14_Relationships_Trust

toc Class 14 May 13, 2010 = Future of email: Ron Chung =

Does email still play a role in personal communications today, with the emergence of the social networking sites?
Advantages: Disadvantages:
 * Archived Internet communications, with persistent data storage and search capability
 * Emails can be split into folders for information organization
 * Activities are stimulated via email notifications, such as Groupon purchases.
 * Accumulate information from all sites rather than visit them individually.
 * Lacks audio, video (Skype), and social (Facebook, LinkedIn) attributes.
 * Emails are a continuous stream, and there is less sense of priority.
 * Emails should be sorted via relevance, urgency, and automatically stored in folders.

Start-ups to assist the email management
[|Xobni]: A Microsoft Outlook plugin to help searching and organizing the inbox. A quick demo for Xobni features: media type="youtube" key="9YhpdKa-NgY" height="385" width="480"

[|Seriosity]: Introduce virtual currency for senders to weigh their emails and for receivers to prioritize the importance. A short demo on how Attent can help you: http://www.seriosity.com/attent_demo.html

[|Rapportive]: A simple social CRM built into Gmail.

= Quora =

Engagement in Quora

 * Content linked to identity allows for better answers**
 * Public identity allows transparency, which in most cases increase the credibility of the answer
 * Users are now held accountable for content they post, meaning inflammatory or irrelevant responses are implicitly discouraged
 * Can make question/answer more of a discussion, rather than a one-way flow of information


 * Social network; discovery through following people**
 * Publicly available identities drive discovery through following people
 * Since interesting people follow interesting topics, Quora encourages discovering new content by following users who are interesting or share the same interests as you
 * A friend recommending a topic to you is more effective than a website recommending a topic
 * Initially, bootstrapping your interests using the interests of your friends is more effective than finding topics yourself


 * Is Quora a discussion site or a reference site?**
 * Quora wants to be a reference, but currently uses a discussion-based interface
 * Seems to lean toward discussion, as replies are linked to identities and therefore encourage directed answers and the mindset that an answer is an opinion, rather than a reference
 * Also due to the fact that most questions on Quora tend toward the opinion type ("Why is _ so popular", "How did become so successful", "What is the best __"); this means Quora can provide answers to questions that aren't purely factual and would not be found on sites such as Wikipedia
 * Lots of unanswered questions currently means a lot of sifting to find an answer, whereas a reference should be quick and direct
 * Check out the topic "How is Quora different from X?" in Quora to see how others are thinking


 * 90/9/1 phenomenon somewhat limits the variety of content**
 * 90% never contributes; 9% contributes a little, and 1% accounts for almost all the actions
 * More details about the [|90-9-1 rule] and a closely related concept, the [|power law] distribution
 * A select group of users seem to drive Quora (co-founders and friends); they have a lot of friends and answer a lot of questions
 * The high quality of answers comes at the cost of variety; a question may have only one or two real opinions
 * Popular notion that wisdom is in crowds; is it that the crowd as a whole has wisdom, or that individuals in the crowd have wisdom? Quora seems to be more of the latter for now


 * Quora encourages a "learning" attitude instead of a "look-up" attitude**
 * Wikipedia hole - inter-related topics, questions, and user networks encourage the Wikipedia hole mindset, where one question indirectly leads to many other questions (and lots of time spent)
 * Services such as Yahoo Answers and Aardvark do not support following topics, which means it is much harder to discovery related content
 * Following topics allow users to discover content they would not have thought to search for


 * Endorsements can bias selected responses**
 * Endorsements on Quora allow users to endorse other users and answers
 * Don't seem to be used all that much
 * Perception that popular users can use endorsements to drive bias of content (popular user upvotes an answer, their friends upvote an answer, etc)

How is Quora Going to Scale?

 * Currently primarily concentrated in Silicon Valley**
 * Quora is currently a closed beta; its user base is primarily composed of co-founders and various degrees of their friends
 * Much of the user base tends to be Silicon Valley, tech, entrepeneurship types
 * Content expertise unbalanced - huge amounts of tech and startup expertise vs very little expertise on other non-professional interests


 * Can Quora expand like Facebook?**
 * Tight networks with loose ties in between networks facilitate effective expansion; Facebook used this model, where universities represented very tight networks and friendships between people at different schools represented loose ties
 * Imitating this model may encourage quick growth, but Quora doesn't have natural tight groupings as Facebook does
 * Topics may represent the best choice for tight groupings, but a particular topic represents a soft grouping (a topic is related closely to several other topics, which are related closely to other topics, etc) and so may not as effective a model as Facebook's universities were


 * Can geography act as a proxy for interest in startups?**
 * Can Quora find geographically-based tech-savvy and intellectual populations in which to concentrate expansion efforts? (New York, Boston, etc)
 * Quora's current users mostly joined because a friend invited them, so spreading by word-of-mouth (and correspondingly, geographically to some extent) is a proven model
 * Geographically relevant content, such as industry or city life, will definitely benefit from local user bases
 * Perhaps geographical regions can represent the tightly knit networks in the Facebook expansion model

Snapshots from class discussions.

= Speed dating = Pairwise 2 * 2 min 3 distinct groups Why am I proud about my project? What is hard about it and how did I cope? = Bit.ly Insights = Hour of a day Day of a week Post your links at 7 am or 7 pm on Thursdays and Fridays.
 * Time matters!**
 * Most people consume data, i.e. click bit.ly links, in the morning when people start working, and at nights when people are off work.
 * The lowest consumption occurs when people are driving back home.
 * Links that survive 7-8 days tend to be generated on Thursday or Friday.
 * Most links that survive 2-3 days are created on Monday or Tuesday.
 * Suggestions for content generator:**

= For next Thursday: Esther Dyson = Listen to mp3 of Digital Exhibitionism [] Post insightful question on FB, top three questions will be answered in class = IdeasProject, Dogfood = = END =

Leo Kung lhkung@stanford.edu Joseph Lau jhlau@stanford.edu