Non-Assertion Convenants

Sun Microsystems holds some patents that might be trespassed on by the folks developing open ID software. So what are they doing about it? They’re promising never to rely on the patents, thus freeing developers from the anxiety of a possible future lawsuit, and that by a particularly large gorilla. Evidently this is not the first time they’ve taken this open source route, and Simon Phipps, one of their own, has blogged about ten reasons why this approach is sometimes the best way to go and why others should adopt it.

The covenant is brief and intelligible:

Sun Microsystems irrevocably covenants that, subject solely to the condition described below, it will not assert any of its U.S. or foreign patents against that portion of a product that implements the OpenID Authentication V1.1 specification by itself or that implements that specification together with OpenID Simple Registration Extension V1.0 (OpenID Implementation).

Condition: this covenant shall not apply with respect to any individual, corporation or other entity that asserts or threatens at any time to enforce its own or any other party’s U.S. or foreign patents against any OpenID Implementation.

This statement is not an assurance either (i) that any of Sun’s issued patents cover an OpenID Implementation or are enforceable, or (ii) that an OpenID Implementation would not infringe patents or other intellectual property rights of any third party.

No other rights except those expressly stated in this Non-Assertion Covenant shall be deemed granted, waived, or received by implication, or estoppel, or otherwise.

The lawyer in me is curious about the exact basis on which this would be enforceable against Sun in any future… misunderstanding. Typically a unilateral promise doesn’t do the trick. I’m imagining that some form of estoppel or detrimental reliance would be the argument (IAALB [I am a lawyer but...]). Does anyone know of a case where such a convenant has been enforced against the maker?

Even so, it’s good to see the wealth being distributed to the developers this way.

Retweet information »


  1. I’m interested to hear your scepticism, Simon. Surely no lawyer but the most suicidally optimistic would likely want to argue that a sophisticated multinational company’s public, written promise to not assert a patent, wouldn’t be enforceable?

    It is widely customary, within the standards organizations context (that is,
    covering specifications rather than code) to issue promises to license – generally only essential claims, sometimes RAND, sometimes RF. No one has ever asserted that those promises are less than adequate, and AFAIK there have been no cases where a company promises to license something RF and then refuses to
    honour the promise. Of course, I am not a lawyer and have not researched this extensively, but that’s one I’ve never heard of.

    One good example is the original SOAP submission to the W3C[1] where the form of words “X Company will grant to any party a license…” is used.

    If this common practice is not acceptable, what is? Do we have to append “… and we really, really mean it”? :-)


  2. It’s not so much a suggestion that it wouldn’t be enforceable as a musing about what the basis would be in law for its enforeceability. This fooling around with the backfill behind obvious results is something lawyers — and even more, law profs — do from time to time. Thanks for the lead to the W3 stuff, Simon.