Re: Unencumbered Checking (was Re: DEPLOY: SPF/Sender ID support in Cou

Anne P. Mitchell, Esq. wrote:

> This brings up a point which has been rolling around in the back of my 
> head for a few days.
> We simply _don't_ know what is in the claim.  In any other area where 
> adoption of some program is to be impacted or encumbered in some way 
> which could cause either a legal liability or an inability to actually 
> implement on the part of the adopter - particularly such a potentially 
> widely used program - it is inconceivable that people would be even 
> considering adopting, or recommending, the program.  In fact, if one 
> had a professional license, and recommended adoption of such a program 
> without a clear understanding of the encumberance, one could be guilty 
> of malpractice.  But I digress...what has been rolling around in the 
> back of my head is this: 
      Regardless of whether software should be patentable (or business 
methods, or ...) and regardless of the fact that trivial "improvements" 
should not be patentable no matter what. We live in a world where they are.
   Aparently we have no idea what parts of the current draft Microsoft 
thinks they have an IP claim on. While that is slightly scary from a 
practical perspective despite aparent IETF policies othewise, I think 
that must mean that Microsoft's claim and their proposed license should 
be given little weight. They are only slightly more significant that the 
possibility that some other entity also has IPR claims against parts of 
the current draft and has not come forward.
Certainly there is much to fear, but fearing the mostly unknow is not 
significantly different from fearing the totally unknown, and almost 
equally pointless.
While I would personally prefer that MARID adopt classic SPF and move 
forward - regardless of other merits it might have Caller-ID is 
Microsoft's baby and I personally beleive that if they truly wish to 
participate in the standards process they need to be open about their 
IPR. They can insist on whatever license they wish and the MARID group 
can decide if that is acceptable AFTER knowing what it is that is being 
licensed. I understand that may not be the policy of the IETF - though I 
beleive it should be. We are discussing changes to a core internet 
technology. I am not personally aware of any internet core technology 
that is encumbered in the way Microsoft is looking to encumber Sender-ID.
   I would advocate for moving forward with Classic SPF, because 
although I beleive that whatever Microsoft's IP that it is not likely to 
stand up to rigorous scrutiny, nor do I think it would be politically 
wise from Microsoft to force the issue - Microsoft has a reputation for 
being aggressively litigous in defense, but with few exceptions it has 
not really been offensively litigous. However I do not beleive the FOSS 
comunity would deploy Sender-ID if a patent did finally materialize, 
until that patent was nullified and that could take alot of valuable 
time. I beleive the FOSS community and a significant portion of the 
comercial SMTP comunity would deploy Classic SPF - even in the face of a 
possible Microsoft patent threat, it is easier and clearer to argue that 
SPF is prior art. There is also a moral and ethical difference, between 
going forward with SPF and going forward with Sender-ID in the face of a 
possible claim from Microsoft. Regardless of the validity of Microsoft's 
patent claim a fight over SPF would be fighting with Microsoft for 
trying to patent what somebody else developed. Fighting over Sender-IP 
would appear more like trying to steal something from Microsoft.
   I would love to see a result that included everyone. Sender-ID, with 
Microsoft on board, and FOSS on board would likely be the most effective 
solution. But the FOSS comunity is not going to sacrifice core 
principles for that and thus far it appears that Microsoft has drawn 
their line in the sand too. Absent Microsoft putting forth a License 
that is compatible with FOSS licenses, or just contributing their IPR 
freely, I think Sender-ID should die here. That is a bad choice - just 
better than the rest.
   I would rather no standard than one that is "owned" by anyone no 
matter how good the terms of ownership.
   Being forced to choose who to exclude I would choose to exclude 
Microsoft before excluding the FOSS community. Though success without 
both is questionable.
   While I personally beleive there is a public value served by killing 
any patent and a high public value for killing stupid patents, there are 
other institutions that serve that purpose. While I think many of the 
IETF's policies on IPR are poor,  in this particular instance wide and 
quick acceptance is critical and the more IPR issues there are the 
longer that will take.