1168
I'm In Danger! (media.piefed.zip)
submitted 9 months ago by to c/programmer_humor@programming.dev
73 comments fedilink hide all child comments
 
[Image of a book from O'Reilly with the title: Vibe Coding. I'm a Developer Now. Followed by an image of Ralph Wiggim at a laptop.]
you are viewing a single comment's thread
view the rest of the comments
[–] 68 points 9 months ago (12 children)

I saw this comment a while ago, and I still stand by it: If vibe coding works, where are all the "million dollar idea" apps?

  • permalink
  • fedilink
  • source
  • hideshow 12 child comments
    •
  • [–] 6 points 9 months ago

    I work in municipal development and we have 20 new "developers" a week trying to get us to buy their permitting apps. All of them are willing to offer us an exclusive discount as an early adopter, and the few I've had meetings with haven't even been able to tell us what backend databases the apps use or understand that there's a difference between an Amending Plat, Site Development permit, and a Building permit.

    And I have to fight the mayor every time because he's all aboard with the AI hype. He tried having all the city ordinances and decelopment manuals re-written by GPT to make them easier to understand, and we had to get the city attorney to explain that not only was it idiotic, but that it would cost a couple hundred grand just to have his firm go over everything and explain the specifics of how dumb it was, and that if a code re-write is needed (and it is), they should spend that money hiring a firm specializing in code review.

    The slop apps are out there - they're just all being pitched to governments and CEOs that have infinite faith in anything that will make people more expendable.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 5 points 9 months ago* (3 children)

    Vibe coding only works for "I need a one time use python script that processes this data and I can't be assed to spend 2 hours remebering how to use Pandas right now."

    Not for million dollar apps.

    Of all the AI use cases, coding is honestly the one that makes sense since like 90% of coding is just copying some other code and massaging it into place. Even as a "Vibe Coder" you kind of still need to know the idea of what code you need and what you want it to do.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 3 child comments
    •
  • [–] 1 point 9 months ago (2 children)

    Even as a “Vibe Coder” you kind of still need to know the idea of what code you need and what you want it to do.

    I've fiddled with it a bit. IM(limited)E, you tend to get the best results if what you tell it you want broadly, and then in a significant level of detail, and especially if you tell it to ask you any questions it has about the design. So something like "I want you to build an app in $LANGUAGE that does $TASK on $PLATFORM. As I see it, I think the interface should look like $DETAILED_DESCRIPTION_OF_INTERFACE and here's what should happen when those elements are interacted with,,, Please ask as many questions as you think are necessary for clarity."

    Also, unless you're setting it to require personal approval for every terminal command, I'd only run one of those in a VM of some kind, where any potential damage from any potential fuckups are limited.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 9 months ago

    I mostly use the web version of Claude. But generally my approach has been to start simple and build bigger.

    "I want a website that does X."

    "Can you add a dark mode toggle."

    "I don'tlike those colors, can you make it more blue."

    "Can you add a log in function with a simple database backend that doesn't use a full SQL install. Users should be able to add a profile image and username."

    "The first user should be an admin and can add or remove other admins."

    "Admins can change the theme colors on a settings page."

    That sort of thing. Start basic, add features, usually one by one, test as you go, see what breaks, if anything.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 1 point 8 months ago

    I feel that as a system analyst I'm pretty well placed to vibe code, as my day job is explaining business requirements in precise enough language that programmers can implement the changes

  • permalink
  • fedilink
  • source
  • parent
  • [–] -4 points 9 months ago (5 children)

    Security issues aside, HAVING the idea is the hardest part, much harder than coding an app. If you have a genuinely good idea, it's pretty easy to find someone to help build it, because that person also wants in on a cash-cow. Lack of ideas is the bigger problem with launching apps today, which a million vibe coders have realized.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 5 child comments
    •
  • [–] 24 points 9 months ago

    Absolutle bullshit. Ideas cost nothing. Implementing something well and marjeting it and building a viable, substainable buisness around it is still insanely hard. Vibe coding just lets you get an MVP THAT YOU DONT UNDERSTAND to market. It doesnt do any of the actual work of the above.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 1 point 9 months ago* (last edited 9 months ago)

    having a good video game idea is the hardest part, which is why every single video game these days are completely unique ideas that didn't steal at all from D&D. Hit points? Hit chance? Random damage? Class archetypes? Character Leveling and advancement? Go into dungeon, loot, avoid enemies, extract and return to safety as a play pattern? Never repeated in the AAA Diablo or Elder Scrolls games or indie games like lethal company or in an FPS like escape from tarkov. It was actually those unique ideas that were never copied that were harder than the bajillions of dollars spent on making them with code and artists. That unique, never-copied D&D idea cost jabillions and bajillions of dollars and time etc actually, much more than the billions spent on development really.

    Baldurs Gate 3 didn't just use the D&D rules and mechanics, it was a complete 100% overhaul of completely new ideas and zero re-hashes of something played in the 70s. It was because of these new innovations that was the hardest part of BG3 and all other games totally not derived of a 70s game people played on their kitchen tables that got jobs in the video game industry

  • permalink
  • fedilink
  • source
  • parent