W3C home > Mailing lists > Public > www-html@w3.org > May 2007

Re: Cleaning House

From: Murray Maloney <murray@muzmo.com>
Date: Sun, 06 May 2007 16:22:41 -0400
Message-Id: <>
To: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>
Cc: www-html@w3.org,public-html@w3.org

At 06:09 PM 5/6/2007 +0100, Benjamin Hawkes-Lewis wrote:

>Murray Maloney wrote:
>>Hmmm. Not so hasty. You can't actually say that it has no meaning.
>>The reason is that you have not examined the profile that is associated
>>with that HTML document -- because I didn't provide a pointer.
>>The thing is, my profile says that when class="ship" the meaning is the
>>name of a ship, which happens to formatted in italic typeface by convention
>>when it is available.
>>Italic text is emphasized.
>Don't you think there is strictly a difference between distinguishing some 
>text from surrounding text with italics and emphasizing text? For example, 
>when italic is used for book, movie, journal, play titles, foreign 
>phrases, taxonomical terms, technical terms, and ship names you wouldn't 
>necessarily apply a verbal stress to the same terms (you might sometimes, 
>but not always). This difference is not some "neo-semanticist" fantasy; 
>it's a common distinction in print publishing. For instance, the Oxford 
>Style Guide (ISBN 0-19-869175-0) introduces italic like this (p. 154):
>"Italic type is a typographic variation of ordinary roman that is used to 
>indicate emphasis or heavy stress in speech; to style titles, headings, 
>indexes, and cross-references generally; and to indicate foreign words and 
>If emphasis really was all italic is for, then everything after "emphasis" 
>would have been superfluous. When the Guide goes on to say (p. 155) that 
>we should "Employ italics sparingly for emphasis, since their unrestrained 
>use can seem startling or precious" it's clearly not trying to discourage 
>us from using italics for ship names!

I have to tell you that I am really having a had time understanding why you 
can't see my point.

I find it burdensome, but I will try an approach that seems to have 
traction in this WG:

For "emphasis"

Webster says: 
1 a : force or intensity of expression that gives impressiveness or 
importance to something
    b : a particular prominence given in reading or speaking to one or more 
words or syllables
2 : special consideration of or stress or insistence on something

Wikipedia says: 
-  accent the appearance, to underline, to put in bold, make something more 
significant or important.
- in typography: visual enhancement a part of a text to make it noticeable
- in discussing the Semitic languages, "emphasis" refers to certain 
phonologically differentiated stop or fricative consonant sounds. The exact 
phonological realization of emphatic consonants varies between languages
See also http://en.wikipedia.org/wiki/Emphatic_consonants

Under "bold", Wikipdia says:
In typography, emphasis is the exaggeration of words in a text with a font
in a different style from the rest of the text—to emphasize them.

It goes on to say:
The human eye is very receptive to differences in brightness within a text 
One can therefore differentiate between types of emphasis according to whether
the emphasis changes the "blackness" of text.

A means of emphasis that does not have much effect on "blackness" is 
printing in italics,
where the text is written in a script style, or oblique, where the vertical 
of all letters is slanted to the left or right. With one or other of these 
(usually only one is available for any typeface), words can be highlighted 
making them "stick out" much from the rest of the text (inconspicuous 
Traditionally, this is used for marking passages that have a different 
such as words from foreign languages, book titles, etc.

By contrast, boldface makes text darker than the surrounding text. With 
this technique,
the emphasized text strongly stands out from the rest; it should therefore 
be used to
highlight certain keywords that are important to the subject of the text,
for easy visual scanning of text. For example, printed dictionaries often 
use boldface
for their keywords, and the names of articles can conventionally be marked 
in bold.

For "italic", Wikipedia says: http://en.wikipedia.org/wiki/Italic_type
Italic type is often used for emphasis to distinguish or otherwise set off 
certain words within text.
In media where italicization is not possible, alternatives are used as 

     * In typewritten or handwritten text, underlining is typically used.
     * In plain-text computer files, including e-mail communication,
        italicized words are often indicated by surrounding them with slashes
       or other matched delimiters. For example:
           o I was /really/ annoyed.
           o They >completely< forgot me!
           o I had _nothing_ to do with it.
           o It was *absolutely* horrible.

Can we all agree, based on these references, that it is completely within 
reason to say:

- The reason that we markup text is to distinguish or emphasize it in some way.
- We often, but not always, employ visual and aural cues to signal those 
- There exists a rich history of typographical practice employed to signal 
- There exists a rich history of vocal practice employed to signal 
- Bold and Italic are forms of emphasis.
- It is widely understood by practitioners that systems may render bold and 
    using other typographic devices if bold and italic are unavailable (or 
    for whatever reason.

>A little earlier on p. 154, the Guide says: "In most contexts, roman type 
>is the standard face used for text matter, though it can be distinguished, 
>for reasons of emphasis, additional clarity, or common convention, through 
>the use of other typographic styles and forms." Note there are three 
>reasons. Many rationales for italic seem to fall into the later 
>categories. As you yourself say above, ship names are italic "by convention".

I know this stuff through 30+ years of experience. I have forgotten the 
titles of all
the books that I have written, let alone all those that I have read on the 
subjects of
typography, technical writing, publishing, and so on. It's not that I want 
to hold out
on you, it's just that I often don't remember where to look to find a 
source that
would satisfy everyone on the list.

When I am in doubt, I usually turn to Liam Quin and Michael Sperberg-McQueen
on matter related to book publishing, and to TV Raman or George Kirscher on
matters related to accessibility.

>>What is the difference between <i>term</i> and <em>term</em>?
>In HTML 4.01 as specified: the first is a term with a font style, the 
>second is a term the author has emphasized. In text/html in the wild, the 
>discernible difference is minimal, unless like Gregory you're in the habit 
>of writing your own CSS.

Hopefull, my exegesis above will allow you to observe that <i> and <b> are 
types of <em>.

>>> > I have been layering semantics onto the CLASS attribute and REL/REV since
>>> > 1993.
>>>   With support from which standard?
>>Let's see, we used REL/REV in 1993 based on HTML 1.0.
>>We used CLASS since it was introduced in 1996, I think.
>>That is when SoftQuad developed HoTMetaL Intranet Publisher.
>The HTML 4.01 standard could be interpreted to authorize this, though not 
>Class may be used for "general purpose processing by user agents".
>User agents may use a profile URI to "perform some activity based on known 
>conventions for that profile. For instance, search engines could provide 
>an interface for searching through catalogs of HTML documents, where these 
>documents all use the same profile for representing catalog entries."
>On the other hand, HTML user agents aren't required to recognize 
>additional semantics from profiles, so Tina's UA would probably be within 
>its "rights" by the HTML 4.01 "contract" to ignore class="shipName" when 
>trying to interpret <i>.

User Agents are not required to recognize any useful semantics from <em>.
In fact, User Agents are not "required" to do much.

I just thought that you were interested in semantics and how to add 
semantics to HTML.
Maybe even today. And the point is that arguing about <b> and <i> is not 
you any closer to adding semantics. It is instead clouding the issue.

If you really think that you get more semantic value out of <em> than <i>,
and you don't understand that you can use CLASS to enhance the semantic
value of any element, then the markup world is in real trouble.

Please tell me it ain't so.

I sure miss having W3C moderators to help bring the discussion back home.
Around now, I miss having the participation of Chris Lilley and Dan Connolly.


Received on Sunday, 6 May 2007 20:50:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:10 GMT