sanarothe
Joined 14 years ago
Joined 14 years ago
If you give it a task, it completes it. It might so a single agent chug, or delegate some parallel tasks..
The architecture is essentially what was agreed on during the planning and spec phases.
So I'm just not understanding like... The codebot produces everything I ask, I use Gemini all day every day and I don't hit limits.
I see that if I were trying to code 10 hours a day the limits might be a problem but... This is a personal plan.
So. When people say that Gemini sucks, or they're hitting limits, I'm not understanding what problems they're encountering...
I guess if you're letting it vibe code huge chunks. I'm doing mostly handwritten code for my current project with a little bit of "I don't want to deal with this, Claude can handle it" and I've spent $1.26 this month for my 446 lines of code.
But yes I suppose at that rate, if Gastown or Beads or whatever is 300,000 lines of code (just to use a project known to be fully vibe coded with rough LOC reported), that would be over $800.
Don't let it vibe code hundreds of thousands of lines of code I guess.
What Dutch OSINT Guy was saying here resonates with me for sure - the act of taking a blurry image into the photo editing software, the use of the manipulation tools, there seems to be something about those little acts that are an essential piece of thinking through a problem.
I'm making a process flow map for the manufacturing line we're standing up for a new product. I already have a process flow from the contract manufacturer but that's only helpful as reference. To understand the process, I gotta spend the time writing out the subassemblies in Visio, putting little reference pictures of the drawings next to the block, putting the care into linking the connections and putting things in order.
Ideas and questions seem to come out from those little spaces. Maybe it's just letting our subconscious a chance to speak finally hah.
L.M. Sacasas writes a lot about this from a 'spirit' point of view on [The Convivial Society](https://theconvivialsociety.substack.com/) - that the little moments of rote work - putting the dishes away, weeding the garden, the walking of the dog, these are all essential part of life. Taking care of the mundane is living, and we must attend to them with care and gratitude.
https://github.com/AntonOsika/gpt-engineer
I believe I have identified the culprit.
"Please don't generate or rewrite code, I just want to discuss the general approach."
Bc I don't know any design patterns or idiomatic approach, being able to discuss is amazing.
Though quality and consistency of responses is another thing... :)
On the other hand, I tried to use powershell the other day for some simple admin tasks and ooooh wow I sure don't have anything nice to say.
Also my living is made on the Windows side, and not the Linux side, so I guess I don't know the pain points of WSL.
We had a 'intro engineering' class that taught orthographic drawing (by hand, about a decade ago, so thankful for it) which included puzzles on what the various 2D views should be given a 3d object, and the opposite direction. Holy crap those were so hard for me, but picked it up with practice.
I have visio open constantly at work. Lots of mind maps, flow diagrams. Take lots of pictures on vacation because the visual memories are basically gone as soon as my body is.
The 'language is a tool for communication' thread from today has some discussion on different types of thinking. For example it was inconceivable to me that someone would visualize words in their head, but I guess that's a thing for some people.
I do have to ask people to draw me a picture occasionally when they're trying to describe something. Don't need a lot of detail, just a rough sketch and I can figure out the idea.
I guess it's better than when we literally injected a sample extract into a bunny to see if he gets a fever. (The old test before LAL)
(We also use the LAL test for bacterial endotoxins in med device for implantables