imhoguy

For some a tent in a forest is enough, others need luxury hotels next to the forest.

It is already happening. E.g. soon F-Droid and any unattested open source Android app distribution will be gone, due in Sep 2026.

Same with GitHub and similar, we have CLAs for a while now for licensing. But I see project maintainers are frustrated with AI generated slop PRs and bad actor contributions. The ecosystem will be closing. You will be able to read code, but forget creating PR without some ID verification (because this is for kids or against terrorism).

Yeah, after recent Samsung One UI and Google icons change...hm.

..and space, try to assemble bulky bed in small bedroom

This is why traditional bare metal and VPS rental wins for any kind of private or small business use.

I just wonder how do you filter out signal from noise - does it self correct wrong memories?

Briar will thrive once EU Chat Control 2.0 passes, P2P E2E encryption is the only way to bypass bullshit laws.

What still works, but ofc depends on what you've made, is to reduce signup friction to minimum or skip it if you don't need user info. Then ensure the data you produce is easily shareable and by the network effect your users can market your service. To bootstrap initial spark you need to share these links yourself in communities which would be interested in your data. Please don't spam. Obviously converting freeloaders into customers is another book to learn.

Clever idea! Although after reading it briefly I see a need for secrets storage.

I've made one Telegram bot hosted on VPS with Docker and cloud LLM. It also interacts with a few other outside services and all credentials are injected via env vars now.

Should I push them as `.env` file for Telegram serverless?

Yeah I thought "wow, some homomorphic encryption* stuff", but then "nah, usual greed".

* https://en.wikipedia.org/wiki/Homomorphic_encryption

What I miss in Claude Code is code completion at prompt line.. ah wait it gets back to IDE and Copilot.

Prompt is executable (interpreted) code with LLM agent harnes now, just the syntax is fuzzy.

"Please print 'hello world'"

Quite similar, but I ask for "/grilling" skill in every prompt so the agent aligns to my decissions before it drifts into own. And I actually ask to make tests before any implementation and let them all fail.

I don't pay. First they learned how to use mute button on rc, then they stopped using YT because of too many ads.

Emotions experienced chart - that is insightful and matches my anecdata.

I think you get bad emotions when you have high expectations about social media and it is your main source of social life. Where positive happen when you have low expectations about social media and it is just addon to your life.

Example of gaps is being lonely, low self esteem, low self worth, no work network, no business network. So you stay glued to FB to build your life, to keep online friends, because you may have not many in life. Or you have no real work network so you need to stay current on LI because your next job is there.

Now imagine blood pump during surgery.

I test workloads with multiple closed and at least one open model now. Good to have a backup on 503s or credits run out.

This is how development looks like for many years now, constant rewrite on the horizon. I think LLM development hype surpassed Blockchain and JS frameworks craze of decade ago.

CLAUDE.md (or AGENTS.md) is not good place to put all code change rules, because these rules would dillute the context and "distract" the agent e.g. during bugs triage or business analysis.

Instead modularize the knowledge with skills and specialized MD files. Agent should lazy load what is needed to do focused work.

Skills have usage description metadata, but with free files you can simply instruct agent with CLAUDE.md to load them, e.g.: "Before you attempt to change any frontend code first load and follow `docs/{JS|HTML|CSS}_coding_rules.md`".

No way! It is a while since I have seen that extension here on HN, I couldn't be mistaken. With AI? Is Pascal still alive?

Except Amazon (Linux, Elastic, RDS..)

Hopefuly I see conscious young people in my surroundings who refrain from using human like AI (voice). They say "it is cringe".

Moreover I think aggresive robocalls and advertising will derail the entire trend in a few years, people will become tired of fake talk.

The worst is that every agent session is generating so many "btw fix this" side-quests that it is really hard to stay in the task focus. I throw some into todo list manually but still it is exploding by the day. Perfect is enemy of good.

In my place I see currently a governance panel effort mandating around LLM agent skills usage, it is so much shit show that I expect productivity is going to fall to 0.5x pre-agents. But not pre-LLM as autocompletion was really helpful in the trenches. The tool in wrong governing hands and you get sand into cogs thrown.

Isn't it how these fulfilment centers already work that way with all these work manuals. Read AMAZON.md /s

I bet drone van is the next thing to be automated.

Literaly they silently refuse to open the pod doors for you.

... and that actually shows - senior engineers have spent actual paid time to train juniors. Plus they used to spent time contributing to open source projects or Stack Overflow, all the stuff which every company benefits from.

It is like with drug dealers, you get some free dose, then cheap one, then you pay an arm and a leg.

Somebody has to finaly pay for these heaps of accelerator hardware.