You can put a BOLT12 in your BIP353 DNS record and get paid by lightning. lnaddress works but it always felt a little hacky. If I have the option I'd prefer to get a proper BOLT12, complete with blinded paths, signatures, etc..

It's really just this line that's confusing, right? Because n isn't really mandatory.

Because later it's more clear with:

If it confused you, I'm sure it's confused others. I think it's worthwhile to fix unclear wording in the spec.

This piqued my curiosity so I hand decoded a bolt11 to check - it had s, p, d, x, c, r, and 9 tagged fields. I guess n is not typically included because you can calculate the public key from the message (bolt11) along with the signature. Omitting it saves space.

Of the tagged fields, only d or h should be included (a description or a hash of the description if it's too long to fit.) p and s are required for payment hash and payment secret, but n is optional. Those still have a mandatory length (even though they have a length field in the invoice) otherwise the reader wouldn't know what to do with them and shouldn't proceed. Keeping these fields with a data length parameter leaves some flexibility for future spec changes I suppose.

That's my interpretation as well.

Excellent reporting as always @anita. I had to check and the ZiG (the currency that just replaced the Zim dollar last year) has lost 65% of its value versus the dollar in the last 12 months. And they want to make it out like bitcoin is the problem because it's down ~6%. They have no concept of what real inflation and debanking problems are like.

I used to wake up 1.5 hrs early for "boot camp" a couple times per week with a personal trainer and about a dozen other folks. Sure, I could work out on my own later, but the accountability plus feeling productive before my day really even started was quite nice. I like my sleep, but I think it was worth it.

I don't think this is anything too controversial. Creating a functional prototype is one thing, but writing well architected software, with robust, clear, concise, and maintainable code is something else. Arguably, if the vibe coder is experienced and good at project management, some of those attributes might be prioritized, but I wouldn't expect any of them to be inherent to an AI generated codebase.

Weigh it down in what way? In terms of the protocol, software and consensus? I think bitcoin has, so far, had a pretty good history of technical merit winning out. That certainly could change, but I'm cautiously optimistic.

In terms of the asset, I do wonder what will happen to adoption and price if there are enough new adopters and they are dead-set on keeping it as 5% of their portfolio vs. equities, real estate or whatever. They see it as an alternative to some tech stock. If that's the case, and they're rebalancing regularly, it will certainly become correlated to every other macro asset in that portfolio. Does it then just end up back in the hands of those with higher conviction over time? I'm not sure if this would stall out adoption or cause a loss of interest due to the price action for long enough to make a difference. At any rate, I wouldn't expect it to act like a debasement hedge or a 21M/∞ asset until the average marginal buyer/seller treats it as such - which could be quite a long time indeed.

I'm more interested to see if there's any blowback from switching to a bearer asset payment from the credit payment model. Arguably chargebacks cause more harm than the fraud they're meant to protect against, but people have probably already been wired to assume a transaction can be reversed now that cash is all but extinct.

If anything, the early adopters of this will be the low-margin, high risk businesses that already have a different cash vs credit price.

I'm not sure what I would do - probably get back into more engineering. I used to spend a lot of time at my local makerspace and was into 3d printing, milling, etc., and making prototypes of anything and everything.

About the original article - I feel like I'm noticing a lot more mops lately. I'm sure there were plenty in previous cycles, but even then it seemed like the new entrants were still interested in an entirely unique technology, or breaking the existing financial paradigm, or at least sympathetic to these interests. I browse the bitcoin subreddit every now and then, and it's changed drastically from even a few years ago. It's full of comments celebrating selling bitcoin to pay off a mortgage or become a homeowner. Because that's the real goal apparently. I've similarly noticed comments about how nobody actually bought and held for >5 years ("those are just bots!!"), and that bitcoin might "go to zero" at any moment. I don't think many people there are interested in using it either, it's just a line item in their robinhood account or something. I one point I would have pushed back against these notions, but at this point it would be spitting into the wind.

Anyhow, it's nice over here on SN!

It the route blinding. CLN adds a fake hop - the "entry node" of the BOLT12 blinded path is actually your own node. Choosing an appropriate fronting node requires knowing how much you'll be receiving, how much inbound liquidity your peer has, how reliable they are, etc.. So for the time being CLN doesn't add a blinded path by default.

I think the point is to reduce conversion per ad. The more people that run this, the less profitable each ad view becomes. Disincentivizing ad spending is the goal.

I hadn't heard of drywells. The permaculturists would use swales to a similar effect - capture surface water until it has time to soak in, and provide organic matter to soak it up like a sponge. In some instances it looks remarkably effective. This sounds like the more industrial scale approach, but I'm sure there are tradeoffs.

Oh, interesting - I was anticipating fingerprinting would be about wallet fingerprinting using heuristics on transactions, but this is about fingerprinting clients. I guess that makes sense given the occasion.

Anything you're particularly excited about coming out of this @schmidty? And thanks for sharing.

I think it was worse at my school - a full 50% attrition in the first year, and we only had STEM degrees, so there was no switching major to something "easy."

I think schools could do a much better job preparing students with good study habits, time management and generally more "grit" before entering university. Sure there are some really bright engineering students, but even many of those faced a setback at some point. The thing most engineering grads share is perseverance and study skills.

Another problem I see, in the US at least, is we turn kids off from entire professions because we seem to have a societal disdain for math. "Oh, math's not important - you won't use that anyway." That should never be a response when a student is struggling. More like: "these are some abstract concepts than can be difficult at first, but that's normal and you can overcome this - let's try again." There are a lot of students who could have been capable, but have given up on math before they even get a chance to study it at a higher level.

Lol, 'trash" and it comes out looking incredible. I love seeing the process behind the scenes. Great content, thanks for putting this together!

Similar, but just is frustrating is the other common occurrence:

Wow, you can see the great recession in US demographics. My son is at the bottom right of this slope, but I felt very fortunate to be gainfully employed at the time. That was not at all the prevailing sentiment at the time.

That's got to be the Tor option, no? I'm certain Uber tracks the hell out of it's customers. The real question is whom they sell the data to.

Reality is getting weird when you've got to take an uber as a VPN.

I've run some of the Mistral instruct and smaller Ollama models myself, but generally rely on trymaple.ai. Can't say I've had this concern.