Thanks for your input. But, given that you don't appear to favor Py vs JS for an ESP8266, I'll stick with microPython I've installed umqtt into the Lib and it seems to function adequately for it's application.
I would also like to see something like a Node.py for Node-RED, but I've experienced a lot of push-back on that suggestion.
It would be nice as TotallyInformation suggested, if there were another flow-base visual tool that was structured flexible enough to allow various Back-Ends to be added, as there will always be new programmings tools in the future.
Can you elaborate what you think node-red would actually gain by writing it in python other than some syntactic differences and curly braces ?
Moderator's comment: I'm wary of this topic going in circles and repeating the same talking points and becoming unproductive. There's little benefit having an endless discussion trading blows over what language is 'better'.
Please consider carefully whether you have something substantially new to add to the discussion regarding a Python runtime for Node-RED.
The rationales for the suggestion have been stated through-out the thread. Ultimately it comes down to a personal preference for Python.
That doesn't preclude anyone else from taking on the substantial task of producing a python runtime - keeping in mind you'd need to not just implement the runtime but a large number of nodes for it to be useful.
It would also potentially get access to all of the science, engineering and data analytics libraries.
I would personally really love to see the power of Python's data management libraries wrapped into a visual programming tool making it relatively easy to analyse massive, complex CSV files for example.
Node-RED is great for IoT, home automation and the like, but it lacks the libraries for data analytics. Not the fault of Node-RED at all. Just that the libraries don't really exist for Node.js.
Once upon a time we did have one really keen person wanting to port Node-RED to Java as they considered it more enterprise acceptable, and I think they did start on it. That was about 7 years ago but I haven’t seen it yet.
In my one of my past posts I explicitly stated that I WASN'T suggesting abandoning JS, but PERHAPS Python could be added as and additional Back-End (e.g. Node.py). The "push-back" was no one seemed to like that idea either. When mdkrieg appeared to be suggesting the same thing I was just replying as to what my experiences were. I hope this clarifies the issue
Oh I see, so not a replacement for node.js, and not something that allows nodes written in python to be installed into the existing node-red.
EXACTLY, sorry if I didn't make my intent clear.
So at least I feel fully confident I selected this path already so long time ago - when I realized my advanced (already developed) Python AI scripts should be kept "as is" and only communicate the results back to NR when they happens. Has been working so well during the years, no need to think differently now
Indeed - if it works - don't fix it.
You wont, of course, expect an Enterprise Architect such as myself to agree to that statement I hope!
Of course not ! It'll always be better after you have spent 24 months and $M on the project ... maybe.
How do you expect people to give you $M's with an attitude like that?
hence why I work in Open Source
How else will I get my kids through Uni