On Wed, 17 May 2006 07:09, you wrote:
> On Wednesday 17 May 2006 6:54 am, Nick Rout wrote:
> > Didn't we have this debate about 3 months ago? (Or was that another
> > list?)
>
> http://lists.ethernal.org/oldarchives/cantlug-0603/msg00245.html
>
> No definitive answer there though - although there were enough doubters
> to give the impression that it is not advisable.

Hello?

There *was* a definitive answer (and I said so at the time).

The standard, RFC2822, states that it is permitted.

However, it seems to cause problems if it's actually used, so the advice is 
not to.

In my opinion the problem is likely to stem from the fact that the "&" 
symbol is the HTML escape character, and "&" itself must be escaped.  
Because of the convergence of email and the web it becomes tricky to ensure 
that a string (such as an email address) should be preserved when typed 
verbatim in a mail client, or cut and pasted from a web page, or vice versa 
(i.e. displaying an email address containing "&" on a web page).

Since it is highly likely that an email address can appear on a web page or 
that an email message can contain HTML in the body it is awkward to keep 
track of whether an ampersand in the byte stream is an actual ampersand, or 
an escape character.

Ampersands cause weird things to happen in other places, for example, in 
dialog boxes or other GUI widgets the ampersand is an escape to mean 
"Underline the next character and treat it as a CTRL- keyboard shortcut".  
In fact, just typing this message (in KMail) I see that the window title is 
correct, but if I click on the menu bar at the bottom of the screen (where 
there is a little KMail icon representing KMail, and another representing 
this message window) the title of the message icon is 'Re: "" in an email 
address', where the second quote is underlined.

I could go on, but it seems that the ampersand is an 'overloaded operator' 
and attempts to interpret its meaning in any particular context are fraught 
with difficulty.

Andy

Reply via email to