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
In the previous cycle we scoped the data grid to a view.
This is a ticket to explore design prototype before proceeding to do the dev work:
to reduce the ambiguity and the guess work during development
to identified potential issues and explore nuances of the proposed direction
to be able to think holistically and then split into coherent smaller iterations when necessary
Starting with:
folder-like-view_zoom-in.mov
where the main idea was to 'travel' through the data like to through folders.
✅ good to have a 'clean' and focused view
✅ supports potentially infinite hierarchy of components
❌ limited bird view: not possible to see the 'batch values' or component values at a glance
❌ 'filters' from one view could conflict with 'filters' in the 'zoomed-in' view (potentially confusing interaction, needs more exploration)
Result:
Instead of pursuing the zoom-in folder-like interaction, we want to go towards another approach, let's call it all-in-one view or column-expand (when talking about zooming-in to a component).
✅ it allows for all-in-one view exploration (zoom out bird view on your data)
✅ more familiar interaction pattern for filter/sorting/group by and all in 1 view
❌ clunky data density when including many components (like columns with empty rows). Which to be fair is also a type of information (like no data) and component columns could easily be filtered out if needed.
❌ high number of the entity/column expands could create usability issues (it would require a more sophisticated styling to help distinguish all the hierarchy visually)
Prototype examples:
explore-tableView.movfilter-tableView.mov
Notes:
how this will be coupled with the timeline? like this:
tree table is still a thing? wouldn't call it a tree table anymore, but some sort of 'merging' similar values would help a lot to scan through the table. Example:
any other properties of the table view? i.e. grouped by? or show/hide instance components? For sure!
Group by:
Show/hide (WIP):
puting archetypes out of scope for now? Actually no! We want to group columns by archetypes, because it makes a lot of sense with our data model. Something like this:
In the previous cycle we scoped the data grid to a view.
This is a ticket to explore design prototype before proceeding to do the dev work:
Starting with:
folder-like-view_zoom-in.mov
where the main idea was to 'travel' through the data like to through folders.
✅ good to have a 'clean' and focused view
✅ supports potentially infinite hierarchy of components
❌ limited bird view: not possible to see the 'batch values' or component values at a glance
❌ 'filters' from one view could conflict with 'filters' in the 'zoomed-in' view (potentially confusing interaction, needs more exploration)
Result:
Instead of pursuing the zoom-in folder-like interaction, we want to go towards another approach, let's call it all-in-one view or column-expand (when talking about zooming-in to a component).
✅ it allows for all-in-one view exploration (zoom out bird view on your data)
✅ more familiar interaction pattern for filter/sorting/group by and all in 1 view
❌ clunky data density when including many components (like columns with empty rows). Which to be fair is also a type of information (like no data) and component columns could easily be filtered out if needed.
❌ high number of the entity/column expands could create usability issues (it would require a more sophisticated styling to help distinguish all the hierarchy visually)
Prototype examples:
explore-tableView.mov
filter-tableView.mov
Notes:
Group by:
Show/hide (WIP):
Resources:
Summary in figma
Next?
To split into tickets together with @abey79
The text was updated successfully, but these errors were encountered: