Cart
[Home] [By Thread] [By Date] [Recent Entries]
On 2/4/2014 1:00 PM, Michael Kay wrote: OK, perhaps that was a flip remark - it sounded good to me when I first wrote it! I was thinking more about discoverability than about usefulness, although now that I think of it is only a little less likely that someone would discover xml:id today than it was a few years ago.On 4 Feb 2014, at 17:21, Michael Sokolov <msokolov@s...> wrote:However, I think it's likely that no one who isn't already using xml:id today is likely to start using it tomorrow.Why? I find it very useful, and have applications where I've stopped using DTD-defined ID attributes and use xml:id instead. It means that id() in a stylesheet works whether or not you do DTD validation, which is much more robust. I guess I also still bear scars from my previous encounter with it (see http://markmail.org/thread/vabspbbgsylrgr6e) when I was trying to get it to work with xpointer and stumbled over less-than-ideal support from the parser (I ended up needing to define xml:id in a DTD). -Mike
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] |
