By now you have built something a real person uses. That is the moment this page becomes possible to write honestly, because every question below is now a question about your own code rather than about technology in general.
Privacy is a design decision
Your program probably stores something. Somebody’s name, a mark, a sign-out time, a phone number “in case”. Each of those was a decision, and defaults are decisions too — nobody has to approve them, which is exactly what makes them worth examining.
Four questions before your program stores anything
- What is the smallest amount of information that still makes this program work?
- Who else can open this file — on this computer, on the shared drive, in a backup?
- Does the person whose information it is know it is being kept, and would they be surprised?
- When does it get deleted, and who does the deleting?
The most common honest answer to the first question is “less than I was about to collect”. A tally does not need names. A reminder list needs a title and a date. Collecting less is not a compromise; it is the version of the program that cannot leak what it never held.
Who cannot use what you built
Software that works for you and fails for somebody else is not neutral — it has quietly picked a user. Test your own project against this list, honestly:
- Can it be used without a mouse, by somebody navigating with the keyboard?
- Can somebody with low vision read it — is the meaning ever carried by colour alone, or by text too small to enlarge?
- Are the messages in plain language? “Invalid input” tells nobody anything; “Please type the number of days, like 14” tells them what to do next.
- Does it assume a fast connection, a recent device, or a phone number the person may not have?
- Does the name field accept apostrophes, hyphens, accents, and names that are one word or five? Systems that reject real names are a long-standing, entirely avoidable insult.
Accessibility is not a feature added at the end. It is a series of small decisions, each cheap while you are making it and expensive afterwards.
Whose rules are in the code
A program is somebody’s judgement written down. When a sign-up form requires a permanent address, or a scheduling tool assumes everyone is free after school, the rule was probably never argued for — it just arrived from whoever wrote it. That is the ordinary kind of bias, and it is far more common than the dramatic kind.
The same question scales up. The emerging research areas the course asks you to look into — machine learning, computer vision, security and cryptography, human–computer interaction — are all places where somebody’s judgement is being encoded and then applied to people who never met them. A face-recognition system that works less well on some faces than others is a design decision showing up as a failure rate. Read the claims made about these systems the way Tech Headlines teaches: ask who benefits, who was tested on, and who is left carrying the error.
The physical cost of computing
Computing is not weightless. It has three costs that are easy to overlook because none of them appear on your screen:
- Manufacturing and e-waste. Devices are built from mined materials, and a replaced laptop does not disappear. Electronics are the fastest-growing kind of waste in many places, and much of it is handled badly.
- Electricity. Every device, server, and search draws power. A room of machines left running overnight is a real, measurable choice.
- Paper. Printing files and emails that nobody reads is the smallest of these and the easiest to stop.
There is a health cost too, and it is not abstract at Grade 11: hours at a badly set-up desk produce genuine musculoskeletal problems and eye strain, and a life arranged entirely around a screen costs sleep, activity, and time with people.
The response is unglamorous and it works: power settings that actually sleep the machines, lab and school policies about printing, repairing and reusing a device rather than replacing it, and passing on hardware that still has years in it. Outside the school there are organisations that exist for exactly this — municipal electronics depots, charities that refurbish computers, and cartridge return programs. Finding out which ones serve your community is a short piece of research with a concrete outcome.
And computing pays some of this back. Modelling a design in software instead of building it saves the physical materials; monitoring networks track water, air, and energy use; routing and scheduling programs cut kilometres driven. The honest position is neither “tech will fix it” nor “computers are the problem” — it is that a computer is a lever, and levers do what the person holding them wants.
Argue the hard cases in When Code Hurts and Should It Exist, and take the four questions above into your own project before you hand it over.
Curriculum connection
D1.1
describe the negative effects of computer use on the environment (e.g., creation of e-waste, excessive use of paper resulting from unnecessary printing of files and emails, heavy power consumption) and on human health (e.g., exposure to radiation, musculoskeletal disorders, eye strain, mental health problems resulting from social isolation, various health consequences of reduced activity levels);
Link to original
D1.2
identify measures that help reduce the impact of computers on the environment (e.g., lab regulations, school policies, corporate and government policies promoting paperless workplaces and computer recycling and reuse) and on human health (e.g., ergonomic standards);
Link to original
D1.3
describe ways in which computers are or could be used to reduce resource use and to support environmental protection measures (e.g., computer modelling to reduce use of physical resources; management of natural resources);
Link to original
D2.1
demonstrate an understanding of emerging areas of research in computer science (e.g., cryptography, parallel processing, distributed computing, data mining, artificial intelligence, robotics, computer vision, image processing, human–computer interaction, security, geographic information systems [GIS]);
Link to original