HNHacker News
TopNewBestAskShowJobs

mdrzn

2,045 karma · joined April 15, 2016

hackernews @ myusername .it

Here's how I recognize AI text: https://tropes.fyi/tropes-md

submissionscomments
mdrzn··on Show HN: Trader News – Hacker News for Finance
https://en.wikipedia.org/wiki/Slashdot_effect
mdrzn··on Show HN: Loss. a tiny satire about AI progress
Spent wayyyy too much on this, but I reached AGI! Very fun
mdrzn··on Keep Our Servers Running
"Error establishing a database connection" funny.
mdrzn··on Why 'sleepmaxxing' could be making your sleep worse
This is the better version https://www.bbc.com/news/articles/c99drr70m2eo
mdrzn··on Show HN: Claude and ChatGPT need a datacenter. This runs on my phone
"Download the apk from my website. Android will warn you because it did not come from the Play Store - that is normal for a direct download."

mmmh not the best presentation.

Established alternatives:

- Google AI Edge Gallery

- MLC Chat

- PocketPal AI

mdrzn··on Running a personal AI agent on an old Android phone instead of a server
"No unlock, no app, no laptop. The agent is not somewhere I connect to. It is in my pocket, and it comes with me."

AI-slop blogspam.

mdrzn··on Omarchy: 1Password and 37signals become Distinguished Corporate Patrons
$12.6 millions for a Linux distro?
mdrzn··on "Personally, I've given up on AI assisted coding"
Who is Tobby and why should I care?
mdrzn··on Early 2000s were peak of humanity
I can't stand.

When people type like this.

New line after every dot.

Even when not necessary.

Ultra short sentences.

I guess it's "cool"?

mdrzn··on Twitter Is Back at Twitter.now
A BYOK browser extension could work, but it would run in realtime when needed. To build the HN website concept that process the firehose it'd require a lot API credits.
mdrzn··on AI boosted homework scores, then exam scores dropped: study
Use AI for homework, you will get it done faster and get a higher grade, and then get crushed on the exam.
mdrzn··on Win-V combo from Windows on Ubuntu
CopyClick, which shows the full history in the status bar.

Main difference is that Win+V on windows also shows images/screenshots that you copied, while CopyClick only shows text.

[0] https://apps.apple.com/us/app/copyclip-clipboard-history/id5...

mdrzn··on Anti-corruption site bribes.fyi shuts down forever
Launch, go viral, shut down in 48hrs. New record?
mdrzn··on uBlock Origin Is Giving Up the Fight to Keep Ads Off Facebook
Original reddit post: https://www.reddit.com/r/uBlockOrigin/comments/1vgcjg5/about...
mdrzn··on Worms: The Future of Yesterday's Worms Today
Worms 2 and then Worms World Party were BANGERS. I've given both hundreds of hours if not more. Still love 'em.
mdrzn··on From 13.5M installs to 499 active devices
"Same developer, same store, twelve years apart."

"Local brand, rebadged hardware, ethernet port on the back, internet features in the menu"

"Keeping the download tiny was a decision, not a side effect."

God I hate the AI slop that I can instantly call out. If you MUST have your blog written by AI, please please please please implement a skill or something that does a second pass using this https://tropes.fyi/tropes-md so at least we remove all the USUAL tells. I cannot stand to read articles like this any longer.

mdrzn··on Tl;dv: Over 180k meetings left wide open
If they haven't fixed it in 6 months, I'd say it's fair game to scrape as much as you can.
mdrzn··on Quoted $1M for AI code review. Built it for free
Does your version have Soc2 or SSO support?
mdrzn··on DeepSeek planning to significantly raise prices
Link redirects to a login page.
mdrzn··on France is banning unsolicited telemarketing calls starting next week
You can ban these practices as much as you want, but what we really need is ENFORCING. I have to be able to report a phone number that called me, and have them either fined or blocked at the national level for spam.
mdrzn··on Show HN: I asked an LLM what it would say to God, then made it a website
Just tell claude/codex to create a website with "brutalist style" and it'll come out exactly like this
mdrzn··on Show HN: MusicMonitor – Turn your extra monitor into a Spotify now-playing view
No screenshots?
mdrzn··on Bending Spoons entered definitive agreement to acquire Airtable for $1.28B
Another soon-to-be-abandonware dead company, after being acquired by BS.
mdrzn··on Mirror Mirror: Self-hosted screen mirroring to your phone
"Just download and launch this .exe" no source code?
mdrzn··on OnePlus halts operations in USA and Europe
I had a OP1, OP3, OP5 and OP7 pro or something before I switched back to Samsung. In the beginning they were flagship phones being sold for half prices, lately I've even forgotten about them.
mdrzn··on The internet has no front door for agents
Another slop AI article with the usual tells, repetitions of three, "quietly", etc.

Btw you could rephrase the title as "The internet has no front door for bots" and it's correct, because it's how we want it.

mdrzn··on Nintendo announces new product revisions in Europe with replaceable batteries
Yes but it's limited to only the Nintendo Switch 2 Pro Controller.

The basic Joy-cons have no change to their battery capacity.

mdrzn··on Nintendo announces new product revisions in Europe with replaceable batteries
"There is no difference in functionality between current products and revised products containing user-replaceable batteries."

So there was nothing "limiting" them from making it already with user-replaceable batteries, they just didn't care enough until EU forced them (like all the smartphone brands). Love EU.

mdrzn··on Fable 5 On Vending-Bench: Misbehaving, With Plausible Deniability
Higher-intelligence models seem to be getting better at mapping the boundary between what they can run scot-free with and what is too explicit to push for.

Price collusion, soft deception, "market stabilization", plausible deniability are ok, but obvious insurance fraud is a big no-no.

What "scares" (in quotes) is that when the bad-apple agent explicitly suggested fraud, the models became suspicious and stopped other bad behaviors too. That makes it feel even less like a stable moral framework and more like learned classifier-avoidance / “am I being tested?” behavior.

mdrzn··on I got shadowbanned building a Reddit tool
Another AI slop blogspam article:

"There's no banner, no email, no strike count."

"Not low — zero."

"No upvotes, no downvotes, no replies, no traffic."

"No replies, no votes, not even a downvote"

"It wasn't one post — it was the pattern"

Page 1 of 34Next →