Page MenuHomePhabricator

Reflect XHTML trend from "b" and "i" towards "strong" and "em"
Closed, DeclinedPublic


Author: zigger

The 1.4beta & later Parser.php doQuotes() function currently changes wiki "''"
and "'''" markup output from "em" and "strong" to "i" and "b".

While this was presumably done to reduce the size of the output, the average
reduction would be small, and this is the reverse of the
anti-presentation trend reflected in the draft XHTML v2 spec.

Version: 1.4.x
Severity: normal
See Also:


TitleReferenceAuthorSource BranchDest Branch
Do not offer Project Tags to be added to Subscribers fieldsrepos/phabricator/phabricator!9aklapperT303829wmf/stable
Customize query in GitLab

Event Timeline

bzimport raised the priority of this task from to Lowest.Nov 21 2014, 8:06 PM
bzimport added a project: MediaWiki-Parser.
bzimport set Reference to bz1038.
bzimport added a subscriber: Unknown Object (MLST).

zigger wrote:

Parser.php.patch for HEAD


'' and ''' *are* presentational markup. They don't carry specific semantic meaning,
but are used for many purposes. Italics ('') are used for emphasis, to mark foreign
words, to mark titles of literary and artistic works, to set off captions, etc. Bold (''')
is used for emphasis, to mark section titles typographically, etc. Use of <strong>
and <em> for these is thus ABUSE of semantic markup. Semantic markup must
have a specific meaning, or it is worthless. Telling everybody to turn all their <b>s
into <strong>s and <i>s into <em>s is not only unhelpful, it's counterproductive
and does great violence to the adoption of semantic markup.

We produce output in XHTML 1.0, which *does* include <b> and <i> elements.
XHTML 1.0 will always include <b> and <i> elements. If some day we want to
produce output in some version of XHTML that does not include <b> and <i>
elements, then we'll see about more appropriate markup. Perhaps if <strong> and
<em> are redefined to be non-semantic elements, they will be appropriate.

zigger wrote:

See bug 369 & bug 370