social-network-portability-faq

From Microformats Wiki
Revision as of 11:48, 17 September 2007 by AndyMabbett (talk | contribs) ('')
Jump to navigation Jump to search

Social Network Portability FAQ

This page documents frequently asked questions (FAQ) about social network portability.

Do microformats or any standard make all my social info portable

Do microformats or any standard make all my social info portable across social network sites?

  • No, nor is it a goal to do so. The hCard microformat can represent the 80% of common information across social network sites like name, birthday, URLs, interests etc., and just being able to share that common 80% is quite valuable. In addition, many social networking and social media sites add their own distinct value for the problems they solve, and it's unrealistic to expect a standard (any stanadard) to try to represent/share all that info in a 100% interoperable matter.

Is the Web itself a social network

Isn't the Web itself a social network minus some proprietary software?

Precisely, the Web (plus a few open data formats) is itself a social network platform.

The key is, what are the minimal simple open data formats to describe the 80/20 of that social internetwork.

When XFN was introduced back in 2003, the choices of relationship values were based on real world uses of what relationships people were already publishing in their visible blog rolls on their blogs. The hypothesis being that blogs + blogrolls already presented us with a giant distributed social network (in contrast to the dominant social silo of the time, Friendster).

With hCard, one can easily capture, publish and share the 80/20 of common profile information across social networks.

With the combination of these two, most of the problem is solved.

We'll know what problems to solve next once more sites have implemented hCard+XFN and we start to take such support for granted, much as we take syndication support for granted in web sites today.

Does OpenID solve the re-enter your personal profile problem

Does OpenID solve the re-enter your personal profile info problem?

  • No it does not. OpenID is fundamentally about proving to one site that you own or control another particular URL. Nothing more. All the profile stuff is extra and even then the specific property set is unspecified in OpenID. That's where hCard comes in. hCard specifies a vocabulary of personal profile info (name, email, birthday, URL etc.) based on industry standard vCard. And in fact that's all you need to solve the "re-enter all your personal info" problem for public sites - no need to authenticate public URLs via OpenID, just read them and parse their hCard(s).

Does OpenID create a globally unique identifier for people

Does OpenID provide people with a way to create a globally unique identifier for themselves?

  • OpenID does not create a globally unique identifier for a person. It is URL (by way of DNS) that creates a globally unique identifier (whether or a person or other purpose), not OpenID. What OpenID does do is add authentication on top of having a globally unique URL that is itself a globally unique identifier, that is, OpenID is a way of asserting that you "own" or otherwise control a particular URL to another service/site. The essential quality of being a globally unique identifier is present in a URL even without OpenID. Thus, no, it is not OpenID that creates a globally unique identifier, rather, simply using a URL creates a globally unique identifier. A URL by itself can be used as a basis for portable social networks, since it provides a way of referring to a single person that makes sense across the whole Web. And as such, people can (and will) create multiple URLs that represent them, so a method of enabling people to tie their URLs together (e.g. XFN's rel="me") becomes important. However, by adding authentication, OpenID can enhance existing globally unique identifier URLs for people, by permitting authenticated access to (perhaps more information in) profiles and social networks represented at those URLs using microformats.