Hacker News Reader

whistle650on A few random notes from Claude coding quite a bit last few weeks

To understand the impact on computer programming per se, I find it useful to imagine that the first computer programs I had encountered were, somehow, expressed in a rudimentary natural language. That (somewhat) divorces the consideration of AI from its specific impact on programming. Surely it would have pulled me in certain directions. Surely I would have had less direct exposure to the mechanics of things. But, it seems to me that’s a distinction of degree, not of kind.

whistle650on Ask HN: Any example of successful vibe-coded product?

Sorry, just a vercel free tier project used by two people… Perhaps this is too much success. Seems to be a rate limit on one of the free data sources I use. Should be vibe-fixed now.

whistle650on Ask HN: Any example of successful vibe-coded product?

Vibe coded this with my son. Something I always wanted since we often record sporting events and want to know if it’s worth watching. So, successful in that sense and works in real time.

https://www.donttellmethescore.com/nfl

whistle650on The entire New Yorker archive is now digitized

I’m a longtime New Yorker lover myself. I think there is some truth to this though: https://open.substack.com/pub/persuasion1/p/how-the-new-york...

whistle650on Bots are getting good at mimicking engagement

That’s true. But you probably can’t. At least any more than others. It’s a systemic issue in the ad network ecosystem which you don’t have much control over. If you can figure it out, odds are lots of others can too. People do assess the quality of traffic sources and do check the return on ad spend. It’s that system wide process that keeps the return on ad spend roughly constant.

The point here, for me, is that a microeconomic perspective on this whole question is more salient than a purely technical one.

whistle650on Bots are getting good at mimicking engagement

And imagine how much more you’d have to pay for each of those clicks if everyone could stop those fraudulent clicks. In equilibrium it shouldn’t change the total ad spend.

whistle650on Bots are getting good at mimicking engagement

This is the key point. Ads and clicks etc are priced in a competitive market. If they don’t deliver the ROI because of bots, then people (including the allegedly hopelessly confused e-commerce retailers) would pay less for the same amount of traffic. It may be annoying (and the cost of dealing with that annoyance would further drive down the price paid for the traffic). But what matters is that an e-commerce site is profitable (enough) after the ad spend, period. If they are not, why do they spend what they spend on the ads?

whistle650on Show HN: Timelinize – Privately organize your own data from everywhere, locally

I thought you could set up an automatic Takeout export periodically, and choose the target to be your Google Drive. Then via a webapp oauth you could pull the data that way. Frequency was limited (looks like it says the auto export is “every 2 months for 1 year”). So hardly realtime, but seems useful and (relatively) easy? Does a method like that not work for your intentions?

whistle650on Will AI be the basis of many future industrial fortunes, or a net loser?

Ok, so how does general anesthesia work? How does ketamine work for depression? The recipes for those are well-known.

whistle650on Will AI be the basis of many future industrial fortunes, or a net loser?

Agreed, and I did mention medicines as examples of things that work but we don’t understand. But they weren’t “made” by us in quite the same way imo.

whistle650on Making LLMs Cheaper and Better via Performance-Efficiency Optimized Routing

It seems they use 70% of the benchmark query-answer pairs to cluster and determine which models work best for each cluster (by sending all queries to all models and looking at responses vs ground truth answers). Then they route the remaining 30% "test" set queries according to those prior determinations. It doesn't seem surprising that this approach would give you Pareto efficiency on those benchmarks.

whistle650on How three years at McKinsey shaped my second startup

Looking at the home page of Meanwhile only made me think of how life insurance is such a different thing than, say, a mortgage. With life insurance, counterparty risk matters. You don't care about your mortgage counterparty. I'm not going to buy life insurance from an insurer with Youtube videos of Anthony Pompliano on their home page. Know your enemy.

whistle650on Gemini 2.5 Flash

Have you tried the Gemini Live audio-to-audio in the free Gemini iOS app? I find it feels far more natural than ChatGPT Advanced Voice Mode.

whistle650on Ask HN: Physics PhD at Stanford or Berkeley

I don't know much about what it's like to do a PhD in physics at Berkeley, but many years ago I did a PhD in physics at Stanford starting out working in experimental quantum optics. I wound up doing something completely different, and felt supported in changing what I worked on. Stanford felt small in a good way, the grad student admin staff was wonderful. Stanford definitely has a different more suburban isolated vibe. Summers felt like you worked at a country club or something.

Who you work with really matters (obviously) and different PIs and labs can have very different cultures which you may or may not feel comfortable with. That alone can make your decision if you are very sure about what you want to do and who you want to work with.

Outside of that, I would say Stanford is a really great place to do graduate work, especially if you're not entirely sure what you want to do.

All of this is with the obvious caveat that my experience is from quite some time ago.

whistle650on Classifying all of the pdfs on the internet

Interesting read with lots of good detail, thank you. A comment: if you are balancing the classes when you do one vs all binary training, and then use the max probability for inference, your probabilities might not be calibrated well, which could be a problem. Do you correct the probabilities before taking the argmax?

whistle650on Newswire: A large-scale structured database of a century of historical news

https://chatgpt.com/share/13f553a8-5cff-42a1-be95-4a9d33cd10...

May also be easy to correct a lot of it:

“For better safekeeping, Russia’s $24,000,000 collection of crown jewels, probably the finest array of gems ever assembled at one time,”

whistle650on MLC-LLM: GPT/Llama on consumer-class GPUs and phones

This is a great project thank you. I've installed the TestFlight app. FYI, right now it's saying in response to "Who was the president in 1973" that it was "Gerald Ford" which is wrong.

whistle650on Diagnosing cancer by profiling the immune system

Wow, thank you for sharing that link. It was rational and humane and I found it very moving. I didn't know who he was, but I'm glad to know of him and his work now.

whistle650on GPT Unicorn: A Daily Exploration of GPT-4's Image Generation Capabilities

In the talk he specifically mentions the very interesting fact that as they improved "alignment" it affected the unicorn output (negatively if I remember correctly). So as long as "alignment" is changing, the output should change. Not sure but ongoing RLHF, changing "system" prompts etc can and do change while the underlying foundational model need not.

whistle650on Auto-GPT Unmasked: The Hype and Hard Truths of Its Production Pitfalls

It is a strange article, speaking as if AutoGPT were not completely nascent, which it is. So the critiques aren't even really wrong. The most valuable observation is that vector DBs are overkill (does LangChain have a stake in Pinecone?)

whistle650on What I Worked On

This feels like a very sterile view of science and it's actual history and practice. I was recently remembering how Marconi's puzzling success in sending a transatlantic wireless signal stimulated the discovery of the ionosphere.

whistle650on Searching the Creative Internet

Totally agree with this. Though I often want something similar but not necessarily “creative”. I’d like to be able to ask “show me blogs or discussions that are substantial about x”. x could be a scientific paper or something.

It is a question of search / discovery mechanisms. Mostly this kind of query is “satisfied” by things like Twitter. But I wish there were a good blog / discussion search engine. Those died a long time ago. As you say the results I am looking for only show up on lower pages in Google. Maybe there is a better search engine for that I don’t know about.

whistle650on Asking Apple for all my data

Google Takeout may at least partly have been the result of (some) employees thinking it was the right thing to do. It’s gotten a lot better over time, it used to be something much more crude that did seem like an internal grassroots thing....

whistle650on Asking Apple for all my data

Partly as a result of regulation it seems all companies are offering ways of giving you your data. E.g. Google Takeout and Facebook’s data download etc. One thing that’s missing are software tools to allow you to do interesting things with your own data (in private). E.g. Google’s MyActivity interface could be a lot better. My interest is partly out of frustration with things like Evernote’s search, and also my desire to be able to ask questions of my data: When was I last at Costco? What was that web page I found from Hacker News during the morning last Monday or Tuesday? When did I last FaceTime my cousin?

Maybe some tools could be natural language based maybe others would be more visual.

It seems to me it would make a great open source project to build such a suite of tools.

It would also be a first step towards bringing individuals more in control of the value of their data.

Does anyone else agree (or disagree)?