readitalready

8 ms·

1.
by readitalready 4mo ago
How do you think the tests were generated? You don't actually think I look at the code, do you?
2.
by readitalready 4mo ago
I don't micromanage it. I let my projects custom linter micromanage it. Every project should have a custom linter for their tech stack. It would check for not just syntax errors, but architectural choices as well as taste guidelines. W
3.
by readitalready 4mo ago
I let agents break things 30 changes down the line. If something breaks, I add a check to my project validator and start over, with the validator providing instructions on what was wrong and how to fix it. It's all automatic, and now I
4.
by readitalready 4mo ago
This flow is really for LLM consumption, since Markdown spec documents are for LLMs anyways. And you can always write a JSON-to-markdown converter for human use (actually, LLMs remember Markdown content better than JSON, so you should use t
5.
by readitalready 4mo ago
I moved from Markdown to JSON for all spec writing about 9 months ago. Although not HTML, it still has the same benefits. Claude and the other models are just so much more reliable in a structured format like JSON/HTML/XML. The mo
6.
by readitalready 4mo ago
Not for me. 2 Claude Max 20x accounts here both at high usage on weekly allotments.
7.
by readitalready 4mo ago
This probably started because of Andrej Karpathy's complaint about deployment being more painful than coding itself.
8.
by readitalready 4mo ago
The point isn't to save money. The point is to make money. And all those companies that are failing because of their fabs, they MADE money because of the fabs. Intel is now on its way to become another AI chip company. They have no inc
9.
by readitalready 4mo ago
Your custom linters don't check architectural design? linters statically check code and provide deterministic recommendations. LLMs are used to make judgement. I specifically write my linters for my project to make recommendations for
10.
by readitalready 5mo ago
No fixing the mess definitely does not take longer than writing it oneself. Your linter should identify all issues - including architectural and stylistic choices - and the AI agents will immediately repair them. It's about 1000x faste
11.
by readitalready 5mo ago
Come on Apple at least start the process of building your own fabs. License some of the processes if you must. Otherwise you will always be at the mercy of fabs that are more focused on AI.
12.
by readitalready 5mo ago
I think companies are charged API prices vs individual prices. That alone is 10x for Anthropic. Not sure though.
13.
by readitalready 5mo ago
No, you CAN'T raise prices because no one will buy them, destroying your brand and your business. And you won't have supply anyways. There is no scenario where not having a fab is beneficial. Apple will have to build their own fab
14.
by readitalready 5mo ago
OMM 0910 is the AI Jesus. We've literally seen this movie before.
15.
by readitalready 5mo ago
Apple doesn't own TSMC. TSMC is putting Apple in the back burner in favor of Nvidia and other high margin companies. They are literally being limited by TSMC (as well as the DRAM makers) because they don't have their own fabs.
16.
by readitalready 5mo ago
You can easily end up like AMD or Intel spending years with your own fab that’s uncompetitive. You could. Or you could have no fab and no supply of chips for your business.
17.
by readitalready 5mo ago
They really need to build their own fabs at this point. AI is going to kill their ASIC and DRAM supply chain if they don't. "Real men have fabs." - Jerry Sanders, first CEO of AMD. Actually, AMD, Nvidia, and Apple need to bui
18.
by readitalready 5mo ago
Took getting their asses handed to them in a war to realize this.
19.
by readitalready 5mo ago
lol no. LLMs are infinitely more capable than programmers. People really do think too highly of themselves.
20.
by readitalready 5mo ago
That's like saying "the programmers are so terrible you have to think ahead of them so they don't make mistakes".
21.
by readitalready 5mo ago
All the CAD and modelings tools have their own scripting languages that LLMs can write to, so you can just use that directly without any built-in LLM support. There will probably be someone doing a pelican-on-a-bicycle for CAD.
22.
by readitalready 5mo ago
It's because LLMs will soon start building real-world objects via CAD. This is the first step. Look at things like the Adam plugin for Onshape. Works great with Opus. It built a toy car for me with one prompt.
23.
by readitalready 5mo ago
No, I definitely wouldn't waste the 5 seconds it takes to unfollow someone. I'd rather spend that 5 seconds on the next interesting item instead.
24.
by readitalready 5mo ago
Because you don't want to see posts from stale connections.
25.
by readitalready 5mo ago
Social networks aren't rolodexes. They're newspapers.
26.
by readitalready 5mo ago
Because I don't want to see all her posts on my timeline anymore. I have to actively unfollow her to do that, which is more work. That's more work than even following someone, because it asks for confirmation or pops up a separate
27.
by readitalready 5mo ago
The signals aren't really working though. It's why algorithms are required, because people want relevance instead of popularity. Having fading connections equates relevance with popularity.
28.
by readitalready 5mo ago
Messages group chats are the circles now.
29.
by readitalready 5mo ago
I mean Kim & Selena will always have a certain level of celebrity status but people like Sydney Sweeney are currently a lot more popular. This is in terms of "are they the most popular people right now" as their instagram coun
30.
by readitalready 5mo ago
I really wish more social networks would have a "fading connections" limit. So many social networks suffer from stale connections and networks, and these connections should expire after a year. Otherwise, it will permanently defin

More ›