Knowledge From Crowds - Better with Institutions + Algorithms

   Knowledge  from  Crowds  –  Be1er  with  Ins6tu6ons  +  Algorithms        h1p://goo.gl/q1DNL      Shaun  Abrahamson      @shaunabe      

 Tap  the  crowd  for  learning  

 Where  did  Stanley  come  from?  

 Crowd  data  recycled  into  knowledge  

 Collec6ve  contribu6ons  into  holis6c  understanding  

 Compe6ng  for  be1er  predic6ons  

 Our  signals  into  rankings  

   New  type  of  ins6tu6on  to  deliver  mobile  services  

Labo

r  

 Capita

l      

 Assets  

Influ

ence  

 Data  

 The  crowd  as  the  gateway  to  critical  resources  

Sharing, Commenting, Reviewing  Open Innovation, Co-Creation, Micro Tasks  

Labo

r*  

 Capita

l      

 Assets  

Influ

ence  

 Data  

“Big Data”  

 *  very  often  “knowledge”  work  

Labo

r  /  Kn

owledge  Work      

 Capita

l      

 Assets  

Influ

ence  

 Data  

Institutions                        vs                        Algorithms  

4 PERSPECTIVES!!

OUTCOMES!PEOPLE!TOOLS!

ORGANIZATION!

OUTCOMES!

Ronald  Coase  “Given  that  produc6on  could  be  carried  on  without  any  organiza6on  that  is,  firms  at  all,  why  and  under  what  condi6ons  should  we  expect  firms  to  emerge?”  About  75  years  ago      

   Why  do  we  organize  work  in  a  certain  way?  

   New  Ins6tu6ons  +  tools  

! Sales!

! Marketing! Operations! Production!R + D!

   New  data  +  algorithms  

! Sales!

! Marketing! Operations! Production!R + D!

CASE:!What went wrong at Wikipedia?!

Source: Mary Meeker presentation at All things D. !    The  end  of  all  other  encyclopedia  business  models  

   Why  aren’t  editors  staying?  

Active E

ditors R

eten

tion

afte

r 1 y

ear (

%)

PEOPLE!!

FINDING AND MOTIVATING THE MOST IMPORTANT KNOWLEDGE RESOURCES!

   What  skills  might  you  tap  into?  

175 million people on !

   Why  will  they  par6cipate?  

Money!

Experience!Attention!

Good!

Stuff!

   What  data  by-­‐products  might  you  rely  on?  

   Why  did  you  start  contribu6ng  to  Wikipedia?  

   What  kind  of  work  environment  do  you  want  ?  

ORGANIZATION!!

WHAT INSTITUTITIONS ARE CRITICAL TO BENEFIT

FROM CROWD LABOR + INFLUENCE!

Elinor  Ostrom  “It's  a  problem,  it's  just  not  necessarily  a  tragedy  ...  The  problem  is  that  people  can  overuse  [a  shared  resource],  it  can  be  destroyed,  and  it  is  a  big  challenge  to  figure  out  how  to  avoid  that.”  About  2  years  ago      

   Organizing  to  resolve  the  “Tragedy  of  Commons”  

   Who  owns  what?  Brand  vs  IP  vs  Confiden6ality  

   Cheap  access  to  dispute  resolu6on  

Collec6ve  choice  processes  

Sanc6on  bad  behavior  -­‐  Don’t  feed  the  trolls  

   Nes6ng  to  scale  

   How  Wikipedia  community  sees  itself  

The  community's  role,  as  some  kind  of  nebulous  science-­‐fic6on  super-­‐en6ty,  is  to:          +  Organize  and  edit  individual  pages        +  Structure  naviga6on  between  pages        +  Resolve  conflict  between  individual  members        +  Re-­‐engineer  itself  -­‐-­‐  crea6ng  rules  and  pa1erns  of          

 behavior    There  are  other  roles  J    Source  -­‐  h1p://meta.wikimedia.org/wiki/The_Wikipedia_Community    

TOOLS!!!

COLLECTING DATA AND CREATING NEW UNDERSTANDING!

   Making  it  easier  to  contribute  

   Understanding  individual  contribu6ons  

My Klout! My Giving (Crowdtwist)!

My Creative Impact(Jovoto)!

   Understanding  collec6ve  health  and  performance  

   Rocket  Science  vs  People  Science?  

 Making  sense  of  all  that  data  

   Wikimedia  founda6on’s  focus  People  +  Tools  

CASE: EDX.ORG!!

NEW EDUCATIONAL INSTITUTIONS !

+ !DATA TO GET SMARTER

ABOUT EDUCATION!

   Content  +  Community  =  Learning  

 Represents  about  40  years  worth  of  classes  at  MIT  

155,000 registered!23,000 tried the first problem set!9,000 passed the midterm!7,157 passed the course!

   Ricardo  +  Arthur  doing  “online  learning”  

   Community  not  just  content  

“One of the best things about 6.002x was the community built by the students themselves. The atmosphere was great: people shared their enthusiasm and knowledge, and lended a hand to those like me who didn’t have the basics for the course.” - Arthur Amaral, 18 years old, Brazil Source: http://blog.edx.org/

   Ins6tu6ons  -­‐  Nes6ng  +  Collec6ve  Choice  

Anant  Agarwal  “We  can  watch  how  many  a1empts  students  made  before  they  got  an  exercise  right,  and  if  they  got  it  wrong,  what  they  used  to  try  to  find  a  solu6on.  Did  they  go  to  the  textbook,  go  back  and  watch  the  video,  go  to  the  forum  and  post  a  ques6on?”  About  1  month  ago      

   Data  to  learn  how  to  teach  

CASE: GIFFGAFF [TELEFONICA]!

!NEW INSTITUTIONS!

TO BENEFIT FROM THE KNOWLEDGE OF !

CUSTOMERS!

   Social  produc6on  for  a  complex  service  

   Homepage  hints  at  how  this  works  

   What  tasks  can  be  performed?  

 Who  is  doing  what  on  GiffGaff  

Value Created!

Income/Expenses!

! Sales!

! Marketing! Operations! Production!R + D!

 From  sales  +  support  to  new  app  development  

 Encouraging  Par6cipa6on  +  Rewarding  Behavior  

 Self  policing  mechanisms  

   Gaming  the  system  -­‐  posts  from  users  who  you  suspect  are  abusing  the  payback  system  by  using  mul6ple  accounts  to  give  themselves  solu6ons  or  kudos.    Tou0ng  for  SIMS/Kudos  -­‐  posts  which  are  ac6vely  asking  for  kudos  or  solu6ons,  it  is  fine  to  have  this  in  your  signature  but  not  to  ask  in  a  post/topic.    Incorrect  Accepted  Solu0ons  -­‐  if  you  spot  an  accepted  solu6on  which  is  incorrect  or  if  a  user  has  accepted  one  of  their  own  responses  as  a  solu6on  unjus6fiably.    Incorrect  Tags  -­‐  If  you  see  that  a  post  has  been  tagged  with  an  irrelevant  or  inappropriate  tag.    Inappropriate  Content  -­‐  Posts  which  are  disrespecmul  to  other  users,  profanity,  adver6sing,  naming  and  shaming  and  generally  causing  discord  or  disharmony  on  the  forum.  

How  is  this  growing?  Also  NPS  =  73  (Apple  =  79)  

Labo

r  /  Kn

owledge  Work      

 Capita

l      

 Assets  

Influ

ence  

 Data  

Institutions                        vs                        Algorithms  

Anant  Agarwal  “We  can  watch  how  many  a1empts  students  made  before  they  got  an  exercise  right,  and  if  they  got  it  wrong,  what  they  used  to  try  to  find  a  solu6on.  Did  they  go  to  the  textbook,  go  back  and  watch  the  video,  go  to  the  forum  and  post  a  ques6on?”  About  1  month  ago      

 Data  +  Algorithms  for  Knowledge  Management  

Elinor  Ostrom  “It's  a  problem,  it's  just  not  necessarily  a  tragedy  ...  The  problem  is  that  people  can  overuse  [a  shared  resource],  it  can  be  destroyed,  and  it  is  a  big  challenge  to  figure  out  how  to  avoid  that.”  About  2  years  ago      

   Community  +  Institutions  for  Knowledge  Management  

   Knowledge  from  Crowds  –  Be1er  with  Ins6tu6ons  +  Algorithms        h1p://goo.gl/q1DNL      Shaun  Abrahamson      @shaunabe