[whatwg] The blockquote element spec vs common quoting practices

[Jukka K. Korpela]

2012-02-12 8:36, Nils Dagsson Moskopp wrote:

> Why do you hate the cite attribute?

I don’t; it’s just useless, and it does not in any way satisfy the 
legal, moral, and scholarly requirements for specifying the source.

>> Seldom does an author wish to quote an entire section. It is not even
>> legal to quote more than is required to fulfill the acceptable
>> purpose of quoting.
>
> Elaborate?

“It shall be permissible to make quotations from a work which has 
already been lawfully made available to the public, provided that their 
making is compatible with fair practice, and their extent does not 
exceed that justified by the purpose.”
http://www.wipo.int/treaties/en/ip/berne/trtdocs_wo001.html#P144_26032

> And I don't think I have ever had a need for providing credits that
> went beyond having a URI in the cite attribute and a corresponding
> hyperlink in the surrounding prose.

By the Berne convention, when a work is quoted, “mention shall be made 
of the source, and of the name of the author if it appears thereon.”

The cite attribute, in addition to being practically unsupported, does 
not mention anything. A reference in the “surrounding prose” is a 
completely unstructured way, regarding HTML markup, and not suitable for 
most quotations. It is not reader-friendly at all to provide a 
bibliographic reference (author and full title at the minimum) inside text.

>> If we start from the semantic and logical concept of a quotation,
>> then it should be obvious that the element should have a subelement
>> for providing source information (“credits”), normally at the end of
>> the element.
>
> That would needlessly complicate parsing the contents of a blockquote
> element quite a bit.

Is the comfortability of parsing crucial here? If you want semantic 
markup, you should be prepared to facing any technical difficulties that 
may arise, rather than let the technicalities dictate rules for markup.

> why would it not be
> “obvious” ro have a “for” attribute for the cite element?

The <cite> element is, in practice, just some authors’ way of writing 
<i>, assumed to be more semantic, when the text is a book title, a movie 
name, or something similar. It has really nothing to do with quotations. 
The work mentioned might be quoted, too, in the context, but that’s 
coincidental.

>> Since in current usage, <blockquote>  means just “indent” more often
>> than not, browsers and search engines should not and will not imply
>> any specific semantics for it. Thus it will be pointless to use it.
>
> Riveting tale, chap. Can you provide proof?

Actually the burden of proof is on those who think that <blockquote> has 
some useful support.

Regarding on what authors actually use <blockquote> for, I’ve seen quite 
enough of pages that use nested <blockquote> elements to achieve 
different amounts of indentation.

Discussion forums sometimes use <blockquote> for quotations (though 
surely not for quotations of sections), but that’s just because the 
authors of forum software found that markup suitable, as quotations are 
to be indented. Those who use table layout or style sheets often don’t 
bother using <blockquote>.

>> So leave <blockquote>  as legacy markup and recommend it to be used,
>> in new documents, only for indentation in rare situations where an
>> author much prefers indentation even in the absence of CSS.
>
> How do you propose to treat legacy content?

The common treatment of <blockquote> has been well documented.

> An alternative might lie in using some kind of framework … for
> description … of resources! Are you reasonably sure that Dublin Core or
> similar vocabularies can not help you with this use case?

No, I am absolutely sure that Dublin Core and friends has nothing to do 
with this. (Besides, DC is an old specification which has been casually 
used on web pages for many years, and turned out to be write-only 
metadata. All the recent efforts on the metadata front have ignored DC.)

Whatever markup might turn out to be useful for metadata that associates 
a quoting document, a quotation, and the quoted source, it first needs 
some elements to relate to. In order to say, in metadata, something 
about the relationship between a quotation and its source, you need to 
mark up the quotation and a reference to the source at the very basic 
level. Preferably, using something that unambiguously mean “quotation” 
and “source of quotation” and not “indent” or “figure caption.”

Yucca