Re: [VE][html5] <strike> warning should recommend using <s> instead

Jukka K. Korpela, Fri, 28 Mar 2014 23:10:42 +0200:
> 2014-03-28 21:44, Leif Halvard Silli wrote:
> 
>> Indeed. After all, <strike> and <s> are synonyms.
>> 
>> Leif H
>> 
>> j.j., Fri, 28 Mar 2014 19:03:42 +0100:
>>> The validator warns about <strike>:
>>> 
>>>   "The strike element is obsolete. Use CSS instead."
>>> 
>>> It should recommend to use <s> instead.
> 
> In browser reality, and by HTML 4.01, <strike> and <s> are 
> synonymous, but by HTML5 CR, which is what matters here, they are 
> not. The <s> element is conforming, whereas <strike> is not, and 
> <strike> has no definition.

Well, I don't see that that make them not synonyms. HTML5 in fact says:

”strike
Use del instead if the element is marking an edit, otherwise use s 
instead.
”
http://www.w3.org/html/wg/drafts/html/master/obsolete.html#strike


> In the HTML5 logic, which is not very logical, but by which 
> validators should work, <strike> is an obsolete, nonconforming, ugly, 
> naughty, impure, cursed presentational tag, whereas <s> is a proud 
> Semantic tag. By that logic, <s> should be used instead of <strike> 
> only in the very special case where the some semantics was meant to 
> be implied by <strike> and this accidentally happens to coincide with 
> the semantics of <s>.

Yes, it is true that del should be used if strike was used to signify 
del semantics, while s should be used if strike was used to signify s 
semantics.

I suppose that HTM5 expresses itself the way it does because it does 
not make any assumptions about whether those who use <strike> has used 
its correctly (that is: as a <s>)  or not. This makes some sense since 
the semantics of <s> perhaps can be said to have been altered.

Leif

Received on Saturday, 29 March 2014 11:34:18 UTC