1. Agenda

1. Review of the actions from the last meeting
2. Review of the KPIs
3. Workplan in view of the next Telco (in October)

2. Salient Points

The team reviewed the different existing KPI points:

A. Compliance to WMO Core Profile and its mandatory elements

The existing inputs needs to be reviewed and more work needs to be done to derive the KPIs from the rubrik checker. Frederic and Herman are going to tackle that task.

B. Good quality title C.Good quality abstract

These two elements have been provided by Anna and reflect well what can be done. As she suggested in order to improve the title and abstract quality a minimum and maximum length should be checked.

The NOAA guidance should also published along side the KPI monitor.

In addition the group decide to check on the possibility to define a set of dictionary of term that should be present for increase the overall quality scoring. Which dictionary should be used (WMO Code registry, Gemet Thesaurus, an adhoc dictionnary ?).

D. Valid Station Identifier and M.

To be checked by Kate

E. Bulletin metadata availability

The group discussed the time information that needs to be checked and present in the metadata records: A list of the necessary time informatin needs to be established and it needs to be checked if it is present in the WMCP standard and mandatory ?

• Data publication frequency
• .... TBD

F. Link for WMOEssential data

This part describe the presence off an online resource for WMO Essential data. It should be verified and always present. In addition the group indicated that it should always be present for all datasets and it should be a valid link.

G. Bulletins have an ongoing temporal extent

Touching on the time KPI, to be checked with Kate.

H. Data Parameter Keyword

The presence of the Data Parameter is recommended but not mandatory. Its presence should be verified and the guidance examples did not seem to be right. From which thesaurus should the parameter keyword list derived.

Marta is going to continue the analysis for the next meeting.

I. GraphicOverview for non bulletins metadata records

This KPI is straight forward and complete. It should be added in the list of the monitor.

J. Broken links

Same Broken links should be continuously verified for all metadata records and contribute to the individual score.

K. Data Policy Possible combination

KPI needs to be created to evaluate the quality of the information regarding the DPolicies. Marta and Jan are going to work on establishing from the guidance a matrix of the possible combinations and what ISO XML elements should be filled depending on the chosen data policy. From that point it can be evaluated how to create KPIs.

3. Actions

The following actions for the next meeting have been defined:

• [Herman, Frederic] - Work on A. as described in Salient points.
• [Anna] - Enhance B & C and investigate how built dictionaries could be used to define a score.
• [Kate] - Define KPIs for the station ID and derive KPIs about the time information as it is linked to metadata availability time and publication time
• [Marta, Manuel] - Define the list of potential thesaurus and propose a thesaurus (or a solution for building a thesaurus) for the Data Parameter keyword
• [Marta, Jan, Manuel] - Define a matrix of all the possible DPolicies combinations to then derived KPIs
• [Guillaume] - Derive additional KPIs from the guidance.