Closed joshmoore closed 5 days ago
I thought about "Movement of the ROI updates the cropped view" before, but couldn't think how to store the link since each panel doesn't have a unique ID. It would need quite big code changes to allow communication between panels (currently the UI components for each panel only listen to changes in the corresponding panelModel).
Certainly the "create a new panel from Rectangle" should be easy enough and be useful. But sometimes you'll create the 2 panels first, before drawing the ROI (as in the e.g. above).
Drawing connecting lines could be done manually if we add support for drawing lines and labels on the page itself (long requested) but they wouldn't be "smart / linked". That would be much harder!
User adds a potentially ephemeral ROI to an image Likely from the ROI edit dialog (since ROIs aren't clickable in the main UI) or potentially from the "Add labels" dialog, user chooses "Use as inset"
The two above points signal probably that the ROIs are too "well" hidden functionality, with some workflow caveats. This is also obvious when you try to add ROIs "from OMERO" from a long list. I guess @jburel was suggesting a revisit of the ROI workflows.
Movement of the ROI updates the cropped view
but, @joshmoore , as you said, the ROI cannot be moved or manipulated from the "main UI". This either means that the ROI workflow should be rehauled (see above) or you are looking at a laborious "jumping" between Edit ROI dialog (a specific dialog which edits only one image at a time atm) and the "main UI". This would in fact invalidate the workflow as suggested.
Or, somehow, we would have to introduce "ephemeral ROIs" which are editable from the main UI. Is that an easy thing to do ? In such case, why not to expand the "ephemeral ROIs" to be the "true Figure-ROIs" and ditch the clanky "ROI Edit" dialog (e.g. rehaul the ROI concepts in Figure, see above)
@joshmoore Actually, partially, your workflow can be achieved in iviewer.
I do not suggest that this is a replacement of your workflow, rather wish to start a discussion to understand better your intended workflow.
Could you maybe add a motivation of your workflow to let us better understand the purpose ? Is it IDR motivated ? Public user viewing figures ?
This is the recent drawing that I was producing that I would have liked to have figure-ified:
So not particularly IDR-centric nor public data centric. Just "OMERO users wants to easily show insets" (or "zoom-levels" if you wish)
Aha, thank you. Maybe the "interactivity" is not so crucial for your workflow then ? I was almost thinking about "magnifying glass" type workflow, but possibly this is not so crucial (although you probably want to adjust "a bit" after you have once drawn the ROI). But I guess this is a false hope, you prbably need it all, including the connecting lines, because your images are large and zomplex and multi-z on top of that, so you cannot allow for too cumbersome jumping between dialogs.
So I guess you finally used the Copy and Paste in combination with the "zoom in" in the pasted panel and/or the "Crop" button in Figure to get the desired "insert" ? And then added the lines in another software ?
Maybe the "interactivity" is not so crucial for your workflow then ?
I think it is. This is exactly what happened to me. Late in the process the choice of which tile to "inset" was changed.
"magnifying glass" type workflow
I think this is a good way of thinking about it. I'm happy for this to be disconnected from "ROI" (though it may all be the same under the hood)
And then added the lines in another software ?
Yes. It was all done in the end in Google Drawings.
The linking-lines feature has been requested at https://forum.image.sc/t/omero-figure-improvements/72158
This issue has been mentioned on Image.sc Forum. There might be relevant details there:
e.g. https://omero.lifesci.dundee.ac.uk/figure/file/364219
cc: @pwalczysko @jburel