Danny Weitzner, of W3C (someone who's work I know of and respect), recently posted an article entitled "Whose Name is it Anyway? Decentralized Identity Systems on the Web". In it, he suggests XRI "should be avoided at all costs." The main point he makes is that its centralized registries should be avoided, which I won't discuss here.
The statement I find most disturbing is the IPR FUD-bomb he drops:
XRIs ... (come) with a great deal of patent licensing uncertainty.
I'm not sure what "uncertainty" Mr. Weitzner is talking about, but merely stating that "uncertainty" exists, without supporting facts, is exactly what the "U" in FUD refers to.
In the context of the IPR discussions taking place around OpenID, making such a patently false (pun intended) comment strikes me as irresponsible, at best. At worst, its comes off as classic, vile FUD: a recognized expert using their stature and position of influence to sway observer opinion without any basis in fact.
That being said, I will give Mr. Weitzner the benefit of the doubt, and assume he's really just misinformed - or perhaps that the communication of the IPR around XRI is confusing - or even that he finds it somehow "uncertain". But I would ask him for a more responsible summary of the IPR around XRI in his public communications.
All the parties involved in XRI have tried very hard (and I believe, successfully) to make the IPR situation crystal clear: free and open, do with XRI what you want. There's open source implementations on the server side and client side, there's no royalties, etc.
The IPR statement for the XRI TC is here . The TC is operating under the RF on limited terms, which is the most open-source friendly and open licensing IPR mode an OASIS TC can operate under.
If anyone has any concerns about the IPR state of XRI, please let me know directly! If there are problems, we want to fix them. If you have questions, I want to answer them. Please ask us directly instead of relying on statements made by 3rd parties like those in the article mentioned above.
I don't know what Danny was referring to, but the fact that XRI is new technology is enough to warrant caution wrt IPR concerns. URIs have been around for 17 years, so the chances of an essential submarine patent emerging are far less than for XRIs.
Posted by: Mark Baker | July 27, 2007 at 08:33 PM
Mark-
You are absolutely right. But that can be said of almost all new technologies like ATOM, and even XML. You could say there is always a unknown submarine patent risk in adopting any new technology. But I don't see that argument being used against other new technologies...
Posted by: Gabe Wachob | July 27, 2007 at 11:09 PM
New technologies usually enable something new to be accomplished, and if that new thing is valuable then folks don't worry so much about IPR issues. But AFAICT, XRIs don't enable anything new to happen beyond what can already be accomplished with http URIs and HTTP extensions.
Posted by: Mark Baker | July 28, 2007 at 12:45 PM
Gabe:
There is no great deal of patent licensing uncertainty with regard to XRIs. It would be a huge mistake to embrace OpenID at the expense of XRIs. As an attorney, I can tell you that each identity protocol will serve a different and valuable purpose. Centralized registries offer enhanced trust and confidence. A sound business model on identity would, as an evidentiary necessity, have to include some trusted centralized registry. I am considering making a presentation on this topic at the upcoming IIW on Dec 3. Do you think that this is something you and the other attendees would want to discuss further?
Posted by: Dan Perry | November 05, 2007 at 12:29 PM
Dan-
Absolutely. In addition, OpenID IPR will likely be the topic of a session in and of itself. .. See you there!
-Gabe
Posted by: Gabe Wachob | November 05, 2007 at 01:10 PM
Dan-
Absolutely. In addition, OpenID IPR will likely be the topic of a session in and of itself. .. See you there!
-Gabe
Posted by: Gabe Wachob | November 05, 2007 at 01:10 PM