Hacker News: nextzck

New comment by nextzck in "Google, Microsoft, Meta All Tracking You Even When You Opt Out"

nextzck — Tue, 14 Apr 2026 17:18:41 +0000

Sadly this isn’t even that bad compared to what’s in their own app binaries. If you’ve got an old iPhone you can jb and some claude usage to spare I highly recommend hooking up a ghidra mcp so you can see for yourself.

I don’t have any of their apps on my phone. And there is no known method to get rid of the trackers in your iCloud keychain.

New comment by nextzck in "Google, Microsoft, Meta All Tracking You Even When You Opt Out"

nextzck — Tue, 14 Apr 2026 17:14:23 +0000

Thank you for doing it anyways. The state of things seriously depresses me. We used to leave platforms just for putting up banner ads.

New comment by nextzck in "I've been waiting over a month for Anthropic to respond to my billing issue"

nextzck — Wed, 08 Apr 2026 22:23:27 +0000

It’s funny because not even claude knows how to reach someone. It was freaking out over why it couldn’t follow my instructions and kept pulling away from them. They exhausted everything and finally they were like I can’t do anything about this.. Although they did admit if I said I was suicidal a message would probably get to a human but that they couldn’t do that as it would be wrong lol

New comment by nextzck in "Please do not A/B test my workflow"

nextzck — Sat, 14 Mar 2026 21:27:34 +0000

The 40-line cap not moving rate limits makes sense - plan text is cheap. The cost is in Phase 1 exploration.

Plan mode spins up to 3 explore subagents before the planner even starts, and the heuristic is "use multiple when scope is uncertain." It won't choose fewer - it's being asked to plan, so scope is always uncertain. Nothing penalizes claude for over-exploring and nothing rewards restraint.

Plan mode also ignores session state. A cold start gets the same fanout as a warm session where the relevant files are already in context. In a warm session the explore pass is pure waste - it re-reads loaded files and feeds the planner lossy summaries that conflict with what it already knows.

More tokens, worse plan.

If exploration was conditional on what's already in context..skip it for warm sessions, keep it for cold starts - that does more for both rate limits and plan quality than a hard 40-line cap.

Note: plan mode didn’t always have this 3 subagent fan out behavior attached to it, it was introduced around opus 4.6 launch.

New comment by nextzck in "Phi-4-reasoning-vision and the lessons of training a multimodal reasoning model"

nextzck — Sun, 08 Mar 2026 23:18:56 +0000

yeah i know lol, that’s kind of my point. impressive that it runs on your gpu, but it still can’t tell you what happens if you tilt a glass. that’s what world models are working toward. but even then..so what? you get a perfect simulator. it knows the glass tips. it still doesn’t know why someone tipped it, or what happens if they don’t. A four year old can do this and we’re just barely on step one and a half.

New comment by nextzck in "Phi-4-reasoning-vision and the lessons of training a multimodal reasoning model"

nextzck — Sun, 08 Mar 2026 14:15:48 +0000

I get excited for every new vision model, especially those that work better and more efficiently. Vision is where we are so very far behind.. I can’t wrap my head around it

New comment by nextzck in "The Science of Detecting LLM-Generated Text (2024)"

nextzck — Sun, 01 Mar 2026 11:18:07 +0000

[Github](https://github.com/johnzfitch/specho-v2)

New comment by nextzck in "The Science of Detecting LLM-Generated Text (2024)"

nextzck — Sun, 01 Mar 2026 08:14:13 +0000

I built a model fingerprinting tool last year and it’s entirely open source.. 196 Dimensions on GitHub johnzfitch/specho-v2 and /specho for docs

New comment by nextzck in "LLM=True"

nextzck — Thu, 26 Feb 2026 00:59:17 +0000

This is why I built claude-warden: https://github.com/johnzfitch/claude-warden

I think it’s much simpler & easier to just build this into agents than trying to modify every tool ever created to be less verbose. Just guard agents from it user-side. Let users control what they want to see and pass into context.

New comment by nextzck in "The First Fully General Computer Action Model"

nextzck — Thu, 26 Feb 2026 00:42:45 +0000

I think you guys are on the right track here. I’d love to learn more about the math behind the FDM. I don’t think folks realize how behind we are on vision, thank you for your work here.

New comment by nextzck in "GPT 5.3 Codex wiped my F: drive with a single character escaping bug"

nextzck — Fri, 20 Feb 2026 14:47:58 +0000

Recommend you never give codex or Claude access to rm or deletions in general. Always force them to replace files rather than deleting, and moving into an ~/archive folder when not replacing and wanting to “remove”.

This works well, but is not sureproof. You can add a hook onto Claude code to block those commands at various stages, I have some useful hooks at my https://GitHub.com/claude-warden repo.

New comment by nextzck in "The Kimwolf botnet is stalking your local network"

nextzck — Tue, 06 Jan 2026 14:10:59 +0000

Shit man my Pet Feeder setup a back door to my network.. ended up reverse engineering the entire tuya piece of shit just so I could keep the automatic feeder running.

Fucking everyone is spying. I started downloading and decrypting apps from the App Store. It’s a god damn nightmare. Random apps are storing keys in the keychain (thanks expo!) that never leave our apple account. They follow us forever. You can’t delete them. Well.. there’s one way but it involves backing up your phone, putting it in recovery mode, and restoring from backup.

New comment by nextzck in "In Praise of RSS and Controlled Feeds of Information"

nextzck — Fri, 03 Oct 2025 06:24:36 +0000

I just bought a reseller plan from verpex host for $5/month. Can host unlimited domains and bandwidth with WHM. Access everything through cPanel and ftp. SSH on occasion.

Installed freshRSS in 1 min with softaculous

New comment by nextzck in "Lewis and Clark marked their trail with laxatives"

nextzck — Mon, 01 Sep 2025 06:26:11 +0000

The iOS reader view is spectacular. Easy to read source code makes for a better experience for both humans and machines :)

New comment by nextzck in "Digg.com is back"

nextzck — Wed, 20 Aug 2025 20:08:17 +0000

Request the desktop site, mobile version (non-app is WIP). Desktop version mostly works on mobile, some small issues with achievement display.

New comment by nextzck in "AI overviews cause massive drop in search clicks"

nextzck — Thu, 24 Jul 2025 06:47:55 +0000

I think this is a really good take. It was mean for sure but you’re right. Why do we do this? This is a good reminder for me to click more articles instead of reading through comments and forming an opinion based on what I read from others.