Skip to content

adding options for filtering/export #168

@RiiaC

Description

@RiiaC

Question/Goal:

Can we give users the option to download only filtered data, or the filtered data plus any additional data associated with the sites (or possibly even nearby locations) that contain filtered data?

Background:

During the 2026-02-07 meeting in Uppsala, when Phil was doing a demo of the SEAD interface he made a comment that I jotted down in my notes as "We should have a checkbox to specify if you want overlapping data (e.g. same site), or only specific filtered data (e.g. just the things actually dated)".

The following week I discussed this with Johan, who replied:

If you really only want to get the data that you filtered on (and I think that's probably a good model that would be more intuitive to most people), then I think we need to start with reworking the filter system so that it uses a sample, rather than a site, as the basic unit. Then we could have everything in the result section also be samples instead of sites and then we could also refine the exports

To which I replied

If I understood it correctly, the thought was to be able to choose at the export level if you want to see only the kind of data that you just filtered for, or if, having found a list of sites with that sort of data, you want all of the data for those sites (which might be more useful for big picture questions) perhaps this should be a topic for discussion at a technical meeting, to better define what we would want. I wasn't clear when Phil made the comment if this is something he has long thought would be a good idea, or if it is something that just came to him as he was doing a demo.

To which Johan replied

Sorry I didn't explain this very well - I should start from the beginning. The problem is that the whole filtering system works on sites, which means that what I get into the result section (as the output from the filters) is a list of sites. So the client doesn't know what the user has filtered on - only that it resulted in this list of sites.
 Of course, the client could manually check in the UI what filters have been applied, and based on that perform an in-browser filtering of the data based on that, but that feels very wrong to me since we then have parallel filtering systems (server- and client-based). It would make far more sense if the server instead outputted to the client a more precise result. 
 Currently it's a list of sites, I think going to a list of samples would be a big improvement, but it might be worth discussing if it should be something else as well, such as a list of datasets. We just have to carefully consider what we want here and why we want it.

To discuss at the next SEAD Technical Meeting that isn't already earmarked for another discussion:

Is this a change we want to add? If so, what is the best way to go about it, and exactly what changes do we want to make, and why? If we can reach a consensus, then this idea can be added to an actual to-do list once it is better defined.

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions