Aidan Heerdegen

I'm not sure how useful a lot of these ideas are any longer, thought it was a good idea I had to add a `model` field in the DB, that...

Yeah I wasn't sure it would be an error so much as not something we're explicitly expecting/supporting. I do know the explorer doesn't handle it though, which is a separate...

The explorer displays the metadata from the "first" experiment when there is more than one experiment with the same name. This is less than ideal. It places a support burden...

If the same variables exist in different files there needs to be some way to disambiguate them. In some cases they may have different time resolution, so can be differentiated...

The old way of disambiguating variables was to specify the filename, e.g. `ocean.nc`, `ocean_month.nc`. These have no informational value, are completely arbitrary and unintuitive for those not familiar with how...

For a well constructed experiment I agree @jmunroe, however there is no restriction on saving the same variable with the same frequency in two different files. I'm not advocating we...

It is possible to have pretty much anything. It may be that we need to define what we would regard as a well designed experiment, and that is that cosima...

Good point @jmunroe. I agree that `cell_methods` is required for disambiguation in that case. I guess it needs to be added to the DB @angus-g

I am taking a look at this as I need it for data discovery purposes.

I gave this a crack. Doesn't seem to make much difference: ``` %%time cc.querying.getvar('01deg_jra55v140_iaf', 'salt_global_ave', session, start_time='1958-01-02 00:00:00', end_time='2019-01-01 00:00:00', frequency='1 daily') CPU times: user 36.4 s, sys: 40.2 s,...