Re: Using DOM to replace media attribute in the link tag on page load

David Perrell wrote:
>  My criteria is: when there are UAs that no longer accept script type
>  as "text/javascript" OR when there are no longer UAs in common use
>  that do not recognize the RFC4329-sanctioned script types.

The first criterion is effectively saying “never”; I don’t expect 
text/javascript to go anywhere even if the application/* media types 
become popularly used simply due to backward compatibility concerns. I 
don’t expect the latter to happen for years, particularly if Internet 
Explorer 8 or later end up not supporting RFC4329.

David Perrell wrote:
>  You ask if you can use the new MIME types without hurting anything
>  while I ask "Why should I change what works to *partially* satisfy a
>  specification that doesn't address current conditions?"

I’m not quite sure what you mean by “partially” or “doesn't address 
current conditions”. If, by those, your message is that something 
shouldn’t be used until it can be freely used in all commonly used 
browsers (i.e., “doesn't address current conditions”) and that 
workarounds (i.e., “partial” adherence to RFC4329 due to the IE 
workaround) mean that a solution shouldn’t be used, I simply have to 
disagree.

Working around lack of browser support and/or being unable to fully 
support a standard until support improves or an alternative is offered 
is nothing new on the Web. (Of course, I realize that that latter part 
can be used to justify both of our chosen methodologies; I guess we’re 
simply in disagreement over how to tackle this issue. Still, I don’t 
think that pointing out that X is deprecated and suggesting how one 
might use Y despite lacking browser support is a bad thing.)

Anyway, I suppose that I should answer your question: “Because you 
prefer to follow published standards (assuming that adherence is 
practical).”

David Perrell wrote:
>  That doesn't mean one of us is wrong (CCs aren't evil!), but I don't
>  think assuming that all future versions of IE will support
>  "text/ecmascript" is good idea.

I maintain a strong doubt that Microsoft would drop support for the 
text/ecmascript MIME type considering their continued emphasis on 
backward compatibility. As I mentioned previously, the occurrences of 
*/ecmascript could be changed to */javascript if preferred; doing so 
would eliminate this particular concern altogether.

David Perrell wrote:
>  I would AT LEAST wait until a conforming version of IE is released
>  and limit the IE-specifics to lesser versions.

I don’t see what difference waiting would make. You’ll still need the 
CCs (or some other mechanism) to deal with non‐conforming versions of 
Internet Explorer, so the amount of work required would not be 
significantly lessened.

— Patrick Garies

Received on Sunday, 27 July 2008 20:41:40 UTC