I have one, and I find the writing part nearly perfect. But there are a couple of downsides mentioned in the article that mean I barely use it:
- Searching and organisation is really difficult. I wish everything would be OCRed transparently and you could instantly search for it (while keeping the original graphics)
- It's unfortunate that the OCR runs on their cloud. It would be really perfect if it was a web app that you could install on your own servers, if you have hightened security requirements for example
- In the UI, there is an on screen keyboard and you have to press keys. Why can't you just write in the text fields?
I think most of these problems come down to the fact that the OCR is some "secret sauce" provided by a third party. I wonder if there is any viable free handwriting OCR one could use instead to build a better experience (open source, source available, or even just some research papers)?
- Searching and organisation is really difficult. I wish everything would be OCRed transparently and you could instantly search for it (while keeping the original graphics)
- It's unfortunate that the OCR runs on their cloud. It would be really perfect if it was a web app that you could install on your own servers, if you have hightened security requirements for example
- In the UI, there is an on screen keyboard and you have to press keys. Why can't you just write in the text fields?
I think most of these problems come down to the fact that the OCR is some "secret sauce" provided by a third party. I wonder if there is any viable free handwriting OCR one could use instead to build a better experience (open source, source available, or even just some research papers)?