Oddbean new post about | logout
 People criticizing nostr apps for their quality are making multiple category errors:

- Individuals (or very small teams) can't produce the same level of quality as large teams, but teams can't exercise as much creativity as individuals
- Optimal UX comes from a need for growth, stemming from a need for profit. Grantees and hobbyists do not have this motive. But for-profit businesses won't be principled about putting the protocol first, while grantees and hobbyists may be.
- Good UX partly comes from experience, and existing best practices. Very little of this is established yet for nostr, both from a design and engineering perspective. We're making it up as we go along.

If you want something new, you have to take the bad with the good. When I started this, my expectation was that it would be a ten year project with a 0% chance of success. Two years in, I'd say we're doing extremely well.

I don't care about growth, and won't for a while. I'm not in it for user numbers or zaps, I want to use software to give my kids a better life. Drop the high time preference, and dig in, because this is going to be a long ride.

With all that said, I do feel a new wave coming in the next year or so, as best practices crystallize, and as existing projects reach a point of maturity where their developers recognize their own limits and need for help. I look forward to seeing teams coalesce to push forward what the creatives started.

This might take the form of more for-profit businesses, but I hope that devs (including myself) will be able to swallow their ego and pitch in on projects that don't belong to them without having to get "hired". The difficulty of this on nostr is of course that the scope of the protocol leaves so many tantalizing possibilities to work on.

For myself, I remain focused on my original mission of serving real-life communities. However, the longer I work on the problem, the larger it becomes. It turns out that there has in fact been decades of work in the space, and there continue to exist many unsolved problems, even without introducing decentralization. It would be hubristic to think that my first attempt at the problem would be either correct or successful. Iteration, exploration, and education are all necessary.

It's very likely that it's impossible for a single developer to cover even a single use case of nostr satisfactorily. We'll all eventually need help. This is just the nature of the project we've set for ourselves. 
 Well said! 
 Doing great! 
 Nostr needs a capital incentive baked into a product that can deliver a 10x order of magnitude improvement that current technology cannot provide. Clones of existing apps don’t deliver a substantial UX improvement for the vast majority of users. The nostr product that will go viral with normies is a product that is fresh.

In my opinion decentralized prediction markets over nostr create that capital incentive that is an order of magnitude improvement on centralized prediction markets. The improvement is obvious: anyone can ask a question and provide liquidity to bet (with Bitcoin, not a shitcoin). In centralized implementations only the prediction market operator can ask questions and provide liquidity according to LMSR. Decentralizing that opens the flood gates to anyone asking any question they want and enabling people to trade on those outcomes.

The difficulty with prediction markets over nostr is implementing trust minimized contracts/share issuance and redemption in event outcomes. It’s a hard technical problem, which is why we don’t currently have an implementation, but it’s possible to do. 
 Polymarket but on nostr with corn would be amazing. 
 Well said. 
nostr:nevent1qqsp9srmc920m8mhzkvrrzj2pkc3r7uxsslxen7j4gk5g75pddzf0cgpzemhxue69uhkummnw3ex2mrfw3jhxtn0wfnj7q3qjlrs53pkdfjnts29kveljul2sm0actt6n8dxrrzqcersttvcuv3qxpqqqqqqzdkyg8w 
 I guess nostr can be called indie tech. 
 Who gives this much of a fuck about the erroneous ones? Engage more about the criticisms that aren't erroneous. 
 I'm 60 and ready for a new chapter in my life story. Bitcoin is going to be a big part. Been a long time ⌛️  since I was this motivated or hopeful. Patience is a virtue and usually rewarded.  
 Great comment 
I have a friend who just started his own beekeeping business in NZ 
 Is this b/c of the Laeserin drama? 
 I have muted both of her nyms 
 The first two premises are asserted without evidence.

Quality/stability is in constant tension with pace of innovation.  Ensuring stability can slow innovation, while accelerating innovation can introduce more bugs.  The constant in this equation is the number of available man-hours.  Changing the size of the team changes the amount of man-hours you have to work with, but the stability vs. innovation tradeoff remains.

Optimal UX is not necessarily tied to a need for growth.  Better UX is motivated by a concern for user satisfaction.  Satisfying the users is how you retain them, whether the product is free or costly.  Dissatisfied users will flee the product, even if is is FOSS.

I would contend that a concern for the communities and individuals who use Nostr should motivate us to build excellent software with an excellent UX, whether those people are paying us for it or not. 
 Teams require communication, which introduces the need for process, which results in rigidity. Only exceptional engineering teams can balance creativity and stability.

The incentives between altrustic FOSS freedom tech and capitalistic for-profit business are not the same, which leads to different outcomes. That's not to say that nostr won't ever have good UX, but the path to it is much less direct. 
 i think over time the clients will get gradually better thanks to being open source. it will depend on how much people join nostr. more people should make the clients get better faster. it is a matter of time. 
 I only joined this week and I don’t see anything to complain about. I love that my feed is not packed with adds or AI generated posts. I’m looking forward to seeing it grow up. 
 10年応援し続けるよ🧘 
 You know what’s a great Nostr app?

Damus 
 "the longer I work on the problem, the larger it becomes" 
nostr:note1ztq8hs25lk0hw9vcxx9y5rd3z8acdpp7dn8a923dg3agz66yjlss8asl82 
 The more limitations we as developers realize about ourselves, the more self growth.

Many breakthrough softwares are done by individuals or small teams.

Id Software shipped 11 games in 12 months back then.

Because of the gold rush of the past 18 years of the internet businesses, it’s just too easy to get a programming job.

Competency has become a myth and people start to believe that good things have to take an army of people to make.

I’m also guilty of it. I’ve only realized how many skills I should’ve developed but didn’t, after I started my Nostr journey.

But I am learning as much as I can. If every indie developer 2X themselves very year, things will eventually get better. 
 your client has tremendous quality. ive seen first hand how many innovations and thoughtful improvement you’ve pushed out. 
but quality in this context is so subjective. People conflate it with personal value based on personal preference. my most used nostr software is nak, nostreq, and nostcat. almost nobody uses them. 🤷 
 any image based winkies that you could recommend?  
 I am still new on nostr but have been blown away at the experience so far. I have the utmost trust in the abilities of the devs I have come across here. The ideas are mind blowing 
 Right on the money 🎯 I don’t have developer skills but I share your perspective on wanting to build a better world for future generations. I’m on nostr because direct, honest, uncensored discussions will be critical. 
 Some thoughts: 

Nostr is revolutionising the world, with the way it connects digitally. 

The majority of the world are programmed to the current ways of the internet ie letting govt and corporate decide what they should say and how they should think. 

Nostr is changing that censorship behaviour - but subtly. 

Randy Pausch in his last lecture coined the term “head fake”.  He taught his students programming - but through various fun projects and performances - so much so the students didn’t realise they were learning to code. 

Nostr is that head fake. It subtly helps people understand that they have a right to their voice and thoughts. 

And the fun projects or use cases that attract people - are the clients and micro apps

The work that all of you are doing is extremely important for the world, and much appreciated. When everything seems doom and gloom, these are the things that give hope. 

I do believe  user feedback is important. You can’t please everyone, but all these feedback can be consolidated so that devs can see what they want to resolve or improve next 

You are right that this is a long journey. It won’t take 10 years of development, but it might take 10 years of trial and error with users - which is painful and challenging and emotional and tiring - but with it comes immense learning that opens doors for the mainstream market 

In your case, its incredible how you have been very active in exploring large volumes of users. You would have seen many positive outcomes and also many challenges. But more importantly, plenty of learning for you to strategise the next step, and the next. And it slowly gets better.

It might be easier to reach out to the early adopters - those who would fundamentally appreciate Nostr’s long term impact for the world. They could be anybody - other tech folks, global leaders and corporate leaders or anyone who have seen the dirty little secrets of the gov’ts, marginalised folks, curious folks, people who just want to give you a chance etc. 

These groups of people will be kind to you - and you can use this opportunity to understand what they like and dislike, their behaviour and usage - to continually make clients, apps and nostr in general, better.

I think the back and forth of users coming and going can be very disappointing, but it's also an opportunity for everyone to learn, strategise and re-strategise, and get creative with solutions. 

You will make a lot of mistakes - but you will also learn from it and bounce back higher each time. And you will keep getting better. 

I appreciate your courage in putting yourself out there.  I'm sure you have heard of this : The only people who fail are those who do not try, and You miss 100% of the shots you do not take. 

I also appreciate your wisdom with both short term execution and long term goals in place

And in 10 years, you and your products  will be more than ready for the mainstream market when Nostr goes viral - and in 20 year, there might be a generation that will be able to maintain their autonomy - and that is a precious gift to give to the world. 

Keep going, keep moving forward. 

You got this! 
 
 Much appreciated Pam! 🫂 
 One correction - most software is not actually done by large teams. It's often one to five guys in an office. The only difference is they don't have to answer customer support emails nor do sales.

I've been looking at a few projects that I liked and thought they're done by many coders. And then I met the (one) guy who has done the project :).  
 The Long Game 💜 
 nostr is leaky (it can't be partitioned effectively), and nostr needs a localised, affinity based division of its relays and user groupings

the biggest failure in nostr architecture at this point is not recognizing that its perfect for private business to use for internal and public communications 
 Maybe you should go back and re-read my original post 
 cat fight!!!!!!!!!!! yesssssssssss 🍿 😏

https://media1.tenor.com/m/oL2GbjrEzLkAAAAd/chandma-cat.gif 
 i'm agreeing with @hodlbod on this one for sure

i think this is actually a big blindspot in the entire field of software architecture to not understand what one dude can do versus what needs more than one dude

and it should be pointed out that @hodlbod is fielding a helper to do his work because he has hit a limit on what one person can build but honestly

just use coracle.social for a bit and tell me it's not pretty much on point

yeah, it's the best one out there, one dude. 
 and not to leave out the praise i feel, @hzrd149 also is a one-dude situation who has built a really amazing nostr client that does a separate set of things that are equally almost perfect in their execution

this really drills at the heart of why nostr will win in the long run

because it has a unit that is small enough that one dude can encompass it with aplomb 
 love noStrudel  
 Fastest thinker I've ever met. 
Best designer on Nostr.  
 You're self-hosting it, tho.

And he's losing control of the sprawling codebase. It's grown too big and complex, for one person, and the bugs are creeping in. Especially, since he started working on Blossom, in parallel. Those are two, relatively large, popular projects, that need to be constantly updated and actively maintained and monitored, run by one, single person, who works almost entirely alone.

Incredibly brittle structure. That is not how we win. 
 my point is just that this wouldn't exist if it weren't for the level good programmers can get to

to take it to the next level does need a good team and it especially need legit architects, architecture is the hardest skill to learn because software does sprawl far beyond the ability to consider it in one "mind-full" if you know what i mean... beyond a couple of source files it is an elephant and you are a blind man 
 Also, really there needs to be a revenue model... nostr:nprofile1qyghwumn8ghj7mn0wd68ytnhd9hx2tcpzfmhxue69uhkummnw3e82efwvdhk6tcprfmhxue69uhhq7tjv9kkjepwve5kzar2v9nzucm0d5hsz8rhwden5te0wdshgetvd35hgefwdpa8yep3xsujucm0d5hsz9thwden5te0wfjkccte9ejxzmt4wvhxjme0qy88wumn8ghj7mn0wvhxcmmv9uqzqfngzhsvjggdlgeycm96x4emzjlwf8dyyzdfg4hefp89zpkdgz99auacd9  was talking about a premium features subscription then nothing followed 
 Has anyone tried to just be like: 
"Hey guys, I've got this new release here where I fixed this and this bug, made x feature way better, etc etc.... Buuuut, I won't release it until this post gets ⚡1.234.567." 
#zapwalls

 
 Needs UX to give zappers tons of recognition of course.  
 site unseen?   just like anygood hooker $ upfront      (o_O)       wisecrack! 
 @Zapstore this is the sort of UVP that releasing apps on Nostr allows for.  
 What's wrong with a feature subscription though? Works very well for YouTube, protonmail, and many others... I just started using a paid search engine now too, this is the way.jpg 
 I don't know. I have a few, as well, and I like them. Also, just a nice way to support a project you like to use, and allow them to finance a free service for other people and stabilize their revenues.

We're going for the feature subscription thing, over a relay subscription. (Being whitelisted on theforest relay will open up additional functionality or services, and access to additional private relays.) 
 1. Offering hosting and computation services makes sense both for monetization and for providing a reliable product/experience
2. Pay-per-use of those services makes you more adaptive in a highly competitive market. Nothing stopping you from offering pre-paid bundles. 
 
 Grants postpone thinking about that sort of stuff  
 For the record, I pointed out that Coracle is the most-stable web client we currently have, last week. I tested them all the same way, and it came out first, but I still prefer Nostrudel because of the threading UX and some administrative functions like two-level relay management.

He couldn't see that, as he has me muted, but that's his problem, not mine. 
 yeah, this is pretty much my technical assessment of the state of the clients

amethyst gets so much hype but it's rubbish in comparison

anyhow, we have our fucking discussion forum, time for some DMs and group chats ffs lol 
 You're self-hosting it, tho.

And he's losing control of the sprawling codebase. It's grown too big and complex, for one person, and the bugs are creeping in. Especially, since he started working on Blossom, in parallel. Those are two, relatively large, popular projects, that need to be constantly updated and actively maintained and monitored, run by one, single person, who works almost entirely alone.

Incredibly brittle structure. That is not how we win. 
 my point is just that this wouldn't exist if it weren't for the level good programmers can get to

to take it to the next level does need a good team and it especially need legit architects, architecture is the hardest skill to learn because software does sprawl far beyond the ability to consider it in one "mind-full" if you know what i mean... beyond a couple of source files it is an elephant and you are a blind man 
 Also, really there needs to be a revenue model... nostr:nprofile1qyghwumn8ghj7mn0wd68ytnhd9hx2tcpzfmhxue69uhkummnw3e82efwvdhk6tcprfmhxue69uhhq7tjv9kkjepwve5kzar2v9nzucm0d5hsz8rhwden5te0wdshgetvd35hgefwdpa8yep3xsujucm0d5hsz9thwden5te0wfjkccte9ejxzmt4wvhxjme0qy88wumn8ghj7mn0wvhxcmmv9uqzqfngzhsvjggdlgeycm96x4emzjlwf8dyyzdfg4hefp89zpkdgz99auacd9  was talking about a premium features subscription then nothing followed 
 Has anyone tried to just be like: 
"Hey guys, I've got this new release here where I fixed this and this bug, made x feature way better, etc etc.... Buuuut, I won't release it until this post gets ⚡1.234.567." 
#zapwalls

 
 Needs UX to give zappers tons of recognition of course.  
 site unseen?   just like anygood hooker $ upfront      (o_O)       wisecrack! 
 @Zapstore this is the sort of UVP that releasing apps on Nostr allows for.  
 What's wrong with a feature subscription though? Works very well for YouTube, protonmail, and many others... I just started using a paid search engine now too, this is the way.jpg 
 I don't know. I have a few, as well, and I like them. Also, just a nice way to support a project you like to use, and allow them to finance a free service for other people and stabilize their revenues.

We're going for the feature subscription thing, over a relay subscription. (Being whitelisted on theforest relay will open up additional functionality or services, and access to additional private relays.) 
 1. Offering hosting and computation services makes sense both for monetization and for providing a reliable product/experience
2. Pay-per-use of those services makes you more adaptive in a highly competitive market. Nothing stopping you from offering pre-paid bundles. 
 
 Grants postpone thinking about that sort of stuff  
 my point is just that this wouldn't exist if it weren't for the level good programmers can get to

to take it to the next level does need a good team and it especially need legit architects, architecture is the hardest skill to learn because software does sprawl far beyond the ability to consider it in one "mind-full" if you know what i mean... beyond a couple of source files it is an elephant and you are a blind man 
 Also, really there needs to be a revenue model... nostr:nprofile1qyghwumn8ghj7mn0wd68ytnhd9hx2tcpzfmhxue69uhkummnw3e82efwvdhk6tcprfmhxue69uhhq7tjv9kkjepwve5kzar2v9nzucm0d5hsz8rhwden5te0wdshgetvd35hgefwdpa8yep3xsujucm0d5hsz9thwden5te0wfjkccte9ejxzmt4wvhxjme0qy88wumn8ghj7mn0wvhxcmmv9uqzqfngzhsvjggdlgeycm96x4emzjlwf8dyyzdfg4hefp89zpkdgz99auacd9  was talking about a premium features subscription then nothing followed 
 Has anyone tried to just be like: 
"Hey guys, I've got this new release here where I fixed this and this bug, made x feature way better, etc etc.... Buuuut, I won't release it until this post gets ⚡1.234.567." 
#zapwalls

 
 Needs UX to give zappers tons of recognition of course.  
 site unseen?   just like anygood hooker $ upfront      (o_O)       wisecrack! 
 @Zapstore this is the sort of UVP that releasing apps on Nostr allows for.  
 What's wrong with a feature subscription though? Works very well for YouTube, protonmail, and many others... I just started using a paid search engine now too, this is the way.jpg 
 I don't know. I have a few, as well, and I like them. Also, just a nice way to support a project you like to use, and allow them to finance a free service for other people and stabilize their revenues.

We're going for the feature subscription thing, over a relay subscription. (Being whitelisted on theforest relay will open up additional functionality or services, and access to additional private relays.) 
 1. Offering hosting and computation services makes sense both for monetization and for providing a reliable product/experience
2. Pay-per-use of those services makes you more adaptive in a highly competitive market. Nothing stopping you from offering pre-paid bundles. 
 
 Grants postpone thinking about that sort of stuff  
 Also, really there needs to be a revenue model... nostr:nprofile1qyghwumn8ghj7mn0wd68ytnhd9hx2tcpzfmhxue69uhkummnw3e82efwvdhk6tcprfmhxue69uhhq7tjv9kkjepwve5kzar2v9nzucm0d5hsz8rhwden5te0wdshgetvd35hgefwdpa8yep3xsujucm0d5hsz9thwden5te0wfjkccte9ejxzmt4wvhxjme0qy88wumn8ghj7mn0wvhxcmmv9uqzqfngzhsvjggdlgeycm96x4emzjlwf8dyyzdfg4hefp89zpkdgz99auacd9  was talking about a premium features subscription then nothing followed 
 Has anyone tried to just be like: 
"Hey guys, I've got this new release here where I fixed this and this bug, made x feature way better, etc etc.... Buuuut, I won't release it until this post gets ⚡1.234.567." 
#zapwalls

 
 Needs UX to give zappers tons of recognition of course.  
 site unseen?   just like anygood hooker $ upfront      (o_O)       wisecrack! 
 @Zapstore this is the sort of UVP that releasing apps on Nostr allows for.  
 What's wrong with a feature subscription though? Works very well for YouTube, protonmail, and many others... I just started using a paid search engine now too, this is the way.jpg 
 I don't know. I have a few, as well, and I like them. Also, just a nice way to support a project you like to use, and allow them to finance a free service for other people and stabilize their revenues.

We're going for the feature subscription thing, over a relay subscription. (Being whitelisted on theforest relay will open up additional functionality or services, and access to additional private relays.) 
 1. Offering hosting and computation services makes sense both for monetization and for providing a reliable product/experience
2. Pay-per-use of those services makes you more adaptive in a highly competitive market. Nothing stopping you from offering pre-paid bundles. 
 
 Grants postpone thinking about that sort of stuff  
 For the record, I pointed out that Coracle is the most-stable web client we currently have, last week. I tested them all the same way, and it came out first, but I still prefer Nostrudel because of the threading UX and some administrative functions like two-level relay management.

He couldn't see that, as he has me muted, but that's his problem, not mine. 
 yeah, this is pretty much my technical assessment of the state of the clients

amethyst gets so much hype but it's rubbish in comparison

anyhow, we have our fucking discussion forum, time for some DMs and group chats ffs lol