I’ve been generally learning how to build Phoenix apps and making a small utility to help measure Paddle.com revenue. It’s been pretty sweet, learning about the ecosystem!
mattei
code-shoily
Materialize wrapper for SurfaceUI. Started working with it last week. Will share once a good number of components is adapted.
lesliep
I’m finishing up an app and going through exercism, and just generally preparing before putting in for some elixir jobs.
StephanLehmke
Working through Functional Web Development with Elixir, OTP, and Phoenix while trying to use it as a template for a long-term server project I’m working on (migrated from NodeJS to Django to Elixir so far 
strzibny
I am preparing a little template for my Elixir side projects. I am also figuring out how I’ll deploy it. I think I’ll compile it locally and run with systemd services. To that end, I made myself a little Dockerfile to compile releases.
Jcambass
I was thinking about open sourcing this one.
The main reason is that I created it as an experiment and while I currently host it under spotifywall.com where it can be used for free, I don’t plan to monitize nor market it in the future.
Would you be interested in seeing this opensourced so that you can either host it yourself, modify it or simply read the code (don’t except to much 
)?
jswny
Writing a Discord bot with GitHub - Kraigie/nostrum: Elixir Discord Library · GitHub. Have to say it’s been a very pleasant experience. Shout out to the maintainers if anyone sees this
Aetherus
I’m trying to build a chopped-down and twisted version of ecto_query.
Ecto is the greatest database access library I’ve ever used. I really love it. But for my current project which will be an online business intelligence application, I need a more dynamic query builder.
Things I want in Ecto Query:
- Composability
- Elegant macro-based API
Things I don’t want:
- Schema (tables need to be created at run-time using user-supplied data)
- Atom-based field API (like
field(q, ^field_name)wherefield_namemust return an atom)
Things I want but not in Ecto:
- A way to tell if a given query is an aggregate or not.
- More aggregate functions and free-style aggregate (like
fragment, but with hints to tell the developers it’s an aggregate).
So, for example, instead of
select(query, [q], q.foo)
I want
select(query, [q], q["foo"])
and
field_name = "foo"
select(query, [q], q[^field_name])
I know this could lead to SQL injection, but I guess carefully escaping the SQL identifiers could help minimize the danger (for now, I just try to adapt to PostgreSQL).
Currently, the biggest problem is joining and keeping tracking the table/subquery aliases.
Werner
Had a problem with my liveview template and couldn’t find the error immediately… then discovered “liveSocket.enableDebug()” and immediately the error was fixed… wow, am ecstatic!
msukanen
Tinkering. Making various “proof of concept” things (read: proving to myself that I can do this or that), e.g. base64 in C, C++, Kotlin, Haskell, Elixir, etc…