Today is a good day to code

Microsoft IE 8 Multiple Standards Mode

Posted: December 31st, 1969 | Author: | Filed under: Uncategorized | No Comments »

Microsoft IE 8 Multiple Standards Mode

Picture of IrvinThere has been a ton of back and forth about the new IE 8 meta tag to allow developers to target a specific standards mode by the specific IE engine. The wisdom here, if you could call it that, is that a page should, technically, always look like it did when it was initially designed in IE if the user implements this tag. There is an obvious problem with this however.

It clearly subverts the need for developers to ever advance their knowledge of standards. Currently, for example, there is a ton of new functionality being suggested by the HTML 5 working draft. It will take 10 years for this draft to become a final standard document. In the interim, IE 8 may implement tags in a way based on the draft, and those methods could change as the standard advances. If the developer's pages continue to look the same in IE 8, 9 ,10, etc… Why would they ever change their implementations to match the standard. If they don't change from their implementation, how is it even a standard. It would depend on when a developer started developing. If they start at IE 9, they wouldn't know how IE 8 standard worked, they wouldn't know how IE 10 standards worked later, etc… If all browser makers followed this “wisdom” then we would have effectively 50 standards which is no standard at all.

It may be easy for Microsoft's developers to bolt on a new rendering engine with each IE release, and bloat the browser, but it definitely does not make life easier for web developers. In addition, how will mobile IE handle this, you can't put a 100 ~ 300 MB web browser on a portable device. Even if technology rushes ahead at a pace allowing this, the smaller more efficient browsers will still likely be faster and users will gravitate to those.

This is a bad idea and I hate it, I understand that designers want their pages to look the same for all time, and it is arguable that users want this as well, but if we go this direction, the W3C and What WG should just stop with standards and we should just forget about it.