L2/03-275

Date/Time:    Thu Aug 21 06:35:36 EDT 2003
Contact:      peterkirk@qaya.org
Report Type:  Technical Report or Tech Note issues

Feedback on proposed draft UTR#31:

I'm a little concerned at the implications of counting zero width  
characters like LRM and RLM as white space. They can easily find their way  
unnoticed into the middle of patterns e.g. when copying from a text which  
has added these characters to ensure correct directionality. I wonder if it  
might be better to add a new category of ignored characters, such that one  
of these found on its own doesn't count as a separator but it is ignored  
i.e. treated as part of the white space if found adjacent to white space.  
Of course the details of this need a little more thought, e.g. does one of  
these actually count as part of the pattern, but I hope you see what I am  
getting at.

Peter Kirk