2014-05-25
Firefox, DRM, and reality
Today I ran across someone retweeting yet another petition asking Firefox to reject DRM or more specifically to reject EME. As this person doesn't usually seem to be given to supporting quixotic and crazy causes, the whole thing has pushed me over the edge about this issue. Apparently soft honesty is not working on people, so I'm going to try being blunt about the situation.
What I think most people who support this petition don't understand is that by asking Firefox to reject DRM and EME, they are asking Firefox to slit its own throat. At this point the all but certain result of Firefox rejecting EME is Firefox's browser share declining drastically, probably to the small single digits. One consequence of this would be Firefox losing its ability to influence the further evolution of the web (for exactly the same reason that almost no one cares what eg the Konqueror people think about such stuff).
Let's start with a basic fact: to most people, browsers are fungible commodities. Most people don't really care all that much about which specific browser they use provided that it can browse the web well. On the one hand this has been great for enabling alternate browsers like, well, Firefox; if you do the same job but better people can easily switch and will do so (Chrome is an excellent example of this). On the other hand this means that most people will promptly switch to another browser if their current browser stops meeting their needs, especially when good alternatives are widely available.
There are four important browsers today: Firefox, Chrome, Safari, and Internet Explorer. Three out of the four of these have embraced EME more or less wholeheartedly and so serious EME-enabled alternatives to Firefox are available on every single major platform it runs on, including Linux (in the form of Chrome). On all platforms except Linux, an EME-enabled alternative is the platform's default browser, making it especially accessible.
A non-EME-enabled browser does not meet many people's needs. Many people care about watching video and listening to music, sometimes rather a lot (enough, sometimes, to pay for it). If their browser stops doing this very few people will decide to stick to it for other reasons such as intellectual purity; most people will shrug and more or less immediately chuck it out in favor of some other browser that actually works.
The inevitable result of Firefox not supporting EME is that it will no longer be a browser that fully meets many people's needs. Almost all of these people will drop Firefox and switch to alternatives, which almost all of them already have conveniently at hand. Firefox's browser share will decrease down to people who don't care about EME'd content, people who don't have an alternative, and people who are willing to endure pain for the sake of either principles or Firefox's other attractions. I do not think that there are very many of these people.
(As a sign of how much people care about these sort of principles in the face of even weak attraction to an arguably better browser, look at how many people moved from Firefox (genuine open source et al) to Chrome (controlled and driven by a large advertising company that is not your friend).)
The EME DRM battle was lost no later than when the other three browsers embraced EME. To demand that Firefox continue fighting it is to demand that Firefox conduct a suicide mission.
You are of course free to ask that Firefox immolate itself in the name of intellectual purity, although I don't think that this is wise or that you're going to be successful (thankfully). But please understand and be honest about what you're actually asking for.
Computing has two versions of 'necessary'
In various fields of computing we often wind up either saying that something is necessary or arguing about whether it is necessary. One of the things that complicates these discussions is that in computing we have two versions or meanings of 'necessary', the mathematic and the pragmatic.
The mathematical version of necessary is minimalism. At its strongest, the mathematical 'necessary' means that this feature or thing is essential, that you have to have it, that things do not work without it. The pragmatic version of necessary is what I'll call economy of effort, the idea that something is necessary when it is the best way to achieve something. The pragmatic version of necessary is a humanist vision.
A mathematically necessary feature can also be pragmatically necessary; it is great when this happens because both sides get to agree. However it's common for pragmatically necessary things to not be mathematically necessary (at which point they often get called unneeded) and sometimes for the mathematically necessary things to not be pragmatically necessary (at which point they can get called too low-level).
A strong adherence to the mathematical version of necessary drives a lot of what I consider pathologies in computing. But a strong adherence to the pragmatic version of necessary also has its downsides, including clutter and incoherence when carried to extremes (which it often has been). And in general adherents of each side not infrequently wind up talking past each other.
PS: I suspect that you can come up with some examples of the mathematical necessary and the pragmatic necessary on your own, so I'm not going to fan the flames of argument by picking out ones here. There are some very obvious suspects among, eg, computer languages.
(I've touched on this idea before back here, among other entries.)