Open dceejay opened 3 years ago
In the original Design note there are three use cases highlighted.
To do the full IDE experience properly is a big piece of work. There are a number of existing projects that provide a very minimal Electron-wrapped version of Node-RED. There is also Saito-san's project that goes further and does provide a more IDE like experience.
My concern is we would want to make sure we do a proper job of the IDE. Is there value in doing yet another version of the minimal Electron install? I understand some users would feel better using something provided by the "Node-RED Project" itself, and that isn't something we should ignore.
I think there is a huge value to the community to having an "official" IDE from the project. But the question is how to get there. This project needs a vision for what it wants to achieve. Just wrapping Node-RED in Electron isn't good enough.
The suggestion here is to initially focus on the first two use cases. There is huge value in making it easier for users to install Node-RED. This is not Node-RED Desktop. This is about creating node-red-installer
- a single file the users can download to install Node-RED.
However, I'm not sure how much work that would really save. Electron doesn't provide a way to customise the install process. On OSX, you get a DMG that just lets you add the node-red
app to your applications folder - that's it. You can then run node-red
as an application - an electron is designed to be run with a Desktop window.
The suggestion I made on our call this week was to not build the custom UI part - so the user can install the electron-packaged version of Node-RED, but then uses their browser to access it. Having thought about that approach some more, I see a number of problems:
So do we continue with the Node-RED Desktop plan to provide some sort of native Desktop application user experience? If so, there's a lot still to do to get from PR #1 into something we can release. At the very least, we need more UX design work to get that first-run experience right.
Create a node red installer is a great idea, because is necessary in industrial environment. Are common computers in the field that is not allowed to access to internet to avoid cyber-attacks, so turn difficult the installation of the node red in field. Of course, is possibly make the download of the archives from npn in a computer with internet access and paste these files in the computer target in the field. But is necessary download the archives in the same OS and CPU to turn its possible. It's turned the node-red less accessible to beginner users like me for example. so i think that the offline installation functionality is really helpful.
Rather than duplicate functionality of the existing public Node-RED desktop project (Saito-san), It is suggested that initially we could focus this effort on creating a native platform installer to help users get both nodejs and Node-RED installed and configured correctly.
This could include configuration screens so help the user set up settings... such as security, projects, themes and could include useful introduction docs and links like the youtube introduction videos etc.
This is just an initial thought - place for discussion below. Feel free to assign to others. Everyone free to contribute.