I prefer to write the name in full aka `jbarlow83/ocrmypdf-alpine` and I'd also suggest to document this because:
* if you use `docker tag` this AFAIK only tags the currently downloaded (=pulled) version of that image
* in case a new update comes out, the new one will not be pulled automatically and one would have to pull and tag the image locally, again
* This `docker tag` command is easily overlooked, if users just run `docker run ocrmypdf` this may or may not work, depending on how it is resolved.
Also, AFAIK if one could get Docker to register https://hub.docker.com/ocrmypdf then this would suddenly be used instead of your image (currently `podman pull docker.io/ocrmypdf` returns a 404 for me, though)
* It is more common to write at least the user namespace there and the project, to prevent such errors.
Also, default [Docker has many shortcuts for this and e.g. assumes Docker-Hub is always being used](https://stackoverflow.com/questions/37861791/how-are-docker-image-names-parsed). Podman usually does not, that's why I personally prefer to use the very full and clear `docker.io/jbarlow83/ocrmypdf-alpine:latest` e.g. for alpine. This makes it not only clear which version is used, but also where it is pulled from (should one have configured different Docker registries).
I've fiddled/struggled with this by myself, by getting a permission error like this one:
```shell
OutputFileAccessError: Output file location (./output.pdf) is not a writable file.
```
I've loosely followed and found https://github.com/containers/podlet?tab=readme-ov-file#in-a-container and explained the required flags in a similar way, but adapted for this tool (it likely won't be used so much on system files).
I've tested it and it works fine for me. The same issue may be on Docker rootless, but I guess people will get that and I cannot test it here.
