I'm mocking up an iOS Bluesky client and I want to allow users to set a background image for their profile/feed that's visible to others, by uploading a blob and referencing it under a lexicon. I was looking to see if there was an existing lexicon for this and I've not been able to find one, but please let me know if there is one. This feels like a distinct image role that could be standardised in a way that could be useful to other clients too. Banner images are much too wide to repurpose like this and I think a portrait centre slice might just be too random if someone has a banner that's text or a group of people.
I don't think there's a universal aspect ratio that would be suitable for a background on multiple shapes of device. Obviously, that couldn't actually be constrained by a lexicon, but if it's for broader adoption, it would make sense to have some way it's used, so I’m thinking it would make sense to include minimal metadata that helps clients render the image consistently to the user's intent. I'm thinking a good compromise for all devices would be a normalised point of interest coordinate in the lexicon which the client attempts to centre, while still keeping all the image edges outside or touching the bounds of the container (a 'filled' image / "background-size: cover"). If you had a landscape photo with a mountain right on its right edge, you might want the mountain to be the central background in your mobile profile, but you want the full image for a desktop web browser rather than having to crop the image for the portrait background and then having your desktop profile's background be a landscape crop of your existing portrait crop. Instead, the background would just be the reference to the image blob and something like a (0.9, 0.5) point of interest coordinate. Does this approach seem appropriate?
I'm mocking up an iOS Bluesky client and I want to allow users to set a background image for their profile/feed that's visible to others, by uploading a blob and referencing it under a lexicon. I was looking to see if there was an existing lexicon for this and I've not been able to find one, but please let me know if there is one. This feels like a distinct image role that could be standardised in a way that could be useful to other clients too. Banner images are much too wide to repurpose like this and I think a portrait centre slice might just be too random if someone has a banner that's text or a group of people.
I don't think there's a universal aspect ratio that would be suitable for a background on multiple shapes of device. Obviously, that couldn't actually be constrained by a lexicon, but if it's for broader adoption, it would make sense to have some way it's used, so I’m thinking it would make sense to include minimal metadata that helps clients render the image consistently to the user's intent. I'm thinking a good compromise for all devices would be a normalised point of interest coordinate in the lexicon which the client attempts to centre, while still keeping all the image edges outside or touching the bounds of the container (a 'filled' image / "background-size: cover"). If you had a landscape photo with a mountain right on its right edge, you might want the mountain to be the central background in your mobile profile, but you want the full image for a desktop web browser rather than having to crop the image for the portrait background and then having your desktop profile's background be a landscape crop of your existing portrait crop. Instead, the background would just be the reference to the image blob and something like a (0.9, 0.5) point of interest coordinate. Does this approach seem appropriate?