This has a number of advantages over GPG. In general, it provides better default security without having to fumble to pick the right options with GPG.
For example:
In order to get sender authentication in GPG, you have to sign the message. GPG doesn't do this by default when encrypting a message, but you can choose to add a signature. But even if you do sign the message, GPG provides non-repudiation which means that you can no longer deny that you sent the message. Kestrel provides sender authentication while preserving deniability.
There are a number of other strong security guarantees that are provided by Kestrel's use of the noise protocol. There is more info on some of those properties here [1]
I built something similar in the past using TesseractOCR and Apache Tika and PyPDF2 / QPDF. The idea is sound. An API based OCR already exists in Apple / Microsoft / and Google so I am not sure this would be that useful. There would be no way for the user to trust that you are not taking the data you are OCR'ing and using it. If you can apply some type of one way encryption of the content and prove it via open source code (like Whisper Systems does for Signal) which seems like overkill and lots of effort for a free app.
I love this. I was trying to make an infinite indie playlist myself that would keep playing new things based on previous likes similar to thumbprint radio in Pandora. Im going to use this app right away to discover new music. Thanks a lot.
How are you loading the data for all of the public repositories? It looks very similar to the front page of https://hub.docker.com/. Assuming you are scraping / using the API to load all of that data?
I think this a solution in search of a problem. Using some open source text editors these days you can print to ebook formats (ePub, etc.) and both Kindle Direct Publishing (doesn't require ISBN even) and Blurb are two very popular routes to self-publish your works and already come baked in with print-on-demand services at no cost to the author.
Yes, this is precisely the problem we're trying to solve.
- Text Editors - with the exception of some specialised tools, text editors are seldom made for authors, they are either too much specialised or too little. Building a custom word processor precisely for authoring means we can fine tune the writing experience.
- Exports - Almost all of the default export options suck, they create unreadable messes of typography with zero consideration for making something beautiful. There is a reader on the other end of your books, it is our duty to give them the best reading experience. With our own typesetting engine we can create beautiful looking books at the click of a button.
- Selling - This is where most of the competition lives, you can publish on KDP but need to stay in their price bracket or they take 70% of your book price. Setting up blurb is not easy. Even if all this happens, you would still need to provide support and handle returns and refunds.
I believe Authors, Readers and the entire publishing landscape would benefit from a fully integrated platform, but only time will tell.
