spf-discuss
[Top] [All Lists]

Re: Am I nuts?

2004-09-16 14:29:25
Mike Hammer wrote:

On Thu, 16 Sep 2004 16:22:14 -0400, Holm, Mark <mholm(_at_)medrad(_dot_)com> 
wrote: 
  
One more point, the filing date is October 10, 2003. I think that means, if 
prior art is going to kill any part of this, the prior art must be dated 
before
the invention date and the earliest invention date is October 10, 2002. Why? 
Because you have only one year from invention date to file the patent and
prior art is going to have to be before invention date. (I am not completely 
sure on this, but the logic is going to be something like this. 


Read a little more carefully, it references another filing from March 2003: 


CROSS REFERENCE TO RELATED APPLICATIONS 


[0001] The present application claims the benefit of U.S. Provisional 
Patent Application Ser. No. 60/454,517, filed Mar. 12, 2003, and 
entitled "Coordinated Reduction Of Unwanted And Unsolicited Electronic 
Mail Messages". 



Mike 

That is getting a lot more into the depths of patent law than I have any 
knowledge of.  I know the one year limit is pretty rigid.  I don't know if 
there are mechanisms to push it back farther.  It could well be that the 
"provisional" application is a mechanism for getting in under the one year 
limit while hammering out the details of the "final" application.  I know there 
are other proceedural quirks.  I have seen patents that are listed as 
"continuation in part of" a previous application that was "abandoned".  I think 
that means, after the examiner disallows part of your claims, you have to 
rewrite the thing and it gets a new application number.


To my eyes, this application looks like more than one invention, thus really 
should be separate applications.  But, the ways of the patent office are rather 
mysterious.

I see at least two: a challenge, response scheme aimed at IP spoofing;   domain 
verification based on DNS lookup of the purported sending domain.  I haven't 
seen yet a claim that combines the two, but it is very likely I haven't read it 
hard enough yet.


The odd part is I really don't see PRA in it.  The multiple "parameter" 
evaluation appears to be aimed at finding the "actual network address" of the 
sending computer, not the "purported" address or domain.  The whole question of 
determining the purported sending domain doesn't appear (to me so far) to be 
addressed at all.  It is just stated as something the receiving computer does, 
with no explanation how.  "Actual network address" appears to mean (in the case 
of the internet) the IP address assigned to the physical machine from which the 
message was transmitted.  Why you have to look in multiple headers to find that 
is a mystery.  The patent as much as says so when it says it can be got from 
the SMTP conversation.  I got to be missing something.

It almost seems like, in the first part of the claims, they are all concerned 
about verifying the IP address and in the latter parts, they don't quite 
completely make the transition to worrying about the sending domain.

I'm not an expert, but what I have read is that the experts consider successful 
IP spoofing not to be a real threat, therefore, the first part of the claims 
cover something no one would really need to implement.

Mark Holm


<Prev in Thread] Current Thread [Next in Thread>