sml schema regular expression performance

Hi, XML schema experts.
I have question regarding Regular Expression performance. I received the opinion that XML schema using very inefficient engine.
Below is a statement. Is it TRUE or FALSE and WHY


when it comes to regex, XSD has one thing VERY WRONG:

anchoring regexes is NOT a way to make them perform better!



Ever heard of "first character optimization" in regex engines? Pretty much all engines _on earth_ have that nowadays. And they perform _worse_ when the regex is surrounded with .*.



End of citation.

What is your take on this?



-Sergey