Kyle Rankin

Chief Security Officer
PGP ID: 0xB9EF770D6EFE360F
Fingerprint: 0DFE 2A03 7FEF B6BF C56F73C5 B9EF 770D 6EFE 360F
Librem Social

Latest posts by Kyle Rankin (see all)

When I read Shira Ovide’s piece “Apple Watch Is a Private Road” in the New York Times, I was reminded of my honeymoon in Cancun. Like many people who honeymoon in Cancun, we stayed at an all-inclusive resort along the ocean. In addition to enjoying the resort, we also rented a car so we could visit some of the surrounding areas, in particular the famous Mayan ruin Chichen Itza.

When we set off to visit Chichen Itza, the map featured a main route via a toll road and an indirect route that winded through the jungle and passed through a number of villages. I didn’t want to get lost, so I opted for the direct route through the toll road. The road was wide, freshly paved, no stop signs, and almost entirely empty–it was a smooth trip with the only stop being the toll booth we hit when we got on the road.

I had assumed the toll would be $1 or so–everything else up to that point had been relatively affordable in Cancun–but was shocked when I slowed down and discovered the toll was $10! This was about three times what the Golden Gate Bridge charged back then! I felt taken advantage of, yet once we got to the toll booth, there was no easy way to turn around or avoid it, so we just paid the fee and I blamed myself for being a dumb tourist who should have researched things better.

We spent the day in Chichen Itza and on the way back I vowed I would not be taken advantage of again. This time we would take the indirect, free route through the jungle. I was so glad I made that choice as I passed through one village after another and saw local people living their lives. While it wasn’t as fast or smooth a road as the toll road, I felt like less of a tourist on a curated tour of someone else’s property and more like I was seeing what “real” Cancun was like.

This was before GPS navigation was common so I carefully followed the paper map and wooden road signs as I moved from one village to another, always ensuring I would avoid the toll road. As I followed the signs to the final road that would take me to the resort, I found myself back on the toll road! I was infuriated but I realized I was so close to the resort, perhaps the toll booth was already behind me. It didn’t matter in any case because now that I was on this private road there was no way to get off. Sure enough, a mile later there was the toll booth. With no way to turn around and no way to get off the private road, I had no choice but to pay another $10 to get back to my room.

All Signs Point to Private Roads

The crux of Ovide’s article is that diversity and openness in computing has brought us many tech advances, but today so many of tech’s most recent advancements are closed and tightly controlled by the vendor:

Think about the last quarter-century of computers and the internet like a highway. The companies that made gadgets and software systems controlled the roads, and cars made by other companies drove (with some restrictions) on those roads. Computer devices would be meh if we couldn’t have access to a diversity of apps, websites and software — and vice versa.

But newer technologies for interacting online — smart watches like the Apple Watch, voice activated speakers, internet-connected televisions and robot-piloted cars — mostly pull us into digital features the device maker creates or tightly controls. They are more like private roads than the open highways of the smartphone and PC eras.

Ovide is right to point out that the recent trend is toward systems that are increasingly more closed. Unfortunately it’s only the latest in an ongoing cycle throughout the history of computing between open highways and private roads. Each swing in the pendulum moves from public, open, shared innovation that lays the open roads to private companies who use those public roads to build their for-profit toll roads. Those companies fight to ensure that no matter what signs you follow, you end up on their private road.

A Brief History of Open Computing

You can find many examples of this pendulum swing in the history of computing. Open software development in the late 1960s and early 1970s led to closed development on proprietary UNIX operating systems by the late 1970s. This spawned the GNU project to forward the goals of Free Software along with the advent of free BSD UNIX operating system variants. UNIX systems were among the first nodes on the Internet as it was being created with open protocols and standards in a collaborate academic environment. Because of this, even though Microsoft would come to dominate the home PC market in the 1990s, it didn’t share the same dominance on the server side, so when Windows PCs finally connected to the Internet, they had to do so with open protocols like DNS and TCP/IP instead of Microsoft’s proprietary NetBIOS protocol. Eventually even local Microsoft-dominated networks began to speak the open protocols of the Internet.

As more PCs got on the Internet in the late 1990s, many Internet Service Providers (ISPs) such as AOL fought to gain private control over this public network in the form of private roads to the Internet called “portals”–a custom web browser that provided a curated view of the Internet along with proprietary chat clients and games. Customers could chat and play games with each other as long as they used the same ISP. Fortunately outside of these ISPs, email and other web protocols were open, so you could use alternative ISPs and see a complete, unfiltered view of the web and use open protocols to chat and game with whomever you chose regardless of what ISP they used.

This open Internet accelerated collaboration on free software as well, resulting in the full set of GNU tools that led to the Linux kernel and the Apache web server, among other prominent free software applications. This spawned a new golden era of open software development and collaboration throughout the early aughts that laid the foundation of free software libraries and utilities that a majority of our web frameworks and cloud software is based on today. Many startups used these open libraries as a jumping off point for their own tools (even Google used Jabber/XMPP technology in initial versions of Google chat applications) .

As the aughts progressed, because the new tech giants were making their money by selling user data, the focus shifted back into creating portals. This time Google and Facebook were the dominant players and set out to ensure that you saw the rest of the web curated through their websites and chatted with your friends using their proprietary services. This “portalization” got worse as people shifted to using smartphones as their primary computers so that most apps became in essence a closed, mini-portal into the wider Internet you’d otherwise access from an open web browser. Now you have five different incompatible apps on your phone you use to chat with different people. Big tech companies can’t even manage to be compatible with themselves: Facebook alone owns three different incompatible chat apps–Google owns six!

Private Roads Paved With Bad Intentions

Smartphones provided tech companies with a blank slate to re-imagine how they approached software. The iPhone in particular rewrote the rules for how tightly a vendor can control a platform. While Apple has always held tight control over their platforms, in the past it didn’t get as much notice since they were a minority player compared to Microsoft’s dominance of the home PC market. Even with Apple’s tight control over MacOS, third parties could still write an application for a Mac without Apple’s permission and Apple customers could install and run it outside of Apple’s control.

The iPhone changed all of this. From the beginning, a developer must have Apple’s approval before a customer is allowed to use their application. To reinforce this control, Apple has advanced their security restrictions on the phone itself so that with each generation of iPhone and iOS, “jailbreaking” or “rooting” the phone so that you can run the software of your choice becomes more and more challenging. With the recent versions of the phone this is reinforced by custom, proprietary hardware and strong cryptography. Even hardware accessories for the iPhone require Apple’s approval or else you will get a warning that the device is not certified. These measures are always marketed as being for security from hackers and more recently also in the name of privacy, but from the beginning it has always been about ensuring that Apple can control which applications and accessories are allowed on the iPhone, in particular when those applications compete with their own offerings.

Seeing Apple’s success, competitors followed their lead so that now Android employs many of the same restrictions (again in the name of security and privacy) so that they can control the software that runs on Android devices. In the case of Android, this also ensures that cellphone vendors can not only pre-install their own vendor software that customers can’t remove, they have also made a side business out of selling software placement on their phones to third parties who often use the access to harvest customer data.

As Ovide’s article states, this closed approach has defined the next generation of computers (smart watches, smart speakers, smart TVs) already. There’s no attempt by these vendors to build open platforms when they design new technology. It’s rare when these platforms play well with each other. Will your smart watch work with your smart phone? Can you control your smart TV with your smart speaker? The only way to ensure technology is compatible is to buy it all from the same vendor. That’s by design.

Traditional computers are on the same path. Google has already extended this same approach–in the name of security–to Chromebooks to ensure that the only applications allowed on their laptops are those Google explicitly approves. Apple is moving quickly to extend these same security measures to their laptops as well. The goal of each of these vendors is to have no open highways, only private toll roads, leading only to their tourist attractions.

Where We’re Going, We Don’t Need (Private) Roads

Think about the future of computers over the next fifty years. Computers will become even more ubiquitous, not just embedded in all of the things around us, but embedded inside us. With advances in neural-computer interfaces, there is a high likelihood that we will be connecting computers directly to our brains within our lifetimes. Which tech company would you trust to control your neural implant?

If a computer can read and write directly to your brain, does it change how you feel about vendors controlling which software you can use or whether you can see the code? Does it change how you feel about vendors subsidizing hardware and software with ads or selling data they access through your computer? Does it change how you feel about government regulation of technology?

One promise of neural technology is to supplement humans with apps that provide instant skills and knowledge. Imagine Apple and an app company get in a dispute, Apple removes the apps from that company from their neural implant OS, and you lose the ability to speak Mandarin, drive, cook, play guitar, or write software?

We can’t accept being a tourist on tech’s toll road, the future demands open highways accessible by everyone, where you can freely go where you want, how you want. Now is the time to disrupt these closed platforms locked to and controlled by a single vendor. Openness and diversity are advantages, not weaknesses, and the future demands more openness, more collaboration, and more freedom and control given to individuals over their own computers.

The solution is to invest in technologies and companies that are building the open highways we need for the future. With Librem computers and the Librem 5 phone running PureOS, we are working to build platforms founded on free software and open standards that put users back in control. Help us build the future we all want to see.

 

Recent Posts

Related Content

Tags