oh it is a nice version of the duck :3

pkia pika

I'm totally onboard with implementing that if you want, although I'd ideally like some way to show it on the detail page too and don't immediately see a clean way to do that.

Yeah, the only thing I can come up with is to only tie it to Pholio. So if a task has an attached Pholio Mock, it picks the cover image from that.

Specifically, I imagine the implementation would be:

Show A Large Picture of Psyduck on this Task's Workboard Card: [ Show Psyduck         V ]  (default)
                                                               [ Do Not Show Psyduck??? ]

My thinking is something like:

• Drag and drop an image on a card to upload + set it.
• If no one has dropped an image, pick a fallback if we can (attached mock, image in description?)
• Some advanced UI to explicitly set/remove the image for users with Linux Keyboard Interfaces who can not drag-and-drop or are on mobile and really need to set images or whatever.
• Would be nice to not just show up on the card? But I don't see how it can fit easily on the task.

I saw this on Subnautica the other day and it seemed sort of neat:

...but it would be silly to make a placeholder mock just for that, I think.

They're building a game so they have a lot of art, and like half of these images are irrelevant flavor, but the stuff like "model/implement an X object" with a picture of the thing is kind of neat:

https://trello.com/b/yxoJrFgP/subnautica-development

Overall I'd like to move Pholio v2 to be a Column UI as well, so it'd drop Pinboard UI for Cards.

I'd lean towards being non-aggressive in setting a cover image. Pholio Mock attached or Image dragged and dropped seem a good start. Maybe figure out how to get Cover Image editable via a normal task page....

Yeah, I don't particularly like fishing images out of the description either.

Only Pholio mocks seems like the shortest path, too.

I think it's actually easier to do drag-and-drop than any of the Pholio stuff. I don't want to do multiple rounds of querying (e.g., load mocks, load images, load files, load transforms) to load a workboard, and it's a lot easier to update caches on a task after an explicit drag-and-drop (we know exactly what we need to update) than after an attach/detach in Maniphest or Pholio edit.

(Although needing JS would normally instantly make an approach the worst ever, the JS for drag-and-drop files is in pretty good shape and doesn't need to do anything new to handle this.)