London Sprint, December 2014

The London Sprint took place 24 - 28 November at the Institute of Education, 20 Bedford Way, London. 

In the lead up to the sprint, the Advisory Group agreed to three high priority outcomes and one low priority outcome. During the first day of the sprint the work needed to be completed to achieve each outcome was entered into JIRA. You can see these tasks and the current status for each of the tasks on the Sprint Agenda page. Below a summary of the work for each outcome is given along with any future actions and risks.

Outcome 1: Complete tools work needed for January release (High Priority)

In order to produce the specifications, an XMI export is created from Drupal. From the XMI, a transform is needed to create the schema. There was no one at the sprint which RDF expertise, as such the XML output was the focus. Much work was undertaken during the week to make sure that the XMI export was valid and to create the XSD. There were 29 tasks raised that needed to be completed.

The team are very close to having a functioning production framework. They are very happy with the progress that was made during the week. However, it was not possible to do a test run of the production framework during the sprint.

Outcome 2: Complete modelling for packages to be released in January (High Priority)

The packages and views that are scheduled for the next release had been handed from the content modellers to the modelling team. During the sprint, there was a focus on cleaning up the modelling work in Drupal. Twenty two tasks were raised to be completed. A number of validation checks were introduced to automatically check the Drupal content. These errors (eg duplicates, missing cardinalities) were fixed and the packages in Drupal were reorganised in preparation for the release. The idea of collections and correspondences was introduced into the model and all packages scheduled for release were reviewed by the modellers.

There are a few small outstanding actions relating to the Representations and Process Core that need to be checked with the full modelling team, but it is felt that the model is in good shape to be released. These packages will be frozen in Drupal and content modellers will be asked to refrain from making any changes without approval from the modelling team.

Outcome 3: Put together the draft January release for review by the TC (High Priority)

The documentation team discussed what would be in the release. Based on this, a number of changes were made in Drupal to ensure that the documentation needed could be produced. The team has almost completed this work, they are making some final tweaks to the template in Drupal and fixing some minor styling issues. The documentation will be ready for the release.

Outcome 4: How is the production framework managed after the project (Low Priority)

Discussions to complete this work were delayed until a future date in order to complete the high priority work.

Progress to the first release.

At the end of the week, the sprint team were confident that the modelling and documentation will be ready for the release. The work needed in Drupal and the transforms are the critical area to focus on in order to ensure that the release can go ahead. With Christmas holidays and other work commitments, resources will be tight to complete this work in the given timeframe. The tools support team and modelling team will continue to meet in the lead up to Christmas to progress the outstanding work items.

A Checklist for Q1-2015 Release has been created on the wiki which lists out the expected contents of the release.


First Name
Last Name
Email Address
    Open Data Foundation
    Statistics Canada
    Danish Data Archive
    GESIS - Leibniz Institute for the Social Sciences
    University Library Bielefeld
    Centre for Longitudinal Studies
    German Socio-Economic Panel Study (SOEP)

    German Institute for Economic Research
    Australian Bureau of Statistics
    GESIS - Leibniz Institute for the Social Sciences
    Swedish National Data Service (SND)
    ABS - Australian Bureau of Statistics
    Minnesota Population Center (MPC)