plone and OpenSocial API Specification
_robert
are there any plans/ideas/projects concerning the implementation of
googles OpenSocial API Specification into plone.
I would be very interested to hear about them.
I have been looking trough the messages of this group but found
nothing.
I have not spent a lot of time checking the OpenSocial API.
I *think* implementing any social networking features in plone without
trying to have it "compatible" with OpenSocial API would be badly
wrong.
robert
Christian Scholz
we discussed this briefly but decided to not pursue it right now (if I
got this right that is ;-) ).
The main reason is probably that we first need some social structure
for opensocial to work,
e.g. if Plone has no idea about what a contact is OpenSocial does not
make that much sense.
Then it depends what you want to do with OpenSocial with Plone. It's
mostly useful if you are
a big social networking site and want to extend your functionality by
providing people to install
applications on your system.
While I think this might be nice in the long run and it would also be
nice to implement some of
the proposals which are floating around now on the OpenSocial
developers list (REST API etc.)
I think this is a lot of work and should be done incrementally.
At least I don't yet see the big benefit for most sites I would
implement. But this might differ
of course from case to case.
What I also see though is that a lot of confusion is still around what
OpenSocial actually
is. It does not really live up to it's name as it does not directly
allow to connect social networking
sites as this might get you in security trouble (and legal). So for
now it's mostly a tool
for allowing more intelligent gadgets to be run on your site.
Another question is also how to best implement it. One way would be to
use the existing Shindig
code and run it as a server (I think that should be possible). Your
site might then provide the
REST APIs for retrieving and storing the data. But I don't know the
code so I might be wrong.
The problem with this is also that it's not as nicely integrated as it
could be as you'd have to
run a Java server next to your plone site.
Better integration would mean to port the Shindig code (which mainly
is what the Google Gadget
Container is) to Python. This might be some work though.
I might be all wrong with the above as OpenSocial seems to be a fast
moving target right now
regarding what standards they use and how it's setup (REST is quite
new, OAuth support in there is quite new,
lots of discussions about how to handle presence etc.).
So my idea would be to implement the base functionality of a social
network and then look what
OpenSocial is doing and making a plan on how to integrate it.
BTW, "compatible" mainly means that we have to provide the APIs. But I
think these
can be provided in any case as a layer on top of whatever we have.
-- Christian
PS: Somebody please correct me if I am wrong about something above :-)
Dylan Jay
1) we're aiming to have smaller modular components that can be used in an
average Plone site, rather than make a new facebook
2) There are 2 ways to be open social compatible. What you're referring to
is being an open social container which lets other peoples social apps sit
inside Plone. IMO its more useful to more people to help Plone content be
used in other peoples opensocial sites... ie be compatible with the
opensocial gadget api. So that would probably come first.
3) There's no hurry. I don't think what we're doing is conceptually
incompatible like Christian says. And introducing it now would add too much
complexity. But sure its possible in the future.
Patrick
> 2) There are 2 ways to be open social compatible. What you're referring to
> is being an open social container which lets other peoples social apps sit
> inside Plone. IMO its more useful to more people to help Plone content be
> used in other peoples opensocial sites... ie be compatible with the
> opensocial gadget api. So that would probably come first.
search for... you guessed it "plone and opensocial" :-) ... definitely
like to emphasize the fact that Plone needs to start to be more
connected with the rest of the world, if you really want a future for
this. The whole silo idea that I feel Plone is based upon is (in my
opinion) horrendously outdated. If you publish content, and then...
well then nothing.. then what's the use ? I need to get that content
out there, let other people contribute potentially.. let others take
ownership of it.. and let the whole world see it. How can I do that if
there's no way to plug in my content into other social networks,
widgets etc.. Yeah.. if my content is a simple document.. great.. but
really.. who *cares* about documents these days..
Dylan Jay
So what api's are there to help publish content outside of Plone?
Opensocial is one but its not obvious how to use that in a generic way. We
could make a portlet or a portlet manager into an opensocial gadget perhaps?
But if Plone is so unfriendly to other api's, what are they?
Dylan Jay
Technical Solutions Manager, Pretaweb.com
Skype:dylan_jay P:+61.2.99552830
> -----Original Message-----
> From: plone-socia...@googlegroups.com [mailto:plone-social-
> netwo...@googlegroups.com] On Behalf Of Patrick
> Sent: Thursday, 26 June 2008 2:39 AM
> To: Plone Social Networking
> Subject: Re: plone and OpenSocial API Specification
>
>
>