Exam practice

1. Gears and punched cards (1640s to 1930s): Blaise Pascal; Herman Hollerith and the US government’s census.
2. Electronic giants (mainframes) (1940s to 1970s): the US Army and the University of Pennsylvania (ENIAC); IBM; in Nepal, the government’s census office.
3. Personal computers (1970s to 1990s): clubs of people who built their own computers, and the small companies that grew from them.
4. The internet and the web (1990s and 2000s): a government research network and universities; Tim Berners-Lee at CERN; volunteer communities who write free software and Wikipedia.
5. Mobile (2000s onward): phone companies and big technology companies; in Nepal, the government’s telecom company (1999).
6. Cloud and AI (2010s onward): a few giant companies with huge data centers; governments, which write the rules.

In the era of the electronic giants, governments led: they paid for the first computers (ENIAC for the US Army, Colossus for Britain, Nepal’s census computer), because each cost millions, needed experts, and was used for government jobs like war and the census.
In the personal computer era, governments were no longer the leaders: chips had become cheap, so clubs of hobbyists and small new companies built computers, and families, schools and offices bought them. Governments mainly became customers and rule-makers.
The difference is the cost: when computing is very expensive, only governments and big organizations can lead; when it is cheap, individuals and communities can.

He started as an individual with an idea, and the work was done by a community: villagers, volunteers from abroad and donors working together, later organized as the Nepal Wireless Networking Project. (Any answer naming the individual and the community, with reasons, is right.)
No company had done it because it didn’t look profitable: few customers, steep hills to carry equipment over, no electricity grid, and people with little money to pay. A company builds where it expects to earn; a community builds where it needs something.

An intended effect is what the technology was made to do; an unintended effect is something it does that its makers didn’t plan, which can help or harm.

• Intended: people can make calls without a telephone line.
• Helpful unintended: phones became wallets, so families receive money from relatives abroad and pay shops without a bank branch (or: flood warnings by text message; farmers checking prices).
• Harmful unintended: scams by phone and message (or: cyberbullying; hours lost to endless feeds; accidents caused by drivers using phones).
• Environment: old phones become e-waste that harms soil and water when dumped or burned (or: the electricity they and their towers use; the mining of metals for them).

• Decomposition: breaking a big problem into smaller parts. The nurse splits “vaccinate every child” into one list for each vaccine and age.
• Patterns: spotting what repeats or is alike. She notices that children from one ward often miss their second dose, so she can visit there.
• Abstraction: keeping only the details that matter. For each child she needs only the name, date of birth, ward and which vaccines they have had, not everything about the family.
• Algorithms: exact steps followed every time. Every week: find the children who have reached the age for a vaccine and haven’t had it; send their families a reminder; record each dose given.

Automation is a machine doing by itself work that people used to do.

• Replace: telephone exchanges now connect calls automatically, so the jobs of operators who plugged in every call by hand are gone (or: human computers replaced by electronic computers).
• Change: bank tellers still work, but ATMs count out the cash, so they spend more time on loans, problems and advice (or: a doctor reading X-rays with software that marks places to check).
• Create: people who repair phones, write apps or deliver orders taken by apps have jobs that didn’t exist before (or: people who label photos to train AI).

A repeated task can be written as exact steps, an algorithm, and a machine can follow those steps faster and more cheaply, without getting tired. Tasks that change each time, need judgment or care, or need hands working in an unpredictable place are much harder to write as steps.
Hard to automate, for example: a nurse (each patient is different; the job needs care, judgment and talking to people); a plumber or electrician fixing old buildings (every house is different, in awkward places); a farmer on terraced hillsides (small, steep, uneven fields where machines can’t easily go); a teacher (knowing each student and what they need).

Access needs five things together, and a phone in the house gives only the first:

1. A device: the household phone may belong to one person, often a man, or be away in the city; it may be a basic phone with no internet.
2. A connection: in many villages the signal is weak or missing, and there may be no electricity to charge it.
3. Money: data packs cost a lot for a poor family, so the phone is often out of data.
4. Skills: someone who never learned, often an older person, can’t use the apps.
5. Language and ability: apps mostly in English or Nepali don’t serve speakers of other languages; a person who is blind or can’t use their hands needs an app designed for them.

The survey numbers agree (as of MICS 2024–25): about 82 in 100 households had internet at home, but in the poorest fifth only about half the people aged 15 to 49 had used the internet in three months.

For people aged 15 to 49 who used the internet in the last three months (as of MICS 2024–25):

• Place: about 91 in 100 in Gandaki Province and 83 in Bagmati, but about 61 in Karnali and 64 in Sudurpashchim and Madhesh; 78 in towns against 68 in villages.
• Money: about 52 in 100 in the poorest fifth of families, 93 in the richest.
• Age: about 85 in 100 of people aged 20 to 24, falling to about 59 for people aged 45 to 49.
• Women and men: about 73 in 100 women against 78 men; and fewer women own a phone of their own (82 against 90).

The largest gap is by family money (about 40), then between provinces (about 30) and ages (about 25); the gap between women and men is smallest, but it is much larger for using a computer (about 12 in 100 women against 24 men).

Any five, for example:

• Write it in Nepali as well as English (and the area’s other languages if many parents speak them): parents who don’t read English.
• Let the text grow with the phone’s text size, dark on light: parents with weak eyes, grandparents.
• Name every button and picture for screen readers: parents who are blind.
• Make the pages light so they open on a weak signal with little data: parents in villages with a small data pack.
• Offer another way: results by text message, or a printed copy at school: parents with a basic phone or no phone.
• Don’t show pass or fail by color alone; write it: parents who can’t tell red from green.
• Use big buttons and plain words: older parents and people with shaking hands.
• Test it with real parents before it goes live.

Mostly yes. A ramp cut into a footpath was made for wheelchairs but helps parents with prams, porters and people with suitcases. Captions were made for people who are deaf but help everyone on a noisy bus or with the sound off. Voice typing was made for people who can’t type easily and is now used by anyone with busy hands. Large, dark text helps people with weak eyes and everyone reading in bright sunlight.
(A good answer may add that it isn’t always free: making an app in several languages, or offering text messages too, costs time and money, so designers must choose, and people with the least power are easiest to forget.)