M Rad

Just trying to keep up with all you people.

Spam-armored email address: **@**2light.com-salt1337 => sha256 => 608b34b4bb18dcf0d1913ddc3730c83bb225b1669721f3515033d705e462fcee

Brute-force the missing letters

date +%T; alph='a b c d e f g h i j k l m n o p q r s t u v w x y z' ; for a in $alph ; do echo -n $a ; for b in $alph ; do for c in $alph ; do for d in $alph ; do if test $(echo $a$b@$c${d}2light.com-salt1337 | sha256sum | cut -d ' ' -f 1) = 608b34b4bb18dcf0d1913ddc3730c83bb225b1669721f3515033d705e462fcee ; then echo -n ' '$a${b}@$c${d}2light.com' ' ; fi ; done ; done; done; done; echo; date +%T

Last Notes

2026-08-21 18:04:02 UTC

- reply

True E2EE encrypted messaging has been around for a long time, with, for example, Torchat, but those systems make no attempt at facilitating social discovery. Whatsapp, Telegram, and Signal match up your contacts by phone number, but using phone numbers as identifiers creates a giant metadata leak. Of course, Facebook will give you all manner of social connection suggestions, and they bought Whatsapp to make money off the metadata, but adding E2EE messaging combined discovery of the underage with cryptographic evasion of oversight. Facebook's commercial interests run against providing proper parental controls, and the result is predictably toxic. Signal has a similar problem with goverment agents infiltrating to get at the metadata.
That metadata leakage, has, heretofore, paid for the infrastructure, plus monopoly profit for Big Tech, of course. The obvious solution is to not use phone numbers as identifiers (nor any identifier used anywhere else for any other purpose). But how does the infrastructure get paid for? Threema charges a subscription fee for it, direct enough, but now you need a way to make the payment private. Can't say how Dorsey plans to turn a profit on Buzz without enshittifying it, but I hope he fares better than with Twitter in that regard. Nostr users can, of course, pay for relay service, and that solution seems the best I have seen so far for having commercial interests line up with not enshittifying user privacy.
So far, we haven't seen many youngters on Nostr, but its combination of social discovery with E2EE messaging brings the same toxic hazard of exploitation as Facebook. Developers will need to make Nostr clients with proper parental controls, but at least they won't have Big Tech lawyers actively impeding unofficial clients.

2026-08-18 01:25:16 UTC

- reply

I would point out that, historically, these trade-offs have been well-trod, going back to 1990's usenet and email, & IRC. Email spamblock developed in layers: legal action against the most ergregious spam cannons, server-side filtering with realtime blackhole lists, and client side with beyesian and deterministic rules. POW, of course, was first proposed for email but never caught on, but we do have it for NOSTR. Usenet had the concept of "bozo filters" for client-side filtering. IRC, in its early days, had the "eris" server that refused on principle to filter anything, but eventually everyone else stopped federating and EFNet (Eris-free) is now the oldest extant IRC network.
As a general rule, server-side filtering creates a hook for authoritarian censorship, which defeats the whole point of NOSTR, so that is a last resort only to be considered when a client-side is truly impossible.
NOSTR support for POW gives us, in my view, the magic bullet to avoid email's fate. Though email is still quite nice if you take a half hour every month or so to tweak your client's filtering rules, that is not a grandma-level skill. I predict that, as NOSTR's popularity rises, spam will get worse, and eventually everyone will start setting POW thresholds. When grandma gets crapflooded, you just can tell her to bump up her minumum POW and life will go merrily along.

2026-08-13 14:22:39 UTC

- reply

I would add that this is a key problem with hyperscaler data center build-outs. Key inputs have distorted price signals. Policy makers hide the cost of enviromental-ish goals (and conceal patronage) in electricity rates, and water policy more resembles tribal warfare than functioning markets. No one (besides foreign-financed shills) would protest these data centers if we had credible price signals to have them (auditably) pay their way.

2026-08-12 02:42:34 UTC

- reply

Well, when I look around, the project most congruent with BIP110's technical ethos is, um, Bitcoin Cash. They retain a hard limit on OP_RETURN and they never implemented Taproot so inscriptions are impossible. (Litecoin has Taproot and inscriptions, so they are right out.) Of course, even BCH can do "colored coin" type protocols where non-transaction activity rides on top of real low-value transactions. Hell, you can do colored coins with paper cash serial numbers so I can't see a way to stop that. Of course, they use standard SHA256 so their mining lives in the shadow of (regular flavor) Bitcoin. I haven't paid much attention to the project so I can't say how centralized their mining has become.
I am not aware of any way to put non-transaction data in Monero, or Mimblewimble-based coins, so those may appeal to the sensibilities of the BIP110'ers.

2026-07-25 13:17:29 UTC

- reply

Accelerating IBD with more intelligent UTXO pruning was the topic of "Accelerating IBD with SwiftSync" at this year's MIT Bitcoin Expo. TLDR...or maybe TLDW...you can get the correct final UTXO set from the network before starting (or compute it without checking sigs), and use that verify only the transactions that destroy UTXO's that never made it to the final set.
Select Day 1 (in the upper left drop-down) and jump to 42 minutes: https://mitbitcoinexpo.org/livestream
The PR on Bitcoin Core got a bit of resistance because it touches the code in so many places: https://github.com/bitcoin/bitcoin/pull/34004
In my experience, much like yours, even with a mere CPU, mass-storage I/O bottlenecks IBD.

2026-07-25 05:46:41 UTC

- reply

BIP110-compliant transactions are also Core-v30 compliant, so, unless an input UTXO is dependent on minority chain's transaction order (very few initially) a miner on the majority chain will snarf up those fat fees. A miner gets nothing from mining the minority chain, no matter how fat the fee, unless the chain is heaviest when presented to the exchange 100 blocks later. Nakamoto consensus at work.
Unless, of course, BIP100 precipitates a BCH-like fork in the *market*. In that case we did see miners opportunistically switch chains for a while, but even there, the BCH market eventually thinned out (and BTC hashpower grew) until any significant shift of hashpower over to BCH would crash its market, making such tergiversation unprofitable.
Of course, the mere presence of a high-fee transaction creates the "goose egg" problem, where miners will want to fork the chain and re-mine the big egg instead of building on the heaviest chain. First described in 2016, we haven't seen it happen yet, but ever-declining coin emission takes us further into untested waters. A disruptive event that creates a large number of high-fee transactions could tip us over into a new, pathological Nash equilibrium.
https://blog.citp.princeton.edu/2016/10/21/bitcoin-is-unstable-without-the-block-reward/

2026-07-25 04:40:35 UTC

- reply

Miners run on thin margins and will likely pick up long OP_RETURN's if they (as is likely) pay more per vbyte. They will switch to whichever exchange will take such blocks, or whichever takes them at the least discount. If the vbyte premium at mining time exceeds the discount at exchange time, BIP110 quietly dies. Whatever happens, price discovery on the risk premium will quickly lay bare how much influence the BIP110-ers actually have. For all we know, the major mining operations have contract terms on what the definition of BTC is, and I have a hard time thinking a court will favor an upstart with less than 1% miner signaling against Bitcoin Core.

2026-07-23 01:25:41 UTC

- reply

A risk-averse miner can hedge, for example, swapping out only 1% of his holdings to merely double his money. Doesn't materially affect the argument. In fact, it would be a gamble to *not* hedge between the two chains. We saw behavior much like this during the blocksize wars, with miners moving hashpower back and forth as prices and mining difficulty gyrated.

2026-07-22 18:42:29 UTC

- reply

Why would you have to be a miner to do this? Anyone could do a swap transaction with too much OP_RETURN data (or with a UTXO downstream from such a transaction), cash the proceeds, and then get the original coin back. One UTXO out of scores entering a WabiSabi coinjoin has 1% of its value reversed by BIP-110, and the whole coinjoin gets reversed. Now what? Every client, swap protocol, L2, and God knows what else, has to be BIP110-aware before miner enforcement, or the most vulnerable users get robbed.
We saw something similar when Overstock conflated BCH for BTC.
https://bitcoinist.com/overstock-glitch-mixes-bitcoin-bitcoin-cash/

2026-07-18 22:21:22 UTC

- reply

Does the thing have a JTAG port? The bootloader had to get on there somehow. Of course, that may be more trouble than the thing is worth.

2026-07-16 18:54:09 UTC

- reply

Well, if you hold the acquisitions forever that would make you a conglomerate, and we know that story doesn't have a happy ending. By my (limited) understanding, the conglomerates of the 70's were a reaction to antitrust law blocking more rational vertical integration and possibly pre-microcomputer economies of scale in inventory management/ERP, while the 2000's version was driven by cheap interest rates. Speaking as a former employee of Dennis Kozlowski's Tyco, I have to wonder how you intend to handle, um, dis-economies of scale. How does Berkshire-Hathaway manage it? Do you intend to stay small? Split into 7 companies like Tyco did? Re-package companies and re-sell them Jack Welch style? (Was an employee of that too...it kinda worked, but his rank-and-yank employee review process is unadulterated evil.) Never, ever, I mean never, like ever, build a headquarters in NYC?

2026-07-14 19:09:10 UTC

- reply

Objections to BIP-110 (not saying I agree with all of them) can fall in several categories:
(1) One my agree with the goal and object to the implementation: The focus on OP_RETURN may just push non-transaction data, more inefficiently, to other forms, and the 1 year sunset seems arbitrary.
(2) Technical conservatism: There is no way to really test if, and to what extent, a change threatens consensus when real money is involved. Nor is it easy to test to what extent wallets or tapscript implementations have come to rely on long OP_RETURN's.
(3) Non-transaction data may be good: Giving miners a source of income outside of transactions can financially diversify miner income, making security more robust against a crash in transaction demand. Bitcoin already subsidizes HODLers (who don't pay any agio) at the expense of transactors (who pay fees) as coinbase awards (and inflation erosion attendant thereto) decline, so such cross-subsidies already exist.
(4) A meta-level objeciton: It is incumbent on the advocates of BIP-110 to answer these objections, and I haven't seen anyone do it in an analytic way driven by data, formal game theory, or even by analysis of monetary/banking history of previous eras. I don't think it is impossible to do that. Thowing out scare senarios (it is spam, nodes will cost too much, bitcoin won't be money anymore) is not sufficient.

2026-07-11 04:34:27 UTC

- reply

The main risk is if they scrutinized you for something else. Customs agents can barely keep tabs on illicit traffic of cultural treasures; to an untrained eye heirloom art looks like mere junk. I personally heard one first-hand account, around 1993, of someone who shipped a bag full of books to China. The local postman knew how to tie a special knot that came undone by pulling the rope in a peculiar way. When the books arrived he could verify that no one opened the bag...via knot-theory message authentication. The only thing I ever had intercepted by Chinese customs was an obsolete computer (I needed an old-style ISA slot) at a time when they were cracking down on improper recyclers.

2026-07-10 16:56:00 UTC

- reply

Loaded that mp3 on my $14 mp3 player for the gym later today. ✅

2026-07-10 14:54:12 UTC

- reply

Link to that interview which doesn't use an oligopolistic centralized service https://talkingbitcoin.com/podcast
Actual not-spy-on-you download
https://traffic.libsyn.com/secure/forcedn/coinstories/070626_Lyn_Alden_Full_Interview_1.mp3

2026-07-07 15:56:35 UTC

- reply

Video of the disicussion (YouTube) https://www.youtube.com/watch?v=1IchG-uewOE
After watching that, I suspect that the mindset behind NOSTR is gradually shifing from "social media without the suckage of centralized power " to "a digital identity adopted widely enought to be useful". If so (a big "if") then we may attain what PGP digital signatures (on Usenet newsgroups for example) promised but never delivered.