Feed

Blood-sucking ticks that trigger a bizarre allergy to meat in the people they bite are exploding in number and spreading across the US, to the extent that they could cover the entire eastern half of the country and infect millions of people, experts have warned.

This happened to a very good friend. He was bitten by a tick while out fishing and it ended up almost killing him. If it had been anyone else, the claim of "I am now alergic to meat" would have been scoffed at.

As I’ve mentioned, while I like the ideas behind RSS, none of the existing RSS readers worked for me. They try to do more than I need. A classical RSS reader fetches full content of the articles, saves it for offline reading and renders the content using an embedded web-browser. I don’t need this. I prefer reading the articles on the author’s website, using my normal browser (and, occasionally, its reader mode). The only thing I need is notifications.

So perhaps résumés as a meaningful signal of candidate interest and qualification are becoming obsolete. And maybe that's OK. When anyone can generate hundreds of tailored applications with a few prompts, the document that once demonstrated effort and genuine interest in a position has devolved into noise.

For most people who want to use AI seriously, you should pick one of three systems: Claude from Anthropic, Google’s Gemini, and OpenAI’s ChatGPT. With all of the options, you get access to both advanced and fast models, a voice mode, the ability to see images and documents, the ability to execute code, good mobile apps, the ability to create images and video (Claude lacks here, however), and the ability to do Deep Research. Some of these features are free, but you are generally going to need to pay $20/month to get access to the full set of features you need. I will try to give you some reasons to pick one model or another as we go along, but you can’t go wrong with any of them.

For most people who want to use AI seriously, you should pick one of three systems: Claude from Anthropic, Google’s Gemini, and OpenAI’s ChatGPT. With all of the options, you get access to both advanced and fast models, a voice mode, the ability to see images and documents, the ability to execute code, good mobile apps, the ability to create images and video (Claude lacks here, however), and the ability to do Deep Research. Some of these features are free, but you are generally going to need to pay $20/month to get access to the full set of features you need. I will try to give you some reasons to pick one model or another as we go along, but you can’t go wrong with any of them.

The arrival of million-token context windows felt transformative. The ability to throw everything an agent might need into the prompt inspired visions of superintelligent assistants that could access any document, connect to every tool, and maintain perfect memory.

But as we’ve seen, bigger contexts create new failure modes. Context poisoning embeds errors that compound over time. Context distraction causes agents to lean heavily on their context and repeat past actions rather than push forward. Context confusion leads to irrelevant tool or document usage. Context clash creates internal contradictions that derail reasoning.

Bases is a brilliant new tool from the Obsidian team. It allows you to create interactive filtered lists of notes. It is user-friendly and extremely powerful, and I think one of the most important additions to Obsidian that we’ve ever seen.

Otroverts place no trust in any group formed around an abstract idea or circumstance of birth, such as ideology, politics, race, economy, religion, and nationality, which exist only in the collective mind. For them, the idea of unquestionable devotion to a group of people linked by a set of tacit criteria agreed upon by the group’s members makes little sense, no matter how venerable that group is in the eyes of the majority.

Developers who misunderstand these terms and assume prompt injection is the same as jailbreaking will frequently ignore this issue as irrelevant to them, because they don’t see it as their problem if an LLM embarrasses its vendor by spitting out a recipe for napalm. The issue really is relevant—both to developers building applications on top of LLMs and to the end users who are taking advantage of these systems by combining tools to match their own needs.

When an agent struggles, so does a human. There is a lot of code and tooling out there which is objectively not good, but because of one reason or another became dominant. If you want to start paying attention to technology choices or you want to start writing your own libraries, now you can use agents to evaluate the developer experience.Because so can your users. I can confidently say it's not just me that does not like Xcode, my agent also expresses frustration — measurably so.