Nostr Web Client

I am not being glib here, Primal is the number 1 argument against Nostr I come across. I don’t really have a response against the argument either, other than hoping that Primal changes or it eventually becomes a marginal client. But at this moment, all we have is promises while it probably is the most popular client out there.

For those not aware, Primal does not use relays on the input side, although it does on the output side. This means that it connects to a caching server, not a relay, and you will only see what is on there. If someone references an event that is not on that server, even though relay hints are included, tough luck; the only redeeming factor is that those outside of this primal box will see your complaint you can’t see the event.

These quirks resulting from Primal being Notes and other stuff transmitted by cashing-server, instead of relays are annoying, but the most worrying part is the fact that its a real manifestation of the biggest fragility Nostr has as a protocol: Clients ignoring the spec and the design philosophy of openness and interoperability while being in a dominant popular position. I can make arguments all day about the theoretical merits of Nostr, but they are easily undermined by pointing to Primal. What am I supposed to say in response? More theoretical notions like people being able to change software, even though we know that is just not going to happen?

I understand people will retain their profile, social graph, posts etc. if they move to another client, but as it stands those just become excuses for Primal to be a bad actor; how about just fixing your shit, so I don’t have to manually broadcast my stuff so people in the Primal-box can see it, but above all I can argue for Nostr without having to concede that the censorship resistant theory is not living up to practice. nostr:nevent1qqspy7pxqwk66dm46yxwetee0m2lrz5zeh2ldcwycrt6fuegyrllqlctrjx2f

BITKARROT 2mo ago

I've always used, since the beginning (2021) when I first got on the Nostr network, a bunch of different clients and many have evolved. So its a constant game of changing clients to find what works and fits best for daily use. R&D by its own nature is rough; and we are still in the R&D phase for everything.

I find myself using primal less and less unless there is some feature that other clients don't support. Having a team and capital to do the work that other projects can't afford was once a big advantage to eat market share, but it maybe soon that AI will even the playing field.

I think that's going to be it, AI tools could potentially level the playing field that we might not need the same amount of capital investment to do the same amount of labor, however it will also be a strategic game in how efficiently one deploys both machine and human resources and time.

Reply to this note

Please Login to reply.

Discussion

Constant 2mo ago

Im not so sure on the AI coding thing, but who knows. Some human labour seems to be a requirement for that to work, because a lot of stuff has to actually be thought about; you need to reason and know what you client needs to do, and then build to software for it to do that thing.

This is because you actively need to ask/query particular things, from particular places/relays. And its the combination of what to ask for where and when that is non-trivial.

Perhaps after there are a number of working examples of this stuff out there, as well as lots of written content about the ideas behind it, that some AI has enough to latch onto and figure it out. But as it stands currently, its not going to figure it out on its own.

BITKARROT 2mo ago

I agree with your sentiment, but I also want to shed light on the fact that using LLMs, MCPs, templates, Agents and sub agents has accelerated my building time 3-5x times. Using the tools is a skill that needs to be cultivated, its not going to happen overnight.

Also doesn't mean we don't need to know the fundamentals; its actually more important than ever before to be able to code review, understand software architecture and be selectively critical.

Maybe I see it differently because I'm knee deep, but I sure feel a massive difference between Nov '24 and Nov '25. Its not like we're in the same decade anymore. Also I can't walk out my front door without hearing someone talk about tokens casually or LLMs; I live in AI valley.

However, I know for sure, whomever isn't using the tools, these people are most definitely going to be left behind.

Constant 2mo ago

Sir. 1 dude made rollercoaster tycoon in assembly on his own. I don't think people like that will be left behind, i think they are the ones everyone else and their robot girlfriendAI depend on. We need those to lead the way actually.

BITKARROT 2mo ago

Leading the way, indeed. But I must ask you to address me as Ma'am as I am a 'she' not a he :D

Constant 2mo ago

My dude, there are no women on the internet

BITKARROT 2mo ago

You’re kidding right?

Séimí Mac Síomón 2mo ago

It's a chick. I've chatted with her. She's cool, don't be a dick.

Constant 2mo ago

Sweet summer child

Vanessa 2mo ago

None

Constant 2mo ago

Never met a bro more frantically looking for a tittybar than you, so i knew you'd get it.

ᴛʜᴇ ᴅᴇᴀᴛʜ ᴏꜰ ᴍʟᴇᴋᴜ 2mo ago

yeah, strategy is everything. this is an arms race. we are building a reusable set of tools and systems that allow us to replace centralized ones and beat them on efficiency and cost. the LLMs will gradually become part of that. little mini-pc boxes are arriving this winter with 128gb that can run full reasoning/visual models. IMO the current projected power apocalypse would be fixed just by decentralizing AI server location. skynet also needs that centralized pattern but it's only a matter of time now before common domestic hardware levels the playing field between us and the big corporates as well. eventually scaling techniques will become better and we will just stack these things on our desk and have our own pocket einstein over ContextVM.

ᴛʜᴇ ᴅᴇᴀᴛʜ ᴏꜰ ᴍʟᴇᴋᴜ 2mo ago

also, we WILL get outside of the DNS box too. just needs a functional distributed data replication protocol for a name registry that uses nostr events, and we are then sovereign down to the IP level.

they will take a bit longer to develop because of funding but once tools have been made, they are out there being used. no tool spreads faster than software.

Constant 2mo ago

For if you missed it:

nostr:naddr1qquhqctr94kkzm3dw35x2tthv43z6tfdw35x2ttydakkz6tw94hxzmt994jx2umfvahz6umsv93k2tfsxshnzvf0xgcrydgzyzamthdqu92k09ulq4p5q77uyqeadu9mkv8hy5f2nqw0mvhsncn5wqcyqqq823cxtshxd