tmhrtly

This is explicitly answered in this post:

Think of it as creating separate Claude identities for different uses: everything, including its memories, will stay scoped to the channels defined by the administrators. For example, a model set up for sales work won’t pass on memories to one set up for engineering; nor will it give engineers access to any sales data or tools. More information about provisioning access is available here (https://claude.com/blog/agent-identity-access-model).

You want as much data as you can get about your health as quickly and as cheaply as possible. In other words, you want a technology optimized for getting as many “megabytes per second per dollar” of information about your body.

This is so far from my vision of what I want from healthcare. I want a healthcare system that is optimised around A) proactively keeping me healthy, and B) reactively helping get back to healthy when I am not. I do not care about the amount of megabytes of data I have about my body.

My concern here is that by gravitating to HTML you lose the ability for a human (you!) to easily co-author the document with the LLM. If it’s just an explainer for your consumption, that’s not a concern - but if it’s a spec sheet for something more complex, I deeply value being able to dive in and edit what is produced for me. With a HTML doc it is much harder to do that than with MD.

Now of course you could just reprompt your LLM to change the HTML - but when I already have a clear idea of what I want to say in my head, that’s just another roadblock in the way.

If this pattern becomes more common I suspect human/LLM co-creation will further dwindle in favour of just delegating voice, tone and content choice to the LLM. I was surprised not to see this concern in the blog post’s FAQ.

I agree, this is the pathway. For me, this is the tool I’ve found that works to nudge me down that pathway by adding extra friction to the routes to cheap, crap dopamine. Often an interruption from this app is accompanied by my brain going “huh, so what do you really want to use this time for?”.

The one thing I’ve found that works for me on my phone is the OneSec app. It hooks into shortcuts (for apps) and a Safari extension (for websites) to prompt you with a small task to do (eg a 20sec breathing exercise) before you access the softblocked content. The time delay + task is enough for me to remind myself that this isn’t what I want to be doing. And in the instances where I actually do consciously want to visit XYZ platform, I can just do the exercise and be granted access.

The only downside is that the Safari extension is granted full access to my web browsing in order to facilitate the website blocking. They say they don’t capture any data and at this point do trust them (you may feel differently). For blocking apps, no private data sharing is required.

Not fully convinced by the "will make fares lighter for everyone else" argument. The economics of planes are heavily weighted towards the passengers up front - business & premium economy make more profit per sqft for the airline than seats at the back. So I'd imagine that a reduction in demand for premium seats could actually increase prices.

The article hints at this, with the following sentence: "Still, few things last forever, and the market for written tech journalism is not what it once was – nor will it ever be again" (emphasis on the written).

I wonder if this is because the purpose of linking to a book is to share awareness of that book’s existence - nobody is about to go and read it then and there to comment on its contents. Whereas the purpose of an article is to discuss it now, in the comments - the consumption horizon and bulk of the content is different.

Second data point. I love that app. Well worth all the money.

I've also customised the automations so I have added friction to opening, for example, Slack after 6PM or on weekends. However it opens immediately during working hours.

I have approx. 0 understanding of how interest rates, inflation rates, etc. all intersect. My economics knowledge is 0. Is there a good fun accessible book to get started?

Power and ground wires alone don't cut it sadly when you're dealing with much higher wattages. There needs to be a level of negotiation between the host and the charger to decide on a specific power (current & voltage) that both can support.

In USB-A this was accomplished through a hodge-podge of different resistances applied across data lines, not officially part of the standard but just done by manufacturers. USB-C is a huge improvement on this.

I do agree however that the cable-labelling situation is awful. Maybe some kinda tier system could help. Every charger, cable and device could have a class. The charging rate is the lowest of the three. E.g. a "Class 5 cable will charge up to 200 watts and has a pink end". If you pair that with a Class 2 charger (say, 50 watts) and a class 3 laptop (100 watts) you'll be limited to charging your laptop at 50 watts.

I wonder how many users. They don’t say, just use the word ‘some’ which makes me wonder if it’s quite a few.

Or you just deposited $2300 into some random person’s account…

Has he given up on scale? He's already got a news article and (many) thousands of eyeballs on him. Maybe the scale is in the publicity. Still, by no means easily reproducible by anyone else trying to do the same.

What if we crash landed into it? That would probably tick the irreversible box. Clearly engineering difficulties quite high though!

I know it's not exactly the be-all-end-all solution to the problem, but it does feel like good quality used marketplaces for fashion (like Depop) could help to counter that flow of 'throwaway products'. In addition, in the long run having a more liquid market for potential resale might encourage people to buy better, longer lasting clothes at the offset (but that might be a pipe dream on my part).

That is a fairly rudimentary viewpoint. People who've put up tens or hundreds of buildings in their lifetime are clearly going to be substantially more experienced and talented at doing that process than you, I or any random teenager. There's so much more work to creating a functioning abode than just evaluating 'will it collapse'. Would you trust a random youth to put up your house?

You could do a lot worse than read about the concept of a Zettelkasten or Commonplace book. Here's my notes on the two. I am currently using the app Bear to take notes on everything and anything, tagging as I go, and trying my best to link notes to other meaningful notes. Less formal than a Zettelkasten but more digital than your average commonplace book.

1. Commonplace Book This is a book of thoughts, findings and general collections of information captured throughout your life and brought together into one place. [How And Why To Keep A “Commonplace Book” | Thought Catalog](https://thoughtcatalog.com/ryan-holiday/2013/08/how-and-why-...) [Commonplace Books Part 3: Choosing a System - GeekDad](https://geekdad.com/2020/03/commonplace-books-part-3-choosin...)

2. Zettelkasten

A different concept, but also with relevance to storing information for a lifetime. From my brief readings, there’s more emphasis placed on two stages of note. The first is a quick jot - an addition. The second is a more thought out, self-contained existence of the note. There’s also focus placed here on _linking_ notes together.

A good example of someone’s Zettelkasten is [[Andy Matuschak’s Note Collection]]: https://notes.andymatuschak.org/About_these_notes?stackedNot...

[Zettelkasten - Wikipedia](https://en.wikipedia.org/wiki/Zettelkasten) [Bear App: A Solid Zettelkasten Solution for iOS Users — Mental Pivot](https://mentalpivot.com/bear-app-a-solid-zettelkasten-soluti...)

I too was baffled by this post. Here's something that goes into a lot more detail: Short version, someone found an exploit that allows them to create NFTs that were allegedly created by a particular artist (but were actually not). https://news.artnet.com/opinion/sleepminting-nftheft-monsieu...

What I love most about this one ('Internet Party Line' for anyone trying to find it in the list) is that it wasn't produced by some random person, it was literally made by Intel and published on their own website. Wild times.

cute product! BTW I'm not 100% sure on the "I missed out on an additional ~$1200 in revenue" claim you make though - increasing the price would probably have dropped some users off who would pay $15 but not $25. I'd still guess you left a bit of cash on the table but not the full sum.

If you're a man reading this, one of the tools you have at your disposal is to pledge not to be a part of a panel comprised of only men. It can help to prompt organisers to assess their biases. Here's one example of such a pledge you could take (either privately, or publicly):

https://www.owen.org/pledge/

I don't know how I feel about the 'avoid retweeting' rule. Admittedly I don't have a huge platform, but retweeting for me has always been a way to introduce my audience to interesting new concepts or people to follow. Ultimately we're stronger as a network, not as individuals.

Also, I'm more likely to take advice on how to 'crush it on twitter' from someone who has achieved that already - this author has around 1700 followers which is fairly modest.

Portable faraday cage over the car and driver?

Agreed - I definitely thought this was for job interviews and was really confused why they needed uber-high quality video recordings of me!

I think you are lightweight endorsing them, not the other way around.

I don’t really see it as shaming people, more just misinforming them. Whether that makes it ethically okay is indeed another question. I imagine the researcher chose the ‘lie’ deliberately so it wasn’t too dangerous.

Four days ago this post hit the top of hacker news [1]. It goes to show how easily misinformation can spread - the correct date was even in the article. I'm as guilty as everyone else for taking this in.

Shoutouts to tobr [2] who noticed - and rurban who posted similarly [3].

[1] https://news.ycombinator.com/item?id=25485651

[2] https://news.ycombinator.com/item?id=25486398

[3] https://news.ycombinator.com/item?id=25487207

Does anyone know if SCMP's coverage has become more pro-China after the recent 'National Security Laws' that were enacted in Hong Kong?

Pieces like this are especially interesting when viewed through a Chinese lens.

My hunch would be that early on in the process of innovation, you're quite a distance from profiting from that research. Unless you have pockets deep enough to go all the way, there's potentially value in encouraging many people to try and get slightly further on solving challenging problems in the open, so that all can learn.