xpe

29 ms·

391.
by xpe 1y ago
If I were to say "Claude Skills can be seen as a particular productization of a system prompt" would I be wrong? From a technical perspective, it seems like unnecessary complexity in a way. Of course I recognize there are lot of p
392.
by xpe 1y ago
The opening line from the video [1] impressed me: > We built an agent builder with true two-way sync between code and a drag-and-drop visual editor. Wow, what a clear pitch. I like it. At the same time, I think about design space betwee
393.
by xpe 1y ago
> which is easy if you have you input source on your dock. correction: on your macOS menu bar (at the top-right of the screen along with WiFi, time/date, etc)
394.
by xpe 1y ago
Poking around System Settings > Keyboard > Keyboard Shortcuts… > Keyboard is a pretty good place to browse if you haven't. There are default keybindings that can changed and switched on/off. For macOS Sequoia 15.6:
395.
by xpe 1y ago
Another tip: lots of useful characters are only an option press away. You can find them by viewing your keyboard [1], which is easy if you have you input source on your dock. Some of my favorites: ⌥k = ˚ (degree) ⌥e a = á
396.
by xpe 1y ago
Here is my high level take: most AI researchers I trust recognize that AI alignment is at least fiendishly hard and probably impossible. This breaks down into at least two parts. First, codifying the values of a group of people is hard and
397.
by xpe 1y ago
Below I'll list each quote and rewrite them with elaboration to unpack some unstated assumptions. (These are my interpretations; they may be different than what the authors intended.) > 1: This is oddly a case to signify there is va
398.
by xpe 1y ago
Cryonics doesn't work currently on the scale of human brains. It works at much smaller scales, though. There was a natural experiment where roundworms were revived after 46,000 years [1] [2]. Assessing how well cryonics might in the fu
399.
by xpe 1y ago
Two things, in the spirit of answering your question and explaining myself. 1. The argument above is sound, but it overstretches my metaphor and sidesteps my point which is: "if there is negligible cost in helping a customer, do it.&qu
400.
by xpe 1y ago
Hello. I've taken some time to think about this. I appreciate your explanation, I care about HN, and I'm open to feedback. > Also, please stop crossing into name-calling and personal attack, as you did here. I see many of my co
401.
by xpe 1y ago
I can see what both of the above commenters are saying. Here is my synthesis: In the US, the powers-that-be are often content to let markets, popular forces, and regulation shape foreign companies. In China (I'm no expert, please weigh
402.
by xpe 1y ago
> It's much easier to point out others' alleged irrational thinking, but the main purpose of books like this is to help you better understand your own thinking. That sounds right. I only can make probabilistic guesses as to wha
403.
by xpe 1y ago
You seem quite interested in right versus wrong. I wonder if you will be intellectually honest if/when I reveal errors, mistakes, oversimplifications, and so on? > I looked up the word in a few different dictionaries and the top ent
404.
by xpe 1y ago
> Now this is ironic. Ironic is the opposite of what you mean, don't you think? By your explanation, Kahneman acted according to his life's work. > The guy who spent his life researching how to live rationally chooses suicid
405.
by xpe 1y ago
I'll answer in various frames: - product development: why make someone "do one extra click" when you can make the extra click unnecessary? - writing: respect your audience's time. - humility: take one minute of your time
406.
by xpe 1y ago
[flagged]
407.
by xpe 1y ago
> Probably nobody at all got confused by that word choice. This is overconfidence; e.g. it "it is clear to me, so it must have been clear to everyone else." Indeed, there is a person in this overall thread [1] saying the use of
408.
by xpe 1y ago
[flagged]
409.
by xpe 1y ago
[flagged]
410.
by xpe 1y ago
[flagged]
411.
by xpe 1y ago
A quote from the above: > It is difficult to reconcile that citizens may have the right to do almost anything to and with their own bodies– from participating in extreme sports to having elective plastic surgery– yet a terminal patient c
412.
by xpe 1y ago
Related, a 5 page page PDF, freely downloadable: Should assisted dying be legalised? Philosophy, Ethics, and Humanities in Medicine volume 9, Article number: 3 (2014) Thomas D G Frost, Devan Sinha & Barnabas J Gilbert https:/&#x2F
413.
by xpe 1y ago
To what degree does this product (or others) look at a collection of documents and offer various possible schema to choose from? This seems like not just a "hard for AI" problem but a "hard for humans" problem. In othe
414.
by xpe 1y ago
$PRODUCT is dead. Long live the $PRODUCT.
415.
by xpe 1y ago
> Please stick to facts here, not hype-filled wishful thinking. You are actively pushing misinformation that makes situations like the OP’s worse. I don't understand what you are talking about. I have to wonder if you posted in the
416.
by xpe 1y ago
> [AI] lacks actual reasoning capabilities. Evals are showing reasoning (by which I mean multi-step problem solving, planning, etc) is improving over time in LLMs. We don't have to agree on metaphysics to see this; I'm referrin
417.
by xpe 1y ago
I'll propose through a simple scenario: An organization wants to compare the risks of deploying a user-facing application backed by an LLM. Let's say they are comparing two LLM options: 1. a self-deployed open-weight LLM (such as
418.
by xpe 1y ago
From page 6 of the NIST "Evaluation of DeepSeek AI Models" report: CAISI’s security evaluations (Section 3.3) found that: • DeepSeek models were much more likely to follow malicious hijacking instructions than eva
419.
by xpe 1y ago
[flagged]
420.
by xpe 1y ago
I'm going to take this slowly and non-controversially in the hopes of building a foundation for a useful conversation. There are no gotchas or trick questions here. 1. Deploying any LLM where a person can use them (whether an employee

More ›