Re: The mapping of phi

Robert Miner wrote:
> Sure.  I think most of us understand about Unicode.  
> 
> But since the whole point of having two code points is that the difference between a straight phi and a curly phi carries semantic meaning in mathematics, it doesn't really help the reader if the 03C6 sitting there in the DOM is unambiguously declared to be the phi symbol character by Unicode when what the reader actually sees on the screen randomly switches between the straight and curly glyphs depending on the font.  
> 
> Having spent far more of my life than I want to debugging phi problems in publishing workflows, I can assure you it is a real world problem, even if the theoretical foundation is clear.


It is indeed a real world problem, but the problem is in the font, and 
the font is what needs to be fixed. Fixing the non-broken parts like 
Unicode, HTML, or DOM just locks in the brokenness and assures that 
nothing will ever work. If we leave HTML, Unicode, and DOM alone (on 
this issue) then eventually the fonts will get fixed or replaced, at 
least within the limited community where this matters.


-- 
Elliotte Rusty Harold  elharo@metalab.unc.edu
Java I/O 2nd Edition Just Published!
http://www.cafeaulait.org/books/javaio2/
http://www.amazon.com/exec/obidos/ISBN=0596527500/ref=nosim/cafeaulaitA/

Received on Wednesday, 2 April 2008 16:37:08 UTC