site banner
ArjinFerman  Tinfoil Gigachad  1mo ago (text post)   524 thread views

Tinker Tuesday for September 2, 2025

This thread is for anyone working on personal projects to share their progress, and hold themselves somewhat accountable to a group of peers.

Post your project, your progress from last week, and what you hope to accomplish this week.

If you want to be pinged with a reminder asking about your project, let me know, and I'll harass you each week until you cancel the service

2
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

EDIT: I realize that I am not quite following the instructions for this forum. I thought of this is as more of a technical discussion forum and did not want to post this in the CW thread.

Coding agents.

I use Gemini and Claude. I pay $20 for Anthropic Pro, so I am not using API tokens for Claude. With Gemini, you can do a lot with just a google account, or at least you definitely could on day 1.

I am hating Gemini lately. If I load the Pro model (default), I get mostly nothing but 503 errors. This was not the case roughly a month ago, when Google was promising a Pro response every 60 seconds and "you won't run out of tokens" or whatever. Fine, it's free, I get it. So now I use gemini-2.5-flash mostly. But this motherfucker is constantly undermining me. If I can get flash into a groove, we do fine, but I fight this guy a lot. I also get a vibe of "petulant laborer". It kind of cracks me up.

Claude is great, super friendly and helpful. I almost exclusively use Sonnet, and Sonnet has a case of the dumbs. It's extremely manageable, and I get more work done with Claude than Gemini. I can switch to Opus, and I will run out of my Pro credits and then open a Gemini session.

When I get stuck, I go to chatgpt.com and copy/paste. This usually unsticks me or gives me a new direction. I see this for 3 reasons:

  1. a new model
  2. a web prompt is a more direct interface to the model than the coding agent
  3. OpenAI sauce

I guess I'll use the OpenAI cursor thing sooner or later. I am CLI-only, headless linux for dev. I have a graphical environment for browsing, etc.

I rarely have the agent make commits, and I have regretted every session where the phrase "vibe coding" popped into my head. I try to discuss more with the agents than make edits, but I do have them make a lot of edits. They are more edit-happy than I would like. The main value-add for the agent, for me, is just the read-access to my project. That they can edit files directly is nice, but also kinda scary and has gone wrong for me. I'm much more comfortable with code generation than code editing. I love these guys for code generation, but I always edit their code, which usually works but ugly. I ask for review more than generation more than edits.

I have read that Claude and/or Claude Code has gotten dumber lately, due to "quantization", which I think makes a lower resolution model that is cheaper to operate. So that may be why I feel like Sonnet has a case of the dumbs. I rarely interact with Opus or Gemini Pro these days, on the CLI.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats