Proposal to developers - a DP proof of concept

0 views
Skip to first unread message

Jon

unread,
Jan 25, 2008, 6:35:10 PM1/25/08
to DataPortability.Action.Implementation
Team,
I'm gauging the interest within the Implementation Action Group for
any developers who would be interested in building out a DP enabled
project. This project would, as you guessed, adhere specifically to
the DP Technical Blueprint and involve the entire DP technology stack.

The proof of concept would be something that all action groups could
be able to cite and allow us hands on experience working with the
entire technology stack.

Jon

Stephen Adkins (spadkins)

unread,
Jan 26, 2008, 10:02:05 AM1/26/08
to DataPortability.Action.Implementation
Hi,

At the SharedUniverse Project (http://www.shareduniverse.net), we
intend to integrate
a full Free Software stack of virtual world services, and we intend to
implement
every standard that DataPortability proposes.

We are starting with a PHP/Drupal core.
We plan to Javascriptify the UI with ExtJS-over-JQuery.
However, we envision ourselves primarily as integrators of disparate
Free Software packages
rather than authors of everything.

We are also in the early organizational stages, and additional
volunteers are needed.
If you are a Free Software developer and want to contribute to a real
implementation that
will track DP standards, this would be a good place.

That having been said, it wouldn't hurt to know all of the similar
Free Software projects/sites
and make sure we interoperate with all of them.

Stephen

richard.pendergast

unread,
Jan 26, 2008, 10:12:04 AM1/26/08
to DataPortability.Action.Implementation
jon. funny you should mention it. thats something ive been working
with several others on setting up for about a week now. i didnt want
to publicise it til the supporting infrastructure was in place, and
wed clarified the responsibilities for the implementation group first.
id love to discuss this further though - maybe skype? sounds like we
have very similar ideas.

my understanding so far (and this is something i am clarifying at the
moment) is that primarily the implementation action group is about the
support of existing projects, with new projects being something that
would typically be initiated externally. as the new project becomes a
reality though, then it becomes an interest for the implementation
group... so in theory its not too much of a stretch.

i guess my concern is that until we know what else is out there, it
would be very easy to step on an existing project, by starting
something new, getting the resources of dp behind it, and forcing the
existing project into redundancy (or, even worse parallel
development).

the two initiatives i was really looking to kick off tonight within
the implementation group discussions, were the filling out of the code
listing (the dp toolbelt), and the creation of an equivalent listing
for the capturing of knowledge relating to existing projects. i see
the separation of the two as important. the code listing is a listing
of code snippets, libraries, etc. that can be used to aid the
implementation of dp. the project listing would be a listing of
existing projects, the poeple involved, goals, achievements etc.

ive held off promoting either tonight, as ive had to clarify first
contribution guidelines a little first with the other dp action
groups. this has taken a couple of hours, and i have to sleep, so ill
do this tomorrow. this has worked to our advantage though, as i will
now be working with the members of the evangelism group getting the
process fleshed out and documented. id prefer others arent held up or
intimidated when looking to contribute. a simple set of guidleines
might encourage more activity.

Jon

unread,
Jan 28, 2008, 11:35:57 PM1/28/08
to DataPortability.Action.Implementation
My inquiry was with regards to launching a group project, open to all
developers within the DP Implementation Group.

J. Trent Adams

unread,
Jan 29, 2008, 8:45:36 AM1/29/08
to DataPortability.Action.Implementation

Jon -

This is just one guy's opinion (ie. mine), but I'd suggest we marshal
our collective resources around the documentation and bridge-building
aspects of the implementation project before trying to build anything
new.

Basically, I see a lot we need to be doing right now in (for lack of a
better term) paperwork land. Once we've papered over most problematic
areas, I think we'll be in a much more reasonable spot to kick up some
reference apps.

My $0.02,
Trent

Jacob Chapel

unread,
Jan 29, 2008, 8:49:37 AM1/29/08
to dataportabilityac...@googlegroups.com
At this point, I see no reason why anyone can't be working on a lot of
the research work currently needed. Like we need find all in
production or pre-production applications and plugins that are
available.

I will be more active in these tasks, just been swamped in the other
action groups getting things setup.

Jacob Chapel

Josh Patterson

unread,
Jan 29, 2008, 4:59:31 PM1/29/08
to DataPortability.Action.Implementation
Something that was an earlier effort, which got spun off, is WRFS.
It's more like a client side stack that discovers, translates, and
aggregates data from many different services. Me, Paul, Zef (and
others) have continued to work on that in the WRFS group. It's aimed
at addressing the secure use of our data via our own apps, and third
party apps as well. A few tech demos are being constructed right now,
you are more than welcome to take a look at those. Just an idea for
anyone who is interested.

Josh Patterson

On Jan 28, 11:35 pm, Jon <jon.cianciu...@gmail.com> wrote:
Reply all
Reply to author
Forward
0 new messages