David Baron's Weblog

In a recent interview Andy Clarke said:

The interesting thing, though, and this is the wider issue, is that there’s no strategic plan, either from the browser makers or, most importantly, from the CSS working group, where they plan in a timetable implementation of these new features. Now we’re not talking about the big design of CSS, and when are we going to get new layout features and things like that, but simple things like for example CSS columns? Webkit implements CSS columns, Mozilla implements CSS columns, but they do it independently, they do it when they want to on their own timetable, and what I’d really like to see is for these browser makers to get together and say “You know what in September, we’re going to introduce these columns across the board, and in October, or in our next release, we’re going to implement this across the board.”

I agree that this would be really useful.

The CSS Working Group actually went through a prioritization exercise last year. (You can see the results in the group's new charter. You can also see Mozilla's feedback here; I'd note that this was from a year ago, and there are a bunch of things I'd change if I were responding to the same questions today.) But this exercise had two problems as far as being useful for satisfying the goals Andy set out above.

First, all the feedback was secret: it was provided only to the chairs of the group, who then computed the results, rather than allowing discussion and negotiation. I think this makes it harder for people to accept the results that they don't agree with.

Second, those involved in the prioritization exercise weren't just Web browser makers. They were also companies that use XHTML + CSS as a mechanism for sending content directly from mobile phones to printers, companies that use XHTML + CSS as a mechanism for formatting to PDF, and companies involved in the CSS working group for unknown reasons (which include both software companies and others).

So, given the results of the CSS working group prioritization, I don't know which things that I didn't consider important were marked as more important because other Web browsers want to implement them (which might mean I should increase the priority even though I don't like them, as I did for multiple background layers), and which were marked as important because of the preferences of companies whose priorities Mozilla shouldn't care about coordinating with.

I'd love to have a forum for discussing this prioritization with other browser makers so that our work can be used by Web authors sooner and more reliably, and I think it would be better if it weren't in a pay-to-play environment like the W3C. However, it also needs to be a setting where Microsoft is willing to participate, and the W3C is the only setting I currently know of where they are.