Speaking as a daily TaskPaper user, this looks great! I totally get why you're doing the things you're doing with this, implementation-wise, and what you're able to achieve because of it.
Having said that, you know very well what a tough sell it is the more general purpose a tool is, and this is very general purpose. I hope having it out there spurs some ideas around how the novel features of this infrastructure can be the basis for some more concrete and easier to market solutions.
Has anything improved recently for RM2 around secure sync to a LAN? Would like to experiment with one for my psychiatrist wife to replace her paper chart notes, but anything hitting an external cloud service is out of the question for patient information. Sync via cable would be too obtrusive. And I gather handwriting recognition etc. is done via cloud.
It's just linux. It has ssh installed (it tells you the password in the "copyright and licenses" menu, and you can private keys on the device like normal). It connects to wifi. You can write a script that runs on the device and periodically syncs notes somewhere (I have this setup with restic for backups).
Handwriting recognition is done by cloud, and, I mean, unless you want to set up your own software for handwriting recognition (doesn't sound easy) you would need to give up that feature (I have)...
DIY handwriting recognition is actually really easy using Microsofts SaaS offerings. The results are astonishingly good too - it makes mincemeat of my awful handwriting.
You might be better off with an e-reader running full Android. I use Syncthing Fork, and it lets me do two-way sync between my tablet and a Macbook, which acts as a local server. No middleman cloud service required.
If you run syncthing on android you lose its power management. Also it's a horrible OS for long running services for other reasons, one of the largest being how huge apps are and how it quickly kills things when memory gets low.
I am using Syncthing Fork from F-droid, and you can configure it to respect Android's power saving features. Besides, you can turn sync off entirely if you choose.
Off topic, but one of my CS professors was part of the original Simula group. Had long sinced moved onto more formalized approaches, i.e., denotational semantics. On more than one occasion was heard to say "the only good object is a dead object."
On the other hand, there are so many opportunities for experimental design limiting the applicability of results in mental health experiments, that sample size and distribution doesn't even begin to touch on it. Bottom line: getting a meaningful result that is widely applicable is a horrendously complex undertaking.
Have met lots of people who are depressed and don't have a troubled past or present. Your descriptions of etiologies, how medications function and help, etc. are simplistic and ignorant at best. My opinion of your opinion is not high, but luckily that's not a good basis for judgment.
I'd agree in so far as it makes pulling scams (or innocently giving useless advice) far too easy:
"I had depression, and I tried X, and it worked. You've got depression, so you should try X." Extremely unlikely to be helpful, and often can be harmful.
Reading some of the comments is a good reminder that solid and meaningful mental health research is very tough and expensive, just based on the sheer variability, which can't be meaningfully captured by 99.9% of studies. Each needs to be seen as one piece in a much, much larger puzzle.
Having said that, you know very well what a tough sell it is the more general purpose a tool is, and this is very general purpose. I hope having it out there spurs some ideas around how the novel features of this infrastructure can be the basis for some more concrete and easier to market solutions.