a) The Membership of the W3C have decided they want to work on DRM.

b) Like most industry consortiums, the W3C is ultimately beholden to its (industrial) Membership. (And it's not clear they can refuse an organisation from joining as a Member without opening themselves up to allegations of being a cartel and the legal complications that would involve.)

A lot of this comes down to the relationship between the W3C and its Member organisations, and whether the W3C can refuse to work on something its Members want to.

There's also some of what you alluded to, which I will call c) A number of Member organisations have made it clear that they will work on this in some public forum regardless of what that forum is.

Now, from a purely pragmatic point-of-view, what is gained by the W3C refusing to work on it? Apple, Google, and Microsoft will still ship DRM modules; the web will still start relying on DRM modules existing within browsers. The outcome is entirely unchanged, as ultimately because of c we've ended up with an interoperable API from JS one can use to deal with DRM modules.

Refusing the venue is purely making a political point, it doesn't change the outcome. Now maybe that political point is a goal in and of itself, but given most of the arguments people make against DRM I'd suggest the goal here isn't a political point but rather reduction of reliance of DRM on the web.

By refusing to work on it, you upset the Membership (because you're going against them), jeopardising your own future (because a industrial consortium is nothing without Members), and not changing the outcome.