PunishedD on Nostr: binkle KaiserKitty :ipv4: Then you're back to the MVP problem. If you want any kind ...
binkle (nprofile…m5h7) KaiserKitty (nprofile…zwas) :ipv4: (nprofile…66f2) Then you're back to the MVP problem. If you want any kind of general acceptance, you gotta have HTML 2-5, robust Javascript/ECMAscript support, rock solid encryption support, and probably some basic file handling. It needs to achieve the Holy Grail of cross-platform deployment, obliterating all hardware discrepancies and dealing with OS rendering quirks. And it needs to be secure, sandboxed preferably.
What would you cut out of that to make it a manageable from-scratch project?
You can do some of those reasonably well on your own; text browsers like Lynx cut a lot, even they suck at non-visual things like encryption. But just cutting Javascript kills almost all of modern web development, even though it's one of the biggest implementation headache you'll face. Abandoning encryption is the same, almost all sites require HTTPS now for logins and e-commerce. You could save time by not making it fully cross-platform, but compared to everything else that doesn't seem like a huge saving.
(That's not even the full list of browser features--extensions, alternative protocols, developer/debug tools, text-based file reading, etc.)
Problems like this are what make people give up and experiment with "small web" protocols instead, like Gemini or Gopher or the resurgence in NNTP (Usenet).
What would you cut out of that to make it a manageable from-scratch project?
You can do some of those reasonably well on your own; text browsers like Lynx cut a lot, even they suck at non-visual things like encryption. But just cutting Javascript kills almost all of modern web development, even though it's one of the biggest implementation headache you'll face. Abandoning encryption is the same, almost all sites require HTTPS now for logins and e-commerce. You could save time by not making it fully cross-platform, but compared to everything else that doesn't seem like a huge saving.
(That's not even the full list of browser features--extensions, alternative protocols, developer/debug tools, text-based file reading, etc.)
Problems like this are what make people give up and experiment with "small web" protocols instead, like Gemini or Gopher or the resurgence in NNTP (Usenet).