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

Minutes - please fix the notes.

From: Timothy Holborn <timothy.holborn@gmail.com>
Date: Tue, 16 Aug 2016 16:40:09 +0000
Message-ID: <CAM1Sok071Pjk8aXyhR5-AHdMGKBqJwK9XH3mu-LT7rH-0gjmSw@mail.gmail.com>
To: Dave Longley <dlongley@digitalbazaar.com>, Manu Sporny <msporny@digitalbazaar.com>
Cc: W3C Credentials Community Group <public-credentials@w3.org>
re: [02:01] <dlongley> tim: I'd like to see a new blockchain (current
computationally flawed), with LDP we're looking at how to create human
centric identifiers. WebDHT did a lot of that and that's been put to one
side and there's a resourcing issue. Given that Vint Cerf is looking at it
for preservation, maybe we should be following that up, directly or
indirectly via CG, etc.

I was actually talking about an alternative to blockchain, rather than a
new block-chain.  Perhaps the discrepancy relates to the semantics of the
concept; however to be clear,

The concern is about the method in which blockchain technology purports
internet-scale improvements to security parameters.  Below are some links
[1][2][3] and i believe 'human centric' alternatives exist; whereby
social-relations can work in combination with PKI / Cryptography to create
an alternative solution that is far less computationally burdensome.

also: where it says re 'given that vint cerf...'

the specific quote from an email today is "[02:00] <dlongley> tim: I talked
to Vint Cerf and he said he's in contact with IPFS for interplanetary
filesystem, etc. "

And my point was that I felt it was worth following-up on what
opportunities exist to improve support for works formerly considered by
WebDHT counterparts of previous Credentials CG works; by looking at what
IPFS may support as a counterpart and/or consideration - if not within the
IG/WG scope of works - perhaps within the CG incubation process...

My underlying concern is that i do not see a methodology that responds to a
basic functional requirement of having some sort of root-identifier
methodology that is human controlled about a human; in which other
identifiers and/or services may be affiliated to it,

to produce a 'human centric', alternative deployment methodology.  I also
feel that 'service provider centric' methods are very well developed and
understood within industry; so the difficulty of looking at these
alternative 'human centric' models; is relatively high - but no less
important.

In fact - i'd argue they're more important; but will work towards
incremental improvements as is possible, cooperatively of course...

Regards,

Timothy Holborn.

[1] https://www.youtube.com/watch?v=bi2thGzzNSs
[2] https://www.youtube.com/watch?v=Kjtgp5h-jEY
[3]
https://www.reddit.com/r/Bitcoin/comments/24zwsr/how_many_qubits_would_it_take_to_break_bitcoins/
Received on Tuesday, 16 August 2016 16:40:49 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 24 March 2022 20:24:42 UTC