But what about my initial plan to turn my #documentcamera into a #scanner and #photocopier ?

Initially, I was frustrated because I expected it meant hacking #scanservjs which would be a rocky road for me.

Instead, I did the following:
- using #ustreamer I created a live stream from document camera. Of course, I created a systemd service

- with some help from chatGPT I set up a simple web GUI with three buttons:

* Download current image
* Download as PDF (with text recognition)
* Print

So, basically, the whole magic required to turn #scanservjs into a #photocopier means:

- have a local #printer accessible over the network or directly attached to the #raspberrypi
- #imagemagick to convert the scanned image
- a new pipepline for scanservjs (similar to mine above) added to /etc/scanservjs/config.local.js

I am done with my project. Sort of.

What did not work:
- Letting #sane see my document camera with the proper #backend for #v4l because the latter is simply alpha software and development stopped long time ago

- As a direct consequence, I could not integrate the camera into #scanservjs without major changes in this software. A lack of knowledge kept me from doing so.