Computers and Society
(CPSC 430, Term 2, Session 201, 2020-21)
| Overview | Grading Scheme | Weekly Reading, Writing & Peer Review | Text and Online Resources | Schedule |
Course Description: This course explores the interplay between information technology and society, with an emphasis on ethical issues. Students will come away from the course with greater understanding of the social and ethical implications of computer use and abuse, an improved ability to think critically and defend their decisions logically, and a greater appreciation for alternate points of view. The focus of the course is on reading, writing and discussion; each week students will complete an assigned reading, write a mini-essay in response, and evaluate the written work of others.
Meeting Times: Tuesday, Thursday, 5:00 PM - 6:30 PM
First Class: Tuesday, January 12, 2021
Location: Zoom (access the link via Canvas. Registration in advance via the link is required.)
UBC Web Page: main course; waiting list
Instructor: Kevin Leyton-Brown
Instructor's Office Hours: Tuesdays and Thursdays 6:20 - 7:00 PM, or by appointment
TAs/IAs and Their Office Hours (links available on Canvas):
Hannah Elbaggari (hre@cs.ubc.ca); Mondays 1:00 - 2:00 PM PST
Kevin Huang (khuang21@students.cs.ubc.ca); Mondays 5:00 PM - 6:00 PM PST
Joey Eremondi (jeremond@cs.ubc.ca); Tuesdays 9:30 AM -10:30 AM PST
Hedayat Zarkoob (hzarkoob@cs.ubc.ca); Tuesdays 2:00 - 3:00 PM PST
Unma Desai (unma@cs.ubc.ca); Wednesdays 4:00 PM - 5:00 PM PST
Harshinee Sriram (hsriram@cs.ubc.ca); Thursdays 10:00 AM – 11:00 AM PST
Greg D'Eon (gregdeon@cs.ubc.ca)
Quick Link:
Piazza;
Mechanical TA (Student
enrolment code: b1ca954a. Problems logging in: email
Hedayat.)
Course Topics: History of computing, networking, and information storage; Essay writing; Ethical theories (Kantianism, utilitarianism, social contract theory, virtue ethics); Networked communications (spam, censorship, filtering); Intellectual property; Privacy; Security; Reliability; Professional Ethics; Artificial Intelligence, globalization, and other changes to the workplace.
Prerequisites: The course requires only third year standing and 3 credits of computer science. Beyond these formal requirements, an ability to speak, read and write fluently in English is essential for success in the class.
Wait List Policy: The course professor is not in charge of the class waitlist, and cannot assess your chances of getting into the class. Please direct any questions you may have to the CS Department Academic Advisors. If you are on the waitlist and want to take the course, you should participate from the beginning. (1) Students who attend class will be given waitlist priority (attendance information will be shared with the main office); (2) Students who register for the course late still have to complete all assignments on the same schedule as all other students. Waitlisted students are eligible to submit assignments before being formally registered.
Equity, Inclusion and Wellness: UBC provides resources to support student learning and to maintain healthy lifestyles but recognizes that sometimes crises arise and so there are additional resources to access including those for survivors of sexual violence. UBC values respect for the person and ideas of all members of the academic community. Harassment and discrimination are not tolerated nor is suppression of academic freedom. UBC provides appropriate accommodation for students with disabilities and for religious and cultural observances. Details of the policies and how to access support are available here. Please also see the CS Department's resources on this topic.
Academic Honesty: UBC values academic honesty and students are expected to acknowledge the ideas generated by others and to uphold the highest academic standards in all of their actions. Plagiarism is a serious offence (see the CS Department's statement) and will be dealt with harshly. Plagiarism is the unattributed use of any external source (e.g., another student, a web site, a book) in work for which a student takes credit, or the inappropriate use of any external source whether or not attribution is made. The seriousness of the offence depends on the extent to which the student relied upon the external source. You must cite all external sources that you use, and write in your own words. Any text that you take verbatim from another source must be in quotation marks and followed by a citation. We will use TurnItIn (an online commercial service) to detect plagiarism including the resubmission of essays used in previous offerings of this course.
Grading Scheme
In order to pass the course, a student must receive a passing grade on the final exam. The exact percentage breakdowns shown here may be subject to change.
| Course Element | Fraction of Grade |
| Class Attendance | 10 % |
| Active Participation | 10 % |
| Weekly Essay Questions | 35 % |
| Weekly Peer Review | 10 % |
| Midterm Exam (take-home) | 10 % |
| Final Exam (2.5 hours) | 25 % |
Active participation in the class is a key element of this course; it simply isn't the same to watch recordings later. Thus, 10% of your grade will come from class attendance. This will be achieved by Zoom polls conducted 5 minutes after class begins and 5 minutes before class ends. In recognition of COVID disruptions, you will get full marks if you complete both polls at least ONCE per week; if you must, you can watch the other class by video. (Please note that we will dock points if we notice people abusing the system, e.g., leaving the meeting shortly after class begins or coming right at the end.) If you're in a time zone that makes it impractical for you to attend the class at its scheduled time at least once per week, I urge you to seek a different class.
A second key element is participating in lectures. Your participation grade will assess your verbal contributions in class, along with activity in Zoom chat and on Piazza that involves a substantive comment (your post must contain discussion or argument related to course material, at the TA's discretion; e.g., posting only a link or "me too" agreement doesn't count). We begin by counting the number of classes in which you spoke; contributing once or more to chat or Piazza can substitute for participating in one class from the same week. Given this count, we will then calculate your grade as follows:
| Participate at least this many times... | Get this many points |
| 17 | 10 |
| 14 | 9.5 |
| 12 | 9 |
| 10 | 8.5 |
| 8 | 8 |
| 6 | 7.5 |
| 5 | 7 |
| 4 | 6 |
| 3 | 5 |
| 2 | 3.5 |
| 1 | 2 |
| 0 | 0 |
Weekly Reading, Writing and Peer Review
This course emphasizes making and evaluating arguments about ethics and the social impact of technology. Each week, we'll follow the same format.
Between Thursday, 6:30 PM and Tuesday, 4:45 PM (sharp)
- Do assigned readings of up to one chapter from the textbook.
- Take a multiple-choice quiz online to test your comprehension of the readings. You must get a perfect score but can keep trying until you do; every time you fail, you must wait 30 minutes before trying again.
- After completing the quiz, answer one essay question (some weeks, your choice from two or three given questions). You'll be allowed up to 2500 characters; that's less than one single-spaced page.
- Do at least 3 calibrated reviews if Mechanical TA indicates that you should. We recommend doing even more to improve your grading skills.
Between Tuesday, 6:30 PM and Thursday, 4:45 PM (sharp)
- Perform your own peer review of four randomly assigned students' written questions (and/or calibration questions).
Quizzes You won't receive any grades directly for completing the multiple choice quiz each week. However, you'll only be eligible to perform a given week's peer review if you get a perfect grade on the quiz, which makes the quiz important to your grade. You must get a perfect score but can keep trying until you do; if you fail to get a perfect score, you must wait 30 minutes before trying again.
Calculating Your Essay Grade Your essay will be graded between 0 and 5 in the following four dimensions.
- Was the essay well structured, stating a thesis, supporting it with argument(s) that are clearly related to this point and (if relevant) distinct from one another, and linking these arguments in a logical way?
- Did the essay do a good job of making its case, choosing relevant arguments, backing them up with evidence and examples at an appropriate level of detail, and responding to contrary views as appropriate?
- Did the essay demonstrate a good understanding of the course's subject matter, including both the topic and the wider context?
- Was the essay presented clearly and in correct English?
Each week you'll receive at least 3 peer reviews of your work (and will perform 4 such reviews). These reviews will be double-blind: neither an essay's author nor its reviewer will know the other's identity. Reviews will explain the rationale for your grade in each dimension, using at least 20 words.
Many, but not all, essays will be graded by TAs. Whenever a TA grades your essay, only the TA grade will matter to your essay score, though you will still see your peer grades. Our system will select essays for TA evaluation based on a range of different factors. TAs will always grade alongside students who have not yet demonstrated sufficient reliability (thus overruling the grades they assign). We will spot check when reviewers' scores substantially disagree and also when essays receive unusually high scores. Finally, we will conduct a large proportion of spot checks entirely at random.
When no TA grades your essay, your grade will be computed as a weighted average of the grades assigned by your peers, where each peer's weight is proportional to our assessment of their "dependability": our belief about how much we can trust each score based on each grader’s behavior so far in the course (see below).
If you disagree with the grade you receive on any essay, you can appeal, explaining your concern in 100 words or more. If you do not provide a convincing argument, a TA may reject your appeal without regrading your essay; however, you can refine your argument and resubmit the appeal. If a TA regrades your assignment, you will receive the TA's grade regardless of whether it increases or decreases your score. You can also flag rude or careless reviews and endorse high quality reviews if you’d like to bring them to the attention of TAs without asking for a change in your grade.
Even if your essay was not initially graded by a TA, we may flag it for later review, in which case we will override your peer grade with the TA-assigned grade. There are two key reasons we may do this. First, if our inference algorithm updates its beliefs about your graders' dependability to an extent that your grade would change by a full point or more, a TA will override your peer graders. Second, if we become suspicious of a student’s grading behavior (e.g., because of appeals or other spot checks) we will go back and regrade other assignments this student has handled.
Essays will contribute increasing amounts towards your final grade as the term progresses, as follows:
- the first 3 essays will be worth 2% of your final grade each;
- the next 3 essays will be worth 3% of your final grade each;
- the final 5 essays will be worth 4% of your final grade each.
Calculating Your Peer Review Grade We use Bayesian inference to estimate each student's "dependability" as a reviewer on a weekly basis. The key idea is that calibrations and assignments graded by TAs give us information about which graders are more reliable; we then bootstrap this knowledge to decide how much to trust each grader on assignments that were not graded by a TA. Your dependability score is our estimate of effort * (1/variance), where effort is (1 - the probability that you always assign the class average without considering the essay) and variance is our estimate of your tendency to differ from TA scores. The system starts out with the assumption that all students have low dependability scores (specifically, low effort and high variance). As you grade assignments and perform calibrations, we'll update these beliefs. In particular, observe that doing more calibrations both helps you to get better at grading and gives us evidence of your grading prowess to overwhelm the system's pessimistic initial belief. Note that if you always assign each submission the class average, our model will assign you a low effort probability; you need to properly identify both strong and weak assignments in order to achieve a high dependability score.
Every time a TA grades an assignment that you also graded, they'll make a binary assessment of whether you offered thoughtful comments (whether or not your scores agreed with theirs). We'll use these assessments to update your effort probability and hence your dependability score.
For each student we will maintain both a "realistic" estimate of dependability and a "pessimistic" (lower confidence bound) estimate. Each week, we'll identify those students with pessimistic dependability estimates below a certain threshold and assign them to grade each other; we'll also have a TA grade each such assignment. (Note that this means that if you have a low pessimistic dependability estimate, you’ll likely get noisier peer grades, but this won’t matter because your grade will be entirely determined by a TA.) The remainder of students (those for whom even our pessimistic dependability estimates exceed a threshold) will perform independent peer grading as discussed above. This system ensures that your assignment will either be graded by a TA or will be graded entirely by peers that the system confidently predicts will grade reliably.
If your pessimistic dependability estimate falls below our threshold, you will also be required to perform 3 "calibration" reviews of carefully pre-graded essays from previous years, which will automatically be graded by the system. These calibration reviews are a designed to help you learn how to grade well and also how to write good essays. You're also allowed to do extra calibrations (whether your weekly assignment was 3 or 0); this is a particularly good idea when your dependability estimate is low, because doing extra calibrations will teach you to grade better and can also improve your peer grading scores.
Each week, we'll assign you a peer grading score; overall, these scores will make up 10% of your final grade. These grades will be derived from our "realistic" estimates of your dependability score. (The exact formula is complex and subject to change, but it's strictly monotone in dependability; you can see your current standing in MTA.) Observe that your dependibility score will start low, but you can increase it quickly by performing additional calibrations and by grading well. You'll only be eligible to perform peer review when you have completed the week's quiz. If you complete only x peer reviews in a week when you were assigned y, we'll scale your peer review grade for that week by x/y; if you complete only x' calibrations in a week when you were assigned y' calibrations, we'll (further) scale your peer review grade for that week by (y + x')/(y + y').
We'll scale your peer review grades in a manner similar to your essay grades:
- the first 3 weeks will be worth 0.6% of your final grade each;
- the next 3 weeks will be worth 0.9% of your final grade each;
- the final 5 weeks will be worth 1.1% of your final grade each.
Late and Missing Submissions At the end of the course, we will drop your worst two essay and peer review grades. (That is, your grade will be calculated based on your best 9 essays and best 9 peer review weeks, even though you will be required to submit 11. Each grade will be computed as the sum of the number of points assigned on your 9 best contributions, divided by the total number of points available for these contributions.) Think of this as permission, given in advance, to not submit for two weeks because of illness, travel, starting the course late, family obligations, COVID emergencies, conflicts with other courses, etc. (In past years, we've dropped only one essay and peer review grade; the extra drop is intended to provide extra flexibility for pandemic related dificulties.) No further allowance will be made for failure to submit essays or peer review, except in truly exceptional circumstances such as a prolonged and serious illness.
Text and Instructional Resources
Textbook We will be using the textbook Ethics for the Information Age, 8th Edition, by Michael J. Quinn. It's important that you have a copy, because we'll be reading pretty much the whole thing. You can buy or rent a digital copy for a lower cost than buying the book.
Mechanical TA You will submit your quizzes and weekly essays, perform peer review of others' essays, and read articles on the course blog using our Mechanical TA site. This site requires you to have a UBC CWL account. If you don't yet have one, you'll need to create one. You also will need to know the student enrolment code "b1ca954a".
Additional Resources Your second week of readings, on essay writing, will not come from the textbook. Instead, you'll read the following set of publicly available resources. For your convenience, they're also gathered together into one PDF in MTA.
- Purdue Online Writing Lab: Developing Strong Thesis Statements
- Douglas College writing: The essay
- Douglas College writing: Creating an Outline
- University of Chicago Writing Program: Argument: A key feature of college writing
- Harvard Writing Center: A Brief Guide to Writing the Philosophy Paper
- Purdue Online Writing Lab: Logic in Argumentative Writing (parts a, b, c, d, e, f)
-
UCLA graduate writing center: The writing process
You're additionally encouraged to read these optional articles: - New York Times: The Sentence as a Miniature Narrative
- New York Times: Where do sentences come from
- New York Times: Other men's flowers
- Thou shalt not commit logical fallacies
The Computer Science Reading Room has prepared an online list of resources related to technology and society, which you can access here. Students in the Vancouver area can also request physical materials, which can be sent out for campus delivery or held for curbside pickup at UBC. If interested please contact the Reading Room either via email or via a Zoom drop-in session (see their website).
This schedule may be adjusted slightly as we go through the term. Slides will be made available after each lecture, and may be accessed by clicking on the links under "topic". Applicable chapter numbers from the textbook are also given.