The Future of Cordova

1

What are every ones thoughts on the future road map for Cordova?

I am finding more and more Cordova plugins are not being regularly maintained or updated, sometimes requiring to find an alternative plug-in (if one exists), which in turn can cause having to spend hours and hours to re-write certain functions within the app.

Plug-in conflicts and build errors can be a headache, such as for the Cordova firebasex plug-in for analytics will not build on iOS (no matter what I’ve tried), so I’ve had to ditch it after spending way too much time on it.

Also, with many AI tools and free vibe code app builders now (Google have a few), could Cordova become obsolete?

A possible solution (while I understand it would be an immense amount of work) could be that NSBasic have their own set of fully maintained Cordova plugins, and, allow us to connect our own AI agents to AppStudio to be able to read and edit all the code, not just the current open code window, so it can understand the whole app, not just one function.

I have been trialling with Claude AI for writing certain code, which has helped with certain bits, but Claude cant access my code directly within AppStudio, this feature would be very beneficial.

As many services do not produce their own “official” Cordova plug-in, we have to rely on 3rd party plug-ins to bridge the gap, which can be slow at getting updates, conflict with other plug-ins, or full of bugs. Meaning, is native build and the official SDK’s the only choice left for app development going forward?

I’m interested in your thoughts.

2

I think there are a couple of different issues at play. First is having easy AI-generated code vs being able to support and maintain that code. Sure, we can get Claude to generate code, but can we read, understand, and fix that code? Personally, I think we’re a decade or more away from getting anywhere near 100% AI generated and maintained code, and by then, everything else will have changed so much that apps, etc., may be a relic of the past.

As for the plugin hell that you described, I think the best solution for that is a registry similar to what Joomla has for its plugins and extensions. Once a plugin falls out of date or gets enough downvotes, it gets removed.

As for the Cordova long-term viability, I find this handy little NPM chart to be useful.

3

Wow, a lot of topics here. Let me try to respond…

Obsolete Plugins: this is an issue with with all open source repositories, not just Cordova. Developers come up with a useful idea, put it on the web, and maintain it as long as they are able and interested in doing so.

I think you might have identified the solution. Since all the plugins are open source, you can fork them into a repository of your own. Then use Claude (or other AI tool) to help you come up with a new version. Since it’s open source, other users of the plugin could also pitch in.

AppStudio “adopting” a bunch of plugins isn’t practical. We ran some stats from VoltBuilder about which plugins were more commonly used: There are way more than you think.

AI Coding: We’ve using this ourselves and getting some very interesting results. For AI coding to work, it needs to see your entire project and understand how AppStudio projects work.

We’ve been uploading test projects to GitHub, then using GitHub Copilot on them. It actually works. However, most AppStudio users are not GitHub users: one of the features of AppStudio is that you just have to install it, and not a complete toolchain. It you’re not using GitHub already, it can be a bit daunting to get started with it.

Firebasex plug-in for analytics: Have you tried posting your question on the VoltBuilder forum?

4

Firebasex plug-in for analytics: yes, there are a lot of posts relating to a simular issue with cocopods.

5

Just checked on the Firebase plugin - give it another try. VoltBuilder has been updated.

6

ok I will do, but for now i just need to get the Android build into beta testing on Android.