[OpenID] "declining OpenID usage"

David Recordon david at sixapart.com
Tue Apr 7 20:29:25 UTC 2009


Also on this note, as I said in a comment on your blog (http://blog.crowdvine.com/2009/04/02/declining-openid-usage/#comment-736 
) Tony, I think that you're in a great place to get some actual data  
about how the user experience effects usages of OpenID on your  
conference networks.  Something as simple as trying out RPX on one of  
your next conference sites and seeing if by using more recognizable  
brands and buttons to start sign in versus a box to enter a URL make a  
noticeable difference.  I would be happy to help make this happen.

--David

On Apr 5, 2009, at 12:51 PM, Chris Messina wrote:

> On Sat, Apr 4, 2009 at 2:56 PM, Tony Stubblebine  
> <tony at crowdvine.com> wrote:
> Chris, thanks for CC'ing me on this.
>
> And thanks for following up!
>
> 1. We're completely bought in to the idea of the open stack. In a lot
> of ways, CrowdVine is an anti-eyeballs business. We're judged on the
> number and strength of the connections we make for conference
> attendees. The open stack reduces barriers to making those
> connections.
>
> Good to hear! Would love to hear how we could help make the UX for  
> the Open Stack work better for you!
>
>
> 2. I was inspired to look at the numbers after seeing Eric Reis talk
> at Web2. He's pushing measurement so that you can have a conversation
> that's based on data rather than opinions. A lot of people have told
> me that our numbers are low because of our UI. That's an opinion. It's
> easy to find supporting arguments for that opinion, but it should be
> just as easy to prove the opinion with some numbers.
>
> I generally support this (speaking from real world numbers rather  
> than just perceptions).
>
> I think that we lack a lot of core metrics in the OpenID community  
> and so our conversations fishtail all over the place without being  
> anchored to a clear, distortion-free view of the world.
>
> While we know we're still early stage getting this technology off  
> the ground, there are enough implementations that COULD be sources  
> of data that it's a tragedy that we haven't organized ourselves  
> around making this potential resource available for inspection.
>
>
> 3. Nobody has shared a single bit of data with me, which makes it seem
> that nobody knows what they're talking about. If you're not measuring
> how are you going to know when you hit on a better design? Are there
> any OpenID consumers on this list? What is your adoption rate? Have
> any of your design changes made any bit of difference? If I switch to
> a better design am I going to move from 1.5% adoption to 1.75% or am I
> going to move to 10%?
>
> Janrain is one of the few services that has published numbers from  
> their work with various OpenID relying parties:
>
> http://blog.janrain.com/2009/01/openid-case-study-afi-begin.html
> http://blog.janrain.com/2009/01/why-websites-should-accept-multiple.html
>
> I wish more companies, including the one that I work for part-time,  
> provided more of this kind of information.
>
>
> 4. A lot of people have reacted to our numbers being low, but very few
> people have reacted to our numbers going down. We get to take regular
> snapshots of nearly identical communities and we've mostly seen a
> decline in usage. For example, we had two large conferences that were
> a year apart in the same city. The first year adoption was 8% and the
> second year adoption was 3%.
>
> I can't explain why your numbers are going down — but it would be  
> interesting if we were able to do a study or survey to find out why  
> — especially from folks who used OpenID in the past but didn't use  
> it again.
>
> I could imagine that audiences are becoming more mainstream and are  
> thus overall less familiar with OpenID, but that's just guessing. It  
> could also be that the value for using OpenID wasn't obvious, but  
> that's a separate issue that needs to be addressed.
>
> It is worth pointing to some of Janrain's case studies (again,  
> lacking other corroborating resources) that suggest that OpenID has  
> in fact increased user signups:
>
> http://www.janrain.com/openid/casestudies
>
> Examples: "Mixx observes a ten-fold increase in registrations via  
> OpenID", "PropertyMaps sees registrations increase by 200% after  
> deploying OpenID and today 25% of registrations occur via OpenID",  
> "37 signals now has 15% of it's users logging in with OpenID".
>
> It's also worth pointing out what's required and what's not; today,  
> most sites require email addresses, so that's what gravitated  
> towards, since it's the familiar. If you required OpenID, you'd  
> obviously see a different story, but you may also setup an  
> impenetrable barrier for some.
>
> Now that you've run the numbers, I'm curious to see if you'd be  
> willing to work with us to do measurements over time, using a  
> variety of OpenID-driven interfaces, from RPX to VidoopConnect to  
> other approaches, like the developing pop-up approach? As you said,  
> you're curious as to why the number went down; well, what if we took  
> steps to see if we could bring those numbers back up by tweaking the  
> interface or discoverability of OpenID?
>
> Chris
>
>
>
> --tony
>
> On Sat, Apr 4, 2009 at 10:13 AM, Chris Messina <chris.messina at gmail.com 
> > wrote:
> > It's a little like Wetpaint removing support for OpenID from their  
> site,
> > when the UI looked like this:
> > http://www.flickr.com/photos/factoryjoe/2478951850/
> > ...and was NEVER able to resolve my perfectly valid and otherwise  
> usable
> > OpenID URL.
> > That said, I think that adoption of OpenID largely comes down to  
> making a
> > better user experience and improve how we communicate the value of  
> using
> > OpenID.
> > While you make a valid point Johannes, I think that we've largely  
> failed so
> > far to provide a proper prescription for how to EFFECTIVELY  
> implement
> > OpenID. Even if you're compliant with the spec, as Crowdvine is,  
> their
> > implementation leaves much to be desired — the fault of which is not
> > entirely theirs.
> > Fortunately, we've recently approved the User Interface WG, the  
> first of its
> > kind for OpenID, and from this group, we should get a spec that  
> helps
> > implementors get more out of supporting the protocol.
> > That said, OpenID doesn't sell itself, and every site that  
> supports OpenID
> > today — for better or worse — is a sales opportunity that we  
> should be
> > capitalizing on to tell the story of OpenID.
> >
> > I have some ideas for how to address this but rather than be  
> reactionary in
> > this particular case, I think that we should look at the data that  
> Tony has
> > presented, consider it, think about what percentage of OpenID  
> usage would be
> > a good target and then find other sources of data that either  
> corroborate or
> > dispute his numbers. We unfortunately have been unsuccessful at  
> getting data
> > from the larger OPs (I'm looking squarely at Yahoo, Google, and  
> AOL) on
> > OpenID adoption or usage; on the flipside, Facebook has a great  
> story for
> > adoption of its Connect product with data to boot so we're  
> currently at a
> > data disadvantage to contrast Tony's findings with those from the  
> wider
> > community.
> > Chris
> >
> > On Fri, Apr 3, 2009 at 12:56 AM, Johannes Ernst
> > <jernst+openid.net at netmesh.us> wrote:
> >>
> >> See http://blog.crowdvine.com/2009/04/02/declining-openid-usage/
> >> I would like to propose a concerted effort to
> >> 1. identify negative OpenID experiences by existing adopters,  
> like this
> >> one
> >> 2. "crowd source", to the OpenID community, the analysis why they  
> may be
> >> seeing what they are seeing
> >> 3. engage with the site to improve their implementation.
> >> For example, when checking out their site, I noticed that the two  
> forms
> >> for creating an account (username/pass vs. OpenID) look almost  
> the same. See
> >> screenshots below. This does not communicate to the user at all  
> that there
> >> is any advantage, on that site, to use OpenID. It can clearly be  
> done much
> >> better, and most of us know how, but they apparently didn't... so  
> negative
> >> experiences are published, and that can't be a good thing.
> >> I'm afraid that unless we combat these kinds of things heads-on,  
> a body of
> >> "knowledge" will emerge that OpenID "isn't used".
> >> (It also may well be that the numbers are not good even for good
> >> implementations. In which case we might need to make substantial
> >> improvements to more than just our ability to teach sites how to  
> implement
> >> OpenID well. It would be useful to try to solicit those numbers,  
> too.)
> >> Anybody agree / disagree?
> >>
> >>
> >>
> >>
> >>
> >> Johannes Ernst
> >> NetMesh Inc.
> >>   http://netmesh.info/jernst
> >>
> >>
> >>
> >> _______________________________________________
> >> general mailing list
> >> general at openid.net
> >> http://openid.net/mailman/listinfo/general
> >>
> >
> >
> >
> > --
> > Chris Messina
> > Citizen-Participant &
> >  Open Web Advocate
> >
> > factoryjoe.com // diso-project.org // vidoop.com
> > This email is:   [ ] bloggable    [X] ask first   [ ] private
> >
>
>
>
> --
> CEO, CrowdVine
> 707.953.3868
> www.crowdvine.com
> blog.crowdvine.com
> www.twitter.com/tonystubblebine
>
>
>
> -- 
> Chris Messina
> Citizen-Participant &
>  Open Web Advocate
>
> factoryjoe.com // diso-project.org // vidoop.com
> This email is:   [ ] bloggable    [X] ask first   [ ] private
> _______________________________________________
> general mailing list
> general at openid.net
> http://openid.net/mailman/listinfo/general

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-general/attachments/20090407/3e6de8a8/attachment.htm>


More information about the general mailing list