UX - IDEATION - 3D - VR
Original Concepts and Proposals for Solving Complex Interaction Problems in Virtual Reality
VR Tutorial Functionality
“Tutor-BOT” or “People BOT” would be and actual in-world BOT that the user would interact with. This proposal is just dealing with explaining functions of the people app. Maybe there could be a few different BOT types, each tutoring a specific feature set of Hifi. The BOT could be used and positioned at the discretion of the domain admin.
Since the BOT has a physical 3D presence it can demonstrate all the functions of the people app using itself instead of another user’s avatar. The user would initiate the the tutorial by clicking the BOT. Maybe the user is prompted to click it. The user might be required to be within a set distance from the BOT.
VR Tutor Bot Interaction
The BOT would work in conjunction with the user’s tablet . The BOT would explain how to open the tablet or instruct the user to press the “people” button. Interaction would ideally include prerecorded voice narration (BOT character voice) but could also work with just text. A graphical “pointer” object (example: large, bold, colored arrow) would be used to focus the user’s attention on the specific button or control the BOT is explaining. There are probably some real technical challenges with this approach. It requires that both the BOT and the tablet be clearly in view and at the right size and proportion. There would be many specific interaction details to work out.
Tutorial Flow
The structure of the tutorial and user flow should ideally be made as simple as possible: (1) BOT explains function, (2) the pointer focuses the user’s attention (tablet) , (3) the user does the action (tablet), (4) the user sees/hears the result (in world), (Repeat 1) the BOT explains the next function. The sequence is repeated until all functions are explained or the user breaks off the tutorial session. This Approach would track the user’s action, the BOT’s action, what happens on the tablet and what happens in-world.
Personal Space Bubble
The following proposal can be viewed as a comprehensive safety system of interactive features however individual features could be implemented independently.
Broad   Considerations   and   Objectives
 A.   Each   user   should   feel   safe   and   have   some   degree   of   control   of   interactions.
 B.   When   examining   potentially   aggressive   interactions   the   system   should   be   fair,   balanced   and avoid   rushed   and/or   unnecessary   judgements.
 C.   Admins   are   also   users   and   have   rights   to   make   decisions   and   set   parameters   within   their respective   domains.
 D.   The   designed   solution   should   ideally   be   as   simple   as   possible.
 E.   Since   we   probably   can’t   envision   every   possible   use   case   we   should   anticipate   adding additional   rules   and   functionality   as   it   grows. 
Determination   of   “Bad   Actor”   Behavior
 There   needs   to   be   a   (simple)   set   of   rules   to   determine   “bad   actor”   status. 
Test: Which avatar was moving towards the other?
Test:   How   many   times   has   the   collision   occurred?
 There   should   also   be   tests   to   determine   if   the   contact   was   accidental. 
Test:   Were   both   avatars   in   motion?
 Test:   Were   one   or   both   in   the   process   of   arriving   in   the   domain?
 Test:   Were   both   users   in   conversation?   (probably   not   easy   to   determine) 
With a few simple tests hopefully we can get to 80% Accuracy
Bad Actor Escalation
 If the collision/touch was clearly caused by one of the users and after the initial auto hiding or the user the user continues the behavior then they get a series of messages. Such as:
 Alert 1: “You Seem to be getting awfully close to another user.”
 Alert 2: “You violated their bubble space again!”
 Alert 3: “Please review the good behavior guidelines.”
Space Bubble modal
If   there   have   been   multiple   collisions   in   a   short   time   but   neither   user   is   deemed   to   be   a   bad actor   then   user   is   shown   a   modal   dialog   box.   The   modal   could   be   something   like:   “(name   of other   user)   keeps   bumping   into   you”
             Choice   A:   Hide   and   Mute   this   user   (and   track)
             Choice   B:   Make   them   an   Associate   (Handshake)
             Choice   C:   Add   them   to   my   Friends   List. 
Usually modals force the user to make a decision before continuing. Alternately we can continue to use “Tablet needs your attention” and the user would deal with it there. Users already in My Friends List would not trigger any kind of messaging. There should also be a way for a user to disregard the dialog box. (close [X])
Time   Out    (Penalty   Box)
 When   the   user   gets   the   third   warning   they   are   teleported   to   a   designated   location   within   the domain,   facing   a   large   sign   with   good   conduct   rules   written   out. Some   bad   actors   may   just   be   immature   or   ignorant   of   VR   norms   and   need   to   be   educated   or reminded.   The   sign   could   also   explain   the   escalation/kick   policy   of   the   specific   domain. 
Tracking   Bad   Actor
 Keeping   track   of   number   of   times   issued   timeout   by   the   system. Keeping   Track   of   number   of   times   hidden   and   muted   by   another   user 
Panic   Button
 There   may   be   unforeseen   use   cases   such   as   a   group   ganging   up   on   a   single   user   in   which   the user   feels   unsafe   and   overwhelmed   despite   the   other   safety   protocols.   The   Panic   button   could teleport   them   to   a   predetermined   safe   location   or   it   could   mute   everyone   and   set   everyone invisible. 
Admin   Control
 Domain   Admin   could   set   threshold   limit   (timeouts,   hides)   for   possible   auto   kick.
 Domain   Admin   could   set   the   domain   up   with   no   limit   of   contact,   no   warning,   no   escalation,   no kick.   A   user   in   this   domain   could   still   retain   the   ability   to   hide/mute   another   and   still   have   panic button. 
Additional   User   Controls
 User   could   set   their   own   personal   limits.
 A.   Set   Threshold   Limit   (how   many   collisions   before   triggering   hide/mute   modal) B.   Set   Size   of   Bubble   (as   a   %   increase   of   current   avatar   scale)
 C.   Set   Visibility   of   Bubble 
Comfort Level Slider (concept) [cautious——()————————carefree] Sliding back and forth invisibly adjusts all the other safety parameters.
Notes of some ideas for redesigning the people app:
Analysis and Redesign of People App
Redesign   of   the   people   app   is   a   big   challenge.   The   existing   people   app   combines   several different   sets   of   functionality   in   a   small   space.   In   general   it   probably   has   too   many   disparate functions   and   the   UI   is   confusing. What   do   we   really   need   it   to   do?
1.   Set   visibility   and   volume   of   other   users.
2. Locate an individual within a group or within the domain
3. Manage/Search friends and connections
What   are   some   of   the   existing   problems?
 A.   The   tablet   blocks   some   portion   of   view   of   other   users.
B. UI is confusing: buttons versus status indicators.
C. Which functions can be separated or regrouped?
Sketch : Connections Tab Showing Filtered List of Avatars/BOTS
Revised People App Concept: 3 Tabs
Connections : Has many filters to display any/all avatars or BOTS Availability : Allows user to set personal parameters and display name. Domain Map : 2D overhead view of domain to help locate others.
Me   (and   my   Avatar)   App
 Instead   of   the   “Availability”   tab   there   could   instead   be   a   “Me”   app   separate   from   “people”   app. The   “me”   app   could   might   include   some   functionality   of   the      avatar   settings   in  Menu_Avatar . The   user   could   control   avatar   settings,   adjust   their   availability,   set   their   display   name   etc. 
Sketch : Avatar Selection “Magic Ring”
Magic Ring
 In-world   the   user   could   simply   click   (select)   an   avatar   and   a   ring   would   pop   up   around   that person   allowing   the   user   to   mute   or   hide   them.   The   ring   would   be   color   coded   green   to   indicate “friend”   or   blue:   “connection”   or   clear   for   no   connection.   It   would   also   dynamically   indicate   the other   user’s   audio   level.   Activating   the   people   app   would   display   all   avatar’s   rings   in-world. 
