App Store control - the escape hatch

Google has suspended Element from the Play Store. This is obviously a wrong call. I like the analogy on their blog: this is like blaming Chrome because it enabled you to browse a dodgy website.

As I write this, the suspension remains in place and only brief details have been reported about how it happened. As a Matrix user I care about this specific case but in the wider picture the details matter less—it’s another instance of an app being removed for the wrong reasons, in a way that never seems to happen to big-business incumbents like Facebook Messenger¹.

The debate about app distribution methods is long and complicated. Some people cling to simple-sounding but myopic solutions like assuming that the free market is producing the best outcome for consumers, or that everyone should just de-Google and de-Apple their phones. A complete analysis of the relevant technical and social trade-offs could fill a book.

I want to pick at one small thread of that debate that is particularly relevant to Element. What’s special about this app is that it’s a gateway between the worlds of centralised and decentralised software.

In a recent article Cory Doctorow described our dilemma as needing to “ally yourself with a warlord.” He was referring to the big tech companies who promise to keep our data safe and our devices secure, so long as we give them complete control over our data and devices. There are privacy variations within these companies, which is often what we’re arguing about, but the fundamental relationship is always the same.

Okay, fine. Most people are stuck with a proprietary warlord on their consumer devices². What apps like Element do is enable us to break out of that environment and participate in a communications system that is decentralised and run entirely on open source software. I won’t rehash here why that’s a good thing. Suffice to say this is a key feature of this app and any others that let you out into the world of XMPP, IRC, and so on.

The app plays another role indirectly, which is at least as important: it offers an escape hatch from the “warlord”. In my younger days, like many hobbyists of the time I frequently flitted between Windows and Linux on my computers³. I quickly got into the habit of using software that was compatible with both the proprietary and free software worlds, like Thunderbird, Firefox and Pidgin so that I had minimal disruption every time I swapped out the underlying system.

There was a brief time when this problem seemed to recede. So many services were now available in web format that it didn’t matter what your OS was. You installed your web browser and got back to work. Now this has reached its inevitable conclusion and we see the same mobility problem where many people are tied to a particular cloud-based vendor.

The power of Matrix is that you can take your entire chat system and move it to a different warlord if the one you’re currently using gets too big for its boots. You can escape the warlords entirely if you’re prepared to wear the personal costs and security risks of that effort. You could even move to a new OS or community-hosted model that doesn’t exist yet. In short, you’re leaving your options open.

It is vitally important that these gateway apps are able to do their job, whatever the OS is. In today’s software environment they’re the only pragmatic solution for the vast majority of users. Google may say that Element’s suspension was a mistake. In truth we’ll never know for sure since it’s an internal decision. We shouldn’t take anything at face value. Gateway apps are under a lot of pressure because of the role they play.

If your chosen warlord forms a pattern of treating these types of apps poorly, perhaps it’s time to use your escape hatch. Hopefully this is unnecessary and the uneasy “alliance” will be stable for now. It’s possible that one day there will be no proprietary platforms willing to take on decentralised messaging apps. If this happens the open source community will be shut out until they can achieve meaningful market share with their own hardware and software platforms. Obviously this is a much harder challenge, especially without creating new warlords.

However we’re not there yet and we may not be for some time. For now, keep an eye on your escape hatches and look after them. You’ll probably want to make a move some day.


¹ I won’t count Fortnite here. A highly public protest against app store economics is a rather different circumstance.

² Or they might have a nett positive experience with the services the warlord offers. That’s a topic for another time.

³ Largely because the free software desktop was cool and enticing but I didn’t have the technical skills to pull it off; also because I was much more of a gamer back then and it was more convenient having everything on Windows.