W3C home > Mailing lists > Public > public-credentials@w3.org > June 2016


From: Timothy Holborn <timothy.holborn@gmail.com>
Date: Thu, 09 Jun 2016 05:08:42 +0000
Message-ID: <CAM1Sok0sQQZYhOBFjorO7mLWvwP-Wr_9GEm8-Rgt+MPeaoCrgg@mail.gmail.com>
To: W3C Credentials Community Group <public-credentials@w3.org>
I've been thinking about it more, and am considering that perhaps it's not
as simple as we've tried to make it.

With the 'inspector' or 'checkpoint' some of these agents will consume the
credential - ie: a banking credential that says - here's access to my
bank-account to trade my funds for your products/services.   In this way, a
record of that transaction needs to be stored somewhere - it's not simply
like a 'check-point'.

But say i want to go to a club or an adult site online - it may have a
'checkpoint' that basically wants to know if i'm over 18.  they don't need
to know who i am, that should be entirely optional - therefore, they don't
need to store any form of transaction record other than session data,
perhaps an IP address and the fact that they've done their job as a
'check-point', which in-turn relates to me presenting a 'claim' - i'm
claiming to be over 18 as an action to a check-point.

But even then - i'm not really claiming it - I'm proving it aren't I?
Verifiable Proofs? https://en.wikipedia.org/wiki/Mathematical_proof

or a fact https://en.wikipedia.org/wiki/Fact

The concept of a fact is broad.  I am over 18, i was not born on the 1st of
july, etc.  these are all facts.

Rather than verifiable - it's perhaps conceptually social - if the term
social relates to humans working both as an agent, having agents work on
their behalf or as self regardless of persona.  Therein the consideration
is that it's not about what a person says about themselves, it's what they
say about or to others.

Therefore the eco-system is in-term producing a "Web 3.0 (ie: linked-data,
semantic web, etc.) Social Facts Secure Document"
capability/format/service/protocol?  I think protocol.

So whilst i'm using particular selectors - IMHO - perhaps that's part of
our problem here?  Perhaps we need to have different forms of agents
specified that in-effect become grouped - which kinda gives bandwidth for
more of us to be 'right' when considerate of context.

With regard to the use of the term 'credential' - that's what we've been
doing.  So, changing names or trying to put modifiers.. well, one idea i
had was

W3-LSD = Linked Secure Document or perhaps Linked Socially Aware Secure

TimBL wrote about the storage layer of the 'socially aware web'
https://www.w3.org/DesignIssues/CloudStorage.html - we're looking at the
statements layer and the means to produce secured documents that fit into
that paradigm.

but without thinking of the 'funny side' of the LSD concept; these things
are essentially secured, decentralised, machine/human readable documents.
 the can be used for all sorts of purposes and perhaps we're under-selling
the work by suggesting it's simply about authoring basic credentialing

This in-turn brings about an array of other scoping issues.

A credential by way of these systems isn't really a claim - it's a verified
document isn't it?  a university issues a credential that says a person has
a degree - or a gov. department issues one that is like a birth certificate
- it's as good as the officially printed document written on secured
paper.  Similar also to bank-notes isn't it?

I've not really thought about it in this way until recently - yet inspired
by the 'secure document' requirements for local identity considerations - i
figured i'd share my thoughts.

I think we agree these things are designed for WWW.  I think we agree that
the document contains linked-data. I think we agree it's a
cryptographically sealed document.

I've been reviewing on wikipedia the definitions of HyperText[1] ,
HyperMedia[2] , Hyperlink[3] but i'm not sure they suit this concept.  Part
of it is about what was called semantic web[4] now linked-data[5].

Perhaps Secure Semantic Web Document?

Secure Web Note?

SIM for SIM-CARD stands for Secure Identity Module.

Whilst i've not got the answer yet, I am thinking that making up a name
might be better, and i'm also thinking about how to do that without being
worried about trade-marks. Is it easier to invent an acronym?

The identity space for the web is innovative on a human-history basis.
These things are the proposed method to have 'things' that people say about
you in a validated way, provided to you for representation / presentation /
use as a utility - for your identity related purposes with others.

I don't think humanity has ever done anything like this. not in context to
the broader eco-system emerging.

So, maybe we work on what it does as to better define it - then invent a


[1] https://en.wikipedia.org/wiki/Hypertext
[2] https://en.wikipedia.org/wiki/Hypermedia
[3] https://en.wikipedia.org/wiki/Hyperlink
[4] https://en.wikipedia.org/wiki/Semantic_Web
[5] https://en.wikipedia.org/wiki/Linked_data
Received on Thursday, 9 June 2016 05:09:21 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 11 July 2018 21:19:29 UTC