4 ms·

bluesky dev here. whoops. as others mentioned, not a hard fix.

• kodah 3y ago
  Hilarious though! I'm guessing this is the kind of stuff the Beta was supposed to find. Any other cool/funny bugs y'all have found?
  • pfraze 3y ago
    I created a rich text system for posts to handle things like links and mentions with the eventual goal of it being the basis for all kinds of rich text (bolding, italics, spoiler tags, etc) the flexibility bit us on the butt. people started faking mentions via the APIs and one user figured out he could pack 1000 mentions into one "@everyone" and cause us all to get notified. pretty predictable in hindsight but I dropped the ball there
    • jesprenj 3y ago
      Any sufficiently complicated C or Fortran program contains an ad hoc, informally-specified, bug-ridden, slow implementation of half of Common Lisp. http://en.wikipedia.org/wiki/Greenspun%27s_tenth_rule http://en.wikipedia.org/wiki/Greenspun%27s_tenth_rule
• gowld 3y ago
  Do you have a public test suite so we can see how solid your protocol is?
• medler 3y ago
  What is the easy fix? Use the .well-known/ standard instead of the current mechanisms, and roll back verification for anyone who’s already been verified with the flawed approach?
• mempko 3y ago
  [flagged]
  • steveklabnik 3y ago
    (not paul) Given that the AT Protocol is not compatible with ActivityPub, I don't see how the first step, let alone the second or third, is an accurate description of the dynamics at play here. > And also how long until Twitter 2.0 or X app uses AT Ironically, pre elon-buys-twitter, it did in fact look like Twitter was going to end up implementing AT, if all went to plan. But then post acquisition, if anything, that looks less likely.
• didip 3y ago
  Can I ask what's your tech stack?