Hacker News: saulpw

New comment by saulpw in "How I Fight AI Brain Fry"

saulpw — Sun, 08 Nov 2026 22:23:23 +0000

It doesn't make sense! AI-generated text is insufferable. It requires so much parsing to get to what it's trying to say. Once I figure out what it's trying to say, I can generally say it with fewer words and less pompous structuring. I bet you could too, and then your blog post would actually be readable!

New comment by saulpw in "How I Fight AI Brain Fry"

saulpw — Sun, 08 Nov 2026 21:17:11 +0000

> That closing is the whole trick.

This is such a short blog post. You read this line (an entire paragraph by itself) and thought "good job Claude, you really captured what I was trying to say here"?

New comment by saulpw in "What do the next 4000 emoji look like?"

saulpw — Sat, 07 Nov 2026 09:27:15 +0000

Let me put it another way: Unicode is for letters, not for words.

New comment by saulpw in "What do the next 4000 emoji look like?"

saulpw — Sat, 07 Nov 2026 06:42:52 +0000

They are not part of any existing writing scheme. I think they could be a different standard like ":+1:" and ":chainsaw:" but it makes no sense for them to be built into text rendering.

New comment by saulpw in "I struggle to believe that AI R&D is ingenious"

saulpw — Fri, 06 Nov 2026 21:36:31 +0000

Truly new ideas are, say, 0.01% of all 'novel' ideas, with the rest being rehashing or recombinations. And even then, the truly new kernel of the truly new idea is some small fraction of the whole thing. Like Einstein's theories of relativity, they were truly fresh, but mostly had to recombine existing work in 'novel' ways.

So it's not there are zero new ideas, but so few compared to the whole of the intellectual domain, that we can say 'there are no new ideas under the sun' and be basically correct to within a small rounding error. And AI is sufficiently ingenious to serve the needs of the vast majority of humanity, even though we'll still need to find/cultivate/inspire some new genius humans if we want to have more truly new ideas in the future.

New comment by saulpw in "Harmony: Chat like Discord. Federated like email"

saulpw — Fri, 06 Nov 2026 03:06:03 +0000

I use an account on a matrix server other than the default matrix.org. My password manager won't fill it in so I have to type this server into the login form manually every time I login.

New comment by saulpw in "You'll need nerves of steel to work with AI agents"

saulpw — Thu, 05 Nov 2026 01:38:49 +0000

At first, maybe not much. But when you're properly set up, it's amazing to watch it cook.

New comment by saulpw in "Ask HN: What's your current AI agentic setup?"

saulpw — Thu, 05 Nov 2026 01:29:59 +0000

I think a lot of people/companies are using 10x the tokens for 10% improvement. If you're willing to spend a modicum of attention/effort looking at what the bots are doing, one max Claude plan with Opus and occasionally Fable is very effective.

New comment by saulpw in "Viral Polls Are Not Representative Polls"

saulpw — Tue, 27 Oct 2026 17:01:12 +0000

There are many polls vs one election. It's a lot harder to manipulate all the polls.

Also, yes, we can and should be volunteering locally. AND ALSO we can have independent statistical review.

New comment by saulpw in "Viral Polls Are Not Representative Polls"

saulpw — Tue, 27 Oct 2026 16:40:34 +0000

How do you confirm the election results itself though, when the election purveyors can't be trusted? Statistical polling has been a third-party check on elections for a century.

New comment by saulpw in "Three new discoveries in mathematics, computer science, and physics"

saulpw — Tue, 27 Oct 2026 01:05:50 +0000

How would that look any different than literally right now?

New comment by saulpw in "Everything"

saulpw — Mon, 26 Oct 2026 17:16:07 +0000

I see no evidence that you're committed to it getting better over time. Tell me again in a year or two when something you've made has actually gotten usable. Until then, build your credibility by making something good, not just making a sloppy copy.

Don't sloppy that copy!

New comment by saulpw in "Why I Am a Programmer"

saulpw — Mon, 26 Oct 2026 00:52:24 +0000

I just produced a restoration/translation of Below The Root last month! Check it out, hope it brings back some memories: https://root.saul.pw.

New comment by saulpw in "The foolishness of "natural language programming" (1978)"

saulpw — Fri, 23 Oct 2026 17:16:27 +0000

Well we've been doing exactly this for 200 years! IMO the use of fossil fuels for manufacturing novelty products that go straight-to-landfill, and daily suburban commuting to bullshit jobs, and diverting water to inhabit cities in inhospitable deserts, is a greater cause of our 'slow genocide' than AI will be. So why would we stop now, when we couldn't be bothered to even slow down before? (Annual carbon emissions aren't even trending downward, much less getting to net negative!)

So while I also foresee and lament humanity's self-genocide, it seems a bit hypocritical to get salty about AI "automation" while broadly accepting our industrial civilization's trajectory.

New comment by saulpw in "Eight Bits of Trust: eMudhra's CA keys cracked in milliseconds"

saulpw — Thu, 22 Oct 2026 17:17:55 +0000

Yeah this was my impression too, I thought it was structurally impossible to have a private key like "4"?

New comment by saulpw in "Facebook Pages require paid subscription to post more than two links per month"

saulpw — Tue, 20 Oct 2026 20:04:33 +0000

Original comment you responded to said "> [Meta] have arguably contributed nothing to society for over a decade" and you responded with a list of tech that is supposed to indicate their contribution. I'm not impressed. I'm also not impressed by fast fashion and child labor phones. Big Tech has been a scourge and 20% less web traffic because of zstd does not absolve them.

New comment by saulpw in "Facebook Pages require paid subscription to post more than two links per month"

saulpw — Tue, 20 Oct 2026 16:59:34 +0000

I also "use" Facebook tracking and ads and I'm not psyched about that either!

New comment by saulpw in "Facebook Pages require paid subscription to post more than two links per month"

saulpw — Tue, 20 Oct 2026 16:31:06 +0000

Yawn, I've used none of these except Linux and I would expect that their Linux contributions have been self-serving and not made it better for regular users like me.

I mean, Hack? Who cares?

New comment by saulpw in "It's still ok to enjoy programming"

saulpw — Tue, 20 Oct 2026 00:39:49 +0000

We were lucky for decades to get paid to do what we enjoy. Virtually no other field gets this. Maybe artists but note that very few artists make even a living from their work. So yes it sucks but as they say, "don't cry because it's over, smile because it happened".

New comment by saulpw in "It's still ok to enjoy programming"

saulpw — Mon, 19 Oct 2026 17:31:05 +0000

"Gathering mental capacity to simply know what next I need to do" is not a waste. That is work. In fact it is the most important work with LLMs. If you spend all day in a hammock dreaming up what next needs to be done (and it's the right stuff), and then type a few sentences into the LLM to generate it overnight, that is better time spent working than spending all day cranking out 20 PRs that are of mid-low value.