clutter55561

This wins the internet this week!

Claude Code is not just a harness. It is a different product. You pick the smallest subscription that allows you to do your work. My “multiplier” on a $100 subscription is 5+.

If you’re using API, on the other hand, there is absolutely no reason to use Claude Code, or Codex.

I use both Claude and Codex, but mostly Claude for planning and coding, and Codex to review Claude’s work.

I follow a sort of waterfall workflow which is verbose but fully transparent.

Anthropic’s $100 subscription works fine for me, but whatever subscription my company has with OpenAI reaches the 5hr limit ridiculously quickly.

ORMs have their place but they are leaky as hell. RDMSs are very diverse, have different languages, and require different optimisation techniques.

ORMs that try to paper over all the differences fail miserably. They become super complicated and generally produce crap SQL.

ORMs also tend to oversimplify database design. They are just tables with primary keys, right? Who needs indices? Who needs to think about collation? God forbid anyone mentions physical organisation of the data!

Having said this, I do use a very small subset of SQLAlchemy (the bits I understand) in data pipelines.

Great stuff, well done and super useful!

Tools that remove the fat seem like a good idea, but I’m highly suspicious of their effect on the LLM’s reasoning.

LLMs were trained in the typical full-fat output found everywhere on the internet, and all of sudden they get a slightly different response that may look like nothing they have seen before.

Does that really save tokens in the long run?

Not exactly. Apple kept it for existing users but stopped offering it to new users. So new users are vulnerable.

I think this is very interesting but you need a better slogan.

Many people here made comments such as “why do I need another SVC since agents are pretty good with git”, which means they barely read your blurb and did not understand your project.

I don’t know if you are a parent of teenagers. Where I live in the UK, teenagers don’t use phone. They only communicate via chat. Relatively large sample (several schools, year groups, ages, etc) but single area.

If you don’t have kids, maybe don’t speak pejoratively about the difficulties in raising children nowadays.

If you do, try to be a bit more empathetic.

I let her have WhatsApp when she was 13. She’s autistic and couldn’t disconnect at all. Her friends got pissed off with her because of something that was said on a chat. She lost her friends at school and ended up being bullied. Masked pretty well too, so when we found out it was too late. Missed years of schooling. I’m pretty sure she will be living with me once she turns 30, so yeah she will be talking to me alright.

None of this prohibition works. Kids will find ways around it, authorities will get stricter, rinse and repeat. A total shitshow.

A lot of people mentioning off-license/booze/tobacco like that was a success story. It isn’t. Outside main/high streets, kids manage to buy stuff just fine. Success requires enforcement, constant vigilance and heavy penalties. Not applicable to Meta at al.

Social media is a drug. Just like crack, making it illegal won’t make it go away. Only education can change this. Unfortunately, we now have multiple generations hooked on it, so I’m not sure this is even possible anymore.

I blocked all social media on my daughter’s phone until she turned 17. I am/was a massive control freak. Guess what happened after that?

I still have control over her apps. I still won’t let install snapchat and every other crap app she asks for. She understands it is for her own good, but none of that matters when “all her friends use it.”

The first iPhone went on sale when she was born. Obama was elected when she was a baby. The world sucks right now.

Rant over.

I got tired of Claude adding their signatures to my commits against my instructions (the settings schema changed at some point), so I added a commit-msg hook that blocks multi-line commits. Easy and works like a charm, and would block this sort of M$ intrusion.

What a despicable behaviour from M$.

I think of my iPhone as a phone plus a mobile browser plus a biometric device. It has a lot of memory and a lot of compute power but that is just because all the crap sites and apps out there, unnecessary animations, etc. One could also claim that a phone is a mobile gaming device, although that is not my thing.

Biometrics is the feature that confers all the power to Apple and Google. All sorts of shady things can be done in the name of security and privacy.

The internet would be a much better place if browsing and biometrics were done in different devices.

Fair enough. My kids are learning to drive and “how fast should I go” is a constant topic in my family right now.

Just yesterday I saw a learner driving at what seemed 10 mph in a 40 mph road, creating a massive queue.

Not American, buddy, so I don’t have skin in the game. But it looks extremely bad when people who got rich in the Silicon Valley engage in politics like this.

Even if one disagrees with a billionaire tax, I’m sure there are better ways to engage with politics besides “protect my capital at all costs”.

It is not so much that they oppose, but how they oppose.

That is pretty sad. People would rather spend money on political activism instead of supporting their home state.

Imagine if they said instead - no need for tax, here is 15 billion to fund the schools.

They could be heroes but choose to be assholes.

Two marathons will never be run in the same conditions, that is the nature of outdoor sports.

Besides weather, there are loads of factors in the performance: shoes, clothes, food, etc. So basically every record gets an asterisk?

Damn was thinking km/h. Thanks for the correction.

~~A car going as fast as him would have gotten a speeding ticket in the residential areas of Wales. Crazy.~~

Edit: I was thinking in km/h and mixed it up. Sorry.

I love programming, but I don’t love working. I’m about 10 years away from retiring and can’t wait. Does that count? ;-)

Exactly what I posted as well!

AI isn’t creating the problem, it is just showing the problem. Those who did not want to learn before AI did so reluctantly, mixing Google and SO. Now they ask AI. An existing problem found a new solution.

Personally, I really enjoy using AI. I have created my own cascade workflow to stop myself from “asking one more question”. Every session is planned. Claude and Codex can be annoying as hell (for different reasons). Neither is sufficiently smart for me to trust them. I treat them as junior devs who never get tired, know a lot of facts but not necessarily how to build.

The same “forgetting pattern” can be said of assembly, hardware, combustion cars, radio, heck, even making fire.

There will always be specialists who can really debug stuff. Mechanics, etc. Time moves on, and we need to move with it.

I’m amazed at this “end-of-world” crap. People use AI to write this shit, to make it even crazier.

This is stupid. Smoking has high social negative externality. It causes cancer to the smoker and to others around the smoker. Who pays for the treatment of those affected? All tax payers.

Want to die? Die fast, not in a way that waste everyone’s money, and don’t take others with you.

All the memories Claude created for me fell in the category remember-to-not-forget, so I disabled it altogether.

Isn’t “memory” just another markdown file that the LLM reads when starting a new session?

I keep two files in each project - AGENTS (generic) and PROJECT (duh). All the “memory” is manually curated in PROJECT, no messy consolidation, no Russian roulette.

I do understand that this is different because the vector search and selective unstash, but the messy consolidation risk remains.

Also not sure about tools that further detach us from the driver seat. To me, this seems to encourage vibe coding instead of engineering-plus-execution.

Not a criticism on the product itself, just rambling.

School was like that back in Brazil in the 80s. Don’t get the grade, try again. And again. And again. Until you do. That is the real “not left behind”.

My thoughts exactly.

A PR, at best, is collaboration, not a transaction.

If they are willing to feed a bug report to their LLM, then perhaps they can also feed a bug report + PR to their LLM and not make a big fuss out it.

Also, at the point they actively don’t want collaboration, why do open source at all?

Strange times, these.