For those who follow our conversations in the forum, there shouldn't be anything surprising in the post - but its useful to have it all set out in one place. I'm sure there will be parts I've overlooked (it was only a 30 minute talk...), but I think it does cover the main areas we want to focus on.
As I say in the post, the roadmap is shaped by the feedback we get from the community. Let us know what you think.
Congratulations for the concise presentation of the Node-RED roadmap.
There are so many interesting ideas... I like the tools for development.
Just some comments:
It is always a challenge to add more features while keeping the same level of performance and ease of use.
With so many nodes available, it is sometimes difficult to manage those installed. For example, how to upgrade the nodes that require so? Selecting and upgrading each of them separately is tedious. Being able to perform it at once would be great. There is the option of npm-gui but it is external. I am sure it is part of the Improving the Developer Experience section but I haven't found it mentioned explicitly.
With micro-controllers going upscale and even performing AI (eg. TinyML on a Cortex-M), does Node-RED consider generating code source we could use to target those MCUs?
There are two great desktop applications for Node-RED: the already mentioned Node-RED Desktop self-contained and perfect for development; and the electron-node-red more suitable for packaging and distributing an application based on Node-RED. I'm using both.
+1 to that. SOMETIMES I want to do them individually, but most often I want to minimize operations interruption and upgrade everything, then reboot once. In those scenarios it would be very nice to be able to upgrade everything without a dozen clicks.
you can always drop to the command line and run npm update in your .node-red directory to update them all. (OK not as nice as via the browser - but certainly scriptable if necessary)
Just a side note - my fork of the natcl original electron version - https://github.com/dceejay/electron-node-red - is somewhat more updated and maintained...
Also note - this is aimed at a different use case to the Desktop Node-RED concept... aiming more for the runnable application with Dashboard rather than a standalone development app.
Excellent. I have only one question (maybe it's been explained before, I couldn't find any reference). Where does the name Node-Red come from? I mean, I get the fact that there are nodes and they are conveniently colored red, but usually there's a story.
st
Thanks for this nice overview!
There was one thing that was new for me:
We donāt plan at this stage to produce an out-of-the-box Distributed Node-RED version - but all the parts will be there to do so.
Although it makes sense to me, since I assume you want people to be able to support all kind of available technologies to implement this (Docker, Kubernetes, ...). Is that correct, or is there another reason?
But do you mean that there will be even no (supported) "reference" implementation (e.g. using Docker), that demonstrates how people can start implementing this for other technologies? Perhaps it is too early to think about this, but I was just wondering ...
It means exactly what it says - we don't plan at this stage to produce an out-of-the-box Distributed Node-RED version. The whole distributed node-red discussion is a long-term idea. It is not something you should expect to see in the next couple of months. It's simply too early to say exactly what we'll do in this space.
@rei_vilo@dceejay let's keep this thread on topic - any discussion around the plan we've published today. This isn't really the thread to propose or discuss minor enhancements to Node-RED - you can start a new thread anytime you want in #development:feature-requests for that.
Nick,
Thanks for the updates. Two take-aways for me:
I like the idea of the Flow Debugger with breakpoints.
Commercial adoption. Like to see more work with commercial entities developing nodes that serve as interfaces to their products or at least providing guidance for using existing nodes to accomplish the task. (I use allot of i2c connectivity, and it's been allot of work each step of the way to just get some of the contributed nodes to install/function and then to configure them for end devices).
As always, I appreciate the work you folks are doing. I wish I were more of a developer type, instead of a power end user. Regardless, I'm building fun stuff with Node-Red despite myself!
Concerning flow debuging
Maybe it makes sense to look into other graphical programming tools and check how they do it. Like Matlab/Simulink, Labview etc.
For example I like how they do it in Labview: when you enable Debug mode it slows down the simulation and starts to visualize messages, moving over the wires and activating different nodes. Itās magnificant.
