The State of “Openness”

Benedict Evans once talked about a sort of “openness Tourette’s Syndrome” that occurs whenever people discuss Apple’s platforms vs. competitors. Basically, it goes like this: someone mentions how good an Apple platform is, and then someone else says, “Yeah, but Android is open.”

There’s a pleasant sort of fiction that is promised with “open” that simply isn’t a viable reality for most people. I’ve heard salespeople use this in retail stores, and I’ve heard IT professionals use this when offering Android to their clients. This demonstrates a fundamental misunderstanding of what Android’s various meanings of “open” are. The type of “open” that people are typically referring to when they use that word is actually conflated with “extensible” or perhaps “has relaxed security”, which are very different things than the “open” that Android was conceived with.

Android’s initial form as a project was open-source, and the Android of today is still technically “open-source” but, due to its reliance on Google’s services and cloud features, the current version of Android that comes loaded on many phones is not nearly as “open” as many would have you believe. Would you like to use another mapping service? How about something other than Google Now? Can you use the features of the home screen without being tied to Google services? Sadly, no.

That doesn’t mean that one couldn’t install the Android Open-Source Project’s fork of the operating system, but it means that the marquee features of the operating system, the things that Google and Android fans like to wave in the air, are inherently tied to Google and make it very difficult to use non-Google-developed operating systems.

Instead of this word “open”, then, let’s use the word “extensible”, since that more accurately reflects the Android OS’s ability to facilitate communication between apps, and to allow developers to build software that adds functionality to the OS or preexisting apps.

The problem with Android up to this point has been that security has not been (or at least hasn’t appeared to have been) a priority for developers or users. While I could try to offer up what I see as reasons for this type of behavior (laziness, “Accept” fatigue). I may be wrong on this, but from what I’ve seen, Android are more than willing to download apps and grant them almost completely exclusive access to their mobile device without really thinking through the ramifications.

Apple has avoided this for many years by sandboxing their apps and keeping inter-app communication on the back burner until they developed a way to allow apps to communicate effectively without sacrificing a user’s privacy or requiring them to grant unnecessary privileges to an app that really shouldn’t require it. Naturally, this came at a cost. For years, iOS users have not been able to install 3rd-party keyboards or send information between apps in a way that was “easy” (to be fair, the iOS “Open In…” functionality has allowed users to send documents and files between applications for some time, but required a degree of savviness from users that was sometimes lacking).

Now that Apple has introduced the ability for developers to create “Extensions”, however, that gap has very quickly been bridged, and iOS 8 will allow developers to create new ways for their apps to interact. Some may argue that Apple’s approach may differ from Google’s, but the end-user result is basically the same: a person will be allowed to install and use third party keyboards, send information between apps, and interact more directly with the data in other apps.

What I’m interested in seeing now, however, is what the conversation will center around now. For many years, Android users have told me that Android has been superior because of its customizability. When I would press these users to provide me with more information about what “customizability” means, they would often say two things: support for third party keyboards and home screen widgets.

These two “features” of the operating system, in my opinion, are not very important, and would often open a user’s device up to instability and/or unnecessary resource usage. I have used Android devices, and I have seen the home screen widgets for the apps that I use the most, and there is no version of reality in which the widgets provide a superior experience to using the app. Again, this is my experience, and maybe there are some people who really enjoy looking at two lines of their mail on their home screen underneath a line and a half of their upcoming calendar events, and not really being able to meaningfully interact with either until they open the app anyway.

Third party keyboard support has also perplexed me, but I can understand the utility for people living outside the United States, for whom third party keyboards can offer substantially improved text entry. That being said, none of the Android users that I discussed this with lived outside the United States, so it seems that their argument is a moot point, or at least purely subjective.

Thus, it seems to me that the discussion of Android as an “open” system (again, in the way that most people understand the term “open”) has lost much of its value. Android as an “extensible” operating system has also lost much of its value, as well (at least as a marketing ploy) in light of the new functionality of iOS 8. How, then, should we be defining “open”?

When we look on the post-PC landscape and see two operating systems that allow their users to interact with their data similarly, and enter information into their devices similarly, and allow applications built upon their platforms to communicate similarly, how should a person decide which device to use? Perhaps the discussion shouldn’t be centered around questions like “Which device lets me tinker with system files?” or “Which device will allow me to inadvertently break things if I wanted to?”, but should really be “Which device is better for humans?”