all 9 comments

sorted by: hot top controversial new old
[–] 20 points 1 year ago (6 children)

This is what happens when we remove the science from CS and start vibing.

  • source
  • hideshow 6 child comments
  • [–] [S] 15 points 1 year ago* (5 children)

    I don't think it has anything to do with vibe coding, the problem with MCP is just it needed to be done fast (because otherwise another protocol would appear and get all the hype) and needed to be very easy to use for hobbyists (again: to get hype, Enterprise software is boring)

  • source
  • parent
  • hideshow 5 child comments
  • [–] 10 points 1 year ago* (2 children)

    What you're describing is definitely the AI vibe: full speed at the cost of quality, logic, and all else.

    To (quite badly) paraphrase Woody Allen:

    "The quality of this code is so bad."

    "Well, at least there's a lot of it."

  • source
  • parent
  • hideshow 2 child comments
  • [–] 1 point 1 year ago (1 child)
  • [–] 8 points 1 year ago

    Yeah the principal engineer at my company is just smashing his keyboard every day to implement MCP services and shit and in the meantime is fucking up our repo and I’m trying to clean up the mess but all upper management cares about is “when AI?”

    He’s not vibe coding it, but as with all the MCP specs, it’s ignoring many things about type safety and the like pointed out in this article

  • source
  • parent
  • [–] 7 points 1 year ago

    Why did we need another competing web API standard?

  • source
  • [–] 5 points 1 year ago

    I’m just here for the Seventh Seal picture. Y’all can ignore me.

  • source
  • 2 comments in c/technology@lemmy.world view that post
    [–] 6 points 8 months ago

    Have kind of given up on MCP and just add skills/prompts on how to interact with the API.

  • source
  • [–] 1 point 8 months ago

    MCPs could learn a thing or two from the failures of ActiveX.

    Will they?

    No...

  • source