Tim Bray wrote:
New information has become available. You may feel that the new information is not significant, but I disagree. Working Groups who refuse to revisit decisions on the grounds that they've already been taken produce things like W3C XML Schema. The WG is perfectly within its rights to decide that they want to ignore this problem, in which case I will be perfectly within my rights to write an I-D describing an extension to solve it, which I will make as consistent as possible with the rest of the APP. -Tim
Now that you've invoked the "W3C XML Schema" name I feel like, while this issue deserves some discussion, we shouldn't use that comparison.
In that group they went down the road of way too many features to support every single use case they could think of... and so strange practices with the use of DTDs. The result was a great deal of complexity (see redefine). So, W3C XML Schema is a product of both a difficult process as well taking on way to much in terms of both features and how they are represented. In fact, I like W3C XML Schema, but I do so by ignoring all the krusty bits that are full of pits with pointy sticks. It is always a good design principle to have "one way to do something". This does go against that unless we are clever about how we might support alternatives. In the end, there may not be a clever way. ...there I will always vote for simplicity (and tell me if I'm not). :) --Alex Milowski
