W3C home > Mailing lists > Public > public-html@w3.org > April 2007

Re: Proposing <indent> vs. <blockquote>

From: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>
Date: Sun, 15 Apr 2007 15:18:27 +0100
Message-ID: <46223433.2030600@googlemail.com>
To: public-html@w3.org
CC: Mike Schinkel <w3c-lists@mikeschinkel.com>, Murray Maloney <murray@muzmo.com>

Mike Schinkel wants to use <indent> frequently, but only because:

1) He finds the CSS box model too complicated.

2) He is not always authoring in standards mode (and so has to cope with
non-standard interpretations of the CSS box model).

http://lists.w3.org/Archives/Public/public-html/2007Apr/0391.html

Since <indent> doesn't even exist yet and wouldn't be usable in current
browsers, this problem might imply a need for a new CSS
property, block-indent, which would be /within/ CSS width rather than
added to it.

Mike claims that bloggers, commentators, and forum posters might wish
to use <indent>:

http://lists.w3.org/Archives/Public/public-html/2007Apr/0577.html

Other than in Mike's own work, I can't recall ever seeing <blockquote>
misused for indentation in social media. Can anyone point to some recent 
(say past 6 months) examples of this in the wild?

Mike further refined this argument by claiming that there are people who
need to indent but who don't have the time to learn CSS:

http://lists.w3.org/Archives/Public/public-html/2007Apr/0582.html

Now I agree with Mike that people shouldn't have to learn CSS in order
to write on the web and that requiring casual authors to write inline
CSS is a poor solution. But since CSS is for presentation and HTML is
for meaning, there is no reason why they should need to use CSS to
communicate effectively. If <indent> is required to communicate
something, then the author must /mean/ something by that indentation, so
that <indent> would have (ambiguous) semantic import.

When <aside> was presented as a possible solution, Mike wrote:

> Aside gives semantics, and its semantics are not applicable to my 
> use-cases which are call-outs for info that should be inline in a 
> document.

http://lists.w3.org/Archives/Public/public-html/2007Apr/0779.html

The word "call-out" suggests to me a form of emphasis (<em>) and indeed
in his last email Mike wrote:

> Often it [<aside>] means "pay attention to this", so in that case
> it's the same as <em>.

http://lists.w3.org/Archives/Public/public-html/2007Apr/0831.html

So why not use <em>? Or if you need an emphasizing element that can take
block content, propose <em><p></p><p></p></em> be allowed or suggest a
new <blockem> element.

Helpfully, Mike provided three example URLs. I want to assess the
appropriateness of <indent> for each.

1. http://www.mikeschinkel.com/blog/thesirensongofssi/

The first, second, and fifteenth instances of <blockquote> are properly
used for block quotation. The third to the fourteenth uses are all for
code examples, where <pre> and <code> could be used instead:

http://www.whatwg.org/specs/web-apps/current-work/multipage/section-phrase.html#the-code

Using <code> has advantages for theoretical screen readers and voice
browsers that could be configured to switch from not reading most
punctuation in ordinary text to reading all punctuation in code, and for
theoretical visual browsers that could perform syntax highlighting. Note
that code examples are hardly typical content for the sort of
non-technical users <indent> is supposed to benefit!

2. http://www.mikeschinkel.com/blog/businesscardsphotosandpersonalurls/

<blockquote> is used twice where an example class would do better:

http://www.whatwg.org/specs/web-apps/current-work/multipage/section-global.html#class1

(Though I think that class needs to be allowed on elements like div that
can contain block content (i.e. multiple paragraphs). /Or/ <figure>
needs to be able to be used without a caption and without "embedded
content".)

3. http://www.mikeschinkel.com/blog/gtfk/

The first use of <blockquote> is in error for <q>. The second and third
uses are heavily disguised direct speech, as "I christen" and "tell
them" confirm. In other words, they could less confusingly be rewritten
as follows:

<p>Anyway, though I still haven't decided which laptop to get, I
christen thee a new meme in my friends honor while I pay homage to that
soon-to-be bygone era where a few people actually did read the manual:
"<dfn><abbr title="Google The F***in' Keywords">GTFK</abbr></dfn>:
<em>Google The F***in' Keywords</em>."</p>
<p>Just to be explicit, there is a proper context for using <abbr
title="Google The F***in' Keywords">GTFK</abbr>. When someone asks you a
question that requires a long explanation that they could have easily
answered themselves, it is perfectly appropriate to simple tell them:
"<em><abbr title="Google The F***in' Keywords">GTFK</abbr></em>!"</p>

I doubt Mike's idiosyncratic indented rendering of direct speech
constitutes a common use case for <indent>. But a (hypothetical)
<speech> element would be useful in other ways than as a styling hook.
In English at least, so long as authors always use standard quotation
marks to demarcate direct speech, JAWS at least can switch to an
alternate voice. But this convention wouldn't work in Turkish, where
there is no standard punctuation for direct speech. And Mike's example
demonstrates it may not even work in English, since he omits the
relevant punctuation on which JAWS depends. <indent> wouldn't help the
situation at all, because as Mike stresses, no semantic information
should be inferred from it. By contrast, a <speech> element would
provide an accessibility cue to screen readers and voice browsers even
in the absence of punctuation.

<speech> could also be used together with the <dialog> element proposed
in the WHATWG draft:

http://www.whatwg.org/specs/web-apps/current-work/multipage/section-prose.html#the-dialog

In summary, while Mike's misuse of <blockquote> does confirm the utility
of example and suggest a possible use for <speech>, it does not suggest
any use-case for a semantics-free <indent>.

--
Benjamin Hawkes-Lewis
Received on Sunday, 15 April 2007 14:31:54 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:42 UTC