[tor-bugs] #21221 [User Experience]: perform usability testing for the mobile security slider

Tor Bug Tracker & Wiki blackhole at torproject.org
Fri Jan 13 20:37:04 UTC 2017


#21221: perform usability testing for the mobile security slider
---------------------------------+--------------------------------
     Reporter:  lnl              |      Owner:  lnl
         Type:  task             |     Status:  new
     Priority:  Medium           |  Milestone:
    Component:  User Experience  |    Version:
     Severity:  Normal           |   Keywords:  orfox, tor browser
Actual Points:                   |  Parent ID:
       Points:                   |   Reviewer:
      Sponsor:                   |
---------------------------------+--------------------------------
 = Background =

 Tor is broadly aiming to 1) prioritize mobile users and 2) make the
 desktop version of Tor browser and Tor browser mobile (currently orfox,
 but will be rebranded) consistent. The security slider allows the user to
 choose their security level while using tor browser. Since this is not on
 the mobile application yet, adding this feature accomplishes both of the
 aforementioned goals. While adding the feature to the mobile version, we
 made some changes from the desktop version to improve the interface and
 make the UI more mobile-appropriate. If you are curious about how and why
 we made the changes to the security slider you can look at the
 [https://trac.torproject.org/projects/tor/wiki/doc/UX/OrfoxSecuritySlider
 project page].

 = Scope =

 The purpose of this test is to see if the security slider is usable.
 Specifically, the metrics for success are:
 * Control: the user understands how to change their settings.
 * Communication: the user understands what each setting does (literally
 understand--“to make the setting from safe to safer, I slide this button
 over”, not that they understand what each setting means or how it is
 technically implemented).
 * Comprehension: the user knows how the settings will affect their
 browsing (again, this along the lines of “this is what I want,” “this
 makes this feature go away and I do/don’t like that” not that they
 understand how JS is a vector for exploits or even is aware of the
 difference between http and https websites).
 * Decision: the user is able to make the decision that they want.

 The ideal case is that the user will see the interface, and be able to
 say, something along the lines of: “when I move the slider, the javascript
 turns off, which will break some functionality. I don’t want that because
 I visit a lot of sites using Tor as my everyday browser.” | “when I toggle
 this, the browser goes into secure mode, which is the bare minimum of
 functionality to operate. It’s conservative and breaks a lot, but I’m okay
 with that because I want to feel safe.”

 We are not testing: if the Tor browser mobile is easy to find on the app
 store, if people are able to install it on their phone, if it is easy to
 use, or if the security slider is easy to find in the menu.

 = Methodology =

 We are running a small-scale, short, qualitative, open-ended, qualitative
 user test.
 * Small-scale: we are testing with 5 users. Read about
 [https://www.nngroup.com/articles/how-many-test-users/ when it is
 appropriate] if you are curious.
 * Short: aim for a maximum of 5 minute interview. The purpose of the
 interview is to get their impression of the UI (does the user take a
 second before they can explain how to use it? Are they confused about the
 settings?) rather than their understanding (can they explain why we used a
 horizontal slider versus a vertical one? Do they understand what turning
 off javascript does?). The longer the interview, the more it tends to lean
 towards their understanding, which is not what we want.
 * Open-ended: users are free to answer the question in their own words.
 This is to allow the users to speak more freely and to communicate
 naturally. (the opposite would be closed ended, and would involve asking
 the user to rank/rate/label the interface, “how hard is it from 1 to 10?.
 This works better for larger groups, but not small groups.)
 * Qualitative: we are focusing on user attitude and comprehension (do they
 like it, will they use it, are they able to use it). We are not concerned
 with user behavior (how many people use which setting, how many clicks,
 how many seconds) for this phase in testing.

 = Materials =

 What we will give to our partners as background and instructions:
 https://drive.google.com/open?id=1aE1txcRdJ9u5Der2-k9i-
 eq8mzvyPyAdmrPGlVKyPxo

 Interview questions and feedback form:
 https://drive.google.com/open?id
 =1yYCLZtZuvK_0EICCiXbxQTHFJUhRHJ_rJgdMO5AR-uA

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/21221>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list