First off, the Katakana middle dot is distinctive enough that I see no problem with visual confusion. Take a look at:<br><br><font size="4">a ・<a href="http://c.com">c.com</a><br><font size="2">vs</font><br><a href="http://a.c.com">a.c.com</a></font><br>
<br>The width and positioning is far different.<br><br>There are other dot-like characters that are far more visually similar to dot, like Arabic zero.<br><br><font size="4"><a href="http://xn--ngbazb1bc2jd8q.com">عربي٠عربي.com</a><br>
<font size="2">vs</font><br><a href="http://xn--ngbrx4e.xn--ngbrx4e.com">عربي.عربي.com</a></font><br><br>But more importantly, there is a real lack of data presented for these kinds of positions. When excluding characters that are in common use on the basis of visual confusability, such as Katakana middle dot, let's see some real data on what a difference this would make in overall visual confusability of characters. Of all of the visually confusable characters in PVALID, what would be the percentage difference but adding or removing Katakana middle dot? And why do people think this can't be handled by exactly the same mechanisms that programs have to handle the visually confusable characters that <b>are</b> PVALID.<br>
<br clear="all">Mark<br>
<br><br><div class="gmail_quote">2009/4/7 John C Klensin <span dir="ltr"><<a href="mailto:klensin@jck.com">klensin@jck.com</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>
<br>
--On Tuesday, April 07, 2009 07:59 -0700 Mark Davis<br>
<<a href="mailto:mark@macchiato.com">mark@macchiato.com</a>> wrote:<br>
<br>
> On what basis?<br>
<br>
It is a punctuation character. It looks enough like a<br>
label-separating period in contexts where middle-dots are not<br>
used that it would likely be mistaken for a period in those<br>
contexts. Such a mistake is fully as dangerous as a character<br>
that looks like a slash. If it were possible to live with a<br>
restriction that required at least one Asian-Japanese character<br>
(Hiragana, Katakana, or Kanji) in the label, I'd be ok with it,<br>
although a little nervous. But, if abc・def and 10・0・0・6<br>
(if one doesn't see the characters, the indicators are U+03FB)<br>
would be valid labels under the only acceptable rule, then I<br>
think the middle dot is at high enough risk of causing harm to<br>
be DISALLOWED entirely. That saddens me as much as it does<br>
Harald.<br>
<br>
On the other hand, if the reason for this problem is that<br>
Romanji is Japanese, then so, I suppose, is hyphen-minus. I<br>
would suppose that could be used instead and mapped, as a<br>
presentation issue, to middle dot were appropriate. Not a<br>
wonderful situation, but maybe better than the alternatives.<br>
<font color="#888888"><br>
john<br>
<br>
</font></blockquote></div><br>