Documents

Common general documents for the ESSnet

View the Project on GitHub I3S-ESSnet/Documents

I3S Roma Hackathon

Minutes

Agenda

Monday 20 May, morning session

Monday 20 May, afternoon session

Tuesday 21 May, morning session

Tuesday 21 May, afternoon session

Wednesday 22 May to Friday 24 May morning

Three hackathon threads:

Friday 24 May, closing session

Conclusions of the workshop and next steps:

WP1 - work done

ARC: The current architecture and capabilities have been modelled. The description of the main functionalitites and the data model have been delivered.

Relais: The current architecture and capabilities and the to-be business layer have been modelled. The description of the main functionalities and data model have been delivered.

Relais and ARC integration: An initial assessment of the optimal solutions to integrate RELAIS and ARC functionalities has been done.

PX-Web:

Intentions for this hackathon:

WP2 - work done

The deliverable of M6, “Cookbook on integration and use”, is just a draft for the moment (it could be uploaded as such in CROS platform) The first deliverable will be an interactive report/guidance with examples. The version 0.01 has been done from Software & integration architecture recommandations and the M6 deliverable.

Next steps :

During the Rome-Sprint :

WP3 - work done:

Lack of progress because of a lack of material

During the hackathon:

WP4 - work done:

A draft version of a questionnaire has been done: aims to know why people are trying to share services but not so willing to consume shared services. During the hackathon, we will have to validate the questionnaire and choose the dissemination tools

There also has been a questionnaire during the last ESSNet. We must see what has been done to be sure that the new questionnaire is different from the last questionnaire. It would be appreciated to have an introductory part explaining what has been done in the last questionnaire.

The questionnaire tool of Eurostat could be used for collect, il will allow to have the ecas authentification.

WP5 - work done

The deliverables D1.1, “List of services to be developed and reused”, and D5.1, “First monitoring report”, have been delivered and validated. The March reporting and April reporting can be found on Github.

The next deliverables are:

By the end of April, the consumed human ressources are 100 days, whereas the total ressources for the project is 900 days.

The possible events where a communication could be done are:

Arc and Relais integration

ARC is a “service implements extended functionalities for integrating administrative data into statistical processes”. “RELAIS (REcord Linkage At IStat) is a toolkit providing a set of techniques for dealing with record linkage projects.”

Integration points at the business level :

The work hypothesis is that data exchange between ARC and RELAIS is performed by a data model translator service. Following this approach, we could combine and harmonize the following process steps and related modules of ARC and RELAIS :

Presentation of the methodology and use-cases of Relais

It dates back to 2006. Istat needed a tools for probabilistic record linking. RL process is divided into several steps, and different methods are available for each steps.

Presentation on sharing algorithm

Presentation by Matjaz Jug from Statistics Netherland (SN).

The benefits of sharing algorithm are many : transparency, productivity, security, multi-domain approach, standardization and innovation.

CBS has a data strategy based on 4 architectural patterns :

When there are a lot of partners and legal constraints, it is necessary to establish a scalable technical and governance framework which can combine access-restricted data from multiple entities in a privacy-preserving manner.

Answers can be given by the MIT project OPAL :

Another interesting approach is the UN Global Platform Methods Store, with an algorithm pipeline :

Features in the WP3 platform

The discussion is on features of WP3 platform but also on the links between WP2 and WP3. It will focus a little bit on common cloud capabilities to determine which capabilities we want to adopt, but also when it has consequences we don’t want as as CSPA community.

Output from the Toulouse sprint :

Preselection for cloud implementation:

Possible connection to cloud implementations :

Result of the brainstorming on a Minimum Viable Product for the platform :

The discussion proposed a first sketch of a statistical architectural pattern for a “sandbox” or more accurately a “workbench”.

Description of a statistical service generalized architecture

The objectives of a statistical service is to perform the following tasks:

Seen as such, a statistical service can be mapped to GSIM by using the objects “Business Function”, “Business Process” and “Process step”.

Business layer

And at application level :

Application layer

Conclusion of thread 1

Work planned :

Work done:

Next steps:

Conclusion of thread 2

The objective is to agree on how to integrate VTL in ARC.

There is an existing norwegian VTL-tool. The french use cases will be :

The general context is to use VTL to express rules and execute them in web questionnaires and for administrative sources (GSBPM phase: data collection).

At logical level:

At technical level:

Conclusion of thread 3

The questionnaire for analysing the main incentives and impediments for sharing has been validated.

A first version of a all-in-one poster has been proposed. This poster is exploring the questions : how to break barriers from not reuse to reuse ? how to promote from not share to share ?

Content of a communication kit:

Options for wideo presentations:

Eurostat projects