You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm not sure what the search API does when we would like to select options under 1 search facet and some of the datasets being searched haven't defined a value for it. For example, suppose we're searching across CMIP6 and obs4MIPs and we want to limit the CMIP6 search to experiment = "historical", and for both projects we want to limit the search to variable="tas" and Time Frequency = "mon". As I understand it, at present if you select experiment = "historical", you will automatically eliminate all the obs4MIP data sets, since they don't include an "experiment" attribute.
I think users would prefer that if a search facet is "missing" from a data set, then it will be assumed to be satisified (i.e., not filtered out). Perhaps it would be even clearer to users if any search facet that is "missing" were assigned the value "unknown" or "N/A". So, if someone wanted to select all datasets where an attribute is undefined, they could select "unknown" along with other values of interest. So in the example above of searching models and obs (including selection of experiment), the user would select 2 items in the list of experiments: "historical" and "unknown" (or "N/A"). They would obtain both observations and all "historical" experiment results.
Do you agree that this would be a good idea? How difficult would it be to implement?
The text was updated successfully, but these errors were encountered:
I'm not sure what the search API does when we would like to select options under 1 search facet and some of the datasets being searched haven't defined a value for it. For example, suppose we're searching across CMIP6 and obs4MIPs and we want to limit the CMIP6 search to experiment = "historical", and for both projects we want to limit the search to variable="tas" and Time Frequency = "mon". As I understand it, at present if you select experiment = "historical", you will automatically eliminate all the obs4MIP data sets, since they don't include an "experiment" attribute.
I think users would prefer that if a search facet is "missing" from a data set, then it will be assumed to be satisified (i.e., not filtered out). Perhaps it would be even clearer to users if any search facet that is "missing" were assigned the value "unknown" or "N/A". So, if someone wanted to select all datasets where an attribute is undefined, they could select "unknown" along with other values of interest. So in the example above of searching models and obs (including selection of experiment), the user would select 2 items in the list of experiments: "historical" and "unknown" (or "N/A"). They would obtain both observations and all "historical" experiment results.
Do you agree that this would be a good idea? How difficult would it be to implement?
The text was updated successfully, but these errors were encountered: