Re: [Social-discuss] What I think GNU Social's structure should be

[Ted Smith]

On Sun, 2010-03-28 at 14:30 -0700, Jason Self wrote:
> Ted Smith <address@hidden> wrote ..
> > In all these discussions about what the optimal structure of GNU Social
> > would be, my foremost care has been freedom. I don't yet know what a
> > fully free network service would look like, but I think that it would
> > have to have the following properties:
> > 
> >       * Based on only free software (obviously)
> >       * Federated, so that any user can run their own node if they wish
> >       * NOT requiring or encouraging software as a service, or SaaS.
> >       * Users totally control who can see their data.
> 
> Social networking isn't SaaS. [1]
> 
Publication and communication isn't SaaS, and Stallman calls that social
networking. But in this case there is more - namely, managing your
social network (who can see what, etc.). Additionally, more could be
done with the social network than that.

> The model that Matt Lee presented already meets your criteria, and can be done
> with existing (and ubiquitous) tools.
> 
Correct me if I am wrong, but I believe Matt Lee is proposing a
monolithic PHP program. In my model, there is a distinction between the
core, transport modules, and the UI.

> It can be operated in a cheap web hosting environment, as Matt suggested, or 
> on
> one of those cheap wall wart servers like Eben Moglen mentioned [2]. Either 
> way
> the user is in control of their data.
> 
On a web hosting environment the user is not in control of their data -
the operator of the hosting service is. It certainly isn't safe to do
encryption on those machines. 

Additionally, using only PHP means that every time a user wants to have
their own node, they need to install, configure, secure, and maintain an
entire GLAMP stack + GNU Social + the interface (the browser, in this
case). That is an unnecessary and very burdensome level of complexity.

> One thing that I like about Matt's model is that it avoids the issue of what 
> to
> do one one of the nodes is offline: If my friend has unplugged their wall wart
> server and are transporting it to another location, I'll be able to get their
> updates once they come back online again, and their RSS/Atom/whatever feed
> becomes available again. The same could be said of thing that I publish.
> 
So if one of the nodes is offline, there's just no possibility to get
data? Active migration (as I'll call it) isn't exclusive to my
structural model; GNU Social could implement it in a monolithic program
(that operates in a federated p2p network).

signature.asc
Description: This is a digitally signed message part