Their has to be a service where u can do https://somemagicservice/<the original post url> that then redirects to the original post and also provides an image of the post as thumbnail?
Wait what happens if u just post the og link directly to lemmy.
I’ve seen Mastodon image posts also appear on Lemmy, as image posts.
But if they upload multiple images to Mastodon, only the first one makes it here. And they can reply to comments on their post, but any images in said reply don’t appear.
I dunno, maybe. I’ve never really looked in to the ActivityPub protocol. I have heard accusations of Mastodon having an extremely non-standard implementation of it though, so it’s probably a bit of both.
Yea, it probably could be best solved through cooperation but the thing that springs to mind for me is that card responses for link unfurling have a standard way of presenting an image, or text, or an image embedded in text - but there’s no standard (even an unofficial standard) way to clearly respond that the link unfurling should be multiple images in an arbitrary gallery ordering. If multiple images are returned in a card the context reads like a blob of text with multiple images in series.
Link to the post
Their has to be a service where u can do https://somemagicservice/<the original post url> that then redirects to the original post and also provides an image of the post as thumbnail?
Wait what happens if u just post the og link directly to lemmy.
Most Mastodon posts can’t really be read by lemmy, but if a Mastodon user posts to a community it usually appears as a text post.
I’ve seen Mastodon image posts also appear on Lemmy, as image posts.
But if they upload multiple images to Mastodon, only the first one makes it here. And they can reply to comments on their post, but any images in said reply don’t appear.
This feels like a lemmy-side issue and potentially one that would be solved by the rendering/app layer.
I dunno, maybe. I’ve never really looked in to the ActivityPub protocol. I have heard accusations of Mastodon having an extremely non-standard implementation of it though, so it’s probably a bit of both.
Yea, it probably could be best solved through cooperation but the thing that springs to mind for me is that card responses for link unfurling have a standard way of presenting an image, or text, or an image embedded in text - but there’s no standard (even an unofficial standard) way to clearly respond that the link unfurling should be multiple images in an arbitrary gallery ordering. If multiple images are returned in a card the context reads like a blob of text with multiple images in series.