- From: Steven Rowat <steven_rowat@sunshine.net>
- Date: Fri, 2 Jun 2017 15:58:15 -0700
- To: David Chadwick <D.W.Chadwick@kent.ac.uk>, Credentials CG <public-credentials@w3.org>
On 2017-06-02 2:23 PM, David Chadwick wrote: > > On 02/06/2017 19:17, Joe Andrieu wrote: ...>> I think your notion of "authorize" would be more commonly regarded as >> acting on. I would go further and use the term apply. > or we could say, make a decision. Strongly agree. And I'm going to try to summarize, because this thread is large and fast-moving, but I feel like there's a chance for consensus cropping up. :-) But first, here's something I think is directly relevant that appeared in the journal "Cell" yesterday, and was reported in MedicalXpress: an interesting new evidence of a mechanism for how we (monkeys, but certainly us too) recognize faces. By attributes! And more simply than anyone imagined, as the authors say. Here's the MedicalXpress story: "Researchers decipher the enigma of how faces are encoded in the brain" https://medicalxpress.com/news/2017-06-decipher-enigma-encoded-brain.html The key point is that there are 200 neurons, only, that are required, and each neuron measures a *single* attribute in the face, like the distance between the eyes, or the width of the hairline, or the skin tone, or texture. We can assume that these attributes are then pattern-matched, together, against stored previous measures in memory, and a match is 'recognized' for the entire group and so the face is 'identified'. This seems like a fundamental process, and perhaps one that parallels how this thread is showing we should approach 'Identity' through Verifiable Claims as well, both on and off-line. So: maybe these four statements are true? : 1. Identity of 'A' is in the eye of the beholder, B. (Beholder C might not recognize A at all, or might do so via a different set of evidence than B does.) 2. For any beholder such as B, 'recognition' (validation) of the identity is done by measuring, or receiving, a pattern of attributes of A, and comparing them with a pattern of attributes stored in B's memory. When a best fit is found, 'recognition' occurs, by B, of A. This is not an exact process, but rather a statistical, probability-based, process. 3. This recognition, of A by B, is the basis for a functional decision by B for some action B will take. B is collecting these particular attributes purposely, to gather information to help in making the decision. Thus the information is not, in general, something A 'gives B', but something B measures. 4. Privacy is concerned with whether A can prevent B from making the measurement. Steven
Received on Friday, 2 June 2017 22:58:49 UTC