1. If I need a browser to access them, that's very similar to "can't access them". Do you expect my programs to start browser sessions to access files? Will my file manager have to connect to that silly web server?
2. I can't effectively OCR them for many non-English language (at least - unless it's made some advances in OCR for some languages for which it's lacking). Also, if I do OCR anything, I then need to correct the mistakes, which takes a h-u-g-e amount of time even when the OCR works relatively well.
You seem to come at this useful tool with a very negative mindset.
- The scanned documents are stored in the filesystem in PDF format, so you can certainly access them without going through the web application.
- OCR doesn't have to be perfect, because it's just being used to locate a specific document when you need it. So if you have 1788 documents and you want to find your apartment lease, you search for "lease" and there's a very good chance that word will be found in the correct document. That's the whole point - storing documents with a low amount of manual effort, in a way that makes it very easy to find them when you need them.
2. I can't effectively OCR them for many non-English language (at least - unless it's made some advances in OCR for some languages for which it's lacking). Also, if I do OCR anything, I then need to correct the mistakes, which takes a h-u-g-e amount of time even when the OCR works relatively well.