W3C home > Mailing lists > Public > public-wcag-teama@w3.org > September 2005

Guideline 3.2 HTML Tests (resend)

From: Christophe Strobbe <christophe.strobbe@esat.kuleuven.be>
Date: Wed, 21 Sep 2005 16:20:03 +0200
Message-Id: <>
To: <public-wcag-teama@w3.org>

I am resending my previous message with one addition and one correction.

At 20:19 9/09/2005, Chris Ridpath wrote:
I've been looking over guideline 3.2 and created a document that lists the 
HTML tests that are currently associated with it:
Adjacent links with same resource must be combined.

It is possible to link to different sections of the same resource by using 
fragment identifiers (and e.g. using the heading text at the other end of 
the link as link text). Would this still count as the "same resource"? If 
the fragment identifiers are different, the adjacent links should not be 

Link text is meaningful when read out of context

I believe that someone on the general list has serious objections against 
this because the author has no control over a user agent (esp. 
screenreader) that presents the content of a document out of order: if the 
user asks the UA to present certain content out of order and, as a 
consequence, the content does not make sense anymore, that is the user's 
Instead of "meaningful" we could require something that is more precise 
from a linguistic point of view: require that link text contains at least 
one lexical word (in English and many other Indo-European languages: nouns, 
verbs, adjectives, most adverbs) and never only function words 
(prepositions, pronouns, auxiliary verbs, conjunctions, grammatical 
articles) [1]. Unlike lexical words, function words are a category that 
hardly evolves as language changes, so it is almost a fixed list that could 
be used by software tools. But the wording of this requirement needs more 
work, because "click here" is still allowed if you require at least one 
lexical word in link text. Of course, "click here" does not identify the 
link destination (test 197, not 97 as I wrote in my previous mail).

All source anchors contain text that identifies the link destination.

[1] More on function words vs. lexical words can be found at 
(For links about parts of speech, see 


Christophe Strobbe

Christophe Strobbe
K.U.Leuven - Departement of Electrical Engineering - Research Group on 
Document Architectures
Kasteelpark Arenberg 10 - 3001 Leuven-Heverlee - BELGIUM
tel: +32 16 32 85 51

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
Received on Wednesday, 21 September 2005 14:21:31 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:54:12 UTC