
Or I'll be tailing a log file or watching something in a database while using the app.

When I'm testing/debugging, one monitor might be the web app I'm working on, the other is the code I'm stepping through. Usually that monitor is split and also has chat visible. When I'm coding, I generally have code on one, and the other has a browser with documentation, notes, issue tracker, or just used for searches etc. Whereas if you use ssh you can be fairly certain that it's been battle tested by a huge number of people who have a lot more time and resources than yourself. I actually think this is better because for a very small open source project like barrier, that might literally be developed by one person, the workload and time/effort to be ABSOLUTELY CERTAIN you've implemented the crypto libraries correctly is a lot of work and worry. In which the VNC daemon on the remote machine only listens on its own localhost, and I use ssh to form the tunnel then use the vnc client on my workstation to connect to localhost:5902 to access it.

An example of the very tiny shell script that I use for VNC-over-SSH to a remote machine. I've only used it between macos and linux machines, so that's easy. If I had to guess, no, barrier is much like VNC in that it's expected you have a ssh wrapper set up with public/private key authentication.
