Searching for validation
I’ve just implemented Safari’s new search input type after reading 37signals blog article “Searchin’ Safari’ which mentions Bartelme’s implementation. This is of course proprietary code from Apple and in no way does it validate, however, it does degrade well. Have I lost all my marbles and stepped over to the dark side of web development?
This site validates for the most part and because it’s written in xhtml strict 1.0 served as text/html I keep an eye on that it stays that way.
But of course that is folly. Web pages like these do not need to validate, not entirely anyway. Adhering to guidelines is one thing following them religiously is another. So why do I bother at all? Well we know that the cleaner your code is the easier it will be to build and maintain web sites. Some say that ‘interoperability’ is key for documents rendered in (x)HTML. That might be true but how much so?
I try to keep my eye on what happens to the content and how that is stored. Any (x)HTML that is used in the content itself will be tied to content and in a real and practical sense they will be tied together forever. Going back and cleaning up faulty code is the last thing you want to need to do. Conformance and interoperability suddenly become very important. The rest doesn’t really matter that much. Web site templates are of a temporary nature and last roughly up to 5 years, not much longer. The same is true for (x)HTML that is generated by the CMS. These two types of (x)HTML can be fixed and altered and need not validate as long the needs of business logic and user requirements are met.
This view is one of the reasons why I’m so vehemently opposed to WYSIWYG editors. These tools are far to whimsical and inconsistent to be let loose on a website’s main commodity, which is its content.