As a technical editor at A List Apart, I’ve been fortunate enough to have a draft of Aaron Gustafson‘s latest article sitting on my laptop for the past couple weeks. While I wasn’t involved with editing the article, I have been thinking quite a bit about since it went into production. While Eric’s made some sort of peace with the proposed “version targeting” mechanism, I’m still feeling as uneasy about it as I did when I first read the piece.
I'm convinced the premise for version targeting is supremely flawed. If I’m reading Aaron’s article correctly, the argument for it is predicated on a logical fallacy:
DOCTYPEswitch is used on invalid pages.
- Some invalid pages are broken.
- Therefore, the
DOCTYPEswitch is broken.
IE7 didn’t, as Chris Wilson suggested, “break the web,” and neither did
DOCTYPE switching. It’s like Drew said: this myth of broken sites isn’t enough to justify a new paradigm for developing websites, and breaking the promise of forward compatibility.
DOCTYPE switching isn’t broken: it’s the publishing processes and software that allow us to publish invalid, untested HTML to the web.
And here’s the important bit: None of this is Microsoft’s responsibility. Nor Firefox’s. Nor Opera’s. The burden is upon developers and designers to comply, test, and validate. As it stands, version targeting won’t improve the quality of our code, or improve the current state of future proofing. Rather, it’s a software patch intended to fix broken human behavior. We need to fix the problem at hand, and a new
meta element isn’t the answer.
I have other problems with this approach (ranging from implementation, to wondering exactly who will maintain a registry of user agent acronyms, to the WaSP’s involvement in this whole thing), but there’s already a fair bit of discussion along these lines. And frankly, I still need to think more about this: two weeks hasn’t been enough to really flesh out all my concerns. And hell, maybe I’ll come around on this whole issue—I don’t know.
But I do know that at the moment, I’m still uneasy.