Hey pals,

We don“t have to take literally the "Open" from OpenSocial. I believe
that is important to respect the privacy concerns and your approach
sounds nice to me and very reasonable.
Also, considering the Social class, I guess the viewer/owner are still
unclear. In the Twitter application
(http://estradacms.com/opensocial/twitter.xml) when I accessed my
profile I get my data from twitter. So I tried to access the Twitter
from a friend's profile and again I get my data. =)

Cheers,

Robson Eisinger.



On 11/4/07, Fabricio Zuardi <[EMAIL PROTECTED]> wrote:
>
> Hello all,
>
> I have a question about something that doesn't seems covered by the
> current version of the API (0.5) or at least not part of the online
> documentation.
>
> Since this api as far as I understand is a work in progress, and since
> this is an *open* api, it is natural to expect that improvements
> suggested by the community, after some discussions and an agreement
> can land into future revisions before we reach a good stable version
> that can be frozen (1.0 maybe).
>
> I believe that this mailing list is (or if not, should be) the
> main/authoritative place to discuss API improvements and decisions, as
> open as possible, following the good path of other standardization
> efforts in the open source community. So I will prefix this types of
> request with "API Council" in the subject to make filtering easy, I
> don't think we need another list for that just yet :)
>
> Enough talking, my first rant: anonymous or unlogged viewers. What
> should we expect from a container? Is this something we want
> OpenSocial API to cover? And if so, how?
>
> Context
> --------------------------
> For big profile-centric social networks today, like Myspace, Facebook,
> Orkut, Friendster, etc that are closed, and/or invitation-only, due to
> the nature of this business and privacy concerns, an account and being
> logged-in is required in order to view anything, so it is fair to
> assume that a gadget/app placed on someone's profile page will always
> have as the Viewer a known member of that network, so the current
> available checks isOwner and isViewer covers everything.
>
> However, other more public accessible and not so profile-centric
> social networks like Flickr, Deviant Art, Last.fm, LinkedIn to some
> degree and public networks built on top of the Ning platform, not
> necessarily requires login to access and view user pages, and so,
> application developers needs to know what to expect from the api under
> this circumstances, otherwise it will have to test on every network to
> see how that particular container have implemented the "anonymous
> viewer" case and make lots of checks in his code to be able to work
> cross-networks. And we don't want that.
>
> Use case
> ------------------------
> * I am an application developer and want to display the list of
> friends of the viewer in my application following the example of the
> Developer's Guide at
> http://code.google.com/apis/opensocial/docs/javascript/index.html
>
> * I also want to prepare my code to behave different in case the
> viewer doesn't have an account on that network, be it to display a
> message "sorry, but you need to be a member", or to fallback and show
> the owner's friends list for unlogged users.
>
> * Since there is not a clearly defined method to check if the viewer
> is logged, or an standard expected value for the
> newFetchPersonRequest, I have to go and test on all containers one by
> one and adapt my code for the different ways containers might treat
> unlogged users, for example, Ning as of today returns a Person of name
> 'Anonymous' and id 'xn_anonymous'. I am pretty sure this is not a
> standardized value because XN is a ning namespace, not a documented
> return value ;)
>
>
> Proposal
> ----------------------
> - Add a new method to the Person class
> http://code.google.com/apis/opensocial/docs/javascript/reference/opensocial.Person.html
> :
> Boolean isLoggedIn()
>
> - The anonymous user would still be a Person instance, which means
> that a dataResponse.get('viewer').getData(); would return an object
> and not false or null, the reasoning for that is that an anonymous
> user is still a person (jokes aside) and it might as well have a name
> like "Anonymous Coward" on slashdot, a generic avatar, and an id that
> is different depending on the network.
>
> - this is different than just checking if the list of friends is
> empty, because an app developer might want different behaviors for a
> registered user with zero friends and an unlogged user.
>
>
>
> What do you guys think? Does that make sense? Is this approach
> reasonable? I would love to hear your toughts on that.
>
>
> --
> Fabricio C Zuardi
> http://cchits.org
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"OpenSocial Developers" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/opensocial-api?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to