Ademan

Public Key

npub19jescdjr3wk552j3q77f3awwhe4qy2ds24xce773exd28nr7emqsm2pt7r

Profile Code

nprofile1qqszevcvxepcht229fgs00yc7h8tu6sz9xc92nvvl0gunx4re3lvasgpzfmhxue69uhk7enxvd5xz6tw9ec82cs8nnh85

Publishing to

offchain.pub

Author Public Key

npub19jescdjr3wk552j3q77f3awwhe4qy2ds24xce773exd28nr7emqsm2pt7r

Show more details

Last Notes

2024-08-04 23:59:30

We had some sprite left over from independence day and I watered it down with a lemon sparkling water. If it wasn't for the sugar I'd drink this every day.

2024-08-04 20:29:10

@nprofile…6yvm I swear you mentioned that there are only meant to be 100 NIPs at some point recently but I can't find the note anymore. Any idea? (Or maybe more useful, any idea where you heard that in the first place?)
@nprofile…g5cd ?

2024-08-02 20:34:22

- reply

Wait, I just noticed that's the 10 year minus 3 month instead of 10 year minus 2 year
https://fred.stlouisfed.org/series/T10Y2Y#
https://image.nostr.build/30fca047146ad8adda33d7ae588d36bed5b53d6510d34ec33ee6aade86c64b2d.png

2024-08-02 20:30:13

WOW, I didn't realize we were so strongly inverted, for so long... I just kind of assumed since the yield curve inversion had exited the news cycle that it had crossed back into positive territory...
https://image.nostr.build/355cd2631ace3a45512d343b56b60900aee05ab51c9e1acf8226d66854ab6eb5.jpg
Source: https://x.com/BobMurphyEcon/status/1819415480328757306

2024-07-25 06:25:11

- reply

Right during my daily meeting... maybe next time.
I need to harass some people into looking at NIP-96 and telling me if I'm crazy or not though.

2024-07-23 02:06:44

- reply

https://github.com/nostr-protocol/nips/issues/1376

2024-07-22 21:36:05

I can tell this issue on the nips repo is going to stagnate forever 😩

2024-07-22 20:44:40

So is Swan pining for the fjords or is this really (as they claim) cutting their activities they had expanded into, presumably leaving their "profitable" core intact?

2024-07-20 19:48:50

- reply

BTW if you're curious about the contradiction I'm seeing, I (try to) explain it here https://github.com/nostr-protocol/nips/issues/1376

2024-07-20 07:12:58

- reply

https://github.com/nbd-wtf/nostr-tools/pull/413 someone identified the fix 2 weeks ago and I spent quite a few hours to come to the same conclusion lol AGH

2024-07-15 18:13:42

- reply

Thanks! Don't worry about it for the moment, I will review coracle before I waste anybody else's time. Sorry for the noise!

2024-07-15 18:09:07

- reply

Hodlbod confirmed he's using nip-96 after all, so I can reference his code to figure out what's going wrong on my end. Don't waste your time reading this garbage until I've looked over his code lol

2024-07-15 18:02:59

- reply

Ah, you're using `file[]` as the form field? I thought I saw something to that effect in the nostr.build source, but afaict that's off spec? nostr-tools uses `file` and that was my reading of the spec as well.

2024-07-15 17:59:31

- reply

So you did... I'm not sure how I came to the opposite conclusion heh.
Thanks!

2024-07-15 17:56:05

- reply

OK I've created a simple sample page for making NIP-96 uploads the way I am in my larger project. I've also included a variation that generates the Auth header differently (according to NIP-96 where it contradicts NIP-98).
Not sure what I'm doing wrong here...
https://github.com/Ademan/nip96-test
You can run it with
```
npx webpack serve -c webpack.dev.js
```
the "My version" checkbox activates the variant upload, rather than only using stock nostr-tools functions.
@nprofile…v296 any ideas? To the extent I am able as a non-web dev who's very unfamiliar with modern PHP, reviewing the nostr.build source it wasn't obvious to me what is going wrong either.
@nprofile…lwv4 I get approximately the same 400 response with nostrcheck, does anything jump out at you?
I appreciate any help y'all can give, and sorry for the noise. At least, if you're aware of clients that work with your nip96 endpoints specifically, I'd love to see them and compare...

2024-07-14 00:18:05

- reply

This is something I was thinking about the other day when you talked about million TPS. Presumably, an end user won't want to lock a bunch of funds to a service provider with no option to reclaim unused ones.
Do you have thoughts on that? I suppose a timelocked alternate spending path would work, but that becomes troublesome if you have a bunch of timelocked tokens on the verge of expiration, too.

2024-07-14 00:14:26

- reply

Really bugs me how some people are picking fights with ecash pretending this is some dirty secret that ecash devs are hiding, considering how up front you are (and have always been) about it.

2024-07-13 21:08:10

- reply

*possibly* related, although I don't think so:
https://github.com/nostr-protocol/nips/blob/master/96.md says payload should be base64 encoded, but
https://github.com/nostr-protocol/nips/blob/master/98.md says the payload should be hex encoded
This seems to be contradictory?
nostr-tools uses the hex encoding, so if the server is expecting base64 that could be a problem, although like I said, I don't *think* these servers are processing the Auth header at all by the time they send me back a 400.
I'll at least try changing it to a base64 encoded payload hash once I get back from mowing the lawn...

2024-07-13 18:11:29

#nostrdev #asknostr #nostr-tools
I'm unable to complete nip96 uploads across nostrcheck, nostr.build, and files.v0l.io I suspect it's my fault somehow, but I'm burning more time than I'd like troubleshooting. I'm using nostr-tools and afaict based on reading nip-96, examining the nostr-tools code, and looking at the requests, I'm not certain what's wrong here.
Afaict the Authorization form field is non-standard but harmless (hacking it out did not change the responses I'm getting, and reading the void-cat-rs source it looks like it should be ignored anyway).
```
POST /n96 HTTP/2
Host: files.v0l.io
User-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:127.0) Gecko/20100101 Firefox/127.0
Accept: */*
Accept-Language: en-US,en;q=0.5
Accept-Encoding: gzip, deflate, br, zstd
Referer: http://localhost:5000/
Authorization: Nostr REDACTED
Content-Type: multipart/form-data
Content-Length: 24052
Origin: http://localhost:5000
Connection: keep-alive
Sec-Fetch-Dest: empty
Sec-Fetch-Mode: cors
Sec-Fetch-Site: cross-site
Priority: u=4
TE: trailers
-----------------------------11234029581503681683701650224
Content-Disposition: form-data; name="Authorization"
Nostr REDACTED
-----------------------------11234029581503681683701650224
Content-Disposition: form-data; name="size"
22933
-----------------------------11234029581503681683701650224
Content-Disposition: form-data; name="file"; filename="2024-07-06-152337_1000x175_scrot.png"
Content-Type: image/png
... snip ...
-----------------------------11234029581503681683701650224--
```
You can see from the size field that the file is quite small.
I get a 400 response from all three services. (Not a 401 or 403 or anything) I'm pretty sure I'm generating the authorization token correctly (also using nostr-tools) and I glanced at the nostr.build source code and I'm reasonably confident the response I'm getting back is generated before the auth header is checked.
Both nostr.build and nostrcheck have some kind of message indicating a problem with the file upload portion. (files.v0l.io gives me a generic 400 html response)
nostr.build example
```
'{"status":"error","message":"Either no file or more than one file posted. Only one file is expected."}'
```
I'm not a web guy so nothing is jumping out at me, but does anyone notice something obviously wrong with the request?

2024-07-11 19:10:02

What is the currently recommended RAM for a bitcoind node, both for initial block download and steady state? I'll be using -txindex and making infrequent RPC calls, if that affects it significantly.
I'm provisioning a VM on a home server for a new node, so I can always adjust later if need be.
#bitcoin #asknostr

2024-07-09 18:05:08

- reply

Ah shoot, missed that you're on vacation.
IGNORE ME

2024-07-09 00:02:58

@nprofile…gu24 why did you go with nostr.build's proprietary API instead of the nip96 one?

2024-07-06 04:15:08

- reply

Have you considered any kind of affiliate program for authors of clients if their users sign up for your paid services? I'm sure every nostr dev out there is sitting there scratching their head going "how the heck do I monetize this?" and maybe that's an interesting option.
(Not applicable to my situation, of course, but for people who are building *serious* clients lol)

2024-07-05 06:33:01

@nprofile…7hl4 I'm working on a for-funsies nostr client that I'll probably abandon fairly quickly and I'd like to integrate nostr.build
If I actually release it, it'll be FOSS and free to use, but the nostr.build github made it sound like I should contact y'all before integrating it.
Happy to provide more details and/or a preview release if you want. (Give me some time to do some cleanup first in the latter case heh) Like I said in all likelihood I'm going to abandon this pretty quick anyway, but I want to be a good steward.

2024-06-30 23:00:19

- reply

I'm actually more interested in public groups too.
My question is just: Do you really need a NIP-29-aware relay at all, for public groups?

2024-06-30 22:13:49

- reply

Can't clients enforce the moderation rules? For public groups, where everyone can see the posts, you don't actually need relay cooperation do you?
A NIP-29 aware relay could restrict event delivery to group members, and provide some privacy, as a value add, though. I'm not sure how inclined I am to trust such a relay to never leak, and since nostr events are non-repudiable, all it takes is one mole to ruin privacy.

2024-06-29 20:56:13

- reply

An automated service that polls known relays periodically and logs some stats (age of the last N notes to estimate usage), logs unusable relays, and has a simple web frontend would be nice. Any additional diagnostics are a bonus but I'm sure people could come up with a bevy of them.

2024-06-28 21:53:39

Hey @nprofile…qmdu do you have any experience with https://en.wikipedia.org/wiki/Tithonia_diversifolia ? I'm trying to find a perennial to build a privacy barrier, and also to generate compost. (The flowers are a bonus as pollinator attraction)
The problem is, I'm in Zone 8a like you, where Tithonia supposedly will die off completely in hard freezes.
I was also interested in pigeon pea, since that can even be for human consumption, but the hard freeze is a problem with them too.

2024-06-27 16:47:24

@nprofile…5uux any thoughts on what a client should do with a kind 1 with NIP-92 metadata but which has no URLs in the body?
I'm inclined to treat them as "attachments" regardless. It makes the most sense for *my* purposes, but it also makes sense that many clients may just ignore them altogether (and therefore I should probably avoid creating such notes).

2024-06-27 06:04:20

- reply

Yet another time when typescript would have saved me over an hour of debugging.

2024-06-27 05:56:37

- reply

huzzah, success!

2024-06-26 20:51:16

- reply

Can't tell if I got my test client banned from nos.lol or if something's very wrong lol

2024-06-26 18:02:09

- reply

I need a private relay for testing lol

2024-06-26 17:51:36

Test
#test

2024-06-20 22:04:20

I feel like this has been done, but has anyone named a nostr client something like "Nosferatu"? #asknostr
If not I call dibs on "Nosbooratu"

2024-06-20 22:03:37

I feel like this has been done, but has anyone named a nostr client something like "Nosferatu"? If not I call dibs on "Nosbooratu"

2024-06-20 20:51:05

- reply

considering how productive you seem to be, I believe it, and I'm sure there's a time machine involved lol

2024-06-18 20:15:10

- reply

phy too, in case someone wants to snipe my LoRastr idea (because realistically I'll never get to it)

2024-06-18 20:12:48

- reply

But I got it right 🤨

2024-06-18 01:59:09

- reply

Is that 8:00 pacific, 10:00 central?

2024-06-18 01:54:20

IndexedDB actually isn't that bad once you get used to it...
https://image.nostr.build/fd8a848f3c01ff6ff250629b8c846e53f12bccdbe1f9d6374b8636f45eca5bc7.jpg

2024-06-18 01:50:04

#primal doesn't create events with nip-92 metadata? #asknostr

2024-06-17 18:35:21

- reply

The rockies and the sierras (my favorite) are both located in inhospitable states...

2024-06-17 00:02:15

Is anyone else slightly annoyed that we have this rich structured format and NIP-92 uses space separation in strings? I don't really have a better solution but this is fugly.

2024-06-13 01:18:28

- reply

Seems like technical difficulties are to blame.
https://image.nostr.build/ca6182152aa179a8fcafc4b851c70e02dd7a8859020d4af13d0ec6e5bd729373.jpg

2024-06-12 22:15:34

- reply

I was wondering the same, I've been watching a lot of Shaun's stuff recently and was looking forward to this episode. The RSS feed has an old TSP instead.

2024-06-11 23:04:25

- reply

A Christian *and* a communist?!

2024-06-11 23:03:20

- reply

Can I use ditto (so I can use Soapbox) but keep my keys with nos2x?

2024-06-11 18:55:35

- reply

and yet you're not funding my bullshit, curious
/s

2024-06-11 03:17:12

- reply

Maybe I don't have a finger on the pulse of the average bitcoin dev, but I for one quite appreciate SQL as well.

2024-06-11 02:44:24

- reply

Bitcoiners hate SQL? I thought our advice to 99.9% of people was "just use a SQL database for what you're doing" ;-)

2024-06-11 00:00:12

- reply

This is their policy, you need "reliable" secondary sources that interpret and explain things. This is unfortunate because all "reliable" secondary sources are garbage these days.

2024-06-10 18:32:53

- reply

That was it, thank you!

2024-06-10 18:23:58

When sharing a nostr link, most (all?) services won't generate a useful preview for js based clients like coracle or iris (I believe they look at the open graph meta tags in the html result), but I think I remember there was one service that would do server side rendering for your note, had decent previews, and had links to open the note in other clients.
Anyone know what I'm talking about? #asknostr

2024-06-10 07:22:09

#test another test
> it didn't work the first time

2024-06-10 07:16:12

- reply

> imagine testing publicly

2024-06-09 15:57:38

- reply

lmao just noticed the year is totally messed up, oh well, plenty of work to do left

2024-06-09 15:56:19

- reply

I am working on something incredibly stupid
https://image.nostr.build/8272c03c3c8c2f9053e71568be9d68de4788301139c0c92402681cc2a54153f9.jpg
https://image.nostr.build/77318b58e23d3c7555331d30b05d8656ce8982e47504fe54513221cd3838afac.jpg

2024-06-09 04:10:03

#test test2

2024-06-08 07:21:07

#test
This is a test

2024-06-08 01:02:20

- reply

When will they revise it down? lol

2024-06-07 20:55:49

- reply

The liquidity ads spec isn't complete yet is it?

2024-06-07 05:29:35

- reply

Ah nvm it's there but optional https://github.com/nostr-protocol/nips/blob/master/96.md#auth
Not sure how I missed that on my first reading.

2024-06-07 04:37:30

@nprofile…cfex do you have a ToS for https://files.v0l.io/ ? (Or is it really just a demo server, and I should stand up my own https://git.v0l.io/Kieran/void-cat-rs server?)

2024-06-07 04:07:31

- reply

Blossom knows what's up https://github.com/hzrd149/blossom/blob/master/buds/02.md

2024-06-07 04:04:25

Why doesn't NIP-96 have the authorization event commit to the upload hash? It *seems* like it should, but I guess it's not *necessary* because of https? It would actually be cool to not require https, since https (effectively) implies reliance on centralized CAs and DNS.

2024-06-07 03:26:41

- reply

I'm sure he doesn't. Plus nostr is a potential competitor.

2024-06-07 01:32:15

- reply

Ah that's true.
At the very least, supporting *emphasis* **emphasis** and block quotes in addition to the preformatted text you already do (on opt-in posts) would be nice. Since you're fiatjaf anyway you can overrule fiatjaf's fiat ruling on no-rich-text ;-)

2024-06-07 01:02:46

- reply

I mean I think every client is already processing image embeds anyway. markdown tables are human readable (nice to look at in plain text, even!).
Link aliases is probably the one that shouldn't be supported.

2024-06-07 00:59:28

- reply

Plus elon probably hates nostr

2024-06-07 00:16:39

- reply

lol, it's more like a technical shitpost I'm using to dip my toes into nostr dev, but a sufficiently shitty journo could definitely spin it

2024-06-06 23:53:29

- reply

It's definitely no longer plain text, if you interpret it as a code snippet/preformatted text.
But as hodlbod pointed out that line has kind of already been crossed with link previews, event and npub references, so *shrug*.
Might as well go full markdown imho. Would be easy to have a content type flag to say "this was MEANT to be markdown". Since markdown is perfectly readable as plain-text too, I think it's reasonably backwards compatible.

2024-06-06 23:45:09

Now seems like a particularly bad time to release the #nostr software I was working on this past week, it'd just add to the fAr rIgHt fire. On the other hand, no such thing as bad publicity?

2024-06-06 21:34:47

- reply

If you still have your twitter account...
https://image.nostr.build/d82263378f7c9cadec78b7ceab75652fb961b8c45162ef3fa618a3e873bb0ab6.jpg

2024-06-06 21:32:25

- reply

You don't hate "journalists" enough.

2024-06-06 18:30:39

The bot problem on twitter is incredibly funny when you're having a technical discussion.
https://image.nostr.build/e12dfcf73fd63bb58d81ca92258eaa433284609e93408a5bc6d499f0563035d9.jpg

2024-06-06 00:52:47

Presumably relays can set limits on number and complexity of filters and subscriptions, do any implementations do this currently? How do they handle this?
I gave the nips repo a brief look and didn't see anything, but I wasn't very thorough.
In particular, I want to subscribe to a set of pubkeys of interest, is there a practical limit to that set size? What happens if I hit that limit?
#asknostr

2024-06-05 22:28:12

# Test
Also a test
==========
```
this is a
test
```
*this is also a test* plus **this is a test**
_yet another test_
[definitely a test](https://coracle.social)
> finally, this is a test as well

2024-06-05 22:26:33

- reply

Whoa, backticks work on kind 1s in coracle? I thought any kind of rich text in kind 1 was heavily discouraged, or has everyone just decided to do it anyway?

2024-06-05 22:15:23

- reply

Hrm, I guess it could also be
['L', '-#t'], ['l', 'sfw-photography', '-#t']
?
That avoids the problem you mentioned and generalizes to other labels and tags.

2024-06-05 21:39:37

@nprofile…tg2c did you ever consider inverse labels for NIP-32?
Something like
['L', '#t'], ['l', '-sfw-photography', '#t']
to remove an event from a hypothetical #sfw-photography hashtag
I have a couple of use cases but moderation is probably the primary one.

2024-06-05 21:36:28

- reply

I'm only saying that validation depends on validating the event id AND the signature. Only doing half validation is exactly as secure as doing neither.
But regardless, I think validation is mandatory at any performance penalty.

2024-06-05 20:25:24

- reply

I was going through the verification flow in my head before I wrote the OP, trying to come up with a justification, maybe DoS related, but couldn't come up with anything.
I don't think there's *any* value in validating the signature without also validating the signature is actually signing the contents, otherwise you could just steal a valid signature/event id pair from any other event and fool these partially verifying clients.
You're either trusting the relay to have verified everything, or not, partial validation doesn't seem useful at all.

2024-06-05 20:07:29

- reply

That's fair I guess. nostr *does* seems oddly byte conscious in some places though (many single digit tags, integer kinds instead of namespaced string identifiers)

2024-06-05 19:38:55

- reply

oh right #asknostr

2024-06-05 19:32:54

Why do nostr events carry their own id? It's derived from the rest of the event, isn't it?
That's 72 wasted bytes on every event as far as I can tell, which isn't *that* much, but I don't understand why it's there at all.

2024-05-30 22:49:07

- reply

What clickbait, that's not a picture of a vegan steak

2024-05-30 21:55:42

- reply

I'll happily die for my children, sorry journos, can't dissuade me.

2024-05-30 21:50:21

- reply

lol, I went and checked my profile immediately prior to posting that, just to be sure.
... now on this post (and on my profile) it's showing 23...

2024-05-30 21:49:15

- reply

Same, my coracle is showing a WoT score of 0 for myself right now, but earlier it was 21, and before that (past week?) it was ~26.
At the very least WoT scores seem unstable lately.

2024-05-30 18:53:01

- reply

It's worth considering whether someone with +3 -3 is really the same as someone with flat 0. (I tend to think not)

2024-05-30 18:51:36

- reply

They tell you a lot imho. They tell you whether someone is a spammer, or annoying, or just plain new, but they're imperfect signals of course.

2024-05-29 23:15:58

- reply

good ol' anarcho tyranny

2024-05-29 04:00:51

- reply

Yeah, he's also just wrong about the privacy point.
He's right about custody though imho. OP's attempt to show ECash isn't custodial feels tortured.

2024-05-24 22:32:46

- reply

I'm sure there's like 4 different colors from which the rest are derived if I looked at the less/sass/whatever source, but I don't have a way to build it right now.

2024-05-24 22:29:43

Bluacle? I kinda like it tbh
https://image.nostr.build/0e84ebd194ef7425f6dff5fb8090335f7c296077f11db1ac8bcb0922e6ab9ce4.jpg
(I didn't change all of the colors so some of the contrasts are weird)

2024-05-23 21:57:03

- reply

It wasn't always a hard question, Phoenix was great.

2024-05-23 21:32:18

- reply

Application specific. If it needs to be self describing, CBOR.
I think the self-describing, binary format use cases are pretty few and far between, personally. gzipped or bzipped json is generally preferable in most cases.

2024-05-23 21:25:45

- reply

Some people get really mad when other people knowingly make tradeoffs they don't like.
OTOH newbies *do* need to understand the risks and a small minority seem to think ecash is a panacea when the tradeoffs *are* genuinely unacceptable to many people for good reason.

2024-05-23 20:24:44

Seems like Lightning AMP combined with a loop-out service would be potentially pretty powerful privacy tech. Of course you need PTLCs to defeat attackers along the route.