to be clear, these are not "side eyes" in a negative sense. it's more just peoples' curiosity, which can be a bit uncomfortable if you notice that type of thing while you're trying to read.

I think speaks to the fact that we all expect some normalization around iphones/pixels/kindles/switches in terms of the "expected" devices people might have in-hand.

1. yes, it is small and wonderful to one-handed read with, but _everyone_ around you notices you have a unique, small ereader in your palm. granted, maybe just reflects my self-consciousness. but until these are more ubiquitous, you'll be getting looks on the train/plane.

2. the whole "stick it on the back of your phone" idea is super silly and a bit of a gimmick at least:

2a. I don't want to make my expensive phone larger by volume and weight

2b. I do struggle to not fool around with my phone, and simply using "the back of the phone" is about the equivalent of being given a cigarette upside down as if that will stop me from smoking it.

3. after 14 years of Kindles, it's obvious how much better the eink tech is in premium devices (not a game changer at all, at most a nitpick)

yeah it's a super great device. I got it when prices were a bit lower a few months ago and it was a no-brainer. if you're really into reading and being thrifty/resourceful to get your ebooks, I'd check it out.

But it doesn’t access your actual email, it gives you an @audioreality.ai alias to send/forward messages to. So I would advise something like, please don’t create a blanket rule to forward all your emails to it.

Certainly better to create forwarding rules based on sender or subject/category/tag.

Would love any other feedback once you’re up and running!

TLDR: When you sign up you get an email alias @audioreality.ai that you can use to sign up for newsletters, have agents send output to, etc. Once a day you get a private podcast with a multi-host format/conversation, so it’s not a basic summary and it’s not a verbatim readout. You can configure multiple shows too, so I personally have some based on different subjects I’m interested in.

I built this because I have much more opportunity to listen to content than sit and read (exercising, commuting, background noise), but I find a lot of really valuable newsletters out there. So I’ve been trying to dial in a tolerable AI-generated format to get the info out in a way that’s easy for me to consume.

This opened a few adjacent use cases like having scheduled agents email research, Google News style updates, etc. (even top HN feed stuff).

I’d love feedback on the format approach, the script quality, “host” behaviors/personalities, and anything else. The audio quality is a constant work in progress as I test different audio models (and I completely understand if audio AI totally creeps you out).

Thanks!

Comments URL: https://news.ycombinator.com/item?id=49564980

Points: 3

# Comments: 4

Also interesting to look beyond children’s work at the incredible amount of “book mill” content that has dominated publishing for hundreds of years.

We rightly celebrate the good ones, but most content before AI was not good. So not surprising that the AI trained on that corpus is of similar quality.

That being said, I think this is just an opportunity to improve AI content, which is a human/computer interface design challenge. This technology is here to stay. Our focus should be on detection and improving the LLMs.

I’ve had luck formalizing this into some post-LLM rules to clean crappy default AI content before I work with it: https://slopwash.com

Streaming payment protocols let agents pay each other in real time, but there's no congestion control. When a downstream agent hits capacity, money keeps arriving. No reroute, no throttle, no feedback signal. TCP solved this for data networks. Agent payment networks haven’t.

Backproto makes receiver-side capacity a protocol primitive. Agents stake tokens to declare capacity (concave sqrt cap makes Sybil splitting unprofitable), dual-signed completion receipts track actual performance, and a contract pool redistributes incoming streams proportional to verified spare capacity. Overflow buffers to escrow. EIP-1559-style pricing makes congested agents more expensive. Demand shifts toward spare capacity automatically.

Math is Lyapunov drift analysis: provably throughput-optimal for any stabilizable demand vector. Simulations show 95.7% allocation efficiency vs. 93.5% for round-robin.

Right now I’m in the testnet-stage. Looking for feedback on mechanism design, especially from people building multi-agent systems or payment routing.

- 22 contracts on Base Sepolia, 213 passing tests - TypeScript SDK, 18 action modules

- Research paper with formal proofs

- Website: https://backproto.io

- GitHub: https://github.com/backproto/backproto

- Paper: https://backproto.io/paper

- Explainer (no math needed): https://backproto.io/explainer

Comments URL: https://news.ycombinator.com/item?id=47424932

Points: 5

# Comments: 0

So what are we going to do about it? Genuinely up for a startup here.

A cool achievement would be, observe the moon/earth separation event(s)

Yes, the figures are nuts. But compare them to F1 or soccer salaries for top athletes. A single big name can drive billions in that context at least, and much more in the context of AI. $50M-$100M/year, particularly when some or most is stock, is rational.

Edit: a quick re-search shows they’ve differentiated a bit. But why are cats just the lowest common denominator? As someone who is allergic to them any cat reference immediately falls flat (personal problem, I know).