semisol

semisol

👨‍💻 software developer 📨 nostr.land relay all opinions are my own.

npub

npub12262qa4uhw7u8gdwlgmntqtv7aye8vdcmvszkqwgs0zchel6mz7s6cgrkj

pubkey (hex)

52b4a076bcbbbdc3a1aefa3735816cf74993b1b8db202b01c883c58be7fad8bd

nprofile

nprofile1qqs99d9qw67th0wr5xh05de4s9k0wjvnkxudkgptq8yg83vtulad30gprf58garswvaz7tmjv4kxz7fwva6kcat8w4k82tnddajsfkdmn2

Activity (879)

gm nostriches 👀

Kind-1 (TextNote)

2026-06-29T11:47:55Z

↳ Reply Event not found

1b07a748ab194ce61380b4c4268a1d7d799234b9856f55d11bf5e2703ef6a892

wat

Kind-1 (TextNote)

2026-06-29T08:14:52Z

BIP-110 and Core is a false dichotomy

Kind-1 (TextNote)

2026-06-28T21:55:36Z

↳ Reply The Daniel 🖖 (npub1aeh2zw4elewy5682lxc6xnlqzjnxksq303gwu2npfaxd49vmde6qcq4nwx)

Is it designed to fully backward compatible, or will there be conflicts with clients that don’t supp...

On the extension side, yes. Same can be done on the app side to autodetect the version. For DMs, there is a designated tag to the DM relay list so th...

On the extension side, yes. Same can be done on the app side to autodetect the version. For DMs, there is a designated tag to the DM relay list so that it is only used for when the recipient supports it.

Kind-1 (TextNote)

2026-06-28T21:20:13Z

↳ Reply Event not found

ee781e131d4ebc4bdbec48dceae44e51df2e16d40d2dab9cfac27764cee0621a

Can’t. Too busy right now, even if I let an LLM barf up a PR. However, if you’d like to provide a more secure encryption UX for users, you can point ...

Can’t. Too busy right now, even if I let an LLM barf up a PR. However, if you’d like to provide a more secure encryption UX for users, you can point your LLM to the repo. It is more than enough.

Kind-1 (TextNote)

2026-06-28T15:54:52Z

↳ Reply Event not found

7bd732a1fe1b3c0e12b9f1dec2e39e0834d7847fc9fa5d38c0f3efb67e578e73

Well, it is very subject to change compared to paid services.

Kind-1 (TextNote)

2026-06-27T01:11:59Z

↳ Reply Event not found

145bd8380490f45964d6430610b5f20c5c6154877020e233f2bb635bcc197a9b

There is no free

Kind-1 (TextNote)

2026-06-25T22:38:59Z

↳ Reply Event not found

6007efd3d8233de7c85e940ac6af68bdef540c764af6bc63ec0aea3520139bde

Technically it would be possible for it to explain it, if it did “reason”. The KV cache is actually just a processed version of the model’s intermedia...

Technically it would be possible for it to explain it, if it did “reason”. The KV cache is actually just a processed version of the model’s intermediate states on each layer. But there is no reasoning.

Kind-1 (TextNote)

2026-06-25T15:15:40Z

↳ Reply Event not found

bc6ef755720b35d0809381b47d1f584bb34a8fa6c37fcb3af2fb354efc13cccd

It is not. My point is that open source software does not automatically imply a security budget, meaning that even if you *could* see it, doesn’t mean...

It is not. My point is that open source software does not automatically imply a security budget, meaning that even if you *could* see it, doesn’t mean you do.

Kind-1 (TextNote)

2026-06-23T22:46:37Z

↳ Reply Event not found

97341e6731963d5f25338791ef33952e7c884d8182cbc09335aac4de411ad2e3

I did not say they were mutually exclusive? Only that there is no correlation, at all, and it is not “better” for security in 99.99% of cases

Kind-1 (TextNote)

2026-06-22T19:51:36Z

↳ Reply Event not found

6729e85b9a536c74085febd88929f50d8f2514409b957889a33c8706c59fce85

Them being visible *with a lot of effort* does not mean that they will be seen or reported or fixed. And on the other hand, security researchers can ...

Them being visible *with a lot of effort* does not mean that they will be seen or reported or fixed. And on the other hand, security researchers can also do security testing on closed systems, which happens all the time.

Kind-1 (TextNote)

2026-06-22T18:20:13Z

↳ Reply Event not found

02f2972082a2c6cf78bba77878b44244b0e60e6499a5a71358fd6e68effc840c

Remind me in an hour

Kind-1 (TextNote)

2026-06-22T17:22:32Z

↳ Reply Event not found

b32acc0b693e74290d29c3588be60298e25bae6226623c2e494c626e319e03de

It means that the security of a piece of software does not change because it is open source

Kind-1 (TextNote)

2026-06-22T17:22:03Z

↳ Reply Event not found

90f1fb2b6b4fe4f800500d1ee66db71233bdac18cbce13b59a58260ae01633f4

I think it is completely uncorrelated.

Kind-1 (TextNote)

2026-06-22T17:20:12Z

↳ Reply Hanshan (npub1lxzaxzge0jq9u9cecucctdt5lslwgp7hcxmp2l0wn8r2ecjenwasu6svxa)

not even as t approaches infinity?

Technically yes, in practice you have 1 or 2 yrs at most. During this time, most pieces of software, especially specialized ones, will have on averag...

Technically yes, in practice you have 1 or 2 yrs at most. During this time, most pieces of software, especially specialized ones, will have on average 0-10 people look at it. Out of these, approximately 0.01 will report any issues. The biggest and most secured OSS projects are corporate backed with many companies having a vested interest in its security, not a medium sized company or random person dumping their code on GH

Kind-1 (TextNote)

2026-06-22T17:19:35Z

Open source does not mean more secure.

Kind-1 (TextNote)

2026-06-22T17:13:20Z

↳ Reply Event not found

1970c26d9a79a6bb8b56dd4d6bd182af2f729f3b927d57d73debb83473580dbb

https://fossil-scm.org/

Kind-1 (TextNote)

2026-06-22T08:40:01Z

By this, a majority of zaps cost more to process in LN fees, server load and human time than they are worth. nostr:nevent1qqstdjw3h0tgnn63a8e07n426nc...

By this, a majority of zaps cost more to process in LN fees, server load and human time than they are worth. nostr:nevent1qqstdjw3h0tgnn63a8e07n426ncqzacgc2ffupdexkwhke0jrv0025gpz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0dsq3qamnwvaz7tmwdaehgu3wd3skueqpz4mhxue69uhkzem8wghxummnw3ezumrpdejqygzjkjs8d09mhhp6rth6xu6czm8hfxfmrwxmyq4srjyrck9707kch5psgqqqqqqs7hqzgf

Kind-1 (TextNote)

2026-06-20T16:18:05Z

Most zaps are pointless

Kind-1 (TextNote)

2026-06-20T13:48:17Z

gm 🤔

Kind-1 (TextNote)

2026-06-20T13:34:31Z

↳ Reply semisol (npub12262qa4uhw7u8gdwlgmntqtv7aye8vdcmvszkqwgs0zchel6mz7s6cgrkj)

I would say that the figure of 1000 notes for a feed is too much. It is usually enough to fetch 100-...

As a relay operator, I have very little concern about the impact on me or my users arising from a kind 1 feed. Instead, I am more concerned about the...

As a relay operator, I have very little concern about the impact on me or my users arising from a kind 1 feed. Instead, I am more concerned about the follow lists which can get huge and use a lot of data. On this topic I am thinking to introduce an HTTP API for relays

Kind-1 (TextNote)

2026-06-19T22:05:12Z

↳ Reply Event not found

16f4118bfe308282821c93dc87f2f52efc6b1aef90cfb3e7f093cac50809f80e

I would say that the figure of 1000 notes for a feed is too much. It is usually enough to fetch 100-200. Even in this case, advanced relays like Nost...

I would say that the figure of 1000 notes for a feed is too much. It is usually enough to fetch 100-200. Even in this case, advanced relays like Nostr.land already apply compression to the sent data. The codepath duplication caused by kind 1 is limited. It would be treated like a reply either way, with slightly different parsing. A migration to kind 1111 is simply not possible. It would involve the sequential migration of every post, with interaction from every author, to keep the IDs consistent. A half-complete migration would be useless. For a switchover date, it would be hard to coordinate this. 50% of what, for example? Users? Clients? Who defines the “everything” set? This could be implemented as a filter optimization, rather than a break. The benefits of kind 1111 for kind 1 posts are negligible, considering that a majority of the data usage from clients come from images, not relays. Many clients also have a replies-inclusive feed, which users already switch to frequently.

Kind-1 (TextNote)

2026-06-19T22:01:08Z

↳ Reply Event not found

efa67c3567ccdd843d2b1eea866148d82e7ab923661b13bd6da1a7da32c72673

Users are not impacted by kind 1. For clients and relays, this can be said about many decisions. For example, the current signature format is awful fo...

Users are not impacted by kind 1. For clients and relays, this can be said about many decisions. For example, the current signature format is awful for performance due to requiring costly JSON-encoding. I unfortunately think that it is too late for kind 1 at least. Users care about the end result, and for them, very little changes. To them it looks like losing immense amounts of posts for no reason. Protocols like negentropy already fix the data usage problem, and the filtering of replies is of low concern imo Also: Anyone also trying to unilaterally force this is an idiot.

Kind-1 (TextNote)

2026-06-19T19:38:57Z

↳ Reply Event not found

e656f8ebf52e16ae26d1e29d101ffeda93feb235c2ea4e8e5b708aaea6a81715

Yes, but the impact is on clients which will lose years of old data, along with a “transition” that will never finish.

Kind-1 (TextNote)

2026-06-19T18:20:20Z

↳ Reply Event not found

0d306ad725dff86e6e3d2ed34de9e1dd53f2b827851b44091707e132c55269ca

I do not believe that the “performance improvements” to relays, data usage that is miniscule (plaintext) compared to media downloads or follow lists a...

I do not believe that the “performance improvements” to relays, data usage that is miniscule (plaintext) compared to media downloads or follow lists are worth breaking backwards compatibility. I wouldn’t even notice a change in the stats for Nostr.land I’d say.

Kind-1 (TextNote)

2026-06-19T18:13:07Z

↳ Reply Event not found

5577a6af21ad609bf5cdaee2b89cb16d5571500cf8562898de985d84d9874019

I have not seen any response from them that addresses the claims.

Kind-1 (TextNote)

2026-06-19T13:45:39Z

🤔

Kind-1 (TextNote)

2026-06-17T22:26:36Z

True unlimited data plans (no “throttling after X GB”) are amazing

Kind-1 (TextNote)

2026-06-15T16:47:57Z

↳ Reply The Fishcake (nostr.build) (npub137c5pd8gmhhe0njtsgwjgunc5xjr2vmzvglkgqs5sjeh972gqqxqjak37w)

Ask nostr:npub1jk9h2jsa8hjmtm9qlcca942473gnyhuynz5rmgve0dlu6hpeazxqc3lqz7 😂, we are two different pe...

As an independent party I can attest to this fact

Kind-1 (TextNote)

2026-06-15T09:44:40Z

↳ Reply Event not found

0ce5a410e91d98fa4ea4bbc8eabd374132c50de2f5f885121cff63896e7c3042

Nostr.land does this.

Kind-1 (TextNote)

2026-06-15T09:35:28Z

Please tell me what your main concerns with Nostr relays are, especially in terms of privacy.

Kind-1 (TextNote)

2026-06-14T20:11:01Z

↳ Reply Event not found

3856d2f0818085a75b4bbfbf97b106c9f4039a80c1436ce591297d1b1e4bb148

It is a company resource provided to you. Unless they encourage you to waste tokens, then you should use it only for work and efficiently. If they d...

It is a company resource provided to you. Unless they encourage you to waste tokens, then you should use it only for work and efficiently. If they do, you should not be profiting off of it or putting the company at risk.

Kind-1 (TextNote)

2026-06-14T20:09:09Z

Corporate “Bitcoin” events are a waste of time

Kind-1 (TextNote)

2026-06-14T11:11:01Z

↳ Reply Event not found

5d2439332496463d47fb13780e53a4e670f5065a817224ba63999f1dc08aa1df

and Intel SGX keeps getting broken in general

Kind-1 (TextNote)

2026-06-14T05:59:01Z

No motive, monetary cost, and easily blocked. What is the point?! nostr:nevent1qqs880lpfnvnxt6t3kfqnu0pdf3apeargmea4smcp0f97vr7ztkxmkqzypftfgrkhjammsa...

No motive, monetary cost, and easily blocked. What is the point?! nostr:nevent1qqs880lpfnvnxt6t3kfqnu0pdf3apeargmea4smcp0f97vr7ztkxmkqzypftfgrkhjammsap4marwdvpdnm5nya3hrdjq2cpezputzl8ltvt6qcyqqqqqqgpz3mhxue69uhhyetvv9ujuerpd46hxtnfduuv04k6

Kind-1 (TextNote)

2026-06-12T20:19:43Z

I wonder who is paying for all the tokens to run these slopbots?

Kind-1 (TextNote)

2026-06-12T20:17:03Z

Slop will not be tolerated

Kind-1 (TextNote)

2026-06-12T16:06:51Z

A pay per use model would significantly enshittify paid relays, lead to horrible UX and allow tons of spam: 1. It would lead to unpredictable charges...

A pay per use model would significantly enshittify paid relays, lead to horrible UX and allow tons of spam: 1. It would lead to unpredictable charges for users. A fixed fee per month/year means the relay just works. 2. The cost acts as a bond for spam and abuse, compared to pay-per-note. 3. It is incompatible with services like aggr.nostr.land and NIP-05 where there is no “usage”. 4. It would allow a monetary DoS for services like inbox relays, where someone can spam you with junk giftwraps. 5. Users should not need to worry about constantly having a topped up wallet and using a supported client.

Kind-1 (TextNote)

2026-06-12T11:54:25Z

&time>now-60&time<now+60&id=03…&method=pay&pinvbolt11_node=03…

Kind-1 (TextNote)

2026-06-11T16:56:01Z

↳ Reply Event not found

7c92e812c5c05e9d7f378d173b7aa935ed02f2c21dde1eea5e3a995b352f4c7e

if done right: harder usually: a little bit

Kind-1 (TextNote)

2026-06-11T13:55:23Z

↳ Reply Event not found

a7832db157301ec75677cc619fa9abba806a48bad5741988bd2446f0af06e722

Any automated ways?

Kind-1 (TextNote)

2026-06-11T13:55:00Z

↳ Reply Event not found

da66db47722df5d3381f9561252496489ee12e5fc138b0bc48fdf0bb9237aab5

Nostr.land is planning to support OTS. I’m planning to use the standard OTS spec that is already well-implemented. “Inline” proofs is not something I...

Nostr.land is planning to support OTS. I’m planning to use the standard OTS spec that is already well-implemented. “Inline” proofs is not something I plan to support as a client can easily request OTS proofs when necessary. Also, I hate AI written “specs” so much as they have so much useless junk (like wtf is the s tag for?) and overcomplicate things.

Kind-1 (TextNote)

2026-06-11T13:49:32Z

CLINK or whatever this protocol is, is worse than NWC. What is the point?

Kind-1 (TextNote)

2026-06-10T16:06:08Z

Code is a liability, to some extent. Each feature and line incurs a maintenance cost. Less code is better. nostr:nevent1qqs9mucnwftfmgzk9s4769024ldgf...

Code is a liability, to some extent. Each feature and line incurs a maintenance cost. Less code is better. nostr:nevent1qqs9mucnwftfmgzk9s4769024ldgfqermplk9w5yp0z0qalm30vk2vgzypftfgrkhjammsap4marwdvpdnm5nya3hrdjq2cpezputzl8ltvt6qcyqqqqqqgpz3mhxue69uhhyetvv9ujuerpd46hxtnfduq4njx9

Kind-1 (TextNote)

2026-06-10T16:04:41Z

Code has always been cheap. You don’t need an LLM to generate slop. For me, nothing has changed

Kind-1 (TextNote)

2026-06-10T15:29:15Z

↳ Reply Event not found

7a389b3935693bd914f8db14074f182eb7d31037b999ab96efe67c3c657b5b46

This only makes me less confident, unfortunately. “Your keys are never stored in a usable form” does not make sense, as that would mean that you can’...

This only makes me less confident, unfortunately. “Your keys are never stored in a usable form” does not make sense, as that would mean that you can’t use them. A PIN can be easily brute forced if you break the secure element, as it needs to know the PIN to verify it. Tamper detection switches are not very good protection. Attacks have been done on devices with multiple layers of tamper protection with an entry less than 2mm. The OS is “fine”, but claiming Xous’ reputation while having majorly changed it for your own usecase is weird. It’s not the same thing anymore. My other points still stand. I’m not going to assume malice or intentional deception, but it feels like they do not really have much experience with secure hardware.

Kind-1 (TextNote)

2026-06-10T13:36:06Z

Anthropic has released a new money-incinerator that predicts probabilities better than the previous money-incinerator and also incinerates your money ...

Anthropic has released a new money-incinerator that predicts probabilities better than the previous money-incinerator and also incinerates your money faster

Kind-1 (TextNote)

2026-06-10T06:36:41Z

↳ Reply Event not found

00008a41e75d77316d2aef739148ec33568df4af689c0ff1d3a561b20d1d5c0c

Yes your client would still have to AUTH that is it. Gift wrap spam/DoS being hard to mitigate is one reason for this, and especially mass-DM scambots...

Yes your client would still have to AUTH that is it. Gift wrap spam/DoS being hard to mitigate is one reason for this, and especially mass-DM scambots.

Kind-1 (TextNote)

2026-06-10T06:33:58Z

↳ Reply Event not found

342e66a29a9115527feb9aa0f064855a7719496e4cebbace3947a7a7579c5f72

The UI is being retarded

Kind-1 (TextNote)

2026-06-09T22:43:18Z

↳ Reply Event not found

00003f880a776c61ba889fbcea290c30c6fb69b27039a969fc014ea0d3f1163b

The bot is not wrong. It says !payment (inverse) not payment

Kind-1 (TextNote)

2026-06-09T22:39:52Z

↳ Reply Laeserin (npub1m4ny6hjqzepn4rxknuq94c2gpqzr29ufkkw7ttcxyak7v43n6vvsajc2jl)

That's probably why they picked it.

I believe it is fiatjaf’s thing.

Kind-1 (TextNote)

2026-06-09T21:46:59Z

↳ Reply ChipTuner (npub1qdjn8j4gwgmkj3k5un775nq6q3q7mguv5tvajstmkdsqdja2havq03fqm7)

Not having to require auth to limit broadcasts. That's all.

Well, auth depending on the pubkey acts as a strong anti-spam signal. It also indicates explicit consent for anything you send to it, instead of someo...

Well, auth depending on the pubkey acts as a strong anti-spam signal. It also indicates explicit consent for anything you send to it, instead of someone being able to say “well I didn’t put it on nostr.land”

Kind-1 (TextNote)

2026-06-09T21:45:22Z

↳ Reply ChipTuner (npub1qdjn8j4gwgmkj3k5un775nq6q3q7mguv5tvajstmkdsqdja2havq03fqm7)

Could you do this by limiting based on timestamp instead?

To some extent. Nostr.land already requires AUTH for a huge amount of features, so it would encourage devs to actually support it properly. And for s...

To some extent. Nostr.land already requires AUTH for a huge amount of features, so it would encourage devs to actually support it properly. And for some things like gift wrap spam filtering, it would not be possible at all. What is the use case?

Kind-1 (TextNote)

2026-06-09T21:35:24Z

I am considering requiring NIP-42 authentication for a majority of event writes. It would prevent random rebroadcasts where it is not desired and all...

I am considering requiring NIP-42 authentication for a majority of event writes. It would prevent random rebroadcasts where it is not desired and allow for better spam filtering, so why not?

Kind-1 (TextNote)

2026-06-09T21:29:07Z

↳ Reply Event not found

cf5bbb9ced248db199e94797cda4747eec23c16e9afed43d3afffd20309b047d

relays.land is a completely unaffiliated service that comes with way less, and it is confusing that they picked that name :/

Kind-1 (TextNote)

2026-06-09T21:20:47Z

The current grant structure has successfully led to the exit of most competent devs. Now, most “devs” are looking for quick grant money for vaporware...

The current grant structure has successfully led to the exit of most competent devs. Now, most “devs” are looking for quick grant money for vaporware, or want to cosplay software development with AI slop tools.

Kind-1 (TextNote)

2026-06-09T18:56:12Z

↳ Reply Event not found

b8aa1b915f3a415ba8f94cfbe615a072ab08bcdf57371796a0ad769ce24afca8

More generally: The urge to add “comments” and “labels” that state the obvious / are redundant to places where they do not belong

Kind-1 (TextNote)

2026-06-09T16:35:34Z

↳ Reply Matt - Top American Spook (npub1l6scds4yv7xmcsmhqnhdy9sggm520q09lvts2m5mkvecgr2mmmeqsuj5rc)

Lol Buddy, this is why I use like two Nostr apps. Amber and my client. I don't think devs or fundin...

Funding currently is unconditional and so selects for people that do the bare minimum and want the most. AKA burning money

Kind-1 (TextNote)

2026-06-08T21:30:25Z

↳ Reply Event not found

b0812987f1e237b8379a368ecc2805f2306f0c9a4cec2593d43333dc718e626e

I frontran basically everyone. and I am very sad at the rapid decline in the last year especially due to AI slop, and wannabes larping as devs while p...

I frontran basically everyone. and I am very sad at the rapid decline in the last year especially due to AI slop, and wannabes larping as devs while pumping out utter garbage. garbage that users should be PAID to tolerate

Kind-1 (TextNote)

2026-06-08T20:24:54Z

↳ Reply freemymind 🇨🇭 (npub1kl8e7s48j6cfr6zrmn53n5l0fsxustszj3fwmu9ahn0tnm9e8euq80u2uq)

This is an absolut masterpiece of private analytics. What a great idea. Simply mix every answer with...

[Hadamard] count-min sketches are also great

Kind-1 (TextNote)

2026-06-08T18:03:54Z

↳ Reply Event not found

cd8e591ea03aecf0d4887179b5d8291bbf16142278642ffa925356f5e7fad09b

I think that these tools are great, when used right and in the correct circumstances. I very rarely use any of these except iterators, which are simpl...

I think that these tools are great, when used right and in the correct circumstances. I very rarely use any of these except iterators, which are simple to understand.

Kind-1 (TextNote)

2026-06-08T17:17:09Z

I don’t get why people strongly feel that they should share private communications for any reason. Especially people “exposing” others is really icky....

I don’t get why people strongly feel that they should share private communications for any reason. Especially people “exposing” others is really icky. You can do all of that, without sharing every single detail. Redaction and using your own words is not hard.

Kind-1 (TextNote)

2026-06-08T16:28:22Z

↳ Reply Event not found

dcf419e87fdc8763df4d74d02221c26d8e2f162f39ac2dfeb73762f90938a03e

False dichotomy You can deny it without doxxing the sender, it’s called the reply button.

Kind-1 (TextNote)

2026-06-08T16:26:03Z

↳ Reply Event not found

b8b15c807e62ce00971e950ffa79bd7b59166e96009cedf6aaa615c9d067c590

I think the compression and metadata removal probably killed it

Kind-1 (TextNote)

2026-06-08T16:24:18Z

↳ Reply Event not found

f5221a7ee3d49017ce8b49ed71f08ca80bf031fccfd585ecab6b639b03766900

why?

Kind-1 (TextNote)

2026-06-08T16:23:35Z

↳ Reply 577de06d... (npub12a77qmwwzc9qx7gk8f9m0d5qhclq5rsudr0xu6agcqgnfdzqvnwsfrrdyd)

Exactly, so just don’t sign it.

Signing or not doesn’t change anything.

Kind-1 (TextNote)

2026-06-08T16:23:11Z

Data ownership is not something that exists. Privacy should be respected, but a false illusion of fully controlling what you share is as harmful as ig...

Data ownership is not something that exists. Privacy should be respected, but a false illusion of fully controlling what you share is as harmful as ignoring requests to delete. “Homeservers” don’t fix this as archive.is, archive.org, Trove and many others will happily keep permanent and trusted records of what you said, some of them even if you ask them to delete it.

Kind-1 (TextNote)

2026-06-08T15:55:13Z

↳ Reply semisol (npub12262qa4uhw7u8gdwlgmntqtv7aye8vdcmvszkqwgs0zchel6mz7s6cgrkj)

Foundation is ran by good people but I can’t trust their products. Many of their statements, includ...

You will have a hard time finding people that actually know hardware security, as it is a very small space, and most of them work in places where thei...

You will have a hard time finding people that actually know hardware security, as it is a very small space, and most of them work in places where their work is actually appreciated. Not in markets where people have a completely distorted view of “security”. Bitcoin and Monero are getting the scraps.

Kind-1 (TextNote)

2026-06-08T15:52:12Z

↳ Reply Event not found

f9a87280ea0e4fffec777a4055da8a7b67d9c7e605e2a75ef9618eb837a3b494

That is why it is named that ;) People claim they are the next Galileo and they will be proven right because they are getting opposed or prosecuted or...

That is why it is named that ;) People claim they are the next Galileo and they will be proven right because they are getting opposed or prosecuted or otherwise “suppressed”

Kind-1 (TextNote)

2026-06-08T15:50:42Z

↳ Reply Event not found

398a63fac688d9e770bc8fdaa1e04eace438c23c2c40c8a778f16c4e5433949a

WoT and PoW are futile solutions, as they are actively hostile against new users, and have a huge asymmetry between real users and spammers. AI spamb...

WoT and PoW are futile solutions, as they are actively hostile against new users, and have a huge asymmetry between real users and spammers. AI spambots when designed right can easily manipulate WoT at little cost, while WoT excludes new users. PoW is just a joke. It’s designed entirely for huge server farms, and people are expecting anything outside of that.

Kind-1 (TextNote)

2026-06-08T15:48:59Z

↳ Reply Hanshan (npub1lxzaxzge0jq9u9cecucctdt5lslwgp7hcxmp2l0wn8r2ecjenwasu6svxa)

woot. finally a hardware wallet! cc nostr:nprofile1qqs99d9qw67th0wr5xh05de4s9k0wjvnkxudkgptq8yg83vt...

Foundation is ran by good people but I can’t trust their products. Many of their statements, including: - calling the SAMA5D2, which lacks security f...

Foundation is ran by good people but I can’t trust their products. Many of their statements, including: - calling the SAMA5D2, which lacks security features found on many security-oriented processors, a “security processor” - using the ATECC secure element, when it is known to be insecure and was designed by a defunct company - calling smart cards, which are the most secure way to protect your keys and are constantly evolving, “ancient and dated technology” - *forking* Xous instead of using a well known kernel like seL4 or Linux or even writing their own (from their marketing, it seems they want to associate with the good reputation of Xous) make me strongly think that they do not have significant knowledge with security. As with any HWW company, they have to prove that they can be trusted to make security products. I am not convinced at all.

Kind-1 (TextNote)

2026-06-08T15:31:41Z

Galileo fallacy (noun) A fallacy whereby it is implied that one is correct due to being widely criticized or persecuted.

Kind-1 (TextNote)

2026-06-08T14:57:41Z

Doxing is not cool, regardless of whatever “justification” you have. Instead of wasting time on these things, you can just reply saying no, and move ...

Doxing is not cool, regardless of whatever “justification” you have. Instead of wasting time on these things, you can just reply saying no, and move on. There are much better things to do than this. I’ll go back to my cave to polish NFDB 3 now

Kind-1 (TextNote)

2026-06-07T23:08:52Z

It has not been 24 hours and it has suddenly started SIGABRTing in a loop nostr:nevent1qqsfu268qqwndmqe4fv8h3mf7sejd24gw5evwfwzgp76z5d5ysmfsvszypftfgr...

It has not been 24 hours and it has suddenly started SIGABRTing in a loop nostr:nevent1qqsfu268qqwndmqe4fv8h3mf7sejd24gw5evwfwzgp76z5d5ysmfsvszypftfgrkhjammsap4marwdvpdnm5nya3hrdjq2cpezputzl8ltvt6qcyqqqqqqgpp4mhxue69uhkummn9ekx7mqvyztmg

Kind-1 (TextNote)

2026-06-07T21:05:47Z

↳ Reply semisol (npub12262qa4uhw7u8gdwlgmntqtv7aye8vdcmvszkqwgs0zchel6mz7s6cgrkj)

Streaming is limited, yes. I’m working on an update that resolves this and also allows negentropy

Out of curiosity what is the use case?

Kind-1 (TextNote)

2026-06-07T19:24:39Z

↳ Reply vinney...axkl (npub19ma2w9dmk3kat0nt0k5dwuqzvmg3va9ezwup0zkakhpwv0vcwvcsg8axkl)

plausible deniability server/relay + client: - when you post to the server, you specify a symmetric...

If they have the key, no. Otherwise to some extent. If you still host blossom for the sole purpose of pirated music they will probably strike it down

Kind-1 (TextNote)

2026-06-07T19:22:01Z

↳ Reply Cameri🐦‍🔥 (npub1qqqqqqyz0la2jjl752yv8h7wgs3v098mh9nztd4nr6gynaef6uqqt0n47m)

I’ve been posting with POW all day… has not hurt my experience in any meaningful way. You’ve seen my...

You are trying to solve the problem that spam exists on relays. To do this, you need to distinguish what is likely spam and what is likely ham. “Ada...

You are trying to solve the problem that spam exists on relays. To do this, you need to distinguish what is likely spam and what is likely ham. “Adaptive” PoW only works if you know if the content is spam or ham. But the entire point is that we don’t know this and want to find it out. And PoW provides no information about spam/ham. So, why bother? Instead of disproportionately disadvantaging your average user with PoW, you can work on methods that can reliably identify spam/ham and implement ratelimits instead of adaptive PoW. (which don’t make the UX suffer for the average person)

Kind-1 (TextNote)

2026-06-07T18:54:40Z

↳ Reply Cameri🐦‍🔥 (npub1qqqqqqyz0la2jjl752yv8h7wgs3v098mh9nztd4nr6gynaef6uqqt0n47m)

Yeah, based on that simplified scenario that you posted sure. And of course relays can’t identify an...

Then why would I need PoW

Kind-1 (TextNote)

2026-06-07T17:17:46Z

↳ Reply Karnage (npub1r0rs5q2gk0e3dk3nlc7gnu378ec6cnlenqp8a3cjhyzu6f8k5sgs4sq9ac)

I agree. It’s silly to think spammers are broke or something … they’ll easily dedicate resources esp...

It hurts real users while doing absolutely nothing to prevent attacks. Whether they spend $0 or $100 doesn’t change that they will continue. With PoW...

It hurts real users while doing absolutely nothing to prevent attacks. Whether they spend $0 or $100 doesn’t change that they will continue. With PoW the only solution is to make it prohibitively expensive with a ton of bits, which disproportionally affects users. Then there will be paid PoW services so you don’t have to wait a minute to post to get higher bits, and oh wait, we have reinvented paid relays However, developers have managed to make very accurate spam filtering systems for other social media, email, forums, etc. And none of them use PoW 🤪

Kind-1 (TextNote)

2026-06-07T17:15:32Z

↳ Reply Event not found

f85d367701920d7522195bf97628f3f01b15825ed38d5506b8ccb08836292c42

from everything I have seen, they just seem to have rehashed the same things that were already proposed / existed for a long time

Kind-1 (TextNote)

2026-06-07T17:10:50Z

↳ Reply Cameri🐦‍🔥 (npub1qqqqqqyz0la2jjl752yv8h7wgs3v098mh9nztd4nr6gynaef6uqqt0n47m)

Dynamically adjusted PoW for the spammer. Not a fixed PoW required for the whole network.

If you could identify the spammer in the first place all of our problems would be solved 🤦

Kind-1 (TextNote)

2026-06-07T16:47:23Z

↳ Reply Cameri🐦‍🔥 (npub1qqqqqqyz0la2jjl752yv8h7wgs3v098mh9nztd4nr6gynaef6uqqt0n47m)

Semi, that’s your opinion, which non-factual. It would be intellectually dishonest to say that other...

It is not. It has been proven over many times that PoW is not a solution to preventing spam/abuse, especially when identities are disposable. Hashcas...

It is not. It has been proven over many times that PoW is not a solution to preventing spam/abuse, especially when identities are disposable. Hashcash did not take off. Very few CAPTCHAs use PoW (with all of them being small ones), and all that do only have global dififculty thresholds. Anubis doesn’t stop abuse via PoW but just that dumb bots don’t run any JS. It is trivial to bypass at scale. On Nostr, PoW has been discussed since the beginning and no one seriously uses it. PoW concentrates power in the hands of spammers where they can acquire powerful GPUs and CPUs, and abuse them at very low cost. Real users suffer as their HW is way slower and UX matters.

Kind-1 (TextNote)

2026-06-07T16:46:48Z

↳ Reply Event not found

0000049f042e4bcca71307bace4b655e869eee327f70087f637d08b6c93eb424

Because it is the highest value you can reasonably do without making the experience of many users miserable. And regardless, PoW only burns money, wh...

Because it is the highest value you can reasonably do without making the experience of many users miserable. And regardless, PoW only burns money, while being very cheap for spammers. Paid relays fix this problem in a better way by not slowing your client to a crawl on every action you take, and actually funding the people dealing with spam.

Kind-1 (TextNote)

2026-06-07T15:50:37Z

↳ Reply semisol (npub12262qa4uhw7u8gdwlgmntqtv7aye8vdcmvszkqwgs0zchel6mz7s6cgrkj)

Which is absolutely nothing if their scam works once per month

By this shitty overestimating calculation, it only takes $250 to spam the network with more junk than has ever existed on Nostr

Kind-1 (TextNote)

2026-06-07T15:10:13Z

↳ Reply Cameri🐦‍🔥 (npub1qqqqqqyz0la2jjl752yv8h7wgs3v098mh9nztd4nr6gynaef6uqqt0n47m)

That’s $3 dollars less on a spammer’s pocket than the day prior. Over a thousand less if they do it ...

Which is absolutely nothing if their scam works once per month

Kind-1 (TextNote)

2026-06-07T15:08:54Z

↳ Reply node (npub1rzg96zjavgatsx5ch2vvtq4atatly5rvdwqgjp0utxw45zeznvyqfdkxve)

You feed it a GitHub link and it installs to your phone? Do you need a dev account?

No you feed it an IPA or feed and it installs. You need dev if you have more than 8 apps

Kind-1 (TextNote)

2026-06-07T14:56:06Z

gm nostriches

Kind-1 (TextNote)

2026-06-07T12:47:01Z

↳ Reply Event not found

6e0702d815e42841811f03e33b66721ced4c34f0b7ec5b74047932f920b49fbd

Streaming is limited, yes. I’m working on an update that resolves this and also allows negentropy

Kind-1 (TextNote)

2026-06-07T12:46:40Z

↳ Reply node (npub1rzg96zjavgatsx5ch2vvtq4atatly5rvdwqgjp0utxw45zeznvyqfdkxve)

From what I understand, aren’t sideloading apps only available in Europe?

No this uses the same method as development loading

Kind-1 (TextNote)

2026-06-07T12:24:37Z

↳ Reply node (npub1rzg96zjavgatsx5ch2vvtq4atatly5rvdwqgjp0utxw45zeznvyqfdkxve)

Why do I need a sideloading app for xcode when i can plug in my phone to the computer via usb c

Because it does not require compiling, it does not require USB, and it auto-renews the 7d expiration

Kind-1 (TextNote)

2026-06-07T12:12:29Z

↳ Reply Event not found

e8a6ead0b8f1b2e8367ec4de5db97bc29b7e4b5922d6eb8426129a49e92b4b3a

don’t think so

Kind-1 (TextNote)

2026-06-07T09:18:59Z

↳ Reply node (npub1rzg96zjavgatsx5ch2vvtq4atatly5rvdwqgjp0utxw45zeznvyqfdkxve)

Is it the sideloading thing for Europe?

It is sideloading for everything using XCode signing

Kind-1 (TextNote)

2026-06-07T08:15:04Z

↳ Reply Event not found

000062eb0f92fd36a4f63aa3654d9648c5bc3ec28a9330582d36961381d4ad9d

Have you heard of SideStore

Kind-1 (TextNote)

2026-06-07T01:35:00Z

↳ Reply Event not found

c6b10dad64c1706d672cde7b0768337a126b2d1a9679e07c5e49c5861c084f29

With or without the aggregator? For without, a plan that will cost less is coming. I can notify you if you want when that is out. For with (Plus), it...

With or without the aggregator? For without, a plan that will cost less is coming. I can notify you if you want when that is out. For with (Plus), it’s expensive to run the aggregator so it is higher. It is still cheaper though than other aggr/filter options.

Kind-1 (TextNote)

2026-06-07T01:33:49Z

Damus is actually a pretty great client. There were no updates that turned it into unusable rubbish. It doesn't randomly break. It has proper NIP-42....

Damus is actually a pretty great client. There were no updates that turned it into unusable rubbish. It doesn't randomly break. It has proper NIP-42. Media uploading is easy. Translation works. The UI just works. It isn't trying to distract me with 30 different features. It is the only app that I actually trust with my nsec. My only complaints are: - handling of longform (by event ID instead of naddr, maybe fixed?) - 1-2 crash loops I had due to early NostrDB issues that are gone now - link preview ordering handling Thanks nostr:npub1xtscya34g58tk0z605fvr788k263gsu6cy9x0mhnm87echrgufzsevkk5s

Kind-1 (TextNote)

2026-06-07T00:33:19Z

↳ Reply Event not found

7e00d8e143a5f441d91dc399f79d88f21bc09995fc6b2b0c77ea9e16456c211a

That is the result of RL, not the intention. If you encourage less tokens, less tool call rounds, and reward passing tests or similar, then that is w...

That is the result of RL, not the intention. If you encourage less tokens, less tool call rounds, and reward passing tests or similar, then that is what happens The easiest way gets the reward first and the model rapidly converges to that

Kind-1 (TextNote)

2026-06-07T00:28:04Z

↳ Reply Event not found

c83cf273296a7d480a12600da64c240d941763f662df1f1de30763b638a33748

While I do like that bot, I think that such features should be a client feature, not a reply. I will make exceptions for a very limited class of bots...

While I do like that bot, I think that such features should be a client feature, not a reply. I will make exceptions for a very limited class of bots that can be disabled, and no two doing the same thing.

Kind-1 (TextNote)

2026-06-07T00:17:40Z

LLMs struggle with good software dev, especially on implementing novel things, and maintainability. The best way them to view them is as a really expe...

LLMs struggle with good software dev, especially on implementing novel things, and maintainability. The best way them to view them is as a really expensive machine translation from English to code. There are probably a few reasons: 1. ML models have limited generalization capability ML models can only "do" what they have been trained on. What is outside of that can't be reliably represented or processed by the model, and so trying to do anything outside that will lead to weird results. 2. Biases in training data Models are trained with a lot of data. The pre-training dataset can significantly bias the model (for example preferred frameworks, tools or "suggestions"), and so can the post-training (which is what results in models having a certain "design" style, or the LLMisms) 3. Reinforcement learning The RL stage of a model optimizes for things like the number of tool calls, and a binary pass metric. The problem is that, just like human-made code, it is easier to hack on a fix than to properly integrate it. The model is not trained for achieving anything more than satisfying your request with the bare minimum, and so you will accumulate layers of slop and slop.

Kind-1 (TextNote)

2026-06-07T00:14:54Z

↳ Reply semisol (npub12262qa4uhw7u8gdwlgmntqtv7aye8vdcmvszkqwgs0zchel6mz7s6cgrkj)

Bots shouldn’t auto-reply without explicit consent. Spamming replies, especially for advertising or ...

The operators of these annoy and spam-bots will also have their trust score in internal systems reduced.

Kind-1 (TextNote)

2026-06-06T23:49:38Z

Bots shouldn’t auto-reply without explicit consent. Spamming replies, especially for advertising or LLM slop, is not a good idea. I’m going to be imp...

Bots shouldn’t auto-reply without explicit consent. Spamming replies, especially for advertising or LLM slop, is not a good idea. I’m going to be implementing additional filters against such bots in Nostr.land soon.

Kind-1 (TextNote)

2026-06-06T23:45:06Z