December VTC: notes and comments

IMC VTC Monday, December 10 and Tuesday, December 11, 2012

Monday's comments:

It would be helpful if any reports produced included metrics on dataset revisions not just newly published items.
Is there a mechanism by which we can capture site data not loaded in NIS?
Report metrics will need to change over time as the NIS develops.
All relevant metrics should be included in the site level report.
A report containing site specific data might be useful for site reviewers.
Reports should include revisions (yes or no), increase in sizes, "inserts and updates".
Reports should recognize and report on changes in long term datasets.
Any reporting mechanism should include an API for extracting customized data.

Tuesday's comments:

The report mock-ups contain too many metrics.
It would be useful if reports included download activity.
Deficiencies in the quality checker will have a knock-on effect on the reports. These errors will then aggregate into the higher level reports.
Warnings are more likely to be erroneous and should be treated accordingly.
Warnings should not be reported beyond site level report.
Each metric should have an explanation that puts the metric in context, especially with respect to accuracy.
It might be useful to have a headcount of data in Metacat - simple metrics % of data in metacat that is in PASTA
The PASTA-centric approach is more forward looking compared to the Metacat-centric approach.