- Количество слайдов: 19
United Nations Regional Workshop on Data Dissemination and Communication, Manila, Philippines, 20 -22 June 2012 Strategies for web based data dissemination A strategy is a plan of action designed to achieve a vision - from Greek "στρατηγία" (strategia). Zoltan Nagy – Statistics Division, Department of Economic and Social affairs, United Nations
Existing Strategies Fundamental Principles of Official Statistics “statistics that meet the test of practical utility are to be compiled and made available on an impartial basis by official statistical agencies to honor citizens' entitlement to public information” Handbook of Statistical Organizations National Strategies for the Development of Statistics (NSDS) The Generic Statistical Business Process Model (GSBPM)
Data dissemination = communication Policy making Professionals & public Analysis & research Policy validations Policy decisions N TIIO TO C IICA UN UN MM MM CO C Statistical needs and education CO MM UN IC A TIO N Knowledge based society Policy accountability INDEPENDENT OFFICIAL STATISTICS ECONOMIC AND SOCIAL PROGRESS Policy options Analysis & assessment
The importance of web-based data-dissemination From 2008 to 2013 the number of Internet users grows by 67% Everyone who has access to internet is becoming a potential user of statistics. Forget the last war.
Identifying users User groups Decision makers (government at central and local level, businesses) Academia (institution that use, research and analyze data) Educational (primary, secondary, tertiary) Public at large Tourists Harvesters and (data) Miners
Tourists Novice or infrequent users, and typically make up the majority of individual users. Looking for basic data either out of curiosity, or to inform personal decisions. Want to be able to find and view data quickly and easily, they prefer low levels of complexity and need only limited functionality.
Harvesters Intermediate and fairly frequent users, who are looking for data to inform basic research or economic decisions. They will accept increased complexity if it results in addition functionality and flexibility in the way they can view and download data.
(Data) Miners Expert users, typically small in number, but using large volumes of data on a regular basis, often for detailed research or analysis. They want simplicity, easy downloads functionality and flexibility, take data offline
A new type - Builders Experts that want to reuse statistical data without copying or downloading it. Requesting ability to access data servers at 24/7 and feed data to maps, visualizations and other applications. Web services - interoperable machine-to-machine interaction over a network". Mashups – hybrid web applications Visualizations Mappings Data aggregators
Defining the content Data Topic – domain specific or across-domain Coverage – geographical and time Aggregation level - micro and macro data Nature of the data – tables, tabulations, time-series, datapoints Documentation Metadata (descriptive and structural) Methodologies and standards Classifications Best practices, business processes, etc.
Subscription models Registration No registration required Registration required (provides better tracking, communication etc) Subscription Free (preferred by many countries) For fee (cost recovery, profit, one-time, periodical, service based ) Multi-tier (free basic and for fee premium services)
User management User access (registered vs unregistered users) User support, helpdesk User surveys (online polling) User activity tracking Web server statistics Analytic services (Google analytics) Custom built tracking services Social networking (Facebook, Twitter. . )
Site administration Data Management Data correction facility Data upload facility Data availability Metadata Management Structural metadata Descriptive metadata Data upload calendar Management Reporting
Resource allocation + Data dissemination group (Centralized or Decentralized) + Systems/Application development + Hardware and software requirements + Long-term maintenance + Operation + Helpdesk -------------------------= TOTAL COST OF OWNERSHIP (TCO)
Content delivery Bandwidth conservation Browser considerations
Software platform and architecture Off-the-shelf products Custom development (in-house, outsourcing) Open source platforms Proprietary platforms Self hosting Outsourced hosting
Design considerations Simplicity and ease of use Easy of navigation Bookmarking Searchability Drill down Dimensional search Full text search
Conclusions One size does not fit all Web-based data dissemination should work as a two way communication Focus has to be on users who frequently visit our sites The maintenance of web-based data-dissemination products is a long term commitment We have to be aware of TCO