more breakage.

while david teller quite reasonably analyzes that “web development is just broken”, other folks are trying to break the whole web by breaking the hypertext transfer protocol with so-called hash-bang uris:

“the problem with this is that the response is no longer representative of the content. i will receive response 302 found for something that is completely different than the actual content. this is where the problem lies to me.” — david coallier.

amongst that hash-bang gang is google.com with their hilarious attempt to index javascript-driven sites by adding a “highly intuitive” ajax url rewriting technique.

(well, maybe this technique is not so hilarious at all, at least it requires only one url entry point – well, maybe a few – for satisfying google’s crawler. on the other hand: why is not google using a headless browser themselves?)

anyway, where are we lost in web development and where are we going with it? did we somehow miss an important turn?

jeni tennison summarizes the situation very comprehensively, although i am left with suggestions for good practice that make browser hell sound like heaven again.

all i can see is paramount cluelessness and disapproval, neither a heads-up for current practices nor a de-facto standard, a reliable way of solving these issues.

of course, saying the web is broken actually means the web is undergoing fundamental changes. someday somebody might come and fix this. or, more likely, we will see an evolutionary outcome, the survival of the fittest. (practice, that is.)