Build Your Own Personal Cloud — Your Files, Your Server, Your Rules

What if the “cloud” holding your files was a small computer on your own shelf?

Upload a photograph to an online service and it can appear on your phone, tablet and laptop. That feels almost magical. Yet behind the scenes there are computers, storage devices, network connections and software.

Google, Microsoft and Apple have made remote storage familiar. They have not made it the only option.

With a Raspberry Pi or a suitable old computer, you can build a personal cloud: somewhere to store and access your files, with the storage under your control.

You can also reach it when you are away from home without necessarily paying a remote-access subscription.

For GCSE and A-level Computer Science students, this is a particularly rewarding project. Client-server architecture, IP addresses, authentication and data storage become parts of something you can actually use.

What Is a Personal Cloud?

One home server does not reproduce everything associated with commercial cloud computing: enormous capacity, elastic scaling and sophisticated redundancy.

Here, “personal cloud” means a self-hosted service that makes your files available across devices, including remotely when configured appropriately.

You might use it for:

• Photographs and family documents.
• Revision notes and homework.
• Programming projects.
• Selected science recordings.
• Finished sailing videos or organ performances.

The important change is that your files live on storage you manage.

It is a little like creating your own digital filing cabinet, then giving your authorised devices a way to reach it.

From a File Server to a Personal Cloud

A Samba file server gives computers access to shared folders over a network. That is an excellent starting point.

A personal-cloud application adds another way of working: access through a browser, individual user accounts, synchronisation clients and sharing controls.

For this project, Nextcloud is a useful option. Its self-hosted software provides file storage and synchronisation, with additional applications available for other functions. Nextcloud

Think of the components this way:

Component	What it contributes
Samba	Access to shared network folders
Nextcloud	A web-based service around stored files
Private remote connection	Access for authorised devices away from home
Backup system	Recovery when files or hardware fail

You do not have to install everything.

Begin with one purpose: upload a file on one device and retrieve it on another.

Why I Like This as a Teaching Project

In my teaching, I want students to connect the vocabulary with the machinery.

It is easy to memorise:

“A server provides services to clients.”

It is more revealing to upload a document from a Windows laptop, retrieve it on an iPad and explain which computer handled the request.

My teaching and media work also make the purpose easy to see. Lesson notes, science photographs and finished sailing videos all need organising and accessing.

I would begin this project with demonstration files. Once the system works, we can investigate whether it would be useful for particular real tasks.

The first successful remote download is a satisfying milestone. The next question is even better:

“What had to happen for that to work?”

Choose Your Server: Raspberry Pi or Old Computer?

You do not need an expensive new machine.

Starting point	Why use it?	What to consider
A recent Raspberry Pi with sufficient memory	Compact and suitable for a modest storage project	Power supply, cooling, USB storage and software compatibility
An older desktop	May already be available; often has internal drive connections	Electricity consumption, fan noise and hardware condition
An older laptop	Compact, with a built-in screen for setup	Drive condition, ventilation and sleep settings

An old computer running a supported Linux distribution can be an excellent starting point. A recent Pi running a supported 64-bit operating system is another practical route.

Check the requirements of your chosen installation before committing. File storage is a different workload from running an online office suite, video calls and numerous additional applications.

For either route, I would use:

• Ethernet to the router where practical.
• A healthy SSD or hard drive.
• Separate backup storage.
• A supported operating system.
• Services configured to start after a restart.

An SD card may be useful for a Pi’s operating system, but I would avoid relying on a cheap card as the only home for important files.

If formatting a drive, copy anything you need first. Formatting erases its contents.

Follow a File: Understanding Client-Server Computing

Imagine opening a photograph on your phone.

Your phone is the client. Nextcloud is the application serving the request. The server’s drive holds the file.

The phone contacts the server and submits a request. The application authenticates the user, checks permission to access the photograph and returns the file.

Several computing ideas work together:

Concept	Role in this project
IP address	Identifies a network endpoint
DNS	Resolves a name to an address
Port	Identifies a service endpoint on a device
HTTPS	Protects the web connection using TLS
Authentication	Establishes who is signing in
Authorisation	Decides what that account may access
Database	Holds application information such as users and file metadata
File storage	Holds the actual file contents

Being able to reach the server does not automatically give someone permission to read every file.

That distinction between connectivity and authorisation is central to understanding networked systems.

Build It in Stages

1. Prepare the machine

Install a supported operating system and apply updates.

Connect the storage drive and configure a persistent mount so it remains available after restarting. Record where the application will store its data.

On your home network, a DHCP reservation in the router can give the server a predictable local address.

2. Install Nextcloud using a maintained guide

Nextcloud needs more than a folder. Its application, database and supporting services must work together.

Nextcloud All-in-One is one installation route. Its official documentation covers private/local access and reverse-proxy arrangements. Follow the current instructions for your chosen setup. GitHub

For a Pi, check compatibility with its architecture and available memory. Begin with file storage and add optional features later.

3. Create accounts and test locally

Keep administration separate from ordinary use.

Create a normal account, upload a harmless PDF and photograph, and retrieve them from another device.

Then create a second account and check that it cannot read the first account’s private files.

You are testing permissions as well as file transfers.

4. Add private remote access and HTTPS

Configure the remote connection and the application’s HTTPS address together.

Nextcloud must recognise the hostname being used. A reverse-proxy arrangement also needs the corresponding trusted-proxy and URL settings.

Follow the installation-specific instructions rather than assuming that installing a VPN completes the application configuration.

5. Restart and restore

Restart the server.

Check that:

• The storage drive mounts.
• The required services start.
• Files remain accessible.
• Remote access still works.

Finally, restore a test file from backup.

A useful installation should survive more than its first afternoon.

How Can You Access It from Outside the House for Free?

The beginner route I would choose is Tailscale, used as a private network between the server and your authorised devices.

It creates encrypted connections and normally works without manually forwarding incoming ports on your router. When a direct connection cannot be established, it can use a relay. Tailscale Docs

This is useful when your internet provider uses carrier-grade NAT, or CGNAT, which can prevent conventional incoming connections to your home.

The basic process is:

1. Install Tailscale on the server.
2. Install it on the phone, tablet or laptop you want to use remotely.
3. Join those devices to your private Tailscale network.
4. Configure access to the Nextcloud service.
5. Test the connection from outside your home network.

Your files still live on your server. Tailscale supplies connectivity rather than file storage.

There is still a third-party dependency: its hosted service helps coordinate connections.

What is currently free?

As checked in October 2026, Tailscale’s Personal plan supports up to six users for $0 and is intended for non-commercial use. A personal or family project can therefore be different from a service used to run a tuition business. Check the appropriate terms before using it commercially. tailscale.com

“Free” here means eligible personal use can avoid a remote-access subscription.

It does not remove hardware, electricity, broadband or backup costs.

Give the cloud a private HTTPS address

Tailscale Serve can make a local web service available within your private Tailscale network and provide HTTPS. It uses a Tailscale-provided hostname, so you do not need to buy a domain for this approach. Tailscale Docs

Nextcloud’s current All-in-One documentation includes guidance for secure tunnelling and reverse-proxy arrangements. Follow that configuration so the hostname, proxy and application settings agree. GitHub

For this private project, use Serve. Funnel is a separate feature for public internet access. Tailscale Docs

The Mobile-Data Test: Prove You Have Remote Access

A phone connected to your home Wi-Fi has not proved outside access.

Try this practical test:

1. Upload a harmless file called “Cloud test.txt”.
2. Turn off the phone’s Wi-Fi and use mobile data.
3. Connect Tailscale.
4. Open the configured private HTTPS address.
5. Sign into Nextcloud and download the file.
6. Disconnect Tailscale and check that a fresh connection to the private address fails.

Already downloaded files may remain available offline. Use a fresh request when testing access.

If it fails, investigate one layer at a time:

• Is the server powered on?
• Does it have an internet connection?
• Are both devices connected to Tailscale?
• Do the access rules permit the connection?
• Is the web service running?
• Is Nextcloud configured for that address?

Troubleshooting becomes much easier when “the cloud is broken” becomes a precise question.

Private Access Is Different from a Public Sharing Link

With this design, you can retrieve your own files away from home. Other authorised users can do so too.

However, a Nextcloud sharing link pointing to a private address will not work for an arbitrary recipient who has no route into that network.

If you want to send a worksheet link to any parent and have it open immediately in their browser, you are asking for a public-facing service. That needs a different access design.

For a first project, personal access is a useful and manageable goal.

Could You Use Port Forwarding Instead?

Yes, but it adds responsibilities.

A conventional arrangement uses:

• A reachable public IP address.
• DNS or dynamic DNS.
• Router port forwarding.
• HTTPS.
• A maintained public-facing server.

Dynamic DNS keeps a name pointing at a changing public address. It does not bypass CGNAT or make a server secure.

A self-managed WireGuard VPN is another option for students who want to investigate networking more deeply. It still needs a reachable endpoint or another way through the network.

Do not expose ordinary Samba/SMB shares directly to the public internet.

For this first build, private remote access gives us a clear, useful goal.

How Fast Will Your Personal Cloud Be?

At home, performance depends on the network, storage and server.

Away from home, your home broadband’s upload speed matters because your server is sending the file out.

Using decimal units:

Transfer time in seconds = file size in megabits / transfer rate in megabits per second

1 byte = 8 bits

For a 1 GB file at a sustained 20 Mbps:

1 GB = 8,000 megabits

Transfer time = 8,000 / 20 = 400 seconds

That is approximately 6 minutes 40 seconds, before allowing for overheads, competing traffic and changing connection speeds.

A revision PDF may be easy to retrieve. A large camera recording can take considerably longer.

For my media work, I would start with finished clips and selected photographs, then evaluate performance before moving entire recording archives.

No Subscription Does Not Mean No Running Cost

Measure the server’s power consumption rather than assuming an old computer is free to run.

Energy used in kWh = power in watts x hours / 1,000

A hypothetical machine averaging 25 W continuously uses:

25 x 24 x 365 / 1,000 = 219 kWh per year

At an illustrative electricity price of £0.25 per kWh:

Annual electricity cost = 219 x 0.25 = £54.75

Use your own measured power and tariff. This is an example, not a current electricity-price quotation.

Reusing hardware can be worthwhile, but compare its ongoing consumption with alternatives.

That makes this project an opportunity to connect computing with energy measurement too.

Synchronisation Is Not a Backup

Synchronisation makes files available across devices. It can also propagate a deletion or unwanted change.

A separate backup should let you recover an earlier state.

For Nextcloud, a complete backup needs the database and configuration as well as the stored files, plus relevant custom applications and themes. Use the supported backup method for your installation so that the captured system is consistent. Nextcloud 36 Administration Manual

A useful target is the 3-2-1 approach:

• Three copies of important data.
• Across two types of storage.
• With one copy off-site.

Include a disconnected or otherwise protected copy where practical.

Then prove you can restore something.

RAID can help maintain availability after certain drive failures. It does not replace a backup.

A backup you have never tested is an assumption waiting to be challenged.

You Become the Administrator

Self-hosting gives you choices. It also gives you maintenance work.

Keep the operating system and applications updated. Use strong passwords and enable multi-factor authentication where supported.

Limit network access to the people and services that need it. Retain Nextcloud’s own login even inside the private network.

Encrypted connections protect data in transit. They do not automatically encrypt the server’s disk, prevent an authorised user making a mistake or protect a compromised device.

Start with demonstration files before trusting the system with important material.

Turn It into a Computing Investigation

For GCSE students, trace a request from client to server, explain authentication versus authorisation, and compare local access with remote access.

For A-level students, extend the project:

• Measure transfer speed for different file sizes.
• Compare Ethernet and Wi-Fi.
• Investigate DNS, NAT and private addresses.
• Examine logs and access permissions.
• Explain the roles of the file system and database.
• Design and test a recovery procedure.
• Evaluate self-hosting against a commercial service.

A particularly useful question is:

“If the server’s drive failed tonight, what would I restore, from where, and how would I know the restoration had worked?”

That requires understanding rather than memorised definitions.

Conclusion: Build Something Useful, Then Understand It

A personal cloud makes computing visible.

The files have a physical home. Requests travel across networks. Accounts have permissions. Performance has limits. Backups have to work.

A Raspberry Pi or an old computer can become a useful service, and eligible personal use can gain private remote access without a subscription.

Begin with one file. Retrieve it from another device. Retrieve it again over mobile data. Finally, restore it from backup.

Those four achievements tell you far more than an installation screen saying “success”.

At Hemel Private Tuition, I want students to understand the systems they use—and gain the confidence to build systems of their own.

 

How Much Water Can a Superabsorbent Polymer Hold?

A tiny pile of powder, an astonishing transformation — and a pinch of salt that changes everything

Pour water onto a small quantity of white granules and watch what happens.

The particles swell, the mixture thickens, and water that flowed freely moments earlier becomes held within a soft gel.

Then add salt.

Depending on the amount added and the polymer used, the gel can shrink and release some of its water. The material that seemed so impressive suddenly performs much less well.

How can the same substance be so absorbent in one set of conditions and so much less absorbent in another?

This is an excellent example of science beyond a familiar classroom definition. We meet polymers in bottles, clothing and packaging. Here, we can explore a polymer whose structure produces a striking, measurable response to its surroundings.

For me, the appeal is that the demonstration immediately invites a better question. “Look what it does!” becomes “How much does it hold, and what controls that amount?”

Meet sodium polyacrylate: a polymer that swells

The material used is cross-linked sodium polyacrylate, a superabsorbent polymer. Suitable grades can absorb hundreds of times their own dry mass in water under favourable conditions. Superabsorbent materials are used in nappies and other products where holding liquid matters. edu.rsc.org

The qualification “under favourable conditions” is essential.

There is no single absorption figure that applies to every product, every liquid and every test. Water purity, polymer formulation, swelling time and applied pressure all affect how we should interpret a result.

A spectacular demonstration in purified water therefore raises a useful question:

Does this tell us how the material will perform in everyday use?

The secret is a connected molecular network

A polymer consists of long molecular chains. In this material, cross-links connect chains into a three-dimensional network.

Picture a flexible fishing net that can expand while its strands remain joined. This is only an analogy, but it helps explain why the material swells rather than simply dispersing as separate chains.

Water enters the network and interacts with its water-attracting chemical groups. The chains contain negatively charged carboxylate groups, written using ordinary text as COO-. Mobile positive ions, including Na+, help balance those charges.

The fixed charges and the distribution of mobile ions contribute to swelling. Meanwhile, stretching the connected network creates an elastic resistance to further expansion.

The gel reaches a balance between processes that favour taking in water and the network’s resistance to being stretched. Wiley Online Library

The water has not disappeared. Much of the swollen gel’s mass is the liquid it contains.

Why does salt reduce absorption?

Table salt dissolves to produce sodium and chloride ions:

NaCl -> Na+ + Cl-

Those ions change the environment inside and outside the gel.

Increasing the surrounding salt concentration generally reduces the ionic contribution to osmotic swelling pressure. Added ions also screen electrical interactions between charged groups, reducing the repulsion that helps the network expand. Wiley Online Library

The balance shifts towards a less swollen network.

A dry sample therefore takes up less liquid from a salt solution, while an already swollen sample can release liquid when salt is added. The Royal Society of Chemistry uses this reversal in its superabsorbent-polymer demonstration. RSC Education

For younger students, “salt changes how strongly the gel swells” is a useful starting point.

At A level, this opens discussion of osmotic pressure, charged polymers and ionic strength.

An important distinction is that the gel is not a cell surrounded by a perfect selectively permeable membrane. Small ions can move through it; its fixed charges affect their distribution.

Adding sodium chloride does not ordinarily destroy the polymer chains. The visible change is principally a change in swelling.

Start with the surprise: make the gel release water

For an opening demonstration:

1. Place about 0.5 g of suitable cross-linked sodium polyacrylate in a clear beaker on a tray.
2. Add 50 ml of distilled or deionised water and stir gently.
3. Allow the particles to swell. The time required depends on the grade and particle size.
4. Ask students to predict what adding salt will do.
5. Add about 1 g of table salt, stir gently and observe over the next few minutes.

Some samples show a striking release of liquid; others change more gradually. Pilot the demonstration with your own material before a lesson.

Use a second, identical hydrated sample without added salt as a comparison.

The question I would ask is:

“If liquid reappears, what does that tell us about where the water was?”

This is a demonstration of deswelling. It does not, by itself, measure maximum absorption capacity.

Turn the demonstration into a fair investigation

The investigation question

How does sodium chloride concentration affect the mass of liquid taken up per gram of dry polymer?

Use fresh dry polymer for every test. Adding progressively more salt to the same gel makes the sample’s history another variable.

Equipment

• Cross-linked sodium polyacrylate from an educational or laboratory supplier.
• Distilled or deionised water and sodium chloride.
• A balance reading to 0.01 g.
• Measuring cylinders, stirring rods and labelled vessels of about 1 litre capacity.
• Fine nylon mesh that retains the hydrated particles.
• A collection vessel, weighing dishes, timer, tray and eye protection.

Use a known material. Unidentified plant-water crystals or powder extracted from nappies may contain other polymers, fibres or additives, making results harder to interpret.

Prepare a concentration series

The following is a suggested classroom investigation, to be piloted with the chosen polymer.

Sodium chloride concentration	Salt needed for 500 ml of final solution
0 g/litre	0 g
1 g/litre	0.50 g
5 g/litre	2.50 g
10 g/litre	5.00 g
30 g/litre	15.00 g

For each solution, dissolve the salt in less than 500 ml of water, then make up to a final volume of 500 ml.

The zero-salt sample is the control.

Suggested method

1. Weigh 0.50 g of dry polymer into each labelled vessel; record the actual mass.
2. Add 500 ml of the appropriate solution.
3. Stir each sample gently in the same way, then leave it for a fixed period, initially 30 minutes.
4. In a pilot test, check whether allowing longer changes uptake appreciably. If it does, extend the time for all samples or describe the result as uptake after 30 minutes, rather than equilibrium capacity.
5. Drain each sample through the same type of mesh for a fixed time, such as two minutes. Do not squeeze or press it.
6. Transfer the retained gel to a tared weighing dish and measure its mass. Minimise losses and weigh promptly.
7. Repeat each concentration three times and calculate the mean.

There must still be free liquid available at the end of swelling.

If the polymer takes up all the supplied liquid, the result is limited by the amount provided. Repeat with more solution or a smaller, accurately weighed polymer mass.

Keep temperature, polymer batch, starting mass, solution volume, stirring, swelling time and drainage procedure consistent.

Calculate how much liquid was retained

For the purified-water sample:

Water uptake (g per g) = (mass of drained swollen gel - mass of dry polymer) / mass of dry polymer

Suppose, as an illustrative calculation:

• Dry polymer mass = 0.50 g.
• Drained swollen gel mass = 100.50 g.

Then:

Water uptake = (100.50 - 0.50) / 0.50 = 200 g per g

That sample has retained approximately 200 times its dry mass in water under the test conditions.

These are example numbers, not experimental results or a promised outcome.

For salt solutions, call the measurement apparent solution uptake. The retained liquid includes dissolved salt, and the gel can partition ions differently from the surrounding solution. Subtracting the dry polymer mass does not isolate the exact mass of water alone.

Also distinguish uptake from total swollen mass. In this example, the liquid-to-polymer mass ratio is 200, while the total swollen-mass-to-dry-mass ratio is 201.

Draw a graph that reveals the pattern

Put:

• Sodium chloride concentration, in g/litre, on the horizontal axis.
• Mean apparent solution uptake, in g per g, on the vertical axis.

Plot the measured points and show their spread, using ranges or error bars if appropriate.

Expect uptake generally to decrease with increasing salt concentration, but do not assume a straight-line relationship or force the graph through the origin.

Useful questions include:

• Where does the largest reduction occur?
• Are repeated measurements close together?
• Could inconsistent drainage explain an unusual result?
• Does the trend support the prediction?
• What evidence would justify calling a result “maximum capacity”?

The graph turns a memorable demonstration into evidence that students can evaluate.

The measurement problem is part of the science

The most interesting difficulty may be deciding what “held by the polymer” means.

Liquid can sit between gel particles or cling to their surfaces. Too little drainage overestimates uptake. Pressing the gel creates a different test by applying a load. Losing gel during transfer underestimates uptake.

A mesh blank can help assess liquid retained by the apparatus: wet and drain an identical empty mesh using the same procedure and measure its retained liquid.

It will not solve every source of uncertainty, but it makes the limitations visible.

For me, this is an excellent teaching opportunity. A fair experiment requires a clear operational definition:

The liquid retained after our specified swelling and drainage procedure.

That is a defensible result, even when it differs from a manufacturer’s figure obtained by another method.

Why nappies need more than a purified-water test

A material used in a nappy encounters dissolved substances, repeated additions of liquid and pressure from the wearer.

The design question is therefore broader than:

Which powder absorbs the most pure water?

Absorption speed, retention under pressure and the way liquid spreads through the product also matter.

This gives students a useful engineering lesson:

The best material is the one that performs well in its intended conditions.

A classroom comparison could ask whether the ranking of two polymer grades changes when both are tested in purified water and in the same salt solution. Use prepared liquids rather than biological fluids.

Take the investigation further

Tap water versus purified water

Compare identical samples using different water sources. Dissolved minerals may change swelling, so record the source and avoid assuming every tap-water sample will behave alike.

Speed versus capacity

Use separate samples stopped at different times.

A polymer that swells quickly is not necessarily the one with the highest final uptake.

Can the gel swell again?

Move a salt-treated gel into successive large volumes of fresh purified water.

Recovery can be investigated, although residual salt and handling losses complicate quantitative comparisons.

Do all ions behave alike?

Advanced students could compare carefully selected sodium chloride and calcium chloride solutions.

Calcium ions can produce stronger changes through specific interactions with the polymer. Research on sodium polyacrylate gels shows substantial volume changes following calcium addition under controlled conditions. ACS Publications

Equal masses of different salts are not equal chemical comparisons: molar concentration and ionic strength must be considered.

Practical care and disposal

Check the supplier’s safety information for the chosen product. Wear eye protection when handling the dry material and avoid raising dust. Keep polymer and gel away from food, young children and animals; do not taste them.

Work over a tray and clean spills promptly.

Collect gel and polymer residues for disposal according to the supplier’s instructions and local arrangements. The Royal Society of Chemistry’s technician notes advise eye protection, avoiding dust and disposing of gel residue with normal waste. edu.rsc.org

Do not wash polymer or gel down the sink or release it into the garden or waterways.

A pinch of salt, a much bigger lesson

The first impression is a material doing something astonishing. The deeper lesson comes when we change its surroundings and measure the consequences.

A few granules connect polymer chemistry with electrostatic interactions, osmotic pressure, experimental uncertainty and product design.

The investigation also challenges a familiar habit: accepting a headline performance figure without asking how it was measured.

At Hemel Private Tuition, this is the kind of practical I would use to move from curiosity to explanation. In the home laboratory, close-up views can reveal the changing gel while measurements give students evidence to analyse.

How much water can a superabsorbent polymer hold?

The scientifically useful answer is:

Under which conditions, measured by which method?

And that is where the real investigation begins.

 

At What Temperature Does the Air Start Making Water?

Measuring the Dew Point — When Does Invisible Water Appear?

A cold glass becomes wet on the outside. Where did that water come from—and can we measure the temperature at which it first appears?

Bring a cold drink into a warm room and droplets soon gather on its surface.

The water has not leaked through the glass. It was already present in the surrounding air as invisible water vapour.

With a shiny metal container, some ice and a thermometer, we can investigate the moment that invisible vapour begins becoming visible liquid.

This gives us an estimate of the dew point: a temperature with connections to misty mornings, wet grass, cloud formation, steamed-up windows and the comfort of a summer afternoon.

The title poses a deliberately intriguing question, but the distinction matters:

The air is not making new water. We are revealing water that was already there.

Why “The Humidity Is 70%” Does Not Tell the Whole Story

“The humidity is 70%” sounds informative.

But 70% at what temperature?

Relative humidity compares the actual water-vapour pressure with the saturation water-vapour pressure at the same temperature:

Relative humidity (%) = 100 x actual vapour pressure / saturation vapour pressure

The saturation value increases with temperature. Consequently, two days with the same relative humidity can contain very different amounts of atmospheric moisture.

Dew point asks a different question:

To what temperature would this air need to cool, at constant pressure and without changing its water-vapour content, to reach saturation?

At a given pressure, a higher dew point indicates more water vapour. It can therefore be more useful than relative humidity alone when comparing how moist different air masses are. Relative humidity remains useful, particularly alongside temperature. weather.gov

What Is Actually Happening?

Water molecules continually move between liquid water and its vapour. Cooling changes the balance between evaporation and condensation.

When a surface becomes sufficiently cold, water vapour in the neighbouring air can condense onto it.

At first, we may see only a faint misting of the metal. Later, small droplets become obvious.

That first persistent misting is the observation we want.

The whole room does not have to reach its dew point. A cold surface can develop condensation while the room air remains much warmer.

This also provides a useful opportunity to distinguish invisible water vapour from visible liquid droplets. The white cloud above a kettle consists mainly of tiny droplets, rather than the invisible gaseous water itself.

The Experiment: Find the First Trace of Dew

Suitable level: GCSE upwards, with extensions for A-level students.
Time: About 20–30 minutes for an initial investigation; longer for repeated measurements.

Equipment

• A clean, thin-walled, shiny metal cup or beaker.
• Water initially close to room temperature.
• Small pieces of ice.
• A digital thermometer or temperature probe.
• A stirring rod or spoon.
• A dry cloth, tray and notebook.
• Optional: a contact surface-temperature probe and a second thermometer for room temperature.

A smooth metal vessel makes the first change in appearance easier to see. Use a manufactured cup or beaker rather than a cut tin with a sharp rim.

Step-by-step method

1. Measure the room temperature. Keep the thermometer away from direct sunlight, radiators and the cooling vessel.

2. Prepare the vessel. Half-fill it with room-temperature water. Clean and dry the outside completely.

3. Choose an observation area. Watch the side below the water level, where the metal is being cooled by the water.

4. Position the temperature probe. Ideally, use a suitable contact probe against the outside near the observation area, leaving adjacent metal visible. For the simpler method, immerse the thermometer in the water without touching the wall or bottom.

5. Add a little ice. Stir gently without splashing. Allow the temperature to settle before adding more.

6. Slow down near the transition. Record the temperature at regular intervals and inspect the surface using good side lighting.

7. Record the boundary. Note the last temperature at which the surface is definitely dry and the first temperature at which a fine mist persists.

8. Repeat the investigation. Let the vessel return to room temperature, replace its water as necessary, dry the exterior and complete at least three trials.

Avoid breathing directly onto the observation area: your breath changes the air you are trying to measure. Handle the vessel as little as possible during readings.

Use a tray, wipe up spills and keep water away from electrical equipment.

Are we measuring the water or the surface?

The surface temperature is the relevant quantity.

An immersed thermometer gives an approximation. In a thin metal vessel with well-stirred water and slow cooling, the water and wall temperatures should be reasonably close, but they will not necessarily be identical.

Describe the result as an estimate of dew point, rather than claiming laboratory precision.

The metal-can-and-ice method is also described by the NOAA National Weather Service. Dry Bulb/Web Bulb/Dew Point

A Worked Example: The Moment Matters

These readings are illustrative, rather than measurements from my laboratory.

Surface temperature	Observation
15.0 C	Dry
14.0 C	Dry
13.0 C	Dry
12.5 C	Definitely dry
12.0 C	First persistent misting
11.5 C	Clear droplets developing

The transition lies somewhere between the final dry reading and the first misted reading.

A sensible report would be:

“The dew point was approximately 12 C, with condensation first observed between 12.0 C and 12.5 C.”

That interval describes the observation boundary. It is not a complete uncertainty calculation: probe accuracy, temperature differences and difficulty judging the first mist also matter.

If the room temperature was 20 C:

Dew-point depression = air temperature - dew-point temperature

Dew-point depression = 20 - 12 = 8 C

This tells us how far the room air would need to cool to reach saturation under the stated conditions.

A window surface could already be colder than that, even while the room remains at 20 C.

What If the Experiment Does Not Behave Perfectly?

Some of the most useful learning comes from understanding why an experiment gives an uncertain or unexpected result.

Condensation appears almost immediately

The starting vessel may already be below the dew point.

Bring the vessel and water closer to room temperature, dry the outside and restart.

Different parts of the vessel mist at different times

The wall may not have a uniform temperature. Ice pressed against one side can create a particularly cold patch.

Stir gently and observe a consistent area away from direct ice contact.

The surface is wet before cooling

Splashes can be mistaken for condensation.

Begin with a dry surface and add ice carefully.

No condensation appears

In sufficiently dry air, the dew point may be below the temperature you can reach using plain ice and water. A mixture containing both ordinarily settles near 0 C.

Record the lowest temperature reached and the absence of visible condensation.

That is a useful limitation of the method, rather than a failed investigation. For this introductory experiment, avoid adding salt to push the temperature lower.

The thermometer displays lots of decimal places

A reading of 12.37 C does not establish that the dew point is accurate to 0.01 C.

The observation and measuring system must justify the precision reported.

A precise-looking number is not necessarily an accurate measurement.

A Thermal-Camera Trap Worth Demonstrating

With a FLIR camera available, it is tempting to point it at the shiny vessel and read its temperature.

However, shiny metal has low emissivity and reflects infrared radiation from its surroundings. Its apparent temperature can therefore be misleading.

A suitable high-emissivity tape patch, allowed to reach the wall temperature and measured with appropriate camera settings, can improve the reading. Keep a separate bare-metal area for observing condensation and compare with a contact probe. Flir

This turns a measurement problem into another lesson:

An instrument produces a reading, but we still have to understand what it is measuring.

From a Metal Cup to the Morning Weather

Dew: a surface reaches the threshold

On clear nights, surfaces can lose energy by radiation and become colder than the air above them.

Grass and car roofs may reach the dew point even when the reported air temperature remains higher.

That helps explain why one surface is wet while another nearby is dry. Met Office

Fog: droplets suspended in air

Dew forms on surfaces. Fog consists of tiny droplets suspended near ground level.

A small difference between air temperature and dew point means that relatively little additional cooling is needed to reach saturation.

However, that difference alone does not guarantee fog. Wind, moisture supply and the way the air cools also matter. Radiation fog, for example, develops as the ground and neighbouring air cool, with light winds favouring its formation. weather.gov

Students could compare morning visibility with their measurements, while avoiding the conclusion that every small temperature gap must produce fog.

Clouds: the connection goes upwards

Rising air expands and cools. If it reaches saturation, condensation onto suitable particles can produce cloud droplets.

Our vessel provides a cold surface; the atmosphere can provide cooling through expansion. Both connect cooling with condensation, although the mechanisms and conditions differ. weather.gov

Frost: sometimes the water becomes ice

Dew can freeze after forming. Hoar frost can also develop when water vapour deposits directly as ice on a sufficiently cold surface.

For advanced students, distinguish saturation over liquid water from saturation over ice. Dew point and frost point are related, but are not identical below freezing. weather.gov

Why Does the Bedroom Window Get Wet?

Imagine room air with a dew point of approximately 12 C.

The window’s inner surface cools to 9 C.

That creates conditions for condensation on the glass, even if the room air is still 20 C.

Three useful questions follow:

• Can we reduce the water vapour being added to the room?
• Can appropriate ventilation remove moisture?
• Can we keep the surface warmer?

Cooking, showers and drying washing indoors can add moisture. Insulation and improved glazing can affect surface temperatures.

Simply warming the air lowers its relative humidity if its moisture content stays the same. It does not, by itself, remove that water vapour or substantially lower its dew point at unchanged pressure.

This makes an excellent connection between a classroom measurement and understanding a home.

Condensation is one possible source of damp; leaks and other causes need separate investigation.

Why Do Some Summer Days Feel Sticky?

A high dew point indicates moist air. That generally makes sweat evaporation less effective, reducing the cooling it provides.

Dew point helps explain why a warm day can feel oppressive even when its relative humidity is lower than on a cool morning. weather.gov

It is not a complete measure of comfort or heat stress. Temperature, sunshine, wind, clothing and activity also matter.

Dew point and wet-bulb temperature describe different things; they should not be treated as interchangeable.

Turn One Lesson into a Weather Investigation

Over several weeks, students could record:

Record	What to include
Date and time	Use a reasonably consistent time
Location	Indoors or outdoors; exact observation position
Air temperature	Temperature in C
Estimated dew point	Temperature in C
Temperature gap	Air temperature minus dew point
Conditions	Cloud, wind, fog, dew or window condensation

For outdoor trials, work in shade and avoid rain or direct sunlight on the equipment.

Useful investigation questions include:

• Does a higher dew point always accompany a warmer day?
• Is morning dew more common after clear nights?
• Does indoor dew point change after cooking?
• How closely do repeated estimates agree?
• Does the hand-measured estimate agree with a weather-station value?

In my home laboratory, this would make a useful link between a visible experiment, PASCO temperature logging and the Davis weather station.

We should compare like with like. An indoor cup measurement and an outdoor station reading describe different air. Even outdoors, location, timing and sensor uncertainty can account for differences.

A simple graph of air temperature and dew point over time can reveal much more than an isolated humidity percentage.

What I Want Students to Take Away

After around forty years of teaching, I value experiments that begin with something familiar and lead to a better question.

A wet glass is familiar.

Asking exactly when it becomes wet turns it into an investigation.

I would begin by asking students to predict where the water comes from. Then I would let them decide what counts as the first reliable evidence of condensation.

The most valuable discussion may concern a blurred boundary, an unexpected reading or disagreement between observers.

Those are opportunities to practise scientific judgement:

Identify limitations, improve the method and report evidence honestly.

At Hemel Private Tuition, this is the sort of practical I want to connect with the ideas students meet in their courses. Temperature, particles and changes of state become tools for explaining something real.

Conclusion: Make the Invisible Measurable

Dew point gives us a temperature at which invisible atmospheric moisture can reveal itself.

It connects a metal cup on the laboratory bench with a misty river, dew on grass, a cloud overhead and droplets on a bedroom window.

The equipment is simple. The thinking can go much further.

Next time a cold drink develops droplets, ask: what temperature did its surface have to reach before the water appeared?

Further Reading

• NOAA National Weather Service: Dew Point vs. Humidity — https://www.weather.gov/arx/why_dewpoint_vs_humidity
• NOAA National Weather Service: Temperature — Dry Bulb/Wet Bulb/Dew Point — https://www.weather.gov/source/zhu/ZHU_Training_Page/definitions/dry_wet_bulb_definition/dry_wet_bulb.html
• Met Office: Dew — https://weather.metoffice.gov.uk/learn-about/weather/types-of-weather/temperature/dew
• FLIR: How Does Emissivity Affect Thermal Imaging? — https://www.flir.com/en-asia/discover/professional-tools/how-does-emissivity-affect-thermal-imaging/
• NOAA National Weather Service: Radiation Fog — https://www.weather.gov/safety/fog-radiation
• NOAA National Weather Service: Cloud Development — https://www.weather.gov/source/zhu/ZHU_Training_Page/clouds/cloud_development/clouds.htm
• NOAA National Weather Service: Dew and Frost Development — https://www.weather.gov/source/zhu/ZHU_Training_Page/fog_stuff/Dew_Frost/Dew_Frost.htm
• NOAA National Weather Service: Dew Point Statistics — https://www.weather.gov/tbw/dewpoint

 

Conway's Game of Life — Can Mathematics Become Alive?

Four simple rules. An infinite chessboard. And behaviour nobody explicitly programmed.

Imagine drawing a few squares on a sheet of graph paper, applying four very simple mathematical rules, and watching the pattern apparently come alive.

Some shapes disappear almost immediately. Others remain unchanged indefinitely. Some alternate between different arrangements, while a few seem to travel across the paper as though they have somewhere to go.

Even more remarkably, certain patterns can produce an endless stream of moving structures.

There is no intelligence controlling them. No computer player is making decisions. Nobody has programmed the shapes to move, reproduce or interact.

Everything follows from four elementary rules.

Welcome to Conway's Game of Life, one of the most fascinating examples of how simple mathematics can produce extraordinary complexity.

It is also a wonderful way to introduce GCSE and A-level students to mathematics they are unlikely to encounter in their normal lessons.

1. A Mathematical Game Without Players

The Game of Life was developed by the British mathematician John Horton Conway and became widely known in 1970 through Martin Gardner's Mathematical Games column in Scientific American.

Despite its name, it is not really a conventional game.

There is no opponent, no score and no winning move.

Instead, it is what mathematicians call a cellular automaton.

That sounds complicated, but the underlying idea is remarkably straightforward.

Imagine a sheet of squared paper extending indefinitely in every direction.

Each square, or cell, has only two possible states:

• Alive — represented by a filled or coloured square.

• Dead — represented by an empty square.

Time advances in discrete steps called generations.

At each generation, every cell examines the eight squares surrounding it and changes according to the same rules.

The fascinating part is that the cells do not need to know anything about the rest of the grid.

They only respond to their immediate neighbours.

Yet when thousands of cells follow those rules simultaneously, remarkable structures emerge.

2. The Four Rules That Govern Life

Each cell has eight possible neighbours: above, below, left, right and the four diagonals.

The four rules are:

Rule 1: Underpopulation

A living cell with fewer than two living neighbours dies.

Rule 2: Survival

A living cell with two or three living neighbours survives into the next generation.

Rule 3: Overpopulation

A living cell with more than three living neighbours dies.

Rule 4: Reproduction

A dead cell with exactly three living neighbours becomes alive.

That is the entire mathematical system.

One particularly important detail is that all cells update simultaneously. We calculate the next generation from the existing one, rather than changing squares individually as we examine them.

Otherwise, the order in which we visit the cells would influence the result.

The rules are often abbreviated to B3/S23.

B3 means that a dead cell is born when it has exactly three living neighbours.

S23 means that a living cell survives when it has two or three living neighbours.

Nothing in these rules explicitly says that a structure should move, oscillate, grow indefinitely or behave like a miniature machine.

Nevertheless, all these things are possible.

3. Our First Experiment — Three Squares in a Row

We do not need a computer to start exploring.

Take some squared paper and colour three adjacent squares in a horizontal line.

Represent living cells by X and dead cells by a dot.

Generation 0

.....
.....
.XXX.
.....
.....

Apply the four rules to every square.

The cells at either end of the line have only one living neighbour, so they die.

The centre cell has two living neighbours and survives.

Meanwhile, the empty squares immediately above and below the centre each have exactly three living neighbours.

Both become alive.

Generation 1

.....
..X..
..X..
..X..
.....

Our horizontal line has become vertical!

Now apply the rules again.

Generation 2

.....
.....
.XXX.
.....
.....

We are back where we started.

The pattern will alternate between these two arrangements indefinitely, provided nothing else interferes.

This little structure is called a blinker.

It is an example of an oscillator, a mathematical pattern that repeats after a fixed number of generations.

The blinker has a period of two.

Period = number of generations required for a pattern to repeat.

Notice something else.

In every generation, the blinker has exactly three living cells.

Its position and total population remain unchanged, but its shape changes.

This gives students three different mathematical quantities to investigate: population, position and repetition period.

4. When Nothing Changes — Still Lifes

Not all patterns move or oscillate.

Try starting with four living squares arranged into a 2 × 2 block.

....
.XX.
.XX.
....

Each living cell has exactly three living neighbours.

Therefore, every cell survives.

No surrounding dead cell has exactly three living neighbours.

Consequently, no new cells appear.

The pattern remains unchanged forever.

This is called a still life.

It provides an excellent opportunity to ask students a question:

Can you design a different pattern that remains completely unchanged?

One example is the beehive:

......
..XX..
.X..X.
..XX..
......

Students can investigate whether every living cell has two or three living neighbours and whether any neighbouring dead cell has exactly three.

This turns an apparently recreational activity into an exercise in logical reasoning and proof.

Finding a stable pattern is one thing.

Proving that it is stable is rather more interesting.

5. The Glider — A Pattern That Travels

Now for something genuinely surprising.

Arrange five living cells as follows:

.X.
..X
XXX

This is called a glider.

Apply the rules four times.

The five-cell pattern will return to its original shape, but it will have moved one square diagonally downwards and to the right.

Continue the simulation and it will keep travelling.

The pattern does not literally contain the same five living cells throughout its journey. Cells die and new cells are born.

What moves is the arrangement of living cells.

This is rather like a wave travelling across water.

The water does not have to move along with the wave, but the pattern of disturbance does.

That distinction between the movement of matter and the movement of a pattern is a valuable idea in physics as well as mathematics.

How fast does a glider move?

The glider moves one square diagonally every four generations.

Its displacement per generation is therefore:

Horizontal displacement = 1/4 cell per generation.

Vertical displacement = 1/4 cell per generation.

Students familiar with Pythagoras' theorem can go further.

Diagonal displacement after four generations = sqrt(1² + 1²) = sqrt(2) cells.

Average diagonal speed = sqrt(2)/4, or approximately 0.354 cell widths per generation.

Strictly speaking, the glider's position is changing in discrete steps, not travelling continuously.

Nevertheless, we can use familiar ideas of displacement, direction and average speed to describe its behaviour.

This is a lovely example of GCSE mathematics appearing in an unexpected setting.

6. A Pattern That Produces Other Patterns

The glider is remarkable enough.

However, mathematicians discovered something even more extraordinary.

In 1970, Bill Gosper and colleagues at the Massachusetts Institute of Technology discovered a configuration now known as the Gosper glider gun.

It contains a structure that repeatedly produces gliders.

Every 30 generations, another glider emerges and travels away.

The original structure continues operating.

Given an infinite grid, an undisturbed glider gun can create an unlimited number of gliders over time.

That was mathematically significant because it demonstrated that a finite starting configuration could produce unbounded growth.

Conway had originally conjectured that this might be impossible.

The glider gun disproved that conjecture.

What I particularly like about this example is the way it illustrates the importance of experimentation in mathematics.

A mathematician can develop a hypothesis, explore examples, and sometimes discover a counterexample that overturns the original idea.

That is genuine mathematical investigation.

7. What Happens If We Start With Random Cells?

Instead of carefully designing a blinker or a glider, suppose we fill a large grid randomly.

Perhaps each square has a 30% probability of starting alive.

Initially, the pattern looks like scattered noise.

As generations pass, clusters form, collide and disappear.

Some settle into stable blocks.

Others become oscillators.

Occasionally, moving gliders escape from the surrounding activity.

What began as randomness can produce recognisable structure.

This is an example of emergent behaviour.

Emergence describes situations in which the interactions of relatively simple components produce complex large-scale behaviour.

We can find related ideas in many other fields:

• Flocks of birds, where individuals respond to nearby birds.

• Traffic flow, where simple driver interactions can produce congestion.

• Ant colonies, where local interactions contribute to organised activity.

• Crystal growth, where local physical processes generate large-scale patterns.

• Computer networks, where many independently operating devices form a larger system.

These are analogies, not examples of systems following Conway's exact rules.

Nevertheless, they raise an important question.

How much complexity can arise from simple local interactions?

For students who have been taught to associate mathematics primarily with equations and calculations, this can be a rather different way of thinking.

8. Can Mathematics Really Become Alive?

The title raises an interesting question.

Have we actually created life?

Not in the biological sense.

A glider does not have cells containing DNA. It does not metabolise, maintain a biological internal environment or evolve by natural selection merely because it moves.

The Game of Life is a mathematical model, not evidence that a computer simulation has become biologically alive or conscious.

Nevertheless, it demonstrates something important.

Apparently purposeful behaviour does not always require a central controller.

A glider looks as though it is travelling towards a destination.

A glider gun looks as though it is manufacturing objects.

Some elaborate configurations can even transmit information and perform computations.

Yet every event is determined by exactly the same four rules.

This brings us to a deeper question.

When we see complicated behaviour, how can we determine whether it arises from intelligence, design or simple underlying processes?

That is a question extending far beyond GCSE mathematics.

9. From Coloured Squares to Computers

One of the most extraordinary properties of Conway's Game of Life is that it is capable of universal computation.

With sufficiently elaborate configurations, its patterns can represent information and implement logical operations.

In principle, arrangements of cells can simulate a general-purpose computer, given enough time and space.

The idea is not that every glider is a little computer.

Rather, carefully constructed interactions between patterns can be used to implement computational mechanisms.

This connects Conway's creation with the work of Alan Turing and the foundations of computer science.

It also explains why the Game of Life appeals to students studying A-level Computer Science alongside Mathematics or Further Mathematics.

The connection is much deeper than simply writing a program to draw coloured squares.

The evolving squares themselves can be organised to perform computations.

The same mathematical world that produces a three-cell blinker can also support remarkably sophisticated logical structures.

10. A Classroom Investigation — Can You Predict the Future?

Here is an investigation I would enjoy giving to a student.

Start with the following L-shaped arrangement:

.....
.XX..
.X...
.....
.....

Without using a computer, predict the next generation.

Then predict the generation after that.

For the first generation, the correct result is a 2 × 2 block:

.....
.XX..
.XX..
.....
.....

The pattern then remains stable.

This is a useful exercise because it demonstrates that three living cells do not necessarily behave in the same way simply because the total population is identical.

Their arrangement matters.

Extension question: Can you find two different initial configurations with the same number of living cells that behave completely differently?

Try a horizontal line of three living cells and the L-shaped arrangement above.

One becomes a repeating oscillator.

The other becomes stable.

Same number of initial cells. Same rules. Different outcomes.

That is an important lesson about initial conditions and structure.

11. Building Your Own Game of Life in Excel

Students do not need sophisticated programming software to explore cellular automata.

Excel or another spreadsheet can be used.

A possible approach is to:

1. Create a 20 × 20 grid with living cells represented by 1 and dead cells by 0.

2. Calculate the number of living neighbours surrounding each cell.

3. Apply the Game of Life rules to calculate the next generation on a separate worksheet.

4. Use conditional formatting to colour living cells.

5. Repeat the calculation to observe how the pattern evolves.

For example, if a cell is located at C3 and its neighbours occupy the surrounding 3 × 3 range, the number of living neighbours can be calculated using:

=SUM(B2:D4)-C3

Subtracting C3 removes the central cell from the count.

The next generation must be calculated separately using the original values for all cells.

For an A-level student, the spreadsheet opens the door to investigating populations, proportions, repeated patterns and mathematical modelling.

There is an important practical detail: Excel grids are finite.

A limited grid needs clearly defined boundary conditions.

We might treat every cell outside the grid as dead, or construct a wrapping grid in which the left edge connects to the right and the top connects to the bottom.

These choices can change the behaviour of a pattern.

In other words, even the boundary conditions form part of the mathematical model.

12. An Even Better Challenge — Write the Program Yourself

For students studying Computer Science, writing a Game of Life simulator is a rewarding programming challenge.

Python is an obvious choice, although JavaScript, C# and many other languages would work.

The central algorithm is:

Create a grid
Choose which cells begin alive

Repeat:
    For each cell:
        Count its eight living neighbours

        If the cell is alive:
            Keep it alive if the count is 2 or 3
            Otherwise make it dead

        If the cell is dead:
            Make it alive if the count is 3
            Otherwise keep it dead

    Replace the old grid with the new grid
    Display the result

Notice that we must calculate the complete new grid before replacing the old one.

A common programming mistake is to change cells directly while counting their neighbours.

That introduces an unintended dependence on the order in which the cells are processed.

The challenge therefore teaches more than loops and conditional statements.

It introduces algorithms, arrays, state, simultaneous updating, debugging and software testing.

A particularly good test is to check whether the program correctly reproduces the blinker.

If it does not return to its original position after two generations, something is wrong.

13. A-Level Extension — Mathematics Behind the Simulation

For A-level students wanting more mathematical depth, there are several possibilities.

Population growth and decline

Let P(n) represent the number of living cells in generation n.

We can investigate how P(n) changes with time.

Does it:

• fall to zero?

• eventually become constant?

• fluctuate periodically?

• continue growing?

We can plot P(n) against n for different starting configurations.

A blinker has constant population:

P(n) = 3

A glider also has constant population:

P(n) = 5

However, a complicated initial configuration can produce a much less predictable population graph.

The important point is that a constant population does not necessarily imply a constant pattern.

Probability

Suppose every cell is independently alive with probability p.

For an interior cell on an initially random grid, the probability that exactly three of its eight neighbours are alive is:

P(exactly 3 neighbours) = 56 × p³ × (1 - p)⁵

The number 56 comes from choosing which three of the eight neighbours are alive:

8! / (3! × 5!) = 56

For a new live cell to appear, the central cell must also be dead.

Therefore:

P(birth at a cell) = (1 - p) × 56 × p³ × (1 - p)⁵

This simplifies to:

P(birth at a cell) = 56 × p³ × (1 - p)⁶

This formula describes the probability of a birth in the first update of the randomly initialised grid.

It does not automatically apply to every subsequent generation because neighbouring cells become statistically dependent as the simulation evolves.

That distinction is a valuable extension into probability and mathematical modelling.

Determinism

The Game of Life is deterministic.

If we start from exactly the same configuration and use the same rules and boundaries, the same sequence of generations will occur.

Yet complicated patterns may be extremely difficult to predict without actually running the simulation.

This illustrates the distinction between a system being deterministic and its long-term behaviour being easy to predict.

Students can investigate the sensitivity of particular patterns to small changes in their starting arrangements.

It is tempting to call every complicated outcome chaos, but mathematical chaos has a more specific meaning. Conway's Game of Life is best introduced as a deterministic discrete dynamical system with complex emergent behaviour.

14. Ten Mathematical Challenges for Students

For anyone who enjoys mathematical puzzles, here are ten investigations.

Challenge 1 — Extinction

Find the smallest non-empty starting pattern that completely disappears after one generation.

Challenge 2 — Stability

Construct a stable arrangement containing exactly six living cells.

Challenge 3 — The oscillator

Find a repeating pattern other than the blinker.

Challenge 4 — The glider

Demonstrate that a glider returns to its original shape after four generations, displaced diagonally.

Challenge 5 — Population

Find a pattern whose population increases during its first generation.

Challenge 6 — Initial conditions

Start with a blinker and add one extra living cell. Investigate how the behaviour changes depending on the position of that cell.

Challenge 7 — Collisions

Create two gliders travelling towards one another. What happens when they meet?

Challenge 8 — Randomness

Generate a 30 × 30 grid with 20% of its cells initially alive. Repeat with 40% and 60%. Compare the population after 50 generations.

Challenge 9 — Changing the rules

Modify the survival rule so that cells survive with two, three or four neighbours. How does the resulting world differ?

Challenge 10 — Design a machine

Investigate whether a system of moving patterns could be arranged to transport or represent a simple item of information.

These challenges range from accessible GCSE investigations to questions suitable for enthusiastic A-level Mathematics, Further Mathematics and Computer Science students.

Some can be answered using squared paper.

Others become much more interesting with a computer.

The important part is not simply obtaining an answer but explaining why the patterns behave as they do.

15. Why I Would Teach This at Hemel Private Tuition

Having taught mathematics and science for many years, I have found that some of the most memorable lessons begin with a question that does not initially look like an examination question.

Conway's Game of Life is particularly appealing because it is accessible without being trivial.

A GCSE student can understand the rules in minutes.

Within the same lesson, that student can explore coordinate grids, probability, pattern recognition, logical reasoning and mathematical proof.

An A-level student can go much further into discrete mathematics, combinatorics and dynamical systems.

A Computer Science student can create a working simulation, optimise the algorithm and experiment with more sophisticated cellular automata.

What makes this such an attractive teaching topic is that students can make their own discoveries.

They do not simply follow a worked example.

They ask questions, make predictions, test their ideas and discover that the results are not always what they expected.

There is also room for failure.

A carefully designed pattern may disappear after one generation.

Another may behave in a completely unexpected way.

That is not wasted effort. It is part of the investigation.

Although Conway's Game of Life is not generally a prescribed topic in GCSE or A-level Mathematics, exploring it helps develop skills that are valuable throughout those courses.

These include logical reasoning, clear explanations, systematic investigation, mathematical modelling and the ability to move between different representations of a problem.

At Hemel Private Tuition, I believe there is considerable value in occasionally going beyond the examination syllabus.

Examination success matters, but so does helping students recognise that mathematics is a creative and continually developing subject.

Sometimes the most effective way to build mathematical confidence is to let students explore something that does not have an obvious answer printed at the back of a textbook.

16. Try the Game of Life Yourself

The best introduction is to experiment.

Begin with three cells in a row.

Then try a square block.

Next, construct a glider.

Finally, create a random population and watch what happens.

Try making predictions before running the simulation.

An excellent starting point is the ConwayLife community website, which provides information, examples and links to simulation tools.

Useful resources include:

• ConwayLife: https://conwaylife.com/

• LifeWiki: https://conwaylife.com/wiki/Conway%27s_Game_of_Life

• Game of Life online simulator: https://playgameoflife.com/

For students who enjoy coding, the next challenge is to create a simulator rather than simply using one.

Conclusion — Complexity Does Not Always Need Complicated Rules

There is something wonderfully surprising about Conway's Game of Life.

Four simple rules applied to a collection of squares can produce stability, movement, repetition, interaction and structures capable of computation.

Nobody has to instruct the glider to travel.

Nobody tells the blinker to oscillate.

Nobody directs the glider gun to produce its next glider.

Those behaviours arise from the rules themselves.

That is one of the most important lessons mathematics can teach us.

Complex behaviour does not necessarily require a complex explanation at the level of individual components.

Sometimes extraordinary outcomes emerge from very simple foundations.

And perhaps the most exciting question is not whether Conway's mathematical cells are genuinely alive.

It is this:

If four simple rules can create such an astonishing mathematical world, what other remarkable things might we discover by experimenting with mathematics?

---

Beyond the Syllabus at Hemel Private Tuition

At Hemel Private Tuition in Hemel Hempstead, I teach GCSE Mathematics, A-level Mathematics, Further Mathematics and Computer Science, helping students develop both examination skills and a deeper understanding of their subjects.

My teaching combines worked examples, investigation, technology and practical demonstrations wherever these can help make difficult ideas clearer.

Lessons are available in person and online.

Mathematics is much more than passing examinations. It is learning to think, question and discover.

Philip M Russell — Hemel Private Tuition

https://www.hemelprivatetuition.uk