‹ BackHN Continuity

Thread

MCP was always a bad idea?

335 points · 331 comments · maharshi365

  1. pjmlp · · focus · HN ↗
    Well, SaaS don't do CLIs for extension APIs.

    Plus the performance issues to restarting processes all the time.

    1. fooster · · focus · HN ↗
      Neither of those things are true?

      Also you are seriously comparing process startup time with the latency of a network call, or worse an llm call?

      1. pjmlp · · focus · HN ↗
        Yes, if the SaaS application needs to fork subprocesses for every little step of agentic workflows.

        Welcome to CGIs, a great 1990's technology for Web applications.

        1. fooster · · focus · HN ↗
          We do that, and the cost of forking the cli vs the llm or network call is very small. It works really well.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.