Thanks for the clarifications, Fabrice. I'm apparently way behind the curve on home automation, since I don't have any sensors generating data, but I do at least have a Squeezebox for streaming music. So I think I get this part:
> The FoxBox is the piece that would sit in your home network, gathering > data from sensors and sending them commands, along with managing the > storage part. It may also be able to display content on an attached > device, stream sound to a remote speaker, etc. So if I understand right, FoxBox is a really useful server for your home. When users buy a new laptop or tablet, they know to download Firefox and install it. When those users buy a new NAS or similar home server, hopefully they will pick one with FoxBox on it, or they will know to download FoxBox and install it. And then and lots of good things will happen because of the synergy between the browser, the server and the user's various devices. I'm still not sure I get this part: However, I'm fine with using > 3rd party apps to get visualization & analysis of my data. That means > pulling their code to run it on my data, instead of pushing my data to > them. Also, that means getting the code to run in a sandbox that doesn't > leak back to the 3rd party. It sounds like what you're describing is downloading software and installing it locally, just like we used to do before the Web. But you're also adding a sandbox so that these third party apps can't communicate over the internet (just like the apps in the days before the internet). Is the privacy sandbox the thing that is new here? (And is that what we have now with locally-installed packaged apps that have device storage permission?) Sandboxed apps like this presumably can't be supported by advertising. So an app ecosystem would have to involve users actually paying for their software, right? David On Tue, Dec 15, 2015 at 11:42 PM, Fabrice Desré <[email protected]> wrote: > Indeed, let me try to explain better what this is about... > > A web browser is also called a User Agent because we rely on it to > access documents and code from remote, untrusted sources. It's great > that you can trust your browser that nothing bad will happen when you > load a page that includes a random <script src=...> element. However, > your browser doesn't give you much control on the data that you produce > online. One could even argue that it facilitated the emergence of > massive silos of data that make little case of user control. > So, I believe that beyond the browser, regaining control of our online > activity calls for a new kind of User Agent. It will need to provide > guarantees that your data and the code processing it are loosely coupled > and under your control. It will also need to be designed so that we can > access our data from anywhere, and allow 3rd party to write apps just > like any other website (no store, no gatekeeper of any kind). > > Some use cases: > - I want to store the data from my activity tracker, either on my local > network or on a remote server I control. However, I'm fine with using > 3rd party apps to get visualization & analysis of my data. That means > pulling their code to run it on my data, instead of pushing my data to > them. Also, that means getting the code to run in a sandbox that doesn't > leak back to the 3rd party. > - Once in a while, I create photo collages to print them. I'd like to > use a service that does just that, but will not get access to the raw > data of my pictures. Knowing which camera I use, the exact date, > location etc. which is in the exif data is none of their business. > - If I had a "smart" thermostat, I would like all the historical data to > be stored in my database. And I want to be able to run algorithms on > that data from different providers, to select the best one for instance. > > The FoxBox is the piece that would sit in your home network, gathering > data from sensors and sending them commands, along with managing the > storage part. It may also be able to display content on an attached > device, stream sound to a remote speaker, etc. > We'll build a way to access your data from 3rd party sites, both from > the local network or outside of it. We'll let "background" processes > process the data. > > There are still many pieces to fully figure out, and many choices to > make. Comments and suggestions are welcome! > > Fabrice > > On 12/15/2015 09:59 AM, David Flanagan wrote: > > Fabrice, > > > > I'm intrigued by the concept of a "user agent for your data" and I love > > the name "FoxBox". But I didn't get to see your lightning talk, and I > > don't think I really understand what you are proposing. The slides are > > quite high-level and the etherpad is mostly implementation detail. I'm > > guessing that I'm not the only one who doesn't get it. Could you dive > > deeper into some of the use cases or user stories that you're thinking > > about for this? If I had a FoxBox in my house, what would I use it for, > > and how would I interact with it? > > > > David > > > > On Sat, Dec 12, 2015 at 3:15 PM, Fabrice Desré <[email protected] > > <mailto:[email protected]>> wrote: > > > > During the lightning talks session in Mozlando I presented an idea > > around the concept of "User Agent for your Data". The slides are > > available at https://people.mozilla.org/~fdesre/data_user_agent.pdf > > > > I wrote down some details on how we could implement this platform in > an > > etherpad: https://public.etherpad-mozilla.org/p/data-user-agent > > > > It's still very early, nothing is carved in stone, but I'd really > > appreciate if people interested by this effort could contribute to > this > > document. > > > > thanks! > > > > -- > > Fabrice Desré > > b2g team > > Mozilla Corporation > > _______________________________________________ > > dev-fxos mailing list > > [email protected] <mailto:[email protected]> > > https://lists.mozilla.org/listinfo/dev-fxos > > > > > > > -- > Fabrice Desré > b2g team > Mozilla Corporation >
_______________________________________________ dev-fxos mailing list [email protected] https://lists.mozilla.org/listinfo/dev-fxos
