ULX Teleconference 2010-03-10
ULX Telecon 2010-03-10
Logistics
- Time: 16:00-17:00 Eastern
- Skype: +9900827042954214
- US Dial-In: +1-201-793-9022
- Room Code: 295-4214
Attendees
- Scott Cantor
- Debbie Bucci
- Paul Trevithick
- Valeska O'Leary
- John Bradley
Regrets
Agenda
0) Roll Call
1) Web Mockup Next Steps
In our call of two weeks ago we agreed to make the following changes to the NIH Scenario Web Mockup. Kevin should be starting on these in the next few days:
- After you click a button in the pop up window we need to launch a real browser (so there is chrome around the window, etc).
- Need to add a "Click here to login" hover text over the Connect button
- Change the magnifying glass to some other icon (rationale: we're not searching, we're selecting)
- Add an "x" / close button to the pop up
- Add a "PubMed Password" button to the mockup
- Add a PIV button to the mockup
- Making the Popup design more refined, polished, Apple-like
Paul: Hiccup with contract, changes should be getting done shortly. Changes proposed over last calls summarized above.
Scott: Popup vs full window issue needs discussion, usability/technical testing, accessibility and IdP expectations are factors.
Valeska: re: Connect button, we can design for user behavior, or change user behavior, but tooltip certainly makes sense.
Some joint discussion of search box icon. Instead of changing the button, we need to move the magnifying glass inside the box a la Apple.com's search
Paul: re: offering password as option, should be identified with the app name rather than just labeling as "local account"
Discussion about how to characterize NIH account/login options, Debbie prefers it refer to "NIH login" and leave it up to NIH login site to further elaborate. Much confusion over how to capture "local" option when "local" is complex. Consensus is to farm out both PIV and password to Government Credential, name TBD.
Valeska: attending Information Architecture summit, good place to present mockup if we can polish it up. Very design/UI centric audience. Paul: let's discuss further under item #3 on agenda.
2) Recap of RSA Conversations
Paul showed our NIH Scenario Web Mockup to a bunch of folks who came by the OASIS booth at RSA:
- Eric Sachs (Google) at RSA
- Eric talked about Google's usability research results.
- He said that the optimal RP UX was dependent on the percentage of users coming to the RP site that had federated logins. If the percentage is over 60% then the popup/Nascar is worse than just asking them to type in an email address, figuring out their IdP [ulx:how?] and then redirecting them
- He said that their experience was that 4 was the maximum number of Nascar buttons that should be presented
Various discussion of Eric's feedback, noting the problems with Google steering people back to Google and hijacking (Scott's words) domains.
- Debbie Bucci and others associated with NIH site
- Agreed that a "PubMed Password" button and a "PIV card" button should/could be added to the popup
- Folks from CA (CA SiteMinder is the code behind the NIH RP site)
- One person (forget their name) suggested that we should make a YouTube video of the ULX login to gather reactions. Someone else said we should "ask folks to try to break it (on YouTube)".
Some discussion on YouTube idea, Valeska noted Flickr can host video and seems to engender sound feedback.
-
- The CA developers said it would be great if there was an open source library that implemented the ULX popup (they wouldn't want to use Janrain for this)
- Would be nice if Janrain implemented our popup
Scott noted it could well be something the Shibboleth discovery work eventually provides.
3) Usability Testing
- Assuming Kevin completes the changes mentioned above, and assuming we think we're ready to take this for a spin, how should we expose this to mere mortal users?
Paul noted we thought we might have large companies with usability labs potentially offering test resources. Valeska noted the upcoming summit is a different audience, not users, but professional UI designers, and would be valuable feedback.
Scott suggested that user testing would require the mockup to get "finished" and fleshed out so that all the pieces function in a real world way so that users don't run into dead ends or "this would have happened now" placeholders.
Valeska explained some of the procedures that a real usability lab would go through in running a test.
Most felt we should at least start reaching out to possible assistance on this to find out what we would need, in parallel with finishing up the mockup work.
4) New Wiki Page
- Inputs to the Selection UI is new wiki page to catalog the kinds and sources of input to the RP UI Popup. Scott Cantor to populate.
Paul refreshed Scott's memory, will try to work on this soon.