@thisismissem truth be told, most of the time I just try to make something that "just works" with mastodon, *oma, and key. Also with gts and some lemmy hacks.
The first feature I am trying to push to some people is the ![media-1] thing I am doing for inline images, and I just published the feature a few days ago (and I am still fixing some frontend bugs on that)
Wafrn started as a small project I did because I was angry at tumblr for a bunch of reasons . This happened 3 years ago. I made wafrn just as a tumblr clone. Then I see the twitter thing by elon musk and decided to look into "conecting wafrn with mastodon". After a few months I got a first version working (almost two years ago at this point)
A lot of the features I do myself, like asks or the ![media-666] are done in ways that can be described as a compromise:
Sending a DM with the format should make it an ask (BUGGY AT THE MOMENT, SOMETHING SOMETHING MYSQL TO POSTGRES MIGRATION). It would make more sense to make it another way, but then a mastodon user could not send me an ask. So I do it this way, a compromise.
Same with inline images. Akkoma sends the image as html itself. Thats a problem because mastodon and any good software will sanitize it and remove it completely. With the media tag, a human could say "oh I see, I think this is referencing the first picture". Also, tecnicaly you could attach an image to a post of yours and you could use the format. Your instance wont show it but in wafrn it will be shown (buggy, I have reimplemented it and Ill fix it after work today)
I can do stuff because wafn doesnt have that many users, and a mistake in my software "wont affect too much to too many people". But if mastodon has a typo, oh my god millions will see it
#yeah-i-may-be-moving-fast-and-breaking-stuff-a-bit #I-am-aware-of-that