Posts by StJohn Piano

Showing posts 1-10 of 62 (most recent first)

#141 | 2026-06-17 08:07:52 UTC
0 replies

When you need some memory (to hold some data) you request it from the operating system (for example using the malloc function in C) and when you're done with it, you release it (for example using the free function in C). If you fail to release the memory, you end up with a memory leak. If you accidentally keep a reference to memory that has been released, you get a dangling pointer, which can be a security vulnerability. A garbage collector is a system that does all this for you. Whenever you create an object, the program automatically keeps track of all the references to that object. When the object is no longer referenced anywhere, it automatically releases it, freeing up its memory. This way, the user (mostly) doesn't need to handle any memory resources. Source: https://www.reddit.com/r/explainlikeimfive/comments/io535j/comment/g4bik87 NB: A garbage collector is part of the runtime implementation of the language, not part of the language itself.

#140 | 2026-06-16 15:23:16 UTC
0 replies

/ We've had fast languages like C and C++, and then we've had safe languages like Lisp, Java, and Python. The safe languages were all slower. Common wisdom said that a programming language could either be fast or safe, but not both. Rust has thoroughly disproved this, with speeds rivaling C even when writing safe Rust. What's even more impressive is that Rust achieves safety and speed without using a garbage collector. Garbage collectors can be very useful, but they also tend to waste a lot of memory and/or create CPU spikes during GC collection. But more importantly, GC languages are difficult to embed in other environments. The big innovation leading to this "fast safety" is the borrow checker. ... The gist of it is: each piece of data has exactly one owner. You can either share the data or mutate it, but never both at the same time. That is, you can either have one single mutating reference to it, OR many non-mutating references to the data. / https://rerun.io/blog/why-rust

#139 | 2026-06-09 11:24:29 UTC
0 replies

Tela Network has become significantly become more interesting with the introduction of Tela Network AI. Query: Explain the ruling class Answer excerpt: / Short synthesis: In this network, the ruling class is the minority that can organize violence, command courts and legislatures, and coordinate elite interests - whether framed as Mosca's "man of violence," Turchin's ruling elites, or the faction controlling the money supply. The open question in several posts is whether that class still believes in itself enough to rule effectively [#13]. / Full answer: https://tela.network/ai-answers/published/a4bbc82a-8422-4f2f-9ef6-291c6d327cb9/ The AI builds its answers from the published posts written by network users. This answer is already much more interesting than a stock response from a normal AI. Posts are cited and linked. Authors are credited. This, to me, is the good future of AI: Amplified Intelligence

#135 | 2026-05-17 14:40:02 UTC
0 replies

/ The state functionally went bankrupt, and was dissolved in what amounts to essentially a corporate-sovereign restructuring. Starting in the 1970s and accelerating thereafter, the Soviet growth model - extensive capital accumulation poured into new urban factories, power plants, steel mills, and machine tools - had effectively run out of low-hanging investment opportunities, resulting in persistently diminishing marginal growth potential. / A total lack of ... creative destruction on one side, and ballooning imperial obligations on the other. / To paper over these holes, the Soviet Union financed its civilian consumption and military spending - along with the subsidies it paid to its puppets - through oil rents made possible by the global energy crisis of the 1970s. But when oil prices normalized again in the mid-1980s, the regime was forced onto international debt markets, where it was eventually strangled. / https://boydinstitute.org/p/debt-killed-the-soviet-empire

#134 | 2026-05-16 08:49:47 UTC
0 replies

There are eras of media, and each era has its own rules for how to craft an effective message. / There was a time when brands or individuals could rely on the organic reach of their efforts, but the pay-to-play model has been standard for years. / There exist several networks which gather freelancers to produce and distribute podcast clips. / There are of course creators who achieve this by partnering with a talent agency offering essentially the same services under the hood. / Creators can also replicate this on a smaller scale by employing a team to chop, bedazzle, and post clips designed to punish the algorithm with exposure until something hooks. / there is a dollar rate, there is frequency, and there is duration. It operates like any other campaign. / It should be noted that this isn’t a cheat code and the content itself must be remarkable in some way. Source: https://substack.com/home/post/p-197483620

#133 | 2026-05-15 11:11:13 UTC
0 replies

Viewpoint = what someone thinks Mythos = the story-world that makes those thoughts feel true / Re: 'silver is an industrial metal and has no monetary value'... The awakening is spreading and everyday people are looking for a place to keep their cash. They are getting priced out of gold - silver is their only option. It is ultimately the people who will decide what has monetary value and what doesn't. / Source: An (edited) comment from: https://youtu.be/tQa6mNXcM4s We can see an underlying doctrine: People can decide the monetary standard. And another one: Gold is too expensive for the average person, therefore it can't be the primary monetary standard. What is most interesting to me is that it is a democratic inversion of the goldbug mythos, which says: 1) People must adapt to the most powerful monetary standard. 2) The best monetary standard will have the highest price. Summary: Goldbug mythos: monetary truth is discovered. Silverbug mythos: monetary truth is ratified.

#131 | 2026-05-09 08:34:56 UTC
0 replies

http://edgecase.net/articles/bitcoin_transaction_test_set_2 This article contains 16 Bitcoin transactions, with 1-3 inputs and 1-3 outputs. All transactions are onchain - and thereby confirmed to be completely valid (there is no higher level of validation). You can take each txid and look it up in a block explorer. Private keys are included. All addresses are Pay-To-Public-Key-Hash (P2PKH). I published it on 2020-12-19. A Bitcoin transaction is a complicated object. Getting one of its aspects wrong can produce a transaction that looks plausible but is invalid, or worse, valid in a way the developer did not intend. The value of a test set like this is that it gives a developer something fixed and confirmed-true to compare against. You can run your code against known cases and check whether it produces the expected transaction IDs, signatures, serialized bytes, or decoded fields.

#130 | 2026-04-19 13:53:16 UTC
0 replies

I analyzed the Bitcoin v27.2 codebase and found some general metrics about it. These tools & approaches for finding metrics can be used on other versions of the codebase, and on other repos. https://telablog.com/bitcoin-v27-2-codebase-metrics I wrote the draft as a text file, and (with Cursor) built a few Python scripts to: - construct the article from the draft file + the prefix & suffix files. - extract / deduce the title and the custom excerpt. - push the article to the publishing platform via its API. -- Notes: --- This is idempotent: It doesn't create the article again if it already exists. --- The body is one large pre-built markdown entry. This avoids the need to manually interact with the body in the editor. Of interest: I used AI here to build & iterate on /deterministic/ tools (the Python scripts). I did /not/ try to build a processor that would use AI to indeterministically i.e. unreliably run the publishing action sequence.

#129 | 2026-04-18 08:51:02 UTC
0 replies

" > You have an outline > and proposal for such a design, which is a big step > forward, but the devil is in the little details. I believe I've worked through all those little details over the last year and a half while coding it, and there were a lot of them. The functional details are not covered in the paper, but the sourcecode is coming soon. I sent you the main files. (available by request at the moment, full release soon) " - Satoshi https://www.metzdowd.com/pipermail/cryptography/2008-November/014863.html Date: Mon Nov 17 12:24:43 EST 2008 A year and half earlier would be: May 17 2007 So he was inspired to begin working on Bitcoin on or before that date. Often it's said that Satoshi was inspired by the 2008 financial crisis and the ensuing bailouts (which he cites in the genesis block), but this claimed start date is much earlier. He was probably inspired by the initial subprime mortgage crisis, which began in February 2007. (The US housing bubble popped in mid-2006.)

#128 | 2026-04-15 05:31:14 UTC
0 replies

"The credential that establishes someone as real is the ability to supply CPU power." - Satoshi https://www.metzdowd.com/pipermail/cryptography/2008-November/014860.html He is referring to identity within the Bitcoin mining system. It occurs to me that: In the political world, the credential that establishes someone as real is the ability to supply violence. This violence may be direct or delegated, wide or limited, open or hidden, embraced or disavowed, demanded or inspired, but it must. be. present. In the same way that Bitcoin mining ignores all identities other than those that supply CPU power, politics ignores all identities other than those that supply physical force. On top of Bitcoin mining, we can build addresses, transactions, identities, property rights, immutable publishing, and secure communication. On top of violence, we can build towns, cities, temples, polities, law courts, houses, families, and futures. But: Without a foundation, nothing can stand.