Recent Posts

[Kofi-Annan Centre for Excellence in ICT/Advanced Information Technology Institute]

Kofi-Annan Centre for Excellence in ICT/Advanced Information Technology Institute

Accra, Ghana. 20 – 24 September 2010.

The Connectathon will be held in parallel with the training sessions 22-24 September. Applications expected include:

• Monitoring (DHIS2, CRIS3, CRIS PEPFAR)
• Clinical (OpenMRS)
• Lab (OpenELIS)
• Human Resources (iHRIS)
• Logistics (OpenLMIS)

Goals:

• Build capacity of Ministry of Health staff in the West African (ECOWAS) region related to current and planned implementation of HIS software.
• Build a network of Ministries of Health, their staff, implementers, researchers, developers in the region and beyond to facilitate sharing of experiences and knowledge for HIS strengthening.
• Expand on the functionality of HIS software by facilitiating user feedback and discussion with developers.
• Establish a regional education and research network for health informatics.

Specific Objectives:

• Review the architecture of Open Source applications in use or intended to be used in the computerisation of health information systems (HIS) in the region.
• Train IT professionals on the installation, use and maintenance of Open Source Software in use in HIS in the region.
• Create a network of IT experts for development and maintenance of Open Source HIS applications in the region.
• Make appropriate recommendations on the integration or inter-operability of HIS applications for the region.
• Make proposals for the establishment of training programmes (in-service and pre-service) in Open Source applications for integrated HIS in the region.

For more information contact:
Patrick Whitaker whitakerp@who.int
Johan Saebo saeboj@who.int

The World Health Organization (WHO) has adopted SDMX and implemented it in a variety of WHO and partner applications to facilitate the development of country monitoring systems. The WHO implementation is called SDMX-HD to reflect its use in the Health Domain (HD). The SDMX-HD is distinguished by its use of common and custom attributes, enabling the flexibility required for harmonization of indicators and local interoperability.

Tools are available for the SDMX-HD in .NET and Java. More information on these tools is available at www.sdmx-hd.org.

The WHO Indicator and Measurement Registry (IMR) enables the definition and harmonization of indicators and export of indicator definitions in the SDMX-HD format. A goal of the IMR is to minimize the proliferation of organization-specific DSD in health.

The public version of the IMR is located at apps.who.int/gho/indicatorregistry. Extranet accounts for indicator creation can be obtained from Patrick Whitaker whitakerp@who.int.

Hi All, I wrote a blog post (http://www.spacetimeresearch.com/archives/232.html) on this topic that I thought some of you may like to read. It does seem to me that there is lots of use of SDMX without any prior agreements on data standards. In fact, we are currently looking at leveraging SDMX capabilities for some work with a couple of Australian Govt clients and we probably won't contemplate any provisioning agreements to begin with. What do others think? Does it make sense to assume that we can start down a modest path of SDMX use, and aim to ramp it up to a more formal, sharing arrangement at a later time?

Al, I agree this makes a lot of sense. As you're probably aware (but for the benefit of others), in our SuperSTAR software we treat RSEs as generic SDMX annotations but it would be very useful as you say to have a more specific observation level quality measures. As a vendor, we do find it difficult in some places to come up with the right, generic use of SDMX because sometimes we find that we need to make our own decisions about how what we would consider fairly fundamental metadata should be recorded in this format.

This is an old post - anyone got any more recent info on this they would care to share?

[domain]

The Content-oriented Guidelines (like most other international statistical collaboration) currently take a domain-based view of statistics. E.g. Cross-domain concepts / code-lists. However, statistical production in many countries is increasingly being organised more according to processes than domains. This is supported by initiatives like the METIS Generic Statistical Business Process Model (www.unece.org/stats/gsbpm), and there is interest in exploring how SDMX can be used for data and metadata flows between different parts of the statistical process.

What implications does this have for the content-oriented guidelines? To what extent are the cross domain concepts and code-lists also "cross-process"? To what extent should standard terminology for describing processes and sub-processes be included in the Metadata Common Vocabulary?

These and related questions will be considered by a new working group on SDMX statistical standards later this year, but it would be good to get some input and ideas from the wider community as well.

[validation and interpretation]

Hi.

I'm trying to make a validation and interpretation (reading observation values) of an SDMX cross-sectional data message using .Net programming.

I wonder if it is possible doing this without using the corresponding (domain-specific) schema. 

I thought it would be enough to have the DSD like I read in another SDMX forum (http://www.metadatatechnology.com/forum/showthread.php?tid=60) that the specific schema could be produced from the DSD.

I just don't understand how to be able to manage the OBSERVATION VALUE (and other Observation levels attributes) if it is defined (extended) by the specific schema, if I understood well.

For this issue I attached some sample files from Bank for International Settlements (BIS).  In this sample (CrossSecional_Sample.xml) the observation values are using the "value" attribute which, if I understood well, is an attribute defined in the BIS_JOINT_DEBT_CrossSectional.xsd file. So, how to "know" that without having the specific schema (BIS_JOINT_DEBT_CrossSectional.xsd).


Thank you for any clarification about this.

Dear Nalini

Many thanks to you, and to the IMF in general, for making available an opportunity via the SDMX forum to consider and discuss this DSD in draft form.

In addition to possibly identifying refinements to the design of a specific DSD, to better support the interests of those providing and consuming the related data, it will be excellent if, though opportunities such as the one you have provided, the community around SDMX can share ideas, and learn from each other, more generally in regard to DSD design, application and further development of cross domain concepts and code lists etc   

The attached comments, suggestions and questions arise from an informal round table discussion of the draft DSD for Government Finance Statistics (GFS) within the Australian Bureau of Statistics.  This included a subject matter specialist on GFS and several staff with a more general interest in data definition, data management and the application of SDMX.  The comments do not represent a formal ABS response.

Best regards

Al

Dear All,

We at the UNESCO Institute for Statistics created an SDMX library in C# and published it as an open source project on google code.

The reason we made this effort as an open source project is for a couple reasons. First, we need this library for our own work with SDMX.

Second, we created this library to be generic to the SDMX standard and not specific to our needs only. So we hope those who find it useful will contribute to the development of the library.

The project location is: http://sdmxdotnet.googlecode.com

The project discussion board: http://groups.google.com/group/sdmxdotnet

Thanks,

Duraid Abbas

This is to remind you to post comments on the draft Government Finance Statistics DSD by April 30, 2010.
Thanks,
Nalini

Tasos,

Thanks so much for your reply!

We're hoping to skip the CSV entirely. I'm trying to determine if there is a way to write code in SAS to generate only an SDMX-ML file as the output.
I'm not a database person (or really, even a stats or SAS person!) so I'll check with some others here on that suggestion.

Thanks again,
Christine