Re: [dnsext] Summary WGLC: draft-ietf-dnsext-ecdsa

[Rene Struik <[e-mail]>]

There may be IPR about representing Q:=(x,y) in a compressed format
Q':=(x,y'), where one can *faithfully* reconstruct Q from Q' and the
elliptic curve domain parameters alone (so-called lossless compression).
To my knowledge, any *lossy* compression technique, including
representing Q:=(x,y) by its x-coordinate x only, is not covered at all
by IPR.

Fear, uncertainty, and doubt should not be based on not understanding
the difference between these two compression methods (or portraying
everything under the universe vaguely similar suspect). Sadly, it
apparently is. (But, perhaps, an "favored" way to mute discussion on
technical merit and portraying the world in black-and-white terms (the
nuclear option, so to speak).)

-50

Rene


On 19/01/2012 4:05 PM, Michael StJohns wrote:
> At 02:30 PM 1/19/2012, Paul Hoffman wrote:
>>> I am curious as to why "nobody" supported squeezing down message sizes.
>>> Was this seriously considered? What was the technical rationale for not
>>> supporting my modest suggestions.
>> I cannot speak for everyone, but I can say why I didn't support it: it would possibly involve IPR hassles. The IPR hassles are not worth it here, in my opinion.
> +50....
>
> EC point compression is a known IPR rat hole.  AIRC - The group chose to slap a thick timber across the hole and glue and nail it down.
>
> Mike
>
>


-- 
email: [e-mail]
Skype: rstruik
cell: [phone]
USA Google voice: [phone]

_______________________________________________
dnsext mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/dnsext