Discourse Audio Message

I tried it, and it currently doesn’t work indeed. On paper, what you are achieving with your plugin should be doable with a theme component (assuming Voice Recorder Component does more or less the same thing), which is way easier to install.

Another example of a plugin that was rewritten as a theme component:


Thank you, I’ll look into it.


This Audio Message Recording plugin is working perfectly on my Discourse. Thank You @peter.be !!!

1 Like

FYI there is some discussion here about Plugin vs Theme Component…


It would have been better to PR a fix?


Here is a Vimeo demonstrating usage of this plugin… (and BTW it also works in mobile :clap:)


FYI this plugin breaks the tag groups page on my site. The entire list of tag groups disappears and you cannot modify any of the tag groups. Removed this plugin and the issue went away…

1 Like

I’ve slipped this over to dev while there’s still some discussion over the development aspects of it. :+1:


Guys, it’s great fun to have a go at writing a plugin no-doubt and a very useful learning experience, and even more fun seeing if ChatGPT can help you write it …

… but deploying a plugin to Production and formally sharing it with the public does put the onus on you to support it going forward.

Relying on ChatGPT is all well and good as a productivity assistant but really you need to understand every line of code yourself.

I wouldn’t have a big problem with any of that if we hadn’t already had a Theme Component covering this functional space (albeit not currently functioning correctly)

Deploying a plugin to take its place is a step backwards as you would always want to deploy something as a Theme Component if at all possible so that it can be used more widely by more kinds of installs, are less invasive, and much quicker to deploy and remove.

You need to have a very good reason to revert to using a Plugin (and that reason may exist but would be good to elaborate on and justify).

I still think as a community we should aim to maintain existing assets when they break rather than creating new ones (unless of course, like the Guest Gate Component, there’s a chance to create a Theme Component instead, which would be a worthwhile improvement)

To that end efforts would have been better directed at PRing to imho:

I suspect the main reason that no longer works is because the original maintainer is too busy on other projects, but there is nothing stopping anyone contributing to the repo to fix it, has anyone tried?

There’s a limit to the amount of time members of Pavilion can devote to the numerous public contributions we’ve made over the years and the community really needs to pick up some of the load.

/rant over


Perhaps because it works? And…

That never happened.

From dev point of view I totally understad you. But as an example of end users I would say plugin vs. component is not that big question, because we (or I am) upgrading constantly anyway, and all that matters is way to get job done.

We don’t need applications to record and upload audio. We have devices that can do it (anyway, almost because Apple still dislike MP3), but it is not very streamlined workflow. One button is, no matter if it comes from plugin or component.

And yes. I prefer components too.


Those who are hosted at lower price points are often restricted to Theme Components for customisation, and that’s almost certainly a significantly larger population, so by switching to a plugin you are cutting out a lot of sites.

Of course you and I don’t have this particular restriction.

1 Like

That’s true. Those are cases I don’t remember ever. But they have limited access anyway. So such situation is just another more (which one should avoid if possible, though),

1 Like

it is actuallly a very big deal since so many instances are hosted, and plugins are also further limited by hosting tiers. theme components have some limitations but also offer much more flexibility and ease of administration for those who are not self-hosted.

1 Like

maybe i am misunderstanding you (language issues) but i disagree - hosting is a very viable option and solution for many forums.

1 Like

Anyway this wasn’t meant to be a big pile on and I’m really pleased @peter.be you’ve got into this space and you are sharing your work, but perhaps you can consider these broader points :+1:


Robert, thank you for bringing up your concerns.

When I decided to code my own solution (since none of the existing ones work atm), I didn’t know about the pros and cons of theme components vs plugins. It would be great if the how-to on plugin development contained a disclaimer, maybe? (Developing Discourse Plugins - Part 1 - Create a basic plugin)

I really just needed a simple solution to facilitate the audio recording workflow. If you, as a veteran in this space, feel inspired to transform the plugin into a theme component, I’d be happy to assist you and answer questions about the source code of the plugin.

If you don’t, I might be able to come back to this eventually, I just can’t say when I’ll have the time :pray:


I have tried to repair the broken Theme Component version, but my skills are apparently not yet up to the task.

I was, however, able to fix and improve some things in the Plugin, so doing that was fun and of course I learned some things too, which is always the best reward!

(i.e. I submitted my very first PR on Github!)

I’ll keep at it, because I do agree about the advantages of Theme Component over a Plugin (and it’s always most preferable to have both options, because Plugins do still have some advantages over TCs!)


@davidkingham I cannot reproduce either the /tags or the /tag_groups pages breaking with this plugin enabled.

Is there any additional info you can share?


I’ve fixed the theme component.


And this plugin is broken. Ember-issue I reckon, but didn’t dig any deeper. It gave an error and refused to start so that’s it.

1 Like