idea is
>> patentable, do you?
>
> Scope of uniqueness.  Surely a few people at Apple know whether an  
> idea is patentable scope or not, I think you guys have a few patents.

Yes, there are such people, we call them "patent lawyers". And they  
never give an opinion that something is definitely *not* patentable.  
That would be irresponsible, given recent examples of patents for  
things like linked lists. If we are to review proposed ideas for  
potential patentability, we will need some sort of legal staff for  
the working group. Does the W3C have the resources to provide that?  
If not, who will do it?

All that being said, it's in our interests to make it as easy as  
possible to agree to the patent policy and to get as many people as  
we can to do so, especially W3C Members with large patent portfolios  
who are not yet participating in this group.

>> The latter has happened before, but I don't know of any
>> instance of your suggested scenario in the case of W3C standards.
>
> W3C has never been quite this public before either.  And the W3C,  
> to my knowledge, has CERTAINLY have never had a specification  
> ostensibly owned by them that is in really the sole product of a  
> member who takes a wide variety of inputs to edit the spec, all but  
> the WG of which inputs are not under any patent constraints or  
> indeed even necessarily IP provenance discovery.

I don't think any W3C standards claim that all ideas in them are all  
generated by members of the group. The SVG working group certainly  
did not invent the ideas of bezier curves or alpha compositing. These  
are ideas that came from the field of computer graphics.

>> Applying the W3C patent policy also would not prevent someone from
>> deliberately using a knowing or unknowing third party to inject
>> something into the spec.
>
> True.  I would hope that no knowing party in the group would do  
> this, but it is just a hope.

Note that the W3C patent policy doesn't in any way forbid this (for  
Member organizations it would if there is a Disclosure Request, but  
not for Invited Experts).

> I would hope any UNKNOWING party would say "hey, that's a great  
> innovative idea.  Could you agree to the patent policy and  
> contribute it?  Come on, if 300 people have done it so far it  
> doesn't seem so high a bar."

If someone suggested an idea specifically for the spec, that might  
happen. But not all adoption of other people's ideas works that way.  
For instance, one thing I would hope we do as a group (and that I  
hope to help with) is to examine popular JavaScript libraries to see  
if there are any common ideas that would be more convenient or more  
efficient if supported directly by browsers. But it would be silly to  
require the authors of Dojo and Prototype to sign the W3C Patent Policy.

>> I think the best way to mitigate risk is to
>> get as many companies with large patent portfolios that are likely to
>> affect the spec into the Working Group.
>
> Leaving, still, Microsoft as the first company to get sued by an  
> intentionally-externally-introduced submarine patent.

Apple is a big patent lawsuit target as well, so I sympathize. But I  
think this scenario is a very minor portion of the patent risk in  
today's software landscape, and also hard to defend against effectively.

Regards,
Maciej

Received on Sunday, 15 April 2007 02:02:20 UTC