Chad,



At 01:25 PM 7/3/2006, you wrote:
The vendor recommends a UTF-8 encoded file. So when I create an XML
file (before any XSLT is applied) I get just square boxes where those
math symbols and other items usually appear. I think this is why they
require the use of an entity. Does that make more sense or am I still
missing the point?

They're not mutually exclusive: you're making sense, but still might 
be missing the point. :-> It could be that the character is correct, 
except the interface you are using to inspect it (browser, editor or 
whatever) has no glyph to show the character, so it shows you a box. 
This doesn't necessarily mean the character has become a box 
character in the data; it could still be the correct character, just 
hard to see and hard to check for correctness.



In such a case an entity may be a more legible representation of the 
character, but as you can see there's also a price to be paid for 
demanding it. The ideal would be to use tools capable of showing you 
what the character actually is. In an imperfect world, requiring 
entities or numeric character references might indeed be a pragmatic 
way of evading the problem or the appearance of a problem. Whether 
it's worth the tradeoff is another question.



Cheers,
Wendell