Social Network Portability Group

[Chris Saad]

So the conversation on the Social Network Portability google group is
going a bit crazy..

http://groups.google.com/group/social-network-portability/browse_thread/thread/be1cec4290ffc6c6

The challenge is very simple. Write an import tool to convert 1 type
of personal data from 1 proprietary silo into the open standard's
listed and list it on the wiki.

The point is that the result is open source and usable by all.

Any specific projects to do anything more complicated with the data is
beyond the scope of the challenge. If we can't get this simple thing
right, then the rest doesn't matter.

Chris

[Danny Ayers]

On Dec 21, 12:29 am, Chris Saad <chris.s...@gmail.com> wrote:
> So the conversation on the Social Network Portability google group is
> going a bit crazy..
>

> http://groups.google.com/group/social-network-portability/browse_thre...

>
> The challenge is very simple. Write an import tool to convert 1 type
> of personal data from 1 proprietary silo into the open standard's
> listed and list it on the wiki.
>
> The point is that the result is open source and usable by all.
>
> Any specific projects to do anything more complicated with the data is
> beyond the scope of the challenge. If we can't get this simple thing
> right, then the rest doesn't matter.

I appreciate your intention here Chris - in fact if you point me to a
silo of your choice which offers enough of an interface that useful
data can somehow be pulled out, I'll have a go at opening it up over
the holiday. (A list of potential targets would be good on a Wiki
somewhere).

But there is a problem with the open standards listed. How, for
example, would I express the following information so it would be
portable across systems :

I often shop here:
http://www.thomann.de/gb/index.html
and here:
http://www.fine-tools.com/index.htm
and my Amazon Wishlist ID is 31J4DB064WSNV
and my birthday is 14th Jan (hence the significance of the above ;-)

Cramming this into hCard+XFN+APML+RSS would be like carrying milk in a
coal bucket. If you can't represent the stuff in a portable fashion,
it's a non-starter.

Cheers,
Danny.

[Chris Saad]

Danny I think that that kind information is great, however containers
for it (such as RDF) can be included over time.

The key right now is to get people interested in the idea that
containers and transports exist without over-complicating the issue
with semantic questions.

Also, right now I am just interested in the fundamental/structural
data such as name, age, friends, interests.

Much like RSS, you only need to capture the core data - and then you
can embed structure in the unstructured content.

This is not the most technically elegant way, but it is the the web
itself evolved and how the mainstream picks up things and starts to
play with them.

Also Baker kindly started a page about possible targets - feel free to
add to it:
http://groups.google.com/group/graphsync/web/list-of-services

Chris