Profile avatar
akoutmos.bsky.social
Get your hands dirty 💻 🎙️ Podcast: http://beamrad.io 📙 Book author: http://akoutmos.com/top/books ⌨️ Elixir libraries: http://hex.pm/users/akoutmos 📈 SaaS founder: http://eaglemms.com
34 posts 954 followers 14 following
Regular Contributor
Active Commenter
comment in response to post
Great work putting this together!
comment in response to post
How did I not know about this! AWESOME!
comment in response to post
It was a very odd experience to put on the Clockwork Pi uConsole. Suddenly it is an Elixir-only computer. elixirforum.com/t/project-th...
comment in response to post
K20
comment in response to post
The civic is currently auto, and the new motor comes with a manual transmission so it'll be getting a full conversion!
comment in response to post
I didn't know about this feature! Thanks for sharing!
comment in response to post
I thought I was cool at a whopping 156MB...I now need to recalibrate lol
comment in response to post
Fixed it for you: "He's always motivating*" 😜
comment in response to post
I actually have not run that benchmark to see what's faster. I would suspect that :ets.file2tab is faster....but I have absolutely 0 data to back up that claim haha. Someone should totally put together a benchee comparison of the 2 😄
comment in response to post
For eaglemms.com I have a whole bunch of read only data like year/make/model data for cars. Some of my DETS files are 100+ MB and I load them from DETS -> ETS on application init. It's data that only needs to be freshened up once a quarter and it is a hot path for everything else in the app.
comment in response to post
That is exactly what I am saying! I used to do something similar with CSVs....but my CSVs actually got so large that it was stalling application start up by a few seconds. Loading DETS -> ETS was instant!
comment in response to post
Agreed. That's also why I suggest people stick to monoliths as long as possible 😁. It's so much easier to not have the problems of distributed systems when you are trying to sort out the problems of product-market-fit. Vertically scale as long as possible 😉.
comment in response to post
I think that this blog post that I wrote for Fly covers exactly that topic fly.io/blog/monitor.... Let me know if you need any help :)
comment in response to post
Like Peter said. PromEx builds upon these tools and offers turn key metrics collection and Grafana dashboards for libraries such as Phoenix, Liveview, Oban, Absinthe, etc (hexdocs.pm/prom_ex/read...). If you want to just experiment with Telemetry it's probably more than you need.
comment in response to post
comment in response to post
That's pretty impressive. ChatGPT, Claude or something else?
comment in response to post
The real Turing test 😮
comment in response to post
Good work on the library! I've heard of typst before but never used it. I'll take a look over the weekend and provide feedback as best I can!
comment in response to post
Also if you have any suggestions for libraries that you want to see bindings for from Elixir, let me know!
comment in response to post
haha. Who needs a rubber duck when you have me 😄
comment in response to post
Thanks! We may be biased but we think it is pretty cool too 😁
comment in response to post
Thanks!
comment in response to post
Putting the final touches on this as we speak! Should have something out in the wild next week :)