Flux is dead, Long live Flux(pipe)
Open Source Adoption, just the other way around
When we talk about "adoption" in Opensource we're usually referring to the amount of users leveraging a specific project or solution. Today, it's the other way around and we're talking about Users taking over an Opensource project in dire straits.
Backstory
Flux is InfluxData’s functional data scripting language designed for querying, analyzing, transforming and acting on data. If you're here you likely know it.
Flux perhaps is not the easiest or the most intuitive language but once you've adopted it and its unique benefits, it's hard to move away. Extremely well designed, performing fast and very easy to extend using GO or Flux itself. A magical mix!
So, People invested in Flux and built their products and pipelines on top of it.
Everything was fine and well until InfluxDB sadly decided to drop Flux support from their InfluxDB 3.0 plans due to incompatibilities in technology and more importantly lack of resources. If he was a GitHub member even Stevie Wonder could have seen this coming when the number of commits and releases started decreasing and issues started going unanswered, sadly followed by the departure of the Flux dev team 😞
🏳️ Sad ending? Not so fast.
As Flux contributors and supporters, we just couldn't look away... Flux is good.
Opensource Street
Opensource is quite rarely a two-way street (yet magic when that happens)
Everyone loves OSS projects when it's time to download a perfectly working application or get free support from strangers on the internet. But when it's time to give back, nobody feels like this is their responsibility or duty and most people only contribute complaints on forums and social media when their free toys disappear.
I shouldn't say nobody. Lots of people are willing to do something, but it's really hard for anyone to coordinate unless there's a license war going on and this was not the case. When not even InfluxData seemed to be ready to fight for Flux...
Can Flux be saved?
TLDR; Yes, with the help of InfluxData and the support of their team!
We reached out to the Flux developers and masters such as Nathaniel Cook, Scott Anderson and many others kindly walked us through our first contributions and PRs and later introduced us to Rick Spencer at InfluxData. Rick is an open-minded person and possibly saved Flux by promoting its whole migration to the InfluxCommunity repository under Community governance. Paul Dix himself blessed our initiative project publicly and sealed this pure community deal:
Currently, qxip is the only active Flux sponsor. Help us do this stuff.
Enter the Flux(pipe)
From the hashes of Flux, we created our stand-alone replacement: Fluxpipe
Created to abstract the Flux script execution out of its natural environment, Fluxpipe includes the full Flux library functionality and extends it beyond its original capabilities adding support for FlightSQL, InfluxDB 3.0, ClickHouse and more.
The goal for Fluxpipe is to be a drop-in replacement for Flux execution, adding the freedom to execute anywhere: on-premise, on servers or serverless functions.
Fluxpipe v1.0.0 is out and just works, fully ready for InfluxDB 3.0 - right now!
👉 Here's a Live Demo you can play with or run for free on fly.io
Why Flux? Don't you make Observability Tools?
Time to Switch Over?
Flux is alive and well. You're not alone.
It's time to stop complaining and start migrating and contributing.
Lots of Flux fans and users already reached out confirming the value of this project! Are you the next one? Any type of feedback and contribution is very welcome!
👉 You can Download, Install and use Fluxpipe using our Github repository.
Yes, Fluxpipe runs at 141,6Km/h