• Open Data Scenarios
    • Three different levels of complexity based on the number of datasets available and the frequency of their update frequency
    • Level 1 –less than 100 datasets with less than 10 datasets changing weekly
    • Level 2 –100 to 1,000 datasets with 10-100 datasets changing weekly
    • Level 3 –Over 1,000 datasets with 100+ changes weekly
  • Solutions
    • Based on the scenarios there are various possible technical architectures
    • In some cases we have provided more than one option
    • We have also given some examples of software systems that can be used. Click on each graph to view the software for each component of the arquitecture.
    • Some of these software are open source and others are proprietary.

    Note: The software systems indicated in this presentation are just examples and not in any way endorsed by the World Bank.

  • Level 1: less than 100 datasets with less than 10 datasets changing weekly
    • Build a conventional website using a standard web server
    • Data as raw data files within web server space (or on public cloud storage)
    • Metadata as micro formats on HMTL page
    • Site search using existing tools like Google site search
    • Manual update of data, metadata and content by central IT team
    • Recommended
      • Introduce commenting feature for feedback

  • Level 2: 100 to 1000 datasets with 10-100 datasets changing weekly
    • Front end as conventional CMS-based website
    • Data as raw, manually managed, data files within web server space, public cloud storage , or individual ministry websites
    • Metadata in SQL database, served through dynamically generated query pages
    • Automated checking of broken links –Community facilities essential
    • Site search by text search on database, CMS search or Google site search (GSS)
    • Maintained by central team
    • Recommended
      • Consider caching to take care of load
      • Do not store metadata in the CMS itself

  • Level 3: Over 1000 datasets with 100+ changes weekly
    • Front end needs to integrate different web services (API)
    • Automated management of raw data files (possibly being stored on public cloud storage) (Level 3a)
    • Management of data using a combination of raw data files and data stored in an Open Data System’s database (Level 3 b)
    • Metadata in optimized metadata repository
    • Delegate submission/maintenance of datasets to individual Ministries with custom dialogues, automatic validation, and role-based access control
    • Automated checking of broken links
    • Community facilities essential, including ability to raise issue with the “owner” of each dataset (e.g contact person, web form)
    • Search by structured search on metadata repository, CMS or Google site search on main site

    Level 3a

    Level 3b

  • Information security best practices
    • Data Governance to ensure
      • Authoritative Source
      • Rules of Engagement
      • Sustainability
    • Data Classification Public v/s Restricted to ensure
      • Private/restricted data not released externally
    • InfoSec Controls to ensure
      • Confidentiality and integrity of data
      • Denial of Service (DoS) attacks

Help/Feedback

User Voice