Shannon Entropy, channel protocols, and propensity for noise

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On a conceptual level, maximizing opportunity (uncertainty) to *scale* is the same as maximizing the Shannon entropy of the protocol in terms of the characteristic we want to *scale*.

The following thought can also be generalized to any protocol or aspect of the protocol, not just email or scale.

Some private discussions stimulated me to have a generalized thought which I personally felt is relevant (also profound, yet obvious) and important enough to share with the people who supposedly design some of the internet engineering in the STDs track.

Regarding the concept of a protocol's "ability to scale", as one of Dean's private criticisms of Iljitsch's request to revisit criticisms of the (probably very old) idea of forcing SMTP to accept email only from a SMTP server of DNS record for the domain in From header (I may have misstated the idea, but my thought herein still survives).

My thought is as result of any design to maximize scale entropy, that we expect (as predicted by Shannon's theories) that the channel has the propensity to generate a maximized amount of noise.

I think most of people would agree that spam is noise (that unsolicited is essentially noise).  My thought is on a conceptual level what my previous anti-spam proposal and what Iljitsch's idea (and probably all other protocol level anti-spam proposals) are attempting to do is to reduce the entropy (opportunities, i.e. uncertainties) of the email channel.

The benefits and costs (tradeoff analysis) within such a conceptual paradigm is what I leave here for thought stimulation.  I am not here to argue any point, merely to postulate the conceptual relationship.

Good luck.

I am resigning *again* from the list, so please do not send any response directly to me.  I can always "pull" the archives if I have desire/time to read followup posts in this thread.

Shelby Moore
http://AntiViotic.com

Dislaimer: I am posting this only in terms of sharing a thought (probably not even a new thought), just to stimulate thought (and perhaps some discussion).  So please do not feel obligated to add noise to this thread (and thus list), if your only purpose is to prove this was never thought of before.  If this is not interesting nor important topic to you, then please ignore it.   I have no interest in defending or nurturing this thought.  It is simply stated to stimulate (or not) depending on each individual reader.



[Index of Archives]     [IETF Annoucements]     [IETF]     [IP Storage]     [Yosemite News]     [Linux SCTP]     [Linux Newbies]     [Fedora Users]