Readwise? Unwise!

Someone recommended Readwise to me the other day, as a tool to revisit and learn from your ebook & article highlights. I was intrigued because so far I have resorted to converting all my ebooks to pdfs before reading and highlighting them, as this is the only way I can guarantee proper access to my highlights and notes later. But of course the conversion is cumbersome and sometimes imperfect. Unfortunately, using Readwise is unwise.

Here is the problem. Readwise allows you to import highlights from ebooks and articles (whether on your Kindle, Pocket, iBooks or more) and consolidate them in one place where you can review them, tag and organise them, search them, and even get automatically reminded of them if you want. Now this is a perfectly local problem. This means that an application on your laptop or an app on your phone could do the import and store the highlights in a local database, and offer all the tagging, searching and reminder functionality through the application or the app. You would buy such an application or app once, and perhaps use it forever.

Unfortunately, Readwise choose to solve this perfectly local problem using a centralised, cloud based, solution. Probably because it doesn’t want to sell an app once, but wants to offer a services that you subscribe to a and pay a monthly fee for. (Remember, technically there is no reason for such a cloud based solution.) Not only does this cost you more money. You also pay with your privacy: Readwise gets to collect and analyse your annotations and uses it to offer advertisers or publishers insights on how you interact with your books. And you lose autonomy: should Readwise go out of business, you lose all your annotations.

Now the only reason why a cloud based solution makes sense from a user perspective is when you own multiple devices, and would like to access and search your annotations from each of them. However, this does not mean Readwise itself needs to integrate a cloud in the app or service itself to offer such synchronisation. It would be equally possible to store the local database used by the local application or app on whatever cloud service you already use, and the system would work as desired (but under your own control). Moreover, it allows you to decide which cloud to use, which might be an extra secure, end to end encrypted one.

By the way: Readwise is unfortunately not alone in this. Scores of apps use the same trick to get you to subscribe to them for no reason. Some examples I came across: Camo (an app that turns your smartphone into a webcam; they don’t even use a cloud but force you to subscribe anyway), Any.do (todo and calendar), Evernote (note taking), etc. Again, for most or all of the functionality these apps offer a subscription is (technically) not required.

Only when (the main function) of such apps is to also collaborate in real time with others, it makes sense to integrate the cloud to implement this into the app itself. Tight integration within the app is necessary in this case because most cloud storage systems consider documents to be the smallest unit of storage, meaning that the only thing users can do is upload, overwrite or delete complete documents. For efficient concurrent editing more fine grained access is necessary, for example allowing several users to delete or insert records or lines within the same database or document at the same time.

It would certainly be interesting, and I think feasible, to build a generic cloud offering such a more fine grained access to the databases and documents it stores. Even more interesting would be to implement an end-to-end encrypted version of such a cloud for much more secure and privacy friendly collaboration! The way Cryptpad works (see this paper) might be a good starting point.