W3C home > Mailing lists > Public > www-html@w3.org > April 2008

Re: proposed li:marker pseudo-class

From: Ernest Cline <ernestcline@mindspring.com>
Date: Thu, 3 Apr 2008 22:43:39 -0400 (GMT-04:00)
Message-ID: <10704426.1207277019499.JavaMail.root@mswamui-andean.atl.sa.earthlink.net>
To: Tina Holmboe <tina@greytower.net>
Cc: www-html@w3.org



-----Original Message-----
>From: Tina Holmboe <tina@greytower.net>
>
>On Wed, Apr 02, 2008 at 01:46:51PM +0100, David Woolley wrote:
>
>> Tina Holmboe wrote:
>>
>>>   What I am saying is that there exist no other mechanism in HTML, or
>>>   for that matter in XHTML, for /explicitly/ stating that order is
>>>   significant except OL*.
>>>
>>
>> I think that is because the current HTML specification still has  
>> elements of the old school style of internet specification, in which it  
>> isn't necessary to specify everything with SHALLs and MUST NOTs.  I  
>
>  I would agree, if it wasn't for the fact that such a philosophy would
>  not require an UL *and* an OL, since it'd be common sense* to render
>  list items in the order they were specified.
>
>
>  * Common sense, as has been shown time and time again, is not a good
>    substitute for exactitude.

What gets lost in the current quixotic quest for the semantic web, is that at first HTML wasn't all that semantic.  OL and UL were but two flavors of list markup along with MENU and DIR that were concerned far more with the appearance than the semantics of the markup.  MENU and DIR fell out of favor because most browsers just rendered them the same as UL, altho MENU is making a comeback with HTML 5.  If it weren't for their default expected appearances, OL and UL could be combined into just one element: L (or LIST if you wish to be verbose).

OL is most often used as a means to provide consistently formated reference to a list item (i.e. the 10th item in the list), with very few uses of OL are truly semantic, as making the 10th item the 3rd item and then changing all text references to the 10th item to read the 3rd item would give the same semantic value.  In that respect, it's no different that referring to the 6th paragraph of the 3rd subsection and then having to adjust references when the 4th paragraph of the 3rd subsection is deleted.

What would be useful would be a means of producing text cross references to items, so for example:  given the following two fragments of markup:

"  ... in the < ORDER type="text" href="#foo" /> paragraph ..."

"< DIV >
< P />
< P />
< P id="foo" />
< DIV />"

the first bit would be rendered as:

" ... in the third paragraph ..."

However, the complexities that it would cause for rendering, especially when done on the fly, make anything like that unlikely to ever be adopted.

Ernest Cline
Received on Friday, 4 April 2008 02:44:20 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:13 GMT