HNHacker News
TopNewBestAskShowJobs

scratchyone

307 karma · joined March 22, 2016

submissionscomments
scratchyone··on Qwen 3.8 27B available on Cerebras at 1500 tokens/s
any way to see the tok/s for all the models listed on your homepage? curious which has the best speed/quality tradeoff for me
scratchyone··on Scanning for Pangram Errors
we have this!!

https://www.pangram.com/solutions/chrome-extension

(disclaimer, i am a pangram research scientist)

scratchyone··on FreeInk: Open ecosystem for e-readers
I contribute to Storyteller in my free time and I absolutely love using my Boox with it, honestly my dream reading setup.
scratchyone··on Show HN: Mach – A compiled systems language looking for contributions
fully self hosted without any external dependencies is incredibly impressive, amazing work
scratchyone··on APC–2 – A professional record cutter for producing original playback discs
ooh, this seems super cool. i see that he doesn't print jackets, have you gotten them printed elsewhere for that?
scratchyone··on iPhones with iOS 26 are freezing FaceTime calls when they detect nudity (2025)
This is an article from almost a year ago, last edited July 3, 2025...
scratchyone··on Search engines alternatives now that Google isn't Google anymore
thank you lol, this has convinced me to switch over to fastmail
scratchyone··on Search engines alternatives now that Google isn't Google anymore
As someone who also loves Kagi, what do you like so much about Fastmail? Seeing it compared to Kagi makes me want to give it a try as well haha.
scratchyone··on Search engines alternatives now that Google isn't Google anymore
I've had generally good experiences with Exa, I use it as an MCP in all my various coding tools and use the API whenever I build an AI thing that needs search.

Also an extremely happy Kagi user but can't speak to their API, haven't tried it yet.

scratchyone··on FBI director's Based Apparel site has been spotted hosting a 'ClickFix' attack
To be fair, just because you understand the code you see doesn't mean its safe to copy-paste. Many of these compromised sites will show something that appears to be a benign command but will be something entirely different when you copy them.

Not good to get people into the habit of copying and running code in their terminal.

scratchyone··on Not alive, but not dead: disembodied human brains used for drug testing
They will never remove tissue if you're still alive. This is the reason organ donation is most common in brain-death cases, because the tissue is still alive but you are entirely dead. As you point out, it would be horrible to dismember someone who is still alive and would certainly violate their oath.

I hope this is a comforting answer, I choose to be an organ donor because of these details.

scratchyone··on Incident Report: Railway Blocked by Google Cloud [resolved]
Honestly still insane to nuke a high-volume client's business after a single payment issue. There would be no reason for Google to believe that a single hiccup like that is evidence that they won't get paid and have to cut account access immediately.
scratchyone··on Canvas online again as ShinyHunters threatens to leak schools’ data
https://github.com/instructure/canvas-lms/pulls?q=is%3Apr+is...

haha i went to go check and they haven't merged a PR since 2017

scratchyone··on Canvas is down as ShinyHunters threatens to leak schools’ data
felony murder is pretty widely regarded as a leading factor in incredibly unjust prosecutions and sentencing decisions. perhaps not the best concept to build your ideas on top of.
scratchyone··on Copy Fail – CVE-2026-31431
that's quite smart. i was almost stupid enough to paste it into a terminal to check if it worked before deciding to wait and let others analyze it first haha
scratchyone··on Why AI companies want you to be afraid of them
funny how he didn't care about ethics the moment it was more profitable to release it than to talk about dangers.
scratchyone··on Copy Fail
unfortunately the page can also lie to you haha. it seems people have reviewed the code by now, but running suspicious shellcode you don't fully understand is never a great idea.
scratchyone··on AI Overviews are coming to your Gmail at work
my company intentionally avoids all AI usage and yet our enterprise google workspace is still constantly sending us ai generated summaries of all of our various sensitive documents
scratchyone··on Why AI companies want you to be afraid of them
tbf most companies don't have a potentially world-ending product. only real similar field is defense contractors who typically can't brag about unreleased ideas as they're classified.
scratchyone··on Why AI companies want you to be afraid of them
Don't worry, these companies will make sure we get to experience both nightmare futures.
scratchyone··on Why AI companies want you to be afraid of them
100% agreed. That's part of the issue imo, these companies pretend their new models are "too dangerous" to seem like they care about the world, yet they have no qualms deploying existing models in warfare or bragging about impending mass-unemployment.
scratchyone··on Why AI companies want you to be afraid of them
I mean, it's the BBC and the article doesn't have any typical AI tells, where is this idea that it's AI written coming from???
scratchyone··on Why AI companies want you to be afraid of them
Honestly we should have learned this claim from AI companies was purely fear-mongering back when GPT-2 was "too dangerous to release".
scratchyone··on Warp is now open-source
I agree, it seems like disabling them is quite easy. I'm more speaking to the new user perspective. From looking at their site I would have no clue that that product even had features aside from AI. I would never have a chance to find out I can disable the AI features.
scratchyone··on Warp is now open-source
In all honesty, the people who want to turn off AI won't be downloading Warp in the first place. I know Warp has interesting terminal innovations because I've been familiar with it since before the AI boom, but new users can't really tell.

Homepage header is "Warp is the agentic development environment", only screenshot on the homepage shows what appears to be a product similar to cursor/antigravity/etc AI IDE. Fair if that's your product direction but there's nothing there that tells new users about your terminal UX improvements. Honestly even if I was in the market for a new AI tool, there's nothing on your website that really tells me why I should pick Warp over any of the many competitors.

Fwiw I think Warp is quite cool, I just mean this as hopefully useful feedback from a new customer perspective.

scratchyone··on Show HN: HNswered – watches for replies to your Hacker News posts and comments
also

> A self-contained security audit prompt is available at docs/security-audit.md.

lmfao

scratchyone··on Elevated errors on Claude.ai, API, Claude Code
i notice that as well. most of the time when i see those it has a retry counter also and i can see it trying and failing multiple requests haha. almost never succeeds in producing a response when i see those though, eventually just errors out completely.
scratchyone··on Elevated errors on Claude.ai, API, Claude Code
is GLM genuinely comparable to claude models? haven't had a chance to test it yet.
scratchyone··on Elevated errors on Claude.ai, API, Claude Code
does google actually host anthropic models themselves?? surprised anthropic allows that, given how notoriously crazy they are about distillation or weight leaks or any hints of their models being used in the wrong way.
scratchyone··on Elevated errors on Claude.ai, API, Claude Code
maybe, but the response to GPU shortages being increased error rates is the concern imo. they could implement queuing or delayed response times. it's been long enough that they've had plenty of time to implement things like this, at least on their web-ui where they have full control. instead it still just errors with no further information.
Page 1 of 4Next →