W3C home > Mailing lists > Public > www-style@w3.org > April 1998

Re: Units of measurements

From: Todd Fahrner <fahrner@pobox.com>
Date: Mon, 6 Apr 1998 09:36:33 -0700
Message-Id: <v03102802b14eb0f3d88e@[]>
To: "Frank Boumphrey" <bckman@ix.netcom.com>, "John Ky (HxG)" <z2206460@student.unsw.edu.au>, "STYLE" <www-style@w3.org>
Thus spake Frank Boumphrey:
>      em takes on a different meaning in CSS to the  usual typographic
>term. In Typography, 6em's=1 inch,

You must be confusing em with pica. A pica is 12 points. There are six of
them in an inch. Only in a 12-point font is a pica equivalent to an em. As
for em as a "usual typographic term", there's no end of debate on the
connection to the letter em. Jan Roland Eriksson has produced a nearly
exhaustive treatment here:
http://home2.swipnet.se/%7Ew-20547/stylework/typograph1-en.html#Ch23 ...
yet he disputes any historical connection to the letter M. I've heard
otherwise, and will recycle a little ascii art from usenet:

It's my understanding that there is a relationship to the letter M, or
rather, to the chunk of metal[1] upon which the raised face of the letter
once typically sat:
   | |\    /| ||   |\
   | | \  / | || . | \
   | |  \/  | || | | |
   | |      | || | | |      ---> "iM"
   |          ||   | |
   ----------------- |
    \           \\  \|
     \           \\  \
      ------------ ---

Unlike most of the letters, which sat on narrower chunks of metal, the M's
was square, at least for the most common designs of the letter M at an
early point in printing history[2] (this was somewhat wider than current
fashion has it). Thus the M's width served as a handy horizontal unit
equivalent to the height of all the characters' chunks; i.e., for the
nominal font size. "Em" came to mean, metonymically, this latter thing,
even after the letter M began to be designed more narrowly.[3]

[1] http://www.tiro.com/lead.gif
[2] http://www.tiro.com/aen_rom.html
[3] http://www.tiro.com/pla_rom.html

>      -----Original Message-----
> From:      John Ky (HxG) <z2206460@student.unsw.edu.au>

>      Also, the      ex unit turned out     to be slightly less than the
>width of the 'x'     character.

As Frank noted, ex is universally misimplemented - as .5em. Last time I
checked, NS also gave too large a value for em; i.e., representing 1 em as
something more like 1.1em.

Todd Fahrner
Received on Monday, 6 April 1998 12:30:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:53:54 GMT