If you’re interested in going paperless, you probably think you need a scanner. It’s true that hardware scanners make turning multipage documents into PDFs very simple. But most of us don’t have easy access to a scanner.

What we do have are phones, and those phones have very good cameras. That’s where scanning apps come in.

These apps allow you to take photos of each page of a paper document, crop out the edges of the photo and straighten everything, then combine those photos into a PDF file. A scanning app is handy, but there’s a catch: a lot of the apps out there are a mess.

That’s what makes FairScan stand out. It’s an app for scanning documents using your Android phone that just … scans documents. That’s it.

FairScan creator Pierre-Yves Nicolas wrote in a blog post last year that he had previously tried several Android apps for scanning documents. “All of them exhibited behaviors that I certainly don’t want,” he says. These behaviors included obvious things like ads, hidden privacy violations, and shady practices such as storing your documents in the cloud—then using them to train AI—with only a tiny text prompt notifying you this is happening.

FairScan, which is both free and open source, doesn’t do any of that. It scans.

To get started, simply install the app. And yes, it’s Android-only for now, but you can download it from the Google Play Store as well as F-Droid, the repository for open source Android apps.

Get the document you want to scan ready, placing it on a flat surface in a well-lit room. Then aim your camera at the first page. A green box will surround the page—adjust until it’s surrounding the portion of the document you want to scan. Take the picture when you’re ready.

If you have more pages you can click the plus button to add them; this allows you to repeat the process with the next page. You can do this as many times as you want, allowing you to scan a multipage document.

When you’re ready, you can export the scanned pages to either a single PDF or multiple JPEG files.

There are a few things you need to keep in mind while scanning. First, lighting is going to matter a lot. You don’t want the shadow of your phone to be in the image, so make sure your phone isn’t positioned between your light source and the document you’re scanning. I find the app works best in a room with diffuse lighting, whether that’s multiple lights illuminating your work surface, or several windows letting in a great deal of natural sunlight. It’s also worth trying to get the paper document as flat as possible, to avoid distortions.

As part of a project to enable “seamless” frequency sharing and handover between satellite and terrestrial networks, Rakuten Mobile today announced that its joint proposal with the University of Tokyo for dynamic frequency sharing technology has been selected by the Japan Aerospace Exploration Agency (Jaxa) for its Space Strategy Fund programme.

Jaxa’s Space Strategy Fund is designed to support multi-year technology development projects by private companies, startups and academic institutions in three key areas: transportation, satellites and related technologies, and exploration. The programme aims to expand markets, address societal challenges and “pioneer new frontiers”.

It is also designed to ensure uninterrupted connectivity for applications such as autonomous vehicles and drones, regardless of location. Research and development for the satellite and terrestrial network integration project is eligible for up to ¥11bn (£53m) in funding. The project is planned to run between March 2026 and March 2031, and this period runs from the date of the funding grant decision through to the end of the fiscal year, in which the first stage-gate evaluation is completed.

The R&D behind the project aims to develop frequency-sharing technology for the handover between satellite communications and terrestrial wireless networks through artificial intelligence (AI)-based next-gen satellite communications.

The researchers note that in direct satellite-to-mobile communication, the same frequency bands are frequently shared with terrestrial networks, requiring interference management based on real-time conditions. Advanced correction is particularly crucial for voice and video calls due to Doppler effect and propagation delays.

Through AI-powered integration of satellite and terrestrial network operations, Rakuten says next-gen satellite communications AI will enable “seamless” connectivity, even as users transition between coverage areas. Additionally, it will look to demonstrate “strong” potential to support next-generation services that demand uninterrupted connectivity, including autonomous vehicles, flying cars and drones.

Rakuten Mobile and the University of Tokyo have a history of collaboration in AI-driven satellite communication research. This includes working on a virtualised Open RAN testing environment on the university campus and conducting joint R&D into ultra-wide internet of things (IoT) coverage utilising low Earth orbit (LEO) satellites.

The latter has seen Rakuten Mobile and the University of Tokyo build an IoT network using LEO satellite communications to establish use cases for IoT services using NB-IoT and LTE devices. This makes it possible to provide low-cost IoT services for various industries in locations that would typically be out of network coverage, such as mountainous areas, remote islands or on the sea.

Through the initiative, the two organisations say they will conduct verification tests for integrated satellite-terrestrial network operations using next-gen satellite communications AI.

The R&D will focus on developing next-gen satellite communications AI to monitor and control direct satellite-to-mobile communications and terrestrial networks sharing the same frequency bands. The goal is to avoid interference and improve operations and service quality for satellite direct communication.

The research will investigate AI-driven power control as well as interference management and traffic optimisation. Potential outcomes include development of an AI application to automatically power on and off satellite communication based on coverage data collected from both satellite and terrestrial base stations.

Additionally, the project will explore implementation of an AI application to manage interference mitigation, frequency adjustments and traffic load balancing. This is designed to ensure optimal connectivity during disasters, user transition and handover scenarios.

A short drive from London, the town of Potters Bar is separated from the village of South Mimms by 85 acres of rolling farmland segmented by a scribble of hedgerows. In one of the fields, a lone oak serves as a rest stop along a public footpath. Lately, the tree has become a site of protest, too. A poster tied to its trunk reads: “NO TO DATA CENTRE.”

In September 2024, a property developer applied for permission to build an industrial-scale data center—one of the largest in Europe—on the farmland. When locals caught wind, they started a Facebook group in hopes of blocking the project. More than 1,000 people signed up.

The local government has so far dismissed the group’s complaints. In January 2025, it granted planning permission. The following October, multinational datacenter operator Equinix acquired the land; it intends to break ground this year.

On a dismal Thursday afternoon in January, I huddled around a gate leading onto the farmland with Ros Naylor—one of the Facebook group’s admins—and six other local residents. They told me that they object to the data center on various grounds, but particularly to the loss of green space, which they see as an invaluable escape route from town to countryside and buffer against the highway and fuel stop visible on the horizon. “The beauty of walking in this area is coming through this space,” says Naylor. “It’s incredibly important for mental health and wellbeing.”

As the UK government races to meet the voracious demand for data centers that can be used to train AI models and run AI applications, similarly large facilities stand to be built across the country. For the people who live in closest proximity, though, the prospect that AI might buoy the economy or infuse new capabilities into their smartphone is thin consolation for what they consider a disruption to a countryside way of life.

Bonfire of Red Tape

Since the mid-20th century, London has been hemmed in on all sides by a nearly contiguous patchwork of land known as the green belt, made up of farms, forest, meadows, and parks. Under UK law, construction is only permitted on green belt land in “very special circumstances.” The aim is to protect areas of countryside from urban encroachment and stop neighboring towns from melding into an amorphous blob.

After the present government came to power in 2024, however, the UK introduced a new land classification—grey belt—to describe underperforming parcels of green belt on which construction should be more readily permitted. At around the same time, the government announced it would treat data centers as “critical national infrastructure.” Together, those changes have cleared the way for a raft of new data centers to be built across the UK.

As they attempt to develop models capable of surpassing human intelligence, the world’s largest AI labs are planning to spend trillions of dollars in aggregate on infrastructure. Across the globe, wherever new data centers are being built, developers are encountering organized resistance from impacted communities.

When the local planning authority approved the Potters Bar data center, its officers concluded that the farmland met the definition of grey belt. They also said their decision was colored by the government’s support for the data center industry. The benefits from an infrastructure development and economic standpoint, they concluded, outweighed the loss of green space.

“People have this slightly romantic idea that all green belt land comprises pristine, rolling green fields. The reality is that this site, along with many others, is anything but that,” says Jeremy Newmark, leader of Hertsmere Borough Council, the constituency that encompasses Potters Bar. “It’s a patch of very low-performing green belt land.”