• 0 Posts
  • 19 Comments
Joined 2 years ago
cake
Cake day: July 21st, 2023

help-circle

  • Depends. I made one in tauri and it was a good fit because I already had a web frontend and rust backend. I was able to reuse both of those with minor changes, now the same code builds the app and the web server/frontend.

    I’d probably go native if:

    • you’re only developing for android and don’t care about desktop/ios/etc…
    • UI performance is really important, like for a game
    • You want to minimize app size
    • You aren’t skilled at web front end development

    With tauri, if you need phone apis that aren’t in the toolkit already you’re going to jump through some hoops going from javascript to rust to kotlin and back again. Its a significant barrier, you’re handling serialization/deserialization of function arguments/results in 3 different languages.




  • I’ll add to this that when I first registered at mastodon.social, my default view was some fancied up multi-column monstrosity. I managed to turn most of that off, but it was super confusing at first. Start simple and let users add complexity.

    Also agree about the empty feed. At least default to the explore tab. Way better would be a tool to help build your feed based on interests. There are various tools for that, but a user shouldn’t have to leave the app and google around to discover them.

    A tool to read your phone contacts and search for corresponding mastodon accounts would be intrusive (some would even say unprincipled), but many users would welcome it.





  • I really like the idea of a grassroots Amazon competitor. That said,

    You need to have a high level of trust. A federated network of shady scams that just take your money and send you nothing half the time is not going to fly. Is there a vetting process, who controls that process, how’s all that work. If its ‘good seller’ reviews, how are those stats protected from manipulation.

    You need to have extreme ease of use. UI barriers that seem trivial to developers can sink a platform.

    If there are problems solvable by centralization, maybe that could be done as a cooperative organization which devs and vendors can join and run democratically.










  • I’m intrigued by a tag based rather than subcommunity based system. I have a personal wiki system I use that is tag based, and its great since most notes fit in multiple categories. Different from a social board though.

    Re images. A giant image, or an animated gif in the middle of a text discussion can be disruptive. The same with things like fancy user sigs that are 10x the size of the comment they posted. Do you have thoughts on how to limit this kind of distraction?