Live data from Hacker News

I used o3 to profile myself from my saved Pocket links

noperator.dev

161–170 of 217 posts

Re: I used o3 to profile myself from my saved Pocket links

#161
post #143

What's your Pocket replacement? Wallabag, Hoarder or something else?

reading an article if it's interesting and moving on if it's not :~^) seriously though, i have struggled with tab/bookmark hoarding, it's a huge relief when you recognize it for what it is and quit. IME the bigger/dustier the backlog gets the more vague psychological guilt accumulates, a weight which isn't truly recognized until it's gone.

As a fellow tab/bookmark (and sessionstore.json and other extension-based export) hoarder, I wonder if our lives could be helped by LLM curation and organization of this mess. Like vague queries against this backlog, like is there anything related to XYZ topic or aspect in the pile? What's the overall composition of this heap of links? How does this composition change over time? Can we plot a 2D scatterplot of all the links with proximity based on semantic/topical similarity?

Or maybe we just need to learn how to prioritize better? Or do some kind of stagewise workflow where the superficial ingestion/collection is followed by multiple steps of culling the less relevant stuff (but still without real deletion, just in case for later). Or perhaps we could now write a sentence of why we think the link may be relevant and what future event or future state of some project or development might make this link gain in relevance again? And then again we could declare that this has happened and what are now the links that are relevant?

I could see some LLM product in this space, but I think this market is fairly niche.

Re: I used o3 to profile myself from my saved Pocket links

#162
post #152

Interesting article. Bizarrely it makes me wish I’d used Pocket more! Tangentially, with LLMs I’m getting very tired with the standard patter one sees in their responses. You’ll recognize the general format of chatty output: Platitude! Here’s a bunch of words that a normal human being would say followed by the main thrust of the response that two plus two is four. Here are some more words that plausibly sound human!…

How do you trust the recipe without context? People say they want one thing but then their actions and money go to another. I do agree there's unnecessary fluff. But "just give me the recipe" isn't really what people want. And I don't think your represent some outlier take because really have you ever gotten a recipe exactly as you outlined — zero context – and gave a damn to make it?

The biggest cooking / recipe app in Germany (Chefkoch) works perfectly fine for millions of people without any of the fluff. It's a list of ingredients and cooking steps, that's it. I don't know a single person that cooks who doesn't use it regularly.

Re: I used o3 to profile myself from my saved Pocket links

#163
post #152

Interesting article. Bizarrely it makes me wish I’d used Pocket more! Tangentially, with LLMs I’m getting very tired with the standard patter one sees in their responses. You’ll recognize the general format of chatty output: Platitude! Here’s a bunch of words that a normal human being would say followed by the main thrust of the response that two plus two is four. Here are some more words that plausibly sound human!…

How do you trust the recipe without context? People say they want one thing but then their actions and money go to another. I do agree there's unnecessary fluff. But "just give me the recipe" isn't really what people want. And I don't think your represent some outlier take because really have you ever gotten a recipe exactly as you outlined — zero context – and gave a damn to make it?

> How do you trust the recipe without context?

Ratings or poster reputation.

I often use recipes from a particular chef's website, which are formulated with specific ingredients, steps, and, optionally, a video. I trust the chef since I've yet to try a bad recipe from him.

I also often use baking recipes from King Arthur based on ratings. They're also pretty consistently good and don't have much fluff.

Re: I used o3 to profile myself from my saved Pocket links

#164

Interesting article. Bizarrely it makes me wish I’d used Pocket more! Tangentially, with LLMs I’m getting very tired with the standard patter one sees in their responses. You’ll recognize the general format of chatty output: Platitude! Here’s a bunch of words that a normal human being would say followed by the main thrust of the response that two plus two is four. Here are some more words that plausibly sound human!…

FWIW, o3 seems to get to the point more quickly than most of the other LLMs. So much so that, if you're asking about a broad topic, it may abbreviate a lot and make it difficult to parse just what it's saying.

Re: I used o3 to profile myself from my saved Pocket links

#165
post #66

As someone with a family background of more left leaning Catholics (which I think are more common in the US northeast), it's interesting that it decided that you are conservative based on Catholicism.

Born in Pittsburgh, raised Catholic, pretty darn liberal. We had alter girls in the 90s, openly gay members who had ceremonies in the church, etc. I'm not catholic now but that was a good church in the 80s and 90s.

Catholicism is certainly interesting! It is somewhat similar from my experience. On one hand it was fairly conservative: super against abortion, and old fashioned family values (ie moms should stay home). On the other, there was a huge focus on service and helping the poor both at home and abroad. And not missionary type stuff, just helping no strings attached. Plus a real interest in education with an open mind. So slightly more complicated than a single left/right value.

Re: I used o3 to profile myself from my saved Pocket links

#167
post #154

Actually, I am underwhelmed. I mean, a decade ago, with WAY simpler machine learning algorithms (no fancy deep learning, just shallow singular value decomposition and logistic regression, https://www.pnas.org/doi/10.1073/pnas.1218772110 ), it was possible to predict personality traits from just a few dozen of social media likes. A single like is (nomen omen) likely less valuable than a link saved (as links come from…

> Actually, I am underwhelmed.

LLM understood the verbal assignment and gave an answer "from the top of its head" without performing any specialized analysis.

Re: I used o3 to profile myself from my saved Pocket links

#168

After reading this I realized I also have an archive of my pocket account (4200 items), so tried the same prompt with o3, gemini 2.5 pro, and opus 4: - chatgpt UI didn't allow me to submit the input, saying it's too large. Although it was around 80k tokens, less than o3's 200k context size. - gemini 2.5 pro: worked fine for personality and interest related parts of the profile, but it failed the age range, job role,…

You should be able to use Google Takeout to get all of your YouTube data, including your watch history. This article is a nice example of someone using it: > When I downloaded all my YouTube data, I’ve noticed an interesting file included. That file was named watch-history and it contained a list of all the videos I’ve ever watched. https://blog.viktomas.com/posts/youtube-usage/ Of course as an European it's a legal…

This can give a false sense of what Google (Alphabet) actually knows about you. That above is Google playing the game of 'ok, here is what we know of your activities on youtube when logged in!'

But Google and the rest of the "advertising" (euphemism for surveillance) industry track and create "profiles" based on a basket of data points, from ip/MAC address to the rest of their bag of tricks.

Re: I used o3 to profile myself from my saved Pocket links

#169

What's your Pocket replacement? Wallabag, Hoarder or something else?

Personally I don't mind Wallabag's, well, barebones interface, and having recently jailbroken my Kindle I think its integration with KOReader is its real killer app. I've found the EPUB conversion works really well (modulo some timeout issues on _very_ heavy pages) and it automatically marks articles as read when you sync with the server.
Post reply on HN