Or, to be more accurate, some Microsoft chump patented something that
was pretty obvious to me at the time, and probably obvious to a bunch of
others. More interestingly, he did it years after I open-sourced my
iGlance application *and* presented the exact algorithm at Codecon, the
premiere P2P conference of the time.
It's patent #20080205288, named "Concurrent connection testing for
computation of NAT timeout period". It's abstract is:
> Concurrent testing of NAT connections using different timeout values to
> compute a keep-alive value for the NAT device. Computation of the approximate
> timeout value is accomplished concurrently over multiple test connections
> within about a time equivalent to the actual NAT timeout value. The
> architecture validates the computation of the approximate timeout value by
> distinguishing NAT connection failure from external failure using a control
> connection. Moreover, computation of the keep-alive value is performed only
> once for a given NAT device rather than being an on-going process for that
> NAT device. When one of the test connections fails, it is determined that the
> NAT timeout value is less than the test timeout value associated with the
> failed test connection. Accordingly, a smaller test timeout value is then
> selected as the keep-alive value for keep-alive processing of the NAT device.
This sounds remarkably similar to the discussions we've had on this list
over the years (including very recently), and that is available in my
iGlance application here:
http://www.iglance.com/
Also, note that iGlance has been open source since 2005 -- you can
download a 2006 snapshot of the code tree here:
http://www.iglance.com/code.html
You can also see iGlance in the 2006 CodeCon schedule here:
http://codecon.org/2006/program.html#iglance
Can anybody suggest a good place to record prior art (other than this
list) such that if anybody wants to contest this patent in the future
they'll be able to easily find it?
-david
_______________________________________________
p2p-hackers mailing list
p2p-hackers@lists.zooko.com
http://lists.zooko.com/mailman/listinfo/p2p-hackers
