Secure Data Store Working group - review the charter, meet the Chairs, invitation to get involved.

Secure Data Store Working Group – Review the Charter, Meet the Chairs, Invitation to Get Involved

Wednesday 13J

Convener: Kaliya Young, Dmitri Zagidulin, Tobias Looker

Notes-taker(s): Paul Knowles

Tags for the session - technology discussed/ideas considered:

Discussion notes, key understandings, outstanding questions, observations, and, if appropriate to this discussion: action items, next steps:

Input document from

WE are working towards and interoperable Spec for dropbox/Google

“Competing” with WebDAV

“Competing” with ___

Expecint way to mount a thing to /MNT

Replication and syncing.

From a productization it is hard cut out.

file system - entirely different way of storing data - competing with SQL on some level. Why are we doing this?

Why not focus on one simple aspect.

The engineering choices in order to make work client side encryption. Have implications on all these other layers.

The choices you have to make on encrypting data influence access control and synchornization.

Henus acts of violating layered models.

Tobias - Mattr - New Zealand company. DIF Hyperledger W3C CCG - DID working group. Interested in this technology in general.

Kaliya - worked for 15 years.

Dmitri - favorite subject SDS - living and breathing for 15 years. Vertical stack for

Lead engineer for that - became clear to me - if we are going to have encryption. WE can’t add on to it after the fact.

If going to do access controls on SDS - we have to build it in.

As engineer building on lot of pre-spec implementations.

Earlier today’s SCIM - user management API.

They definitely need to join the conversation.

Make sure the 4 original “camps” communities - that their interests are met. All the decentralized folks. Including the European work MyData, MAIDSAFE, Redecentralized folks.

Out of scope Design or Development non-HTTP

We need an Architecture Reference Model

We need Marchitecture Diagrams

Someone - actually a sellable product - what this product does and how it is different then other produtcs.

Unless there is some common terminology - what the “thing” is and across competitive boundaries.

People will only spend money on a thing if there is actually a category.

What we do is that we are extremely different.

Competing about the details…

What is the market going to be called and the

The money was on in the software.

There was a tone of technologies that have to interoperate.

3-4 letter acronyms.

Play around in so many different ways.

Input documents

Ultimately doesn’t matter what the name is

it matters if people use the same thing.

People use it…

MARKET BUILDING opportunities.

Common language.

The market was defined by the reference architectures.

You are right about this. Telco’s Celular - Mobile. Did grow this way and Microprocessors.

Someone hacks something together in the market and it gets solved.

We are going for something like that here defining standard interface.

Question. List from working group.

Encryption of Metadata.

Can indexing work on client side encrypted data.

Some members require indexing to be only encrypted.

Also capabilities for unencrypted meta data.

techniquest - solar and elastic search do for enrypted search - the search terms are hashed.

I think exactly like you are saying - exact matching encrypted storage is not hard.