On 3 March 2010 20:40, Matija Šuklje
<address@hidden> wrote:
Hullo,
(as on the Libre.FM list) I wonder what the goals of the GNU Social project
are. I've gone through the mailing list's archives and haven't found much
tangible then "trying to be something new".
In the past I've given a bit of thought about what I would like to see in a
social networking thingamajigger, namely:
* tighter integration with the desktop and applications — APIs anyone?
* open standard technology like W3C's RDFa, FOAF[], SIOC[] ...maybe even
NEPOMUK[]
* easy and better controlled access (e.g. from your desktop)
* less bloat — e.g. why reinvent the wheel with albums, IM and e-mail, if we
have perfectly usable protocols and apps for that already! (also Facebook apps
= big no no)
* lotsa lotsa more control over my data
* smarter ontology
This is possibly a more detailed way of saying, give everything a URI and you can scale with the Web.
That means desktop and web should be 99% the same model, you only need NAT busting for the other 1% (which isnt easy)
We had a telecon with timbl on the social web incubator group today. I think this is absolutely spot on.
http://www.w3.org/2010/Talks/0303-socialcloud-tbl/
Not everything is built out yet, but it's moving forward. I think if you want to build yet another facebook clone, stick with elgg, but if you want to build something new, something revolutionary. The blueprint is there. It's harder work, but greater benefit long term.
Personally, I'd love to see it all happen on my desktop with a possible fall-
back on a website — an analogy would be how you can use your favourite e-mail
client to handle your messages over IMAP, but when you don't have access to
that you can still use the webmail.
Yes, why not some UPnP / DynDns solution also
BTW, have you read Steven Pemberton's[6] interview[7] and his talk warning
about "Walled Gardens"[8] and how to avoid it? Well, he points out some very
big flaws of current "Web 2.0" networking sites and how it could be solved. His
idea of a "personal website" that would hold all the content of a person,
which would be then aggregated to specialised websites (e.g. networking site,
photo albums site, music sharing site etc.) makes a lot of sense to me.
Havent read but makes sense, also look at
http://smob.me/ some of us are starting to make personal websites that have social features and are interconnected
Especially if we can have the desktop automatically push data to it and pull
data from others' personal websites. ...just imagine if you wanted to write an
e-mail to e.g. your best friend's brother's girlfriend you could just type
into your e-mail client "best friend -> brother -> girlfriend" and it would
check through the personal websites their FOAF entries etc. to provide you
automatically with the appropriate e-mail address :]
One of a million possible uses. The thing that TimBL said today was:
People have written all these applications on your phone, easy to use, no need to login, and they do what you want. What about making the same style of application for your computer (PC)?
I think he has a point. He said after the first few are built they will be able to inter connect and will be a virtuous cycle. Im not saying ignore web 2.0, im a fan of it, and it has it's place, but if you want to do something new here, and it seems you do, think web scale, which is essentially as simple as giving everything meaningful a URI and allowing them to link together. :)
...just my 0.0146616816948904 € ;)
Cheers,
Matija
-.-.-
[1] Resource Description Framework in attributes — http://rdfa.info
[2] Friend of a Friend — http://www.foaf-project.org
[3] Semantically-Interlinked Online Communities — http://sioc-project.org
[4] Networked Environment for Personalized, Ontology-based Management of
Unified Knowledge [NEPOMUK] is probably the most advanced and the only viable
ontology system for a semantic and social desktop which KDE[4] is nice proof
for (and AFAIK Gnome started using it as well) —
http://nepomuk.semanticdesktop.org
[5] http://nepomuk.kde.org
[6] Steven Pemberton is one of the old cats of W3C and the internet itself —
http://homepages.cwi.nl/~steven/
[7] http://dot.kde.org/2009/11/02/walled-gardens-semantic-data-and-open-web-
interview-steven-pemberton
[8] http://www.w3.org/2009/Talks/10-23-steven-openweb
--
gsm: +386 41 849 552
www: http://matija.suklje.name
xmpp: address@hidden