Most school laptops aren’t replaced because they fail. They’re replaced because software moved on. Here’s why that matters — and what to do about it before the next refresh cycle.
Here’s a question I want you to sit with for a moment. The last time you replaced a fleet of school laptops, did they fail? Did the screens crack, the keyboards stop working, the batteries refuse to charge? Or did someone — a vendor, a department, a Microsoft update — tell you they were no longer supported?
I’ve been working in IT for 25 years. I’ve built secure networks for governments. I’ve worked in 27 countries. And in the last decade — working with Australian schools — I’ve watched the same scene play out over and over again. A trolley of laptops gets wheeled out of a classroom. They’re not broken. The CPUs still work. The RAM is still fine. The hard drives still spin. The hardware is, by any sensible definition, still working.
And yet the laptops are about to be retired. Why? Because the operating system stopped getting updates. Because a critical application now requires a newer version. Because the IT vendor flagged them as “end of life.” Because the software moved on — and the hardware was left holding the bag.
This is what I want every ICT manager, business manager, and principal in Australia to understand: we don’t have a hardware problem in our schools. We have a software one. And once you see it that way, almost everything about how we procure, manage, and retire devices in Australian education starts to look different.
The hardware is fine. The software moved.
Hardware doesn’t get worse with time. A laptop that ran perfectly well in 2020 still runs perfectly well today, in pure mechanical terms. The transistors haven’t degraded. The instruction set hasn’t changed. The clock speed hasn’t dropped. The device is, on any objective measure, the same device it was the day it came out of the box.
What has changed is everything around it. Operating system updates raise the minimum specs. Web applications get heavier — the same Gmail tab in 2026 uses several times the memory it did in 2018. Office suites that once fit on a floppy disk now run to gigabytes and require constant internet connectivity. Background processes — telemetry, indexing, syncing — quietly consume CPU cycles that used to be available for the actual task.
The result is a slow squeeze. The laptop hasn’t got worse. The world it lives in has got more demanding. And eventually a perception sets in — among teachers, among students, among IT staff — that the device is “old,” “slow,” or “tired.” It isn’t. It’s just been asked to do more, by software that wasn’t designed for it.
The device isn’t the bottleneck. The software is. This is the line I keep coming back to
Then comes the bigger problem, and this is where it shifts from inconvenient to expensive. Compatibility. One critical application — your assessment platform, your finance system, your single sign-on — drops support for the older operating system. Or a new operating system release simply refuses to install on the hardware. And the whole device gets retired, not because it failed, but because it was declared unsupported.
What we’re dealing with is not failure. It’s exclusion.
The Windows 10 moment.
If you want to see this dynamic in its purest form, look at what happened in October 2025. Microsoft ended free support for Windows 10. Devices around the world — still working, still capable, still doing the job — were instantly classified as “unsupported.” And the message to schools, businesses and households was simple: upgrade, pay for extended security, or accept the risk.
The numbers are extraordinary. Microsoft itself estimates that roughly 240 million PCs worldwide cannot upgrade to Windows 11 because they don’t meet the hardware requirements — most notably the TPM 2.0 chip [1]. In Australia, that’s somewhere in the order of three to four million machines. A significant share of them are sitting in schools.
These devices haven’t broken. They haven’t worn out. They’ve simply been ruled ineligible for the next operating system. And the path Microsoft offers, in order of preference, is:
- Buy new hardware. Replace the device.
- Pay for Extended Security Updates (ESU). Approximately AU$85 per device in year one, AU$170 in year two, AU$340 in year three — doubling annually [2]. Around AU$595 per device across three years.
- Run an unsupported OS. Which, post-October 2025, raises real privacy and cyber insurance questions.
The unspoken fourth option — run something else on the device — is the one most school IT leads have never seriously considered. And it’s the one that, in my view, changes the economics of school IT entirely.
What the experts are actually saying
Leanne Wiseman, Professor of IP Law at Griffith University, Brisbane: “The end of Windows 10… highlights the broader systemic issue of software-driven obsolescence. Increasingly brands are failing to properly support their products — from laptops to medical devices and assistive technologies. By stopping software support, this forces Australians to replace these products, increasing costs and creating huge amounts of waste.” [3]
This isn’t a fringe view. The Right to Repair movement in Australia, several state governments banning e-waste from landfill, and a growing body of academic work all point in the same direction: the problem isn’t the hardware. It’s the software lifecycle.
What it actually costs to keep replacing things.
Let’s do the maths on a typical Australian school. Imagine 400 students and 50 staff. A blended fleet of, say, 500 devices. Refresh cycle: every four years.
Each refresh cycle costs the school somewhere between AU$500,000 and AU$1 million, depending on device class and procurement model. That’s between AU$125,000 and AU$250,000 per year, just on devices, before you add software licensing, support contracts, ESU fees, or the cost of disposing of the old fleet. And much of that spend is driven not by hardware failure, but by software requirements that have crept upward year on year.
Now think about the carbon. According to the Tech Carbon Standard, an industry reference framework, around 80% of a laptop’s total lifecycle emissions are embodied carbon — generated during manufacturing, transport, and end-of-life, not during use [4]. Dell’s own published Product Carbon Footprint documents put manufacturing alone at 81% for an example laptop model [5]. So when a school replaces a working device after four years instead of running it for seven or eight, it’s writing off the majority of the carbon cost — and most of the financial cost — for no actual benefit in learning outcomes.
80%
Of a laptop’s lifetime carbon footprint is generated before it ever reaches a classroom. Replace too early, and you’re throwing most of that away.
This is the bit that should make every school business manager pause. We are paying — financially, environmentally, and in equity terms — for an upgrade cycle we didn’t ask for. The hardware was never going to fail in four years. The software just decided it should.
The five quiet gatekeepers in every classroom.
When I look at how the gap actually opens up in a school — between students who have access and students who don’t — it’s almost never a single decision. It’s five smaller ones, made over years, that gradually narrow who gets to participate.
1. Compatibility is the killer.
Devices aren’t broken. They’re “unsupported.” One library drops backward compatibility, one OEM stops releasing drivers, one vendor declares end of life, and the whole device is retired. Not because it can’t do the work — because something downstream decided it shouldn’t.
2. The refresh-cycle lottery.
Three years in one school, eight in another. Same curriculum. Same expectations. Completely different capability. In Bring Your Own Device (BYOD) classrooms it gets worse — two students sitting at the same desk can be working on completely different generations of hardware, with completely different software support.
3. “Works” versus “supported.”
A device can function perfectly well — boot, run programs, open files — but if it can’t run the required platform, it’s effectively excluded. The exam software won’t install. The assessment portal won’t load. The student is locked out of the task, not by their ability, but by their hardware.
4. Built for the newest, used by the oldest.
Software vendors test on current-generation hardware. They optimise for the latest CPU, the latest GPU, the latest accelerator. The lowest common denominator disappears from their roadmap. Which means schools running older fleets — which is most Australian schools — find their devices performing worse with every release, even when nothing visible has changed.
5. Updates as gatekeepers.
Every operating system update raises the minimum bar. Every raised bar excludes another tranche of devices. It’s gradual. It’s quiet. It’s relentless. And it’s the single biggest driver of forced hardware replacement in Australian schools today.
What the alternative actually looks like.
The most common pushback I get when I talk about this is: “Sure, that’s the diagnosis. But what’s the realistic alternative?” The honest answer is that there are three, and they’re not mutually exclusive.
First, extend the life of existing hardware by running lighter software on it. This is what governments in Argentina (Conectar Igualdad, Huayra GNU/Linux), Kerala, India (free and open-source software mandated across all government schools, 35 million people), Extremadura, Spain (70,000 secondary school PCs migrated from Windows to Linux, with audited annual savings of around €30 million), North Macedonia (Ubuntu-based desktops deployed nationally), Bolzano, Italy (custom FUSS Soledad Linux for the regional school system), and Mexico (Escuelas Linux, in 528 institutions across 114 countries) have already done at scale. This isn’t theoretical. It works. The savings are documented. The student outcomes hold up.
Second, separate the lifecycle of “the device” from the lifecycle of “the operating system.” A laptop is not the same thing as Windows. A laptop is not the same thing as macOS. A laptop is a piece of hardware that can run many different operating systems, including modern, secure, and well-supported Linux distributions built specifically for education. When the OS lifecycle ends, you change the OS. You don’t have to change the hardware.
Third, build procurement plans around longevity from the start. Buy hardware that can be repaired, upgraded, and reflashed. Buy software that doesn’t force you onto a treadmill. Plan device life around classroom utility, not vendor end-of-support dates.
What we build at Digital Education Systems
BURST.PLUS™ is our school operating system — a low-cost, modern, open-source platform designed to revitalise older school hardware. It’s what lets a school keep a 2019 laptop running like a 2026 laptop, on the same machine, with the same students, doing the same work.
It pairs with Crystal Eye EDU — defence-grade cybersecurity, built for schools, in partnership with Red Piranha — so longer device life doesn’t mean weaker security. See what we build →
What we’re really asking schools to take back.
Underneath all of this — the carbon, the dollars, the equity gap — there’s a more fundamental question. Who decides when a device in your school is finished?
At the moment, the honest answer is: not you. The decision is made by an operating system vendor in another country, a hardware OEM in another country, and a software supplier in another country. The teacher in the classroom, the IT lead in the office, the principal signing the procurement order — none of them are setting the lifecycle. They’re inheriting it.
That’s a strange position for a school to be in. And it’s a position that’s costing Australian education real money, real carbon, and real equity. The student in a low-budget regional school doesn’t have less talent than the student in a well-funded metropolitan one. They have less compatibility. They get bumped off the upgrade cycle earlier. They lose access to the platforms their peers use, not because of anything they did, but because of decisions made far away from them.
The hardware is fine. The software moved. And the question every Australian school can ask, right now, is whether the next refresh has to be a replacement — or whether it could be something else.
If you’re at a school and any of this sounds familiar, I’d genuinely like to hear from you. There’s no sales script on the other end of this email. Just a conversation about what your fleet actually looks like, what’s driving the next refresh decision, and whether there’s a sensible alternative on the table. Email me at it@diged.au.
References & further reading
- The PC Doctor (2025). Windows 10 End of Support 2025: What Australians Must Know. Estimate from Microsoft: ~240 million PCs worldwide cannot upgrade to Windows 11. thepcdoctor.com.au
- 4iT IT Support (2026). Windows 10 ESU pricing and migration: what Australian SMEs should do in 2026. AU$85 / AU$170 / AU$340 per device per year, ex GST. 4it.com.au
- PIRG (2025). Global voices speak out on the End of Windows 10. Featuring Prof Leanne Wiseman, Griffith University. pirg.org
- Tech Carbon Standard. Hardware Life Cycle Emissions. Around 80% of laptop life-cycle emissions are embodied. techcarbonstandard.org
- Tech Carbon Standard. Employee Hardware Example (drawn from Dell’s published Product Carbon Footprint documents). Manufacturing 81.4% of total. techcarbonstandard.org
- Educational Infrastructure Services Australia. Managing Electronic Waste in Australian Schools. Context on Australian school e-waste volumes. eisau.com.au
- EducationHQ (2025). Bridging the digital divide: used laptops helping Oz kids ‘locked out of opportunity’. 44% of Smith Family-supported students lack a connected laptop at home. educationhq.com