> Rather the author seems to be trying to redefine "operating system" to include things that have never been a part of such a term before, like online message routing and mapping services. Since when do we define "operating system" as including libraries for specific online services? User interface libraries, kernels, drivers, multimedia frameworks, sure ... but mapping services?
Mobile devices have an expectation of having more built-in features than other devices, so it makes sense that we would expand what we define as an OS.
Not to mention that the map services are implemented as core pieces of the OS within Android, so it's very much like having a proprietary libc with other proprietary libraries. You could create a program that doesn't use libc, but nobody does that because of duplicated effort (and the fact that Google won't hold your hand if you don't use their libc).
So I would very much argue that Android is on the path to becoming even more proprietary (it always was proprietary, just because of the fact that effectively all phones require blobs in the kernel as well as other firmware that is updateable but is proprietary).
Mobile devices have an expectation of having more built-in features than other devices, so it makes sense that we would expand what we define as an OS.
Not to mention that the map services are implemented as core pieces of the OS within Android, so it's very much like having a proprietary libc with other proprietary libraries. You could create a program that doesn't use libc, but nobody does that because of duplicated effort (and the fact that Google won't hold your hand if you don't use their libc).
So I would very much argue that Android is on the path to becoming even more proprietary (it always was proprietary, just because of the fact that effectively all phones require blobs in the kernel as well as other firmware that is updateable but is proprietary).