[Freedombox-discuss] Software as Data, Transformation as a Service
Dr. Daniel Alexander Smith
ds at ecs.soton.ac.uk
Thu Sep 6 00:18:51 UTC 2012
On 14 Aug 2012, at 16:35, Nick M. Daly <nick.m.daly at gmail.com> wrote:
> In short: we have several groups of folks working on problems of
> standardized, distributed, and data-centric applications. Why aren't we
> cooperating, or at least communicating, about our problems and
> Lionel Dricot , Dr. Daniel Smith [1, 2], Michiel de Jong , Mike
> Macgirvin , and Markus Sabadello  seem to have come to similar
> conclusions: having a bunch of semi-interoperable applications that do
> the same thing but don't share their data is wasted effort and added
> complexity (it's "silly"). When applications can share their data store
> and give the privilege, responsibility, and complexity of storing their
> data to a separate data layer, writing applications becomes a lot easier
> and the applications themselves become more reliable, flexible, and
> under the user's control.
> Here, we have several communities with at least a few dozen people
> working in this problem space, so we should try to communicate and
> cooperate on our shared goals. In this world of limited resources and
> volunteered time, there's no greater sin than duplicated effort.
> I believe that the Unhosted, WebBox, and Friendica projects are furthest
> along in their work, and it'd be interesting to hear which pieces
> they've completed. We could then start getting ideas of how everything
> can interact, cross project. Some FreedomBox tools, like Exmachnia 
> (and, to a lesser extent, FreedomBuddy ), might also be useful here.
> I'd love to hear any thoughts you folks have on this subject.
> Thanks for your time,
> 0: http://ploum.net/post/freasy-future-for-gnome
> 1: https://github.com/danielsmith-eu/webbox
> 2: http://users.ecs.soton.ac.uk/~mvk/webbox-pim.pdf
> 3: http://unhosted.org/manifesto.html
> 4: http://friendica.com/node/24
> 5: http://projectdanube.org/
> 6: Exmachnia is a tool to allow maintainable, system-wide automated
> configuration from any front-end.
> 7: FreedomBuddy is location-ignorant location tool: it lets friends keep
> communicating despite constant motion through both IP- and
> meat-space, without relying on third parties.
Hi Nick, Everyone,
I am indeed the main developer on WebBox. Since the original e-mail I've looked into the other software, and have started on developing compatibility with RemoteStorage. Specifically I've implemented WebDAV (almost entirely, more than is required by RemoteStorage), and WebFinger. I'm almost there with OAuth2, but some important deadlines have stalled this, but it should be there around mid-October.
I agree that fragmentation and duplicated effort is a bad thing, and I want to do what I can to prevent it on the WebBox front.
We have some particular offerings, in terms of the methods of WebBox-to-WebBox sharing communication that seem to be unique, and I will want to push those ideas into some form of standard way to doing machine to machine sharing (all using existing protocols of course).
Freedombox is definitely a driver for us, and has been heavily referenced in our meetings, although we're not as au fait with the hardware (dream plug?) as the mailing list is - we're not involved with any hardware in terms of our own remit, so I'm happy to push to whatever makes sense, although I should say that I do have a couple of Raspberry Pis here that I've been enjoying playing with.
We're almost at a point where we will want to tackle things like WebBox communication through NAT, using changing IPs and NAT-to-NAT broadband type things - so for us, any help in that area will be welcome. If I can avoid burning our own cycles on that kind of development, that would be great, as it's not really an area we're particularly into (and, selfishly, not an area I'm confident in submitting research papers into! I'm doing this under my academic research hat at the moment.)
In conclusion, thanks for the update, I completely agree that we do need better communications. I for one have subscribed to the freedombox list.
I wouldn't mind a new list/organisation that for just for "us" though - perhaps the RWW group as Melvin Carvalho suggested, would be a good place to start? The WebBox history aligns with this quite well, as we did start WebBox in TimBL's office at W3C/CSAIL.
Sorry for the wall of text! I'll keep an eye on this list in the mean time, and I encourage any ideas of how we can merge our efforts (and I'll try to reply in a more timely manner!).
Dr. Daniel Alexander Smith
Scalable Semantic Web Interfaces
Web and Internet Science Research Group
Electronics and Computer Science
University of Southampton
More information about the Freedombox-discuss