slipstream
Slipstream

The lowest levels of the stack are too ossified to have your ideas adopted, while the highest levels are already idea saturated. The most interesting place to innovate is a thin slice in the middle, where small changes can have radical effects. Right now I believe the sweet spot is identity and data storage of small payloads

2024-07-03 01:12:16

- reply

Are the technical details of how this display works public? A patent perhaps?

2024-07-02 15:11:53

- reply

I imagine it _works_ but doesn't scale

2024-07-01 16:32:09

Happy Canada Day! 🍁

2024-06-30 12:18:44

- reply

Sounds like a feature

2024-06-26 17:39:14

- reply

You can hardcode a large list and only 1 need be honest & operational for your client to find more honest peers.
If 99 of 100 hardcoded peers are down, you're still good. And on the next client update you change the hardcoded list to another working set of 100

2024-06-26 17:25:42

- reply

Finding seed nodes is a much bigger problem than nostr, even bitcoin has hardcoded domains for finding the first peer.
In practice I don't think this is actually a big issue

2024-06-26 17:22:41

- reply

There's a lot of reverse proxy infrastructure that won't like this. Would be wonderful if we could at least put the original naddr in a header or something

2024-06-25 05:32:45

- reply

Which client supports nip-95?

2024-06-25 01:03:01

- reply

I like blossom's approach. https://github.com/hzrd149/blossom/blob/user-server-list/buds/03.md
Could do the same with nip96.
IPFS would work as well but I prefer http unless the file is extremely popular

2024-06-23 04:09:21

- reply

Nice job, what is your rating?

2024-06-21 12:52:27

- reply

Pretty neat that there's no MEV, though it looks like it has the same high liquidity requirements of Ark

2024-06-19 18:36:05

- reply

Blastr style storage is what attracted me to build on nostr in the first place. I'm not prescribing it, but the option for the user to choose a Blastr proxy relay is great

2024-06-18 13:12:55

- reply

Are you going to open a nip? Or is your goal to just raise awareness of your tool and hope others carry the flag for a nostr integration?

2024-06-18 12:12:56

- reply

If urgent (it rarely is), push notification. Otherwise email.

2024-06-11 23:15:19

What's your favourite tool to grab the raw json of a nostr event given an ID?

2024-06-11 22:04:58

- reply

How so?

2024-06-11 21:09:31

- reply

Damus does not add PoW to your posts. That's true for basically all notes on nostr right now. There's not really much point adding it since no apps/relays use PoW based spam filtering

2024-06-11 18:24:03

- reply

You can already do this attack with a 2 of 2 multisig where one party is the government. You need their permission every time you transact, and will only approve transfers to another 2 of 2.

2024-06-11 18:02:05

- reply

For apple, the enclave requires a faceID proof before it'll sign. But, afaik no btc wallets use them because the curve the enclave supports is secp256r1 (whereas bitcoin is k1).
I heard op_cat provides enough account abstraction to make it work, but, have not looked into if that's true or not.

2024-06-11 17:05:27

- reply

Right I see the use to compensate for the lack of a built-in secure enclave (like most PCs), so it can't be high jacked from working memory by other malicious software 👍

2024-06-11 12:08:55

- reply

What's even the point of a hardware device if there's no trusted display? Does it offer any security over a software wallet on a device that has a secure enclave like a phone?
If the software side is compromised it'll just show you the address you intend to send to and request the device to sign the malicious transaction

2024-06-10 18:48:51

- reply

I saw that and was confused as well. What could the logic possibly be to see "too many" primary sources as problem?

2024-06-10 17:53:58

- reply

I believe that's within spec for nip96 unfortunately. I've seen some research around using snarks to prove the original hash of the image which could be interesting to apply here

2024-06-10 15:47:12

- reply

It was mostly a joke, but, I can imagine an agency that fronts consolidated capital (at a discount) in exchange for directly taking possession of the private keys.
Fully trusted of course. This would probably only be in the realm of large corporations

2024-06-10 00:50:04

- reply

Probably because the spec is so simple

2024-06-09 14:49:29

What is a DVM?

2024-06-09 13:23:19

- reply

Not quite ready to release yet, but it's a writing platform similar to Obsidian or Logseq. Also supports serving up your content on your own domain as a "blog" of sorts

2024-06-08 18:26:43

- reply

Oh wait I've already used the right kind... has this bud been around longer? Somehow I've already implemented this exact flow in my app

2024-06-08 18:09:18

- reply

Makes sense. That is how I've already implemented the blossom integration in my app (just with a different 'kind' for server list - I'll switch it to standard)

2024-06-08 17:19:55

- reply

Why not define a blossom://<hash> url instead of having clients try to extract the hash?

2024-06-08 13:55:16

- reply

Soon we'll have consolidation firms for UTXOs in the same way we do for debt

2024-06-08 04:31:56

Would be cool if nip96 servers that do image transformations, and therefore serve files with modified SHAs, included a ZKP of the transformation.

2024-06-08 03:54:05

- reply

Honestly the protocol is so simple it's pretty easy to build straight off the nips repo.

2024-06-08 00:48:53

- reply

that's the image

2024-06-08 00:37:55

- reply

Yes, but that's only because you used a gateway url so it's no different from a normal image server. I doubt it would work with a real IPFS url

2024-06-07 04:02:42

- reply

How many nip96 servers does it upload a single file to?

2024-06-05 20:24:08

- reply

What about these notes makes them more tricky for clients to find?

2024-05-30 12:47:31

- reply

How are you calculating the amount of entropy in the lines connecting the emojis?

2024-05-30 03:43:53

- reply

Naw, users shouldn't have to look at block explorers on this level

2024-05-29 18:46:40

- reply

What will that mean for application support? To me, a privacy focused Android derivative sounds more appealing than a custom OS

2024-05-29 18:00:07

- reply

One difference is I set the npub as a dns record so that many sites can all share the same single CDN file

2024-05-29 17:56:32

- reply

Nice! I'm working on something similar

2024-05-27 12:33:35

- reply

curious if it would still honour delete requests or no?

2024-05-25 21:35:06

- reply

based

2024-05-25 14:54:04

- reply

honestly looks great

2024-05-21 22:35:05

Is there a nip96 url scheme?
Something like "nip96://<sha256>"

2024-05-18 14:15:39

Running #blossom

2024-05-03 12:11:50

- reply

Would be good for that too. Eliminates mev as well

2024-05-02 23:19:43

I just want to homestead and write open source software

2024-05-02 22:57:53

- reply

Are you thinking of Mercury?

2024-04-20 01:39:24

What's your guess for the lowest feerate in the halving block?

2024-04-17 16:14:54

Good news is that sequencers are more involved with rollup protocol design than miners are. Therefore the financial incentive for sequencers to capture all MEV and leave none to the miners should sway rollup designs towards those that protect the L1 from centralizing MEViL

2024-04-05 00:46:54

Where can I get nostr stats like DAU?

2024-04-03 12:46:37

- reply

works as well as previous versions... which is to say, super slow on a pi

2024-03-31 22:39:28

- reply

deepweb usually means non-indexed http (normal web pages you can't find in Google), whereas darkweb refers to non-http (web pages you can't even visit without special software, usually Tor)

2024-03-25 13:42:51

- reply

Why do I see this happen to so many people? How do you lose your follows?

2024-03-14 01:23:01

- reply

Tell us how it goes!

2024-03-13 18:31:53

- reply

No trusted display? What is the purpose of this device?

2024-03-13 12:52:56

- reply

Every combination is valid so there's no "wrong" guess

2024-03-05 21:35:04

- reply

I assume you're only testing the censorship of your link to the images (ie the actual in-protocol note), right? Because one of the images themselves is already deleted from the host.
Do you think nostr needs stronger file decentralization?

2024-02-27 23:54:10

- reply

Cold start meaning the cache is empty, and it loads that fast??

2024-02-16 17:42:52

Nothing like going to a library's repo and seeing the latest commit adds the feature you needed and were about to implement.

2024-02-16 00:48:41

- reply

Mixing katakana and hiragana like this is not correct Japanese. To be consistent it should be: サトシナカモト.
But even then, names would normally be in kanji rather than "spelled out" like this. For example the family name Nakamoto is commonly spelled as 中本

2024-02-07 13:20:22

Cautiously optimistic about Citreon. Curious how efficient BitVM actually is in practice.

2024-01-30 20:58:35

- reply

I haven't read this either, but I'm curious now... might pick this up to see how much I forgot

2024-01-30 20:40:56

- reply

That's true, but only part of the story of the blocksize wars. Remember the winning side ("small blockers") also increased block size with segwit. This was much more a story of who controls the chain and what its purpose was than literally "should we tweak some config"

2024-01-28 18:17:13

- reply

Not surprising given there's 157k wBTC locked into Ethereum. There's tons of demand for EVM compatible Bitcoin

2024-01-28 01:42:34

Might be useful to have a standard for discovering nip96 servers. Even just a single event with the domain tagged with a standard way to make it searchable.

2024-01-28 01:37:58

- reply

Do you have any methodology for choosing relays?

2024-01-27 03:47:21

Ever since chatGPT came out, stack overflow has been super slow. What's up with that

2024-01-26 09:39:27

Idea: a "Bitcoin backed relay" (I'll use BBR for short).
A BBR is a nostr relay that stores events on the Bitcoin blockchain using one of the blob store techniques (OP_RETURN, taproot envelopes, etc)
Saving a message here is _expensive_, you would not want to do it often, but comes with the upside of permanent storage on all BBRs (they all share the same backend state; like if nostr did relay gossip with decentralized spam protection)
This could be useful for notes that are very important. Notes that everyone must find. For example, perhaps a key rotation note, or some kind of deletion event. I could also see standard notes being here by people just looking for wider attention or posterity ("vanity notes")

2024-01-25 19:15:26

- reply

Joined! I'm ujo on there

2024-01-25 18:51:31

Any go players on this app? #gostr

2024-01-25 04:00:45

- reply

Where can I find these scripts?

2024-01-24 12:58:55

Self custody is cypherpunk. Building out custodial scaling solutions is not very cypherpunk, but, it is pretty based. I appreciate this community exploring both ecash and covenants

2024-01-20 18:59:59

Any good statechain explainers? I'm having a hard time groking the concept.
#statechain

2024-01-20 17:28:59

Curious on your opinion #cashu #fedi #note1mv3…0ak7

2024-01-20 17:23:32

Seeing a lot of excitement here around custodial scaling solutions like fedi and cashu. Cool!
I'm curious what this group thinks of drivetrain, which is also a custodial scaling solution, except one where anyone can join the "custody federation" without permission (because miners as a cohort are custodians of drivechain funds)

2024-01-19 23:47:22

- reply

Oh I see that now; you're right for that page.

2024-01-19 18:55:35

- reply

Here's what I see: two home buttons, no Damus icon. https://image.nostr.build/f0a2f73e03cfc840f6ca632b4f2ff2c7a5515d1446c032daf88683225009edae.jpg

2024-01-19 18:50:46

- reply

Honestly it sounds like we have different apps or versions because what you're saying makes no sense to me 🤷

2024-01-19 12:41:18

#damus really needs to change the "home" button in the upper left to a "back" button. There's already a permanent home button in the bottom bar so it's fully redundant anyway.

2024-01-19 02:00:43

Let's conquer the universe.

2024-01-18 18:29:48

- reply

You're right that higher interest rates generally reduce liquidity, but there's other liquidity sources counter-acting that. For example nearly 2 trillion dollars have drained from the reverse repo since start of 2023 to fund USG deficit spending.

2024-01-18 12:14:38

- reply

Think liquidity (which already increased), not interest rates

2024-01-10 20:07:41

- reply

1) Holds peg better (less likely to have a discount/premium)
2) Lower costs

2023-08-15 00:06:15

Thoughts on spacetimeDB?

2023-08-14 17:01:16

- reply

Very excited for this one!

2023-08-12 02:51:47

- reply

Is there a standard for npub domain redirection? Something like dnslink?

2023-08-11 13:07:23

- reply

Wonder if this nip could be made compatible with the passkey standard Google/Apple/etc are pushing. Would be a shame to have fragmented public key identity networks

2023-08-09 02:54:35

Is there any dnslink nostr gateways?

2023-07-04 13:28:48

- reply

Yeah many of my most interesting follows are twitter-exclusive still

2023-05-09 12:22:27

It seems like cashu mints are a strict upgrade to custodial lightning services. Any counter arguments? Only downside I can come up with is the tech is nascent.

2023-05-01 17:13:46

- reply

It's gambling / FOMO. Even the website with the protocol docs has a reminder that these are pointless/worthless and something like Taro is a better technical implementation

2023-04-06 01:35:47

Inscribooors are bidding nearly $2000/MB to backup data on-chain instead of using commit hashes or openTimestamps. Even for paying upfront (not monthly) for a forever backup, that's more than I expected

2023-04-05 13:00:48

What are the practical ways to scale UTXO set size now that Stamps are on the scene? I've heard of utreexo, anything else?

2023-03-20 21:08:22

- reply

How much storage do these models take? Considering this from an information/knowledge compression perspective, how small can we make "the internet's knowledge"

2023-03-06 22:48:11

- reply

The best I can do to steelman a blockchain based social media like lens is they do get some improved UX from the strict singleton (no partial truths depending what relays you've got). The obvious argument is the UX is worthless if the rollup can't scale enough (either fees go up or decentralization goes down)