Let's see: https://github.com/bitcoin/bitcoin/blob/master/src/base58.h
Your model with dashes explicitly prevents the feature of selecting the whole address with a double click. Currently, we use bech32m for Taproot, which works quite nicely, and you can just introduce it from the start, with some prefix, matching your coin name: https://github.com/bitcoin/bips/blob/master/bip-0350.mediawiki

If this is the real address, then do you want to use 96-bit addresses? If so, then after around 2^48 operations, you could have collisions.

If you store it as an ASCII string, then you are doing it wrong. But anyway: if your chain would be two times bigger than needed, then it is your problem. And also, are these addresses identical, or not?
Because some users could use A-F, and others could use a-f. If you store it in ASCII, then they would be different, but cryptographically, they could be identical, when parsed to the hex data.

Of course. All users, including Satoshi, should use their real name, instead of some nickname. Let's go full KYC immediately, that would definitely help.

By the way: your idea wouldn't stop the spam, and would only discourage some good users from posting. For example ertil was neutral tagged as an alt account of stwenhao. And now, both stopped posting. Hunt them all, so they will leave the forum, and move somewhere else, where their anonymity will be more respected.

Some recent example, where over 0.22 BTC was sent to a buggy TapScript: https://mempool.space/tx/d69a58b25c81280bb9ef2f72953ec5f495c3cf4b60937cbe92be87bd6f442f43

And then, around 0.02 BTC was burned into fees, probably by some bots, trying to claim it.

It is not that common to see a transaction with 10k sat/vB fee rate: https://mempool.space/tx/69e515c92295d2f1e9d9643a270d5e86acf579b834f8fb22b316d0c9cb3f2a0a

I think you are trying to do everything you can, to be 100% sure, that you won't be a merit source.

Oh, so you suddenly don't like AI anymore?

Maybe the current situation will wake up some people, and they will stop calling people "AI Luddites", and really notice, that a lot of humans are so lazy, so the main use case of AI on forums will be just spamming. Any productive use of AI will require some effort, so many users won't do that, if it would be almost as hard, as making it without any AI.

I guess we will see even worse AI generated cases, where it would be harder and harder to verify their quality. Some people are trying to poison some models, or even seriously think about stopping their contributions to the Open Source world, to not feed huge models with too much content. I already saw some people moving their repos out of GitHub, to avoid them being constantly scanned by bots.

What about this one: https://bitcointalk.org/index.php?action=profile;u=2745308

It depends, what exactly is placed in the output Script. If it is blocked by BIP-110, then it will be invalid on that chain, but still valid on BTC.

I guess the easiest way to check it, is to run some local regtest, produce a transaction with similar Script, but different keys, and then try to spend it, after artificially reaching a given block height or timestamp. Then, as long as BIP-110 proponents don't try to block any public keys, it should give a proper answer.

However, I wouldn't worry too much about it, as long as BIP-110 has much less than 50% support. Currently, they have something like 1%, maybe 2%. As long as BTC community would reject such proposals, you would only lose some altcoins, but not BTCs.

Possible outcomes: https://www.youtube.com/watch?v=P9PQr4Fj5tQ

I guess BIP-110 will simply fail, or become some irrelevant altcoin, but only time will tell.

Is it tested in regtest, that it works correctly with far into the future timestamps? Because there are four bytes per locktime, so after 2038, it can be tricky, because of this bug: https://en.wikipedia.org/wiki/Year_2038_problem

Unless it is simply locked to some future block height, but then, the accuracy of that could be questionable, and instead of being unlocked in 2050, coins could be spendable in 2049 or 2051, depending on the exact hashrate fluctuations. For example: someone in January 2009 could assume 52,500 blocks per year, and assume, that to lock coins for 18 years, and get it unlocked in January 2027, it would need to be a block height of 52500 * 18 = 945000. And then, it will turn out, that instead of January 2027, it would be unlocked in April 2026.

It can be trivially solved:
Which means, that if you have z-value, r-value, and s-value, then you can calculate just "z/(s-r)", and recover the private key in this way. It is unlikely, that there are any such transactions in the wild, because anyone could easily sweep these coins.

This can easily result in banning peers, which you don't want to ban, or not banning those, who you want to target. A better idea would be to observe the traffic, and make decisions, based on that.

Which means banning the client, regardless of the User Agent, just if it consumes too much resources.

Probably the easiest way is to use RPC commands, to measure some things, and then make an external script, which will execute RPC commands on given peers.

Again, it would help only for a while, but users can change their User Agent as they wish, and then, blocking next peers will be harder, if they will do that.

This is what can be easily used. And I guess it should be enough, to get the amount of bytes, sent or received by a given node, and then ban some of them, which use too much resources.

Below 80 bytes.

It was always just an anti-spam tool. Just when there were less merits in circulation, some people thought, that it is about finding "good" posts, or "posts you agree with". But it was simply not the case.

It is a similar mistake, as if someone would think, that Bitcoin transactions are free, because they were in 2009. It is not the case, because the system is not designed in this way. And the same with merits: they are needed only to filter spam, and not to reward great posts.

Sending merit is a simple signal: "this post is not a spam". Nothing more, nothing less.

I can see the opposite problem: there are too many merits in circulation, and the only way to burn them, is to send them to old, banned accounts, like satoshi.

Let's see: https://github.com/bitcoin/bitcoin/pull/8149

If you want to find a commit, where Segwit was introduced, then it is here: https://github.com/bitcoin/bitcoin/commit/d612837814020ae832499d18e6ee5eb919a87907

Only as "immortal" as CloudFlare is. Of course there are backups, but by taking down third party services like that, it can be easily turned off.

People didn't want to agree on merit burning, by sending even zero, so they won't agree on going negative.

Which means, that the only option is to send merits to blocked accounts like satoshi, because other ways of burning excessive merits were rejected.

There are places, where you have more bots than humans. Or, where you have a lot of stupid answers, and even if they are produced by biological species, their answers are still more and more worthless. I guess for some people, AI made more harm than good, and instead of improving, some people started producing worse answers, than they did in the past (maybe because some AI models also degraded, and their 2025 answers can be worse than their 2022 answers for the same prompts, for some reason).

I guess it must be hard to be a merit source today, when you enter some board, and there are no good posts to send merits to.

Similar topic: Is this forum getting worse and worse?

A section literally called "Mining".

Some mining pools are just not taking low-fee transactions, below 1 sat/vB. Because they don't have to.

Exactly. Fully-filled blocks with just 1 sat/vB would mean 0.01 BTC in fees per block. If it will be lowered to 0.1 sat/vB, then it means 100k sats per block. If the basic block reward will be halved, and the fees will be too low, then the end result will be just like in testnet3: 1192 satoshis per block, and low fees, would lead to a low difficulty.

So, it is a question about long-term incentives: if the current fees will be lower and lower, then how they will raise, when it will be needed? Because users will happily pay one satoshi per transaction, so if we assume 250 vB per transaction, and 4k transactions per block, it would mean 4k satoshis as a reward from fees. And then, who will mine a block worth 4k satoshis in the future? Or: how easy it would be to reorganize the chain? Or even: how many confirmations will be needed?

I guess you are trying to send more coins than you have. Because 1.177 mBTC is your balance, which means 0.00117700 BTC. But you are trying to send 0.02 somehow.

However, it seems that you got what you wanted anyway: https://mempool.space/testnet4/tx/90faba7e5fbbe17be50b44f216d039a57fbf611101ad3ad91c5ec95e9c0b2fc0

Sure, that's why the first metric to check, is activity to merit ratio (which is visible on the left, and can be easily checked). So, if you want to avoid punishing people with more posts, then you can replace posts to merit with activity to merit, and then it should work better.

Many statistics on BPIP can be checked for the last 30 days, 90 days, 120 days, or 150 days. Even if there are bubbles, then they can reach more stable values over time. So, to avoid punishing some users, then it can be counted per more posts, or per more days, and only applied then. For example: for lower ranks, for the first 100 posts or so, rules can be less restrictive. But if someone made 100 posts, and didn't receive a single merit, then it definitely means, that there is some problem. And if that user can write hundreds of posts more, and also remains unmerited, then something is at least suspicious.

Of course. But forum is mainly made for longer than shorter posts. For shorter messages, there are other channels, like IRC, Discord, or whatever. And also, forum is prepared for the content, that needs to be indexed.

Yes. But it is a clear signal: "try harder". Or "write better content". People don't have to make masterpieces, rewarded with 50 merits. Much more important thing, is if they can get any merits from anyone.

And it is by design. To get at least a single merit from many sources, it is enough to just get their attention. If the content you make is not a spam, and if they reply to you, by just quoting it, then you should get at least a single merit from them. So, 1:1 ratio is achievable just by "being noticed", and "writing in the always-merited places". And then, going to "1:0.5" is all about "putting some effort". So, picking 1:2 or 1:5 is just "extremely easy". But well, even 1:10 is better than nothing.

Topics like that are blacklisted by many sources, just because of the title. Also, if there is "a thread with more than 1000 pages", then it is usually considered as "the bubble for spammers". In that case, it is good, that we have such topics, because their main purpose, is to keep spammers in their own honeypot, and let other users talk in shorter topics, where things are easier to find and discuss.

Also, topics with a lot of pages, are needed mainly for entertainment. Serious topics are splitted out of such places, and developed separately. But because first, you need a content, and then you can give it a title, it is a good brainstorming tool.

Yes. Usually, reaching something around 1:0.5 is easy. I think the limit should be somewhat like 1:1 or 1:2. Because 1:1 basically means reaching a single merit per post, and 1:2 means one merit per two posts. For 1:0.5, it means two merits per post, and many people send two merits, instead of one, just because it allows sending back one merit by the receiver.

Also, the activity counter can tell something. Usually, it should be 1:1, meaning that your post count is the same as your activity. If it is not, then it is usually a sign of writing simply too much, and then, the quality can be easily increased, by making a delay. For example: if someone has one activity point per four posts, then by just posting 4x less things, the quality could usually significantly go up, by spending that time, to make one better post, instead of four average ones.

That's why alt-accounts are needed: to say, what you want to say, without worrying about merit blacklists. Usually, owning at least two accounts is needed, because then, one account can simply "be nice", and earn enough merits, and another account can "say the truth". Also, it is useful, if you want to argue with yourself, because many people cannot stand the fact, that you can disagree with your old posts, and for that reason, alt accounts are useful, because otherwise, people would believe, that the account changed hands, if you start judging your old arguments.

In many places, removing/editing an old post is blocked. Fortunately, there are many crawlers here, so even if someone destroys "the property of the forum", by making old topics less useful, then the previous content can still be accessed. But in the case of just counting some ratios, removing posts can simply not decrease the counter, and that should solve it. Which means, that if someone produced 1000 posts, and removed 900, then if 1000 will be used, instead of 100, when counting ratios, then it should be hard enough to game the system.

A rule of a thumb is to reach 1:0.5 ratio, which means two merits per post on average. Better results are great, but worse results usually means, that there are too many posts, and their quality should be increased. Another rule of a thumb, is to have 1:1 post to activity ratio, which means sending one post per day on average. Then, by just comparing these two counters, it is easy to determine, if a given post should have a delay of publication, or not. And then, if there are many delayed posts, then there is more than enough content, to moderate it, and make one good post, instead of many worse ones.