On 19 November 2015 at 03:38, Richard Stallman <[email protected]> wrote: > [[[ To any NSA and FBI agents reading my email: please consider ]]] > [[[ whether defending the US Constitution against all enemies, ]]] > [[[ foreign or domestic, requires you to follow Snowden's example. ]]] > > "Complete control over all your data" is very abstract, and very > broad. Are we talking about the data that YOU are keeping? Or > personal data in the possesion of services you use? Or other > data about you that various systems have collected? >
Good point. That sentence wasnt as clear as it could be. What I had in mind was "user generated content". Ideally everyone will store much of their data on their own freedombox, but that's not a world we live in today, wrt free software solutions. So as we transition I think it's valuable to consider both data you store personally, and data you trust others to store for you. > > We should test these proposals against some specific cases to see if > what they would require would be sufficient for those cases. > Absolutely! One thing I'd like to see is more interop among free software solutions. Interop is hard, in the same way that "communication is hard', the reality being that we are short on resources and need to show a benefit for prioritizing work. Id like to argue that offering data freedom is a great way to increase the longevity of systems we build, by leveraging the network effect. One specific recommendation I have is to implement the mime-db project: https://github.com/jshttp/mime-db Github pages recently introduced this, and it works really well. It allows users to store files in any of 3000 different formats, taken from various standardizations efforts and user submitted formats. The gist of is is that from an extension, the data is served with a mime type that the user would like. This would be a major win for data freedom imho, because now we are putting the power in the hands of users, to create, inspect, improve and share data. > > > Looking at the specific points, some seem to be for cases where you > use a service to store your data in, while others seem to be meant > for services you use. > Agree, it's an attempt to match roughly what we see today in our systems. The point I think we are at is that we've taken the first step of allowing users to store limited styles of data. If we can make it open ended, and allow CRUD operations (generally we're not bad at buidling RESTful services), and we've moved forward. With my two further proposals of allowing access control and realtime updates the hope is to turn data freedom into a giant declarative data base on which a turing complete systems can be delivered on a mass scale (sorry, buzz words!). In short, let's try and improve data freedom and watch 1000 flowers bloom! I do appreciate some of this points are quite vague, hopefully I've given an idea on how we can progress, and I welcome discussion to refine and enable us, as a community, to take positive steps forward! > > -- > Dr Richard Stallman > President, Free Software Foundation (gnu.org, fsf.org) > Internet Hall-of-Famer (internethalloffame.org) > Skype: No way! See stallman.org/skype.html. > >
