Comments on: Some challenges in current DataPortability trends

By: Shreyas Doshi

Shreyas Doshi — Sun, 17 Feb 2008 07:58:20 +0000

Hi Chris, I disagree with your contention that the “Sign in with a Yahoo! ID” experience on Plaxo is antithetical to the promise of OpenID. First off, Plaxo has ensured that this doesn’t really preclude users from using a non-Yahoo! OpenID. In fact, you will notice that the familiar OpenID textbox is currently *above* the Yahoo! button. Hence, I don’t see how this is creating a “mess of tightly coupled vendor relationships”. Secondly, as you and the previous commenter have also noted, the button experience removes the complexity with using OpenID. We have really emphasized that we would like to enable mass adoption for OpenID and as has been stated in many places elsewhere, requiring users to remember and type in URLs as their identifiers has been a major blocker to mass adoption thus far. There isn’t much benefit in refusing to adapt user experiences at the expense of mass adoption, is there?

I do agree with your point that user education is paramount. On that front, I believe we at Yahoo! do a better job of educating users about OpenID than just about any OpenID Provider today. We are actively thinking about pursuing other alternatives that make the OpenID experience easier and safer for users. What you see on Plaxo and other OpenID RPs right now is a good start, but I have a feeling we (the OpenID community) are going to evolve much further than that.

By: Khürt Williams

Khürt Williams — Thu, 14 Feb 2008 23:05:04 +0000

Chirs,
The Yahoo Login thing does conern me. I do hope OpenID does not become another microsoft Passport. I can understand Yahoo and Plaxo wanting to me things easier of their users. Asking users to use and existing ID ( which is what OpenID is all about anyway ) to login is easier that explaining what OpenID is. The user may already have an OpenID but just not know it.

By: victor

victor — Tue, 12 Feb 2008 17:11:11 +0000

Hi,
Just want to share my experience on OpenSocial and Social Graph api.
First, you can really write the same simple OpenSocial application for different social networks. I have tried with hi5 and Orkut and it works : excepted the version problem that will be solved when everybody will use the last 0.7 version, you can right once and put it everywhere (if you do not use specific extension . And moreover, my application try to solve a dataportability goal : open and free you social graph and your friends list. I have to had other export features and security policy but I think the main idea is there.
About Google Graph API : yes, with a Social Graph navigator you can really discover some information people perhaps don’t want to share, specific code to inform people about XFN tags is a good idea. We need to be able to decide if the link will be or not tagged.

By: Nick Lothian

Nick Lothian — Mon, 11 Feb 2008 06:14:05 +0000

“Presumably these exist to differentiate each network and encourage developers to write enhanced apps for the environment.”

This makes it sound like this is a bad thing. No two social networks are alike, so of course each one will have features they want to offer that others don’t.

As some of those features become more common I would expect they would push down into the standardized API, but ATM that isn’t appropriate.

I don’t think that comparing it to browser hell is fair at all. Perhaps something more like SQL or RSS/Atom extensions is fairer – you get decent functionality without extensions, but for platform specific extensions you need to use platform specific code. As time passes, some of those platform enhancements become standardized.