The IE8 Standards Default is Wrong, Wrong, Wrong!
Kudos to the IE team for making the HTML5 DOCTYPE invoke "super, super" standards mode without the <meta /> element. Still unknown are the XML and XHTML MIME types, one can only assume those trigger super standards mode too.
The more I think about this idea the more I think about how there could be other ways, and how this idea is the wrong one.
One flaw that I can see, beside the obvious ones I've already talked about is that the new rendering mode switch is also an HTTP header. This being the case, what if proxies, ISPs, firewalls, or HTTP servers decide to, by default, include this HTTP header, X-UA-Compatible: IE=edge? Then none of the supposed benefits of backward compatibility remain because depending on who gives you your content, someone with an agenda could decide what rendering mode is best for you. For example, let's imagine that the Apache software foundation decides to, by default, include this new HTTP header, a scenario that isn't too far off considering how badly received Microsoft's proposal has been.
Another thing that strikes me as wholly unreasonable is how the IE team, and the few people that agree with them, seem to have their heels dug into the sand on this issue, and seem to see this <meta /> tag as the only possible solution to their quagmire. Also unnerving is why only a few industry experts were consulted and the community at large were not? I like Meyer, Zeldman, Koch, WaSP and all, but they don't speak for me, I didn't elect them to be my representatives, and I strongly disagree with their respective stances on this issue.
This idea means that IE7 never dies. Already we have forever set in stone and forever available the backwards compatibility mode that we all know and love, quirks rendering mode. There is no official cut-off date for this rendering mode, even when the web has advanced to HTML 6 or 7, we'll still see web pages from 1993 able to go on about their merry way in quirks compatibility mode.
Technology changes, it's a fact of life. In this technological adolescence, we aren't going to get things right on the first try, or the 2nd, or the 3rd, and we know that nothing will ever be perfect and not everyone will be satisfied. Does that mean that we keep our mistakes and our inferior ideas around and in-use forever? Soon analogue television will no longer exist in the United States. A clear, definitive cut-off date was established by the FCC. By 2009, analogue television will be a relic of the past. Some people aren't happy that this event is imminent, and are fighting it profusely, but the technological improvements and innovations that will come from this obsolete technology being put by the wayside are too great to ignore. And it doesn't hurt that the replacement services will be far more lucrative and efficient than those that came before it.
That is another flaw that I see with Microsoft's proposal, and, in fact, also with quirks rendering mode. There is no cut off date for backward compatibility, no point in time where the old, obsolete way of doing things can be formally abandoned and the world as a whole move forward. We can't say, update your content by this time, or risk it breaking.
With this frozen in time approach to rendering, we'll be teaching and coping with the ways of Internet Explorer past for long into the future, this significantly raises the bar for new web developers, who already must struggle with an immense amount of information.
Additionally, where is the incentive for these websites with horribly outdated code to move towards a more standards-compliant approach? Sure, I understand that some people don't care about standards, but, in my mind, if you can't be bothered to read the blueprints on how to build a website, then why should you be guaranteed your website work forever? The web was designed to be interoperable and platform neutral. Some companies didn't approach the web that way, they thought that the web was Microsoft's, and Internet Explorer's. So, Microsoft, you want to make a way for those vendor lock-in sites to continue to function? Then do it with Active Directory, and leave the web alone. Intranet sites are the anecdote of the day for this compatibility mode... use intranet tools to fix the problem.
For the web at large, if you're really hell-bent on compatibility, design a database just like you did for phishing. Start with your biggest customers, because apparently those are the only people you care about, and the only people who have any say in the future of the web. (Prove me wrong, I want to be wrong about that!) Put those hostnames in a database that triggers "backward-compatibility" mode. Give the hostname owners a way to remove their hostname from your database, or opt into better standards compliance through a code-based override.
I did everything I was supposed to do. I used only conditional comments to target specific versions of IE to hack specific IE bugs. I didn't use CSS parsing hacks because I recognized they were a really dumb idea from the start. I didn't blindly target IE without also targeting the version. I taught others to do the same in my CSS books. And yet Microsoft still expects me to make up for the incompetent assholes that didn't do it right. Fuck that. So, you painted yourself into a corner? Fine, I understand. Find a way to fix it that doesn't put standards in the back seat and that puts a definitive cut-off to the old way.
This approach facilitates vendor lock-in, preserves the old, tired way of doing things perpetually, and punishes the wrong people.
Comments
-
Posted: 02/07/08 10:02 pm Raphael P. Faria
I see no better words for our opinion.