Unstream is now a tips platform, and other things I'm working on
A tip is the smallest thing a listener can do for an artist. You walk by a street musician, pull out some spare cash, and drop it in a jar. I’ve wanted something with that simplicity but online. Instead, we have a lot of tipping/patronage platforms with other cruft and fees built on top, encouraging artists to turn their art into a whole other offering simply to justify charging enough to make it worth their while.
So I’m excited to share that you can now tip directly on Unstream! The premise is simple:
- Artists who claim an Unstream profile can now connect a Stripe account to that Unstream profile, in which to receive direct tips
- No mandatory platform fee – artists can choose to set aside 0-5% of their tips to support Unstream’s development
- Fans will be able to leave a tip directly from your Unstream profile, as well as from the Mac app and browser extensions if they find an artist they like while streaming. There is no need to sign up or go to a different platform - it’s all quite frictionless. They will also have the option to cover the platform and payment processing fees for the artist.
To my knowledge it’s the lowest-friction and lowest-cost way to simply tip a music artist, purpose-built for music artists. Here’s a little idea of what it looks like for the fan viewing my profile:

I’m excited to see how this goes as an experiment. A few artists have already enabled tips as well, check them out:
Upgrading Unstream
In parallel I’ve been making some (occasionally frustrating, but increasingly necessary) updates to Unstream’s backend.
First, I decided to upgrade the underlying database service. It’s currently hosted on a service called Supabase, which has a free offering but, in the past 2 months, I’ve been bumping up against the limits of what their free offering allows. The result was Unstream simply breaking for its (few) users: searches would just fail, profiles wouldn’t load, you could barely even log in, until I restarted the database server manually.
Upgrading fixes all of that for $25/month. My alternative is to migrate the entire database, compute and authentication layer of Unstream to a different service. I’ve been considering this anyway, since there are some aspects of the services Unstream relies on that I don’t love: Supabase is a VC-backed startup whose platform runs on Amazon’s web services; same with Netlify (which handles deploying the app and running background jobs) and Upstash (which hosts the caching service, allowing Unstream to be fast).
I’ve wanted to find some alternatives closer to my values – independently owned or at least accountable to different incentives than the VC structure allows – but challenges abound with that. As I’m based in the US, my options for “ethical” alternatives in this space are a bit limited because of the simply small number of service providers that are not beholden for venture capital or Wall Street. Moving to a EU-based provider is potentially interesting but something that’s harder to manage from the US (I explored Hetzner as one option). I got this close to moving all of Unstream to the Digital Ocean app platform, but that would have taken a TON of stressful and complex work that was in many ways over my head, and would not have gained me much in terms of cost savings or peace of mind.
So I’m still on the same architecture as before, but with a better database that I’m now paying $25/month for. If you like what I’m doing with Unstream and want to help me keep it alive, I’d never turn down a donation on my Ko-fi page.
Down the line I’d like to still pursue a migration to a more ethical architecture for Unstream, or potentially offering a way to self-host some version of it. That’s for later.
Next up: A social promotion toolkit for small artists
In completely unrelated news, my employer Buffer has decided to host its first-ever public build contest, called Commit. We released a public API earlier this year and have invited builders in our audience to try and build truly cool and useful enhancements or entire products on top of that API. As an employee and tinkerer myself I was compelled to try and build something as part of the contest. (I can’t win the contest as an employee, but we have a hilariously-named “Conflict of interest” award for those of us participating.)
I’ve always wanted to build some kind of better social media toolkit for artists. Being on social as an artist is uniquely exhausting: the content that seems to work the best is viral video content on TikTok and Instagram Reels, which is a lot of work; plus you have to be on all the other social networks to even have a fighting chance of being seen by someone who might make sense as a potential fan.
While I don’t at all condone fully automatic social media or posting AI-slop garbage, I’ve started to feel this desire to put some of my content on autopilot: occasional nudges and links to new and older releases that might show up in some potential fan’s feed, punctuating between actual posts of mine that show my personality.
So as part of this Commit contest, I started building the feature I wanted: an autopost engine for release and patronage reminders, built directly into Unstream. You’ll be able to connect up to 3 social profiles for free (or more if you connect a paid Buffer account), set some non-AI-generated post templates for the different kinds of auto-posts you want to have on a recurring basis, and Unstream will randomize a schedule of these posts based on the releases and links you’ve already got on Unstream.
I don’t have much to share yet on this, but it’s coming together quite quickly (perks of being a Buffer teammate).