[WG-OTTO] OTTO WG Agenda Tomorrow 11/10

Mike Schwartz mike at gluu.org
Tue Nov 10 14:33:58 CST 2015


Rainer,

I really like the requirements doc:
   
https://github.com/KantaraInitiative/wg-otto/blob/master/docs/sources/requirements/requirements.md

I think we should discuss this in detail tomorow, especially as it 
relates to JSON linked data. This video was really nicely done: 
http://gluu.co/json-ld-intro-video  And there seems to be a link between 
JSON linked data, and triples as you mention in your doc. I'm guessing 
something like : (entity ABC, belongs, federation XYZ)

I think we all like the idea of the blockchain, its the implementation 
that is so unclear. I'm concerned that if we don't consider the 
blockchain publication method, we may miss something important in the 
design of the schema. But I get caught up on details like "which 
blockchain." And what's the business model.

With the linked data approach more concrete, I think we can map out how 
to connect to a blockchain, probably the Bitcoin blockchain, which is 
the most active. I guess we'll just have to figure out how to do this...

- Mike



On 2015-11-10 14:10, Rainer Hoerbe wrote:
>> Am 10.11.2015 um 18:34 schrieb Mike Schwartz <mike at gluu.org>:
>> 
>> Just a reminder about the meeting tomorrow at 8am PT.
>> 
>> I think we should discuss putting the blockchain idea on the back
>> burner and focus on the schema. I'm hoping this would enable us to
>> work in parallel to resolve some of the many questions around how to
>> best use the blockchain technology.
> 
> I agree to put the implementation questions of blockchains aside for
> the time being. However, we should come to a consensus about the
> higher-level concept of a distributed public ledger.
> 
> To recap: Traditional patterns to establish trust use trusted third
> parties who register and sign factoids, frequently from multiple
> authoritative sources. This proof of correctness is a well-established
> practice, as exemplified in this document [1] that testifies that my
> grand-grandfather had fulfilled his military service by 1847,
> including his name and place and date of birth.
> 
> A public ledger is different in that does provide _continuity_ instead
> of _correctness_. Binding a name to a key, or more general claims, are
> trustworthy because a claim is public and cannot be altered or revoked
> without the public noticing it. Combined with the requirement of the
> claim author and/or subject for an on-going monitoring of their
> respective claims there is a guarantee that facts cannot change
> unexpectedly, and cannot be censored either.
> 
> Besides this difference there is also an advantage for a public ledger
> in that it is not owned by a particular operator. Of course, there is
> the development team and other stakeholders, but this is much more
> transparent and distributed than a traditional CA, notary, bank,
> government register etc. While a TTP has a specific scope in their
> business model, public ledgers can accommodate multiple domains,
> business models, jurisdictions and assurance levels, allowing for
> re-use of assertions.
> 
> A combination of the traditional TTP-model with a public ledger would
> therefore achieve a higher and more scaleable way of sharing trust
> assertions. Third-party assertions will address the problems of trust
> on first use, DoS, lack of monitoring, link to legacy trust sources.
> The public ledger will provide irrefutable proof of continuity.
> 
> I started with a draft requirement document that includes the thoughts
> above:
> https://github.com/KantaraInitiative/wg-otto/blob/master/docs/sources/requirements/requirements.md
> [2]
> 
> Cheers, Rainer
> 
> PS: Apologies for tomorrow: I will be with a customer tomorrow an
> won’t make the call.
> 
> 
> 
> Links:
> ------
> [1] http://files.hoerbe.at/stammbaum/CarlConradMeier18471027.jpg
> [2]
> https://github.com/KantaraInitiative/wg-otto/blob/master/docs/sources/requirements/requirements.md

-- 
-------------------------------------
Michael Schwartz
Gluu
Founder / CEO
mike at gluu.org


More information about the WG-OTTO mailing list