A question for those who may have experimented with building a chat tool on Bubble. What I am interested in is if you have and how was the performance?
Before I embark on building an chat tool I would like to know if anyone has had a good or bad experience with real-time/ synchronous performance when implementing a group chat on Bubble?
@gregjohnkeegan Yes I’ve done an ugly prototype. One catch, the server side is not Bubble, but can use Bubble’s API Connector and API workflow endpoints to communicate from the Bubble server.
For live charting, it might work by sending a signal via websocket, so the page knows when to retrieve the fresh data via database (workflow set RG data source, or some other method to override cache). Or just send the data via websocket.
Edit - live charting is more suited to SSE Server Side events than websockets. I’m assuming you mean charting using a Bubble source … external sources usually offer a javascript sdk.
The orders update in real time - would love to be able to generate my own unique view of this data. I’ve actually done it already with GET requests but it’s only useful if it can update via websocket and not via refresh.
Ha, on that note, I have a larger enterprise app which was built for me and uses Tornado websocket and has never missed a beat in the 3 years I have been using it.
It basically keeping the connection open between multiple groups (client browsers) and sending metrics such as live survey results, chat, activity (who is doing what and when) actions (such as being able to activate certain things in the client browser). It’s fairly complex in its requirements.
We’ve seen pubnub used in many/most video conferencing solutions because these solutions also need a real-time signaling layer (same as chat), so likely a good solution for real-time chat as well.