InformAction Forums

Giorgio Maone wrote:No, it's not a reasonable expectation on public sites.

I did not mean reasonable in the absolute sense, but that it might seem reasonable to a hypothetical dev because without NS it does hold (the exceptions you listed are rare, hardly anyone turns off session cookies, and the third party injection is definitely not the norm). You can't code on assumption of everyone upholding best practices. The web is no doubt full of code that works for the expected case but not for the edge cases. The reality is, a page relying on such an (albeit stupid) assumption will work for the overwhelming majority, until they install NS.

Then the goal should be for bad practices to be adjusted and corrected rather than everyone else succumbing to the same bad habits to make a subset of developers happy. Technology should never be dumbed down to the lowest level of skill developers but rather they should learn to better themselves or suffer the occasional break in their code. Whose to say in the process of learning why it doesn't work they don't get better understanding and write better code?

resolved in the latest dev build. 2.0.3.3rc1 anything earlier still had minor issues.

the goal should be for bad practices to be adjusted and corrected...

Don't I wish. Just look what my SeaMonkey UA is now (& by default - not my default, but by the SeaMonkey developers):

Mozilla/5.0 (Windows NT 6.1; rv:2.0b7pre) Gecko/20100915 Firefox/4.0b7pre SeaMonkey/2.1b1pre

The Kaiser gave in?
Makes me kind of sad, although I'm no SeaMonkey user (anymore).
Still, Gecko remains Gecko!