CS10: The Beauty and Joy of Computing - Su22

​

Agenda

• Social Implications
• Free Time
• Algorithmic Bias

​

Announcements

• Final project is due 8/4
• Make sure you’ve met with your TA and gotten it approved before you start working
• HW5 was due yesterday, with slip days can turn in tomorrow
• In-lab final 8/4
• Paper final 8/10

​

Social Implications

What do we mean by social implications?

• Our collective attempt to remind you that “with great power, comes great responsibility”
• Similar to studying social sciences / history - whenever something big happened (e.g. women wearing pants, industrial revolution), there were effects - we get to study those
• Rise in technological consumption / adoption past 20/40/100 years

​

https://www.visualcapitalist.com/rising-speed-technological-adoption/

​

Which social implications?

• lots of different avenues to explore these affects
• Political
• Economical
• Environmental
• Cultural
• Etc etc.
• We’ll look at a few today

​

Why and how should i care?

• Helpful info as a consumer
• As a computer scientist, understanding your ethical responsibility
• How the work you do affects yourself and others
• Yes, this is in scope for the Final
• Put the rest of what you learn in this class in context
• Relevant for concurrency lecture next week!

​

​

Social Implications: Free Time

Credit to Josh Hug for these slides:)

Free Time

According to the American Time Use Survey [Link], an average American weekday for a full-time employee in 2015 looked like:

• 8.13 hours work.
• 8.16 hours personal care, including sleep.
• 7.71 hours of everything else, including 3.28 hours of “TV, leisure, sports”.

​

Fundamentally, our most limited resource is our time and attention.

Technology and Free Time: Television

Technology has completely changed the way we use free time, sometimes with surprising speed and impact.

• 1939: “The problem with television is that the people must sit and keep their eyes glued on a screen; the average American family hasn’t time for it… Radio can flow on like a brook while people listen and go about their household duties and routine. Television, on the other hand, is no brook; it is more of a Niagara, a spectacle for the eye.” - Orrin Dunlap Jr [Link].
• Today: The average American watches ~5 hours of television per day [Link] (this includes all Americans, not just those working, and also includes weekends).
• One theory: Any advertising driven information technology medium will evolve to consume your attention to its greatest possible ability (more on this later).

​

Technology and Free Time: Smartphones

20 years ago nobody would have predicted teenagers would be using their phone every day to keep up with social media communications.

​

Unlike TV:

• Smartphones are always there.
• Smartphones are watching you back.

​

Possible Repercussions

Only correlation is proven here, not causation!

(Figure from https://www.theatlantic.com/magazine/archive/2017/09/has-the-smartphone-destroyed-a-generation/534198/)

Possible Repercussions

Only correlation is proven here, not causation!

Possible Repercussions

Only correlation is proven here, not causation!

Other Worlds are Possible: The World That Was

As the same article above mentions, this is all in sharp contrast to 1972/1973, when Bill Yates took a bunch of photos of kids hanging out at a Tampa Roller Rink:

• http://lenscratch.com/2014/01/bill-yates/

The World That Is

By contrast (Link):

• “There’s not a single exception. All screen activities are linked to less happiness, and all nonscreen activities are linked to more happiness.”
• “For all their power to link kids day and night, social media also exacerbate the age-old teen concern about being left out.”
• “Today’s teens may go to fewer parties and spend less time together in person, but when they do congregate, they document their hangouts relentlessly—on Snapchat, Instagram, Facebook. Those not invited to come along are keenly aware of it.”

Counterpoints

Counterpoints: "Is Screen Time Really Bad for Kids?"

• “...But other researchers began to worry that such dire conclusions were misrepresenting what the existing data really said.”
• “Earlier this year, Amy Orben and Andrew K. Przybylski, at Oxford University, applied an especially comprehensive statistical method to some of the same raw data that the 2017 study and others used.”
• Their results, published this year in Nature Human Behavior, found only a tenuous relationship between adolescent well-being and the use of digital technology.
• “What makes one study that draws on that data distinct from another is a series of choices researchers make about how to analyze those numbers.”

Counterpoints: "Is Screen Time Really Bad for Kids?"

“For instance, to examine the relationship between digital-technology use and well-being, a researcher has to define “well-being.””

• “The M.T.F. survey, as the Nature paper notes, has 13 questions concerning depression, happiness and self-esteem.“
• “Any one of those could serve as a measure of well-being, or any combination of two, or all 13. A researcher must decide on one before running the numbers;”
• “suppose five ways produce results that are strong enough to be considered meaningful, while five don’t. Unconscious bias (or pure luck) could lead a researcher to pick one of the meaningful ways and find a link between screen time and depression without acknowledging the five equally probable outcomes that show no such link.”

Counterpoints: "Is Screen Time Really Bad for Kids?"

“For instance, to examine the relationship between digital-technology use and well-being, a researcher has to define “well-being.””

• “For the M.T.F., Orben and Przybylski identified 40,966 combinations that could be used to calculate the relationship between psychological well-being and the use of digital technology.”
• “When they averaged them, they found that “digital-technology use has a small negative association with adolescent well-being.”
• “the strength of the association screen time had with well-being was similar to neutral factors like wearing glasses or regularly eating potatoes.”
• “The real conclusion of the Nature paper is that large surveys may be too blunt an instrument to reveal what those risks and benefits truly are.”
• “What’s needed are experiments that break “screen time” into its component parts and change one of them in order to see what impact that has and why,” says Ronald Dahl, director of the Institute of Human Development at UC Berkeley.”

Counterpoints: "Is Screen Time Really Bad for Kids?"

My belief: Screen time isn’t inherently bad, but the applications on our screens are manipulative and could be made much much better.

​

What do you think?

• Society has a pessimistic vibe these days (ecologically, economically). Nice to have something mindless and relaxing to chill out. Can go find funny stuff.
• We live in an age where you can’t escape information. Internet has made you aware of trouble. (Age Of Information by Lil B).
• People who are predisposed to depression, will tend to gravity towards screentime.
• In many societies, the internet opens up new channels of information, opportunities for education.

​

​

​

Social Implications: Algorithmic Bias

A very generalized conceptualization

A lot of what we do in computer science / data science is use existing data to answer questions about the future

​

​

​

If I know something follows the trend of y = x from x = -1 to x = 6, I can make guesses about other data points I haven’t seen yet

Given x = 7, what informed guess can I make about the value of y?

probably y = 7

Is this always going to work?

​

​

​

​

Maybe this is helpful and useful when I want to:

• Make a guess about how much my insurance will be raised if I buy a sports car
• Add a new item to my restaurant menu and need to predict its popularity based on other dishes
• Predict when the housing market will crash again so i can decide whether to invest in property
• Buy stock
• Etc etc

Is this always going to work?

What about when I want to:

• Predict which colleges kids will get into based on their SAT scores
• Hire the next big CEO based off what current big CEOs look like
• Set insurance rates based off what people in a wealthy area can pay
• Use an algorithm to decide whether or not someone is guilty of a marijuana related offense based off who is usually

​

Is this always going to work?

• We are constantly recording, storing, and creating data
• It isn’t over just because discrimination may considered to be not be as severe as it used to be
• Unconscious bias plays a BIG role in this
• Because we use this biased (read: racist, homophobic, classist) data, we continue to make biased (read: racist, homophobic, classist) decisions and more biased (read: racist, homophobic, classist) data

Quick Aside: Types of learning

Supervised Learning

Using labeled datasets to train algorithms to classify data of predict outcomes accurately

E.g. spam classifier for email

Unsupervised Learning

Provide algorithm with a bunch of data to make sense of through patterns/underlying structure

​

​

Reinforcement Learning

Don’t have all the experience ahead of time, then deploy in real world for learning in real-time

E.g. got a reward, so try to repeat that action, or, failed this task so try to do task differently

​

​

​

26

How do we fix this!!

• First step is to acknowledge that it’s happening: where are we grabbing data from and what confounding variables are at play
• How much are these confounding variables having an effect on our algorithm and can we correct for it?
• Take accountability
• No real solution out there yet… We’re still learning…

​

​

https://www.brookings.edu/research/to-stop-algorithmic-bias-we-first-have-to-define-it/

​

​