Hacker News: internetter

New comment by internetter in "GPT-5.6 Sol Pricing Cut by 50%"

internetter — Tue, 18 Aug 2026 01:00:55 +0000

Possibly OAI? If you have OAI tokens you are a captive audience. If you have OpenRouter you are bidding on a free market.

OpenRouter attributes this promotion to OpenAI https://x.com/OpenRouter/status/2089416739398254662

New comment by internetter in "How Bluesky draws its logo on screenshots"

internetter — Mon, 17 Aug 2026 22:51:04 +0000

Nobody in the comments talking about snapchat where like one of the core pillars of what "sets their service apart" is the difficulty of taking a screenshot without notifying the other party

New comment by internetter in "Ask HN: Alternatives to GitHub"

internetter — Mon, 17 Aug 2026 19:20:33 +0000

> but there’s a reason 99% of the world’s devs does not follow that workflow.

Is there?

New comment by internetter in "Incident with Github.com [resolved]"

internetter — Mon, 17 Aug 2026 18:39:04 +0000

That is essentially your number of followers. Pretty hard to have a system that cannot be gamed. I think followers are probably loosely correlated with tech ecosystem impact in some capacity, but the second there is an incentive to accumulate them, well...

New comment by internetter in "Anthropic Risk August 2026 [pdf]"

internetter — Fri, 14 Aug 2026 21:55:49 +0000

"all traffic through our systems for collecting human feedback data from contractors evaluating our models ran without blocking biological classifiers"

"totaled around 133M exchanges."

While this wound up being relatively benign, I still find this concerning, amidst numerous sandbox escapes, and previously, unreleased models being accessible via a custom URL. I don't think these companies are giving the responsibility they possess enough weight. How many more issues like this exist?

New comment by internetter in "Anthropic Risk August 2026 [pdf]"

internetter — Fri, 14 Aug 2026 21:42:51 +0000

Is fable not just mythos with safeguards?

New comment by internetter in "Compression is prediction"

internetter — Tue, 11 Aug 2026 23:59:48 +0000

common theory. see https://prize.hutter1.net/

New comment by internetter in "Show HN: I spent 2 years designing a mechanical Magic Keyboard"

internetter — Thu, 06 Aug 2026 18:12:11 +0000

also see https://caniuse.com/?search=squircle (though no gradient border)

New comment by internetter in "Plug-in solar is coming. Plug-in batteries should follow"

internetter — Sun, 02 Aug 2026 04:29:34 +0000

Millions of households have plug-in solar. The payoff is something like 1.5-5 years. Of course, this should be tens or hundreds of millions, but this isn’t vaporware

New comment by internetter in "RipGrep musl binaries occasionally segfault during very-large searches"

internetter — Sat, 01 Aug 2026 15:04:19 +0000

The computer seems to have identified source code to blame though? Unless of course it hallucinated which there’s always a non zero chance of

New comment by internetter in ".self: A new top-level domain designed to support self-hosting"

internetter — Tue, 30 Jun 2026 01:32:20 +0000

a sub (level below) domain from a top level domain (.self) would presumably be pavel.self

New comment by internetter in "Canvas online again as ShinyHunters threatens to leak schools’ data"

internetter — Fri, 08 May 2026 01:41:06 +0000

You're being deliberately obtuse. Canvas has many many features. Wikis and discussion boards and quizzes (with some anticheat) and groups and the list goes on and on. Furthermore, while it was never the flashiest thing, it did it better than many of its predecessors. Yes, an individual class may not use all of these features, and yes canvas has suffered feature creep even over my time as a student and yes canvas is not doing anything technically challenging, but there is enough of it that each school rolling their own everything would be a drastic waste of everybody's time and money.

New comment by internetter in "Uber torches 2026 AI budget on Claude Code in four months"

internetter — Fri, 01 May 2026 16:38:33 +0000

https://www.forbes.com/sites/annatong/2026/03/05/cursor-goes...

"Cursor estimated last year that a $200-per-month Claude Code subscription could use up to $2,000 in compute, suggesting significant subsidization by Anthropic. Today, that subsidization appears to be even more aggressive, with that $200 plan able to consume about $5,000 in compute"

New comment by internetter in "Uber torches 2026 AI budget on Claude Code in four months"

internetter — Fri, 01 May 2026 16:36:47 +0000

I know I'm responding to AI right now, but

> which means figuring out if the company can afford this level of productivity at scale.

If it was actually productive, then the revenue would increase and affordability wouldn't be a question.

New comment by internetter in "Copy Fail"

internetter — Thu, 30 Apr 2026 15:48:36 +0000

Technical writeup is also slop I fear

New comment by internetter in "Bitwarden CLI compromised in ongoing Checkmarx supply chain campaign"

internetter — Thu, 23 Apr 2026 15:36:11 +0000

> A password manager does not need a CLI tool.

A password manager absolutely does need a CLI tool??

New comment by internetter in "OpenAI's response to the Axios developer tool compromise"

internetter — Thu, 23 Apr 2026 15:06:18 +0000

Wrapping fetch is not rocket science

New comment by internetter in "OpenAI's response to the Axios developer tool compromise"

internetter — Thu, 23 Apr 2026 04:20:26 +0000

Just slow and convoluted internals due to the accumulation of cruft over time

New comment by internetter in "OpenAI's response to the Axios developer tool compromise"

internetter — Thu, 23 Apr 2026 04:02:06 +0000

This is why people still need to know how to write code and why it is asinine to have an LLM write code without a human reading it. Good developers should know what good code looks like and push back when what they're fed is wrong.

New comment by internetter in "OpenAI's response to the Axios developer tool compromise"

internetter — Thu, 23 Apr 2026 03:55:12 +0000

Any sufficiently competent typescript developer can build out an adhoc wrapper (that just inherits the type definition and passes along whatever it is passed after altering it however needed) in under a hour. It doesn't scale in the sense that you don't expose a configuration, but config as code is king.

(Source: have built out much more scuffed variants of this than the one I just described like https://github.com/boehs/ajar)

I guess a LLM can do as well. Although that's not something I'm quite ready to admit.