COD Cycle

Common Operational Dataset tutorials are found in the IM Toolbox > Tutorials > Common Operational Database tutorials section.


Overview
Effective COD development is composed of six COD cycle functions:
1) plan, 2) collect, 3) process, 4) endorse, 5) communicate and 6) maintain. The endorsement of the dataset by the IM network is the step that sets it apart from all other operational datasets.


Partners and relationships are required to support the COD cycle. An IM Netwok or Information Management Working Group (IMWG) that supports the development and maintenance of a COD may vary depending on the context and participant availability; at a minimum the group must endorse the COD by agreeing that the dataset will be used for preparing and responding to humanitarian situations. Certain CODs may require participation of particular partners based on their expertise and new partners may be identified at any stage of the COD cycle. No CODs is created by any organization alone and the relationships among the IM Network are as valuable as the datasets themselves. See the Partnership and Governance section for more details

The COD cycle should be proactively worked on prior to emergencies in the data preparedness phase in support of the contingency planning process (see the IASC IM guidance). The cycle is to be done yearly, with a review of newly available datasets or updates required at the planning stage. If no update is required, the Collection and Processing phases can be skipped and endorsement should be obtained so that all partners are aware that there is no change to COD. The metadata should be updated to indicate the new endorsement date and the Communicate and Maintenance phases should then be followed.


1. Plan
Development of goals, identification of critical partnerships and available datasets

COD planning involves determining the goals of the IM Network, determining the critical partnerships necessary at each step in the cycle, and identifying data sources. Although some partnerships will be the same for all CODs (those actors involved in the response), each dataset might also have unique partners. Identifying the critical partners can be a challenge in many countries as occasionally different governmental departments will claim responsibility. It is important to keep in mind that CODs are operational and by deciding to use one dataset over another the IM Network is not engaging in a political discussion but is simply determining which dataset is most useful during a humanitarian response.


2. Collect
Acquisition and evaluation of datasets

The collection phase involves the acquisition and evaluation of datasets to identify the dataset that will become the COD. This phase also includes negotiating licenses, as limited sharing possibilities may influence the dataset that is chosen. Metadata should be collected at this point and further developed throughout the COD cycle. Acquisition can be done by OCHA or another organization - ideally the organization with the closest and most sustainable relationship with the data source. It is likely to be OCHA in many cases but when OCHA is not present in a country an alternative data Guardian or Sponsor may be better placed to collect the data from the source (see COD Governance for more information). Consider contacting the HC or RC if a course is reluctant to share a COD (the HC/RC will be part of the COD cycle as they endorse the list of CODs).

The evaluation involves a quick quality assurance check in-country to ensure compliance with minimum standards of data characteristics (spatial and attribute) and metadata. The process identifies potential problems or opportunities with datasets that should be considered when deciding what dataset should be the COD. The end result is a Provisional COD that can be shared (with appropriate metadata that it is being improved throughout the process).


3. Process
Processing and validating datasets

The processing step of the COD cycle involves improving the dataset through cleaning and standardizing so it is in the 'best available' standard possible. The standard should allow it to be used for multiple operational purposes (cartographic, frameworks, and analysis) and for corporate / system purposes (the later is especially true for administrative boundaries). This phase generally takes the longest; it produces a Candidate COD.

The validation step is a final check on the Candidate COD. It includes technical review to make sure the corrections made in the processing phase are agreed to. This phase involves explain to partners what changes took place in the processing to assist in the next phase of the COD cycle.

4. Endorse
Endorsement of official COD (consider RO and CO difference)

The endorsement phase of CODs is the defining moment of the COD cycle as it is at this stage that the IM Network agrees that the provisional COD becomes the Endorsed COD for a country. Acceptance is also required by others in the country / region including: the RC/HC. A third endorsement is done by OCH-FIS who verifies metadata and does a quick and final check on the dataset.

5. Communicate
Sharing datasets and communicating

Sharing CODs in different formats, with appropriate metadata should occur as a preparedness activity or within the first 48 hours of an sudden on-set emergency though the distribution and management mechanisms available and appropriate for the country or context, taking into account national systems and capacities. The format(s) that are provided should satisfy different types of users. Metadata is as important as the datasets. Communication of new datasets / updates to the humanitarian community (local, regional and headquarters level) is critical to this phase especially as changes may impact partner's systems. Many different ways of communicating should be used including: IM network meetings, meetings notes, Skype messages, emails, advocacy material.


6. Maintain
Maintenance and Versioning datasets

Maintenance and versioning are an important steps in the COD cycle. The process of maintenance requires the COD to be examined to verify its relevance and current accuracy.If new datasets or changes are required then the COD cycle must start over again. Versioning helps manage datasets. When a new version is created, updates to metadata and changes to dataset name etc. inform users of the lifetime of the COD dataset. Curation on HDX of the old and new dataset is critical for COD discoverability.


The table below identifies the quality check, the COD term and where the dataset should be placed for each phase of HDX. The images below illustrate the same information visually.

<table to be inserted>


Tips

  • Include data trail (metadata throughout)
  • Make the process transparent
  • Include partners in each phase

Resources