(In reply to comment #0)

> refer to:
> http://oli.jp/example/blockquote-metadata/

Like I explain in public-html [1], Oli is, on that page, effectively using the following pattern:

<figure>
  <blockquote>Quotation</blockquote>
  <figcaption>About the quotation</figcaption>
</figure>

And I think that this bug should be solved by adding text to the definition of <figure> which recommends to use the above pattern when metadata about quotations are needed.

[1] http://lists.w3.org/Archives/Public/public-html/2013Aug/0080

(In reply to comment #1)
> (In reply to comment #0)
> 
> > refer to:
> > http://oli.jp/example/blockquote-metadata/
> 
> Like I explain in public-html [1], Oli is, on that page, effectively using
> the following pattern:
> 
> <figure>
>   <blockquote>Quotation</blockquote>
>   <figcaption>About the quotation</figcaption>
> </figure>
> 
> And I think that this bug should be solved by adding text to the definition
> of <figure> which recommends to use the above pattern when metadata about
> quotations are needed.
> 
> [1] http://lists.w3.org/Archives/Public/public-html/2013Aug/0080

I think this is an interesting solution to the problem but, since <blockquote> is a sectioning root, I think <footer> is more apt for <blockquote> metadata. It also requires less markup and (for me at least!) less congnitive reconfiguration. 

In addition, I think the <figcaption> has a less coupled relationship with its sibling <blockquote> than the <footer> would as a child element.

(In reply to comment #2)

> I think this is an interesting solution to the problem but, since
> <blockquote> is a sectioning root, I think <footer> is more apt for
> <blockquote> metadata. It also requires less markup and (for me at least!)
> less congnitive reconfiguration. 

It is not clear to me why you consider that the section root feature has as consequence that the metadata element should be kept inside (rather than outside) <blockquote>. And, FWIW, <figure> is a sectioning root as well.

(The meaning of 'section root' is: """can have their own outlines, but the sections and headings inside these elements do not contribute to the outlines of their ancestors""".)

> In addition, I think the <figcaption> has a less coupled relationship with
> its sibling <blockquote> than the <footer> would as a child element.

Well, they would then both be generic children of <figure>, and as such they would be on equal footing and 'together'. (It could perhaps be a useful pattern for footnotes?) If we are talking about the generic level, then it can also be seen as a good point to keep them separete. After all, one (blockquote) is a quotation while the other (figcaption) is not a quoation ...

AT the HTML5doctor page, Oli pointed to a reply from Hixie [1]. In this reply, Ian suggested to introduce a new <credit(s)> element rather than changing <footer>.[1] And such a new element seems like a better idea than <footer> if first we want to place metadata inside <blockquote>. Quoting Ian:

"""
I expect we will eventually create a <credit> element that goes inside 
<blockquote>, <figure> or <figcaption>, <caption>, and maybe other 
contexts as well. At the moment, I'm deferring adding it so that we can 
see how <figure> and the other new elements do in the wild.
"""

The pros of <credit> are that it:
  a) avoids the problem of "what if I want to quote a <footer>?",
     and thus is easier to understand.
  b) is a new element - we don't need to change the semantics
     of an existing elemen - plus that it is easier to fine tune
     a new element for this purpose, rather than adding more 
     gotchas to an existing element
  c) can be used for giving credit also inside <figure> and 
     eventually in other elements.
  d) does not have a name that indicates that it should be 
     placed at the foot of the element - a <credit> could
     just as well go at the start of the element.

[1] http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2012-February/034822.html

(In reply to comment #3)
> (In reply to comment #2)
> 
> > I think this is an interesting solution to the problem but, since
> > <blockquote> is a sectioning root, I think <footer> is more apt for
> > <blockquote> metadata. It also requires less markup and (for me at least!)
> > less congnitive reconfiguration. 
> 
> It is not clear to me why you consider that the section root feature has as
> consequence that the metadata element should be kept inside (rather than
> outside) <blockquote>. And, FWIW, <figure> is a sectioning root as well.
> 
> (The meaning of 'section root' is: """can have their own outlines, but the
> sections and headings inside these elements do not contribute to the
> outlines of their ancestors""".)
> 
> > In addition, I think the <figcaption> has a less coupled relationship with
> > its sibling <blockquote> than the <footer> would as a child element.
> 
> Well, they would then both be generic children of <figure>, and as such they
> would be on equal footing and 'together'. (It could perhaps be a useful
> pattern for footnotes?) If we are talking about the generic level, then it
> can also be seen as a good point to keep them separete. After all, one
> (blockquote) is a quotation while the other (figcaption) is not a quoation
> ...
> 
> AT the HTML5doctor page, Oli pointed to a reply from Hixie [1]. In this
> reply, Ian suggested to introduce a new <credit(s)> element rather than
> changing <footer>.[1] And such a new element seems like a better idea than
> <footer> if first we want to place metadata inside <blockquote>. Quoting Ian:
> 
> """
> I expect we will eventually create a <credit> element that goes inside 
> <blockquote>, <figure> or <figcaption>, <caption>, and maybe other 
> contexts as well. At the moment, I'm deferring adding it so that we can 
> see how <figure> and the other new elements do in the wild.
> """
> 
> The pros of <credit> are that it:
>   a) avoids the problem of "what if I want to quote a <footer>?",
>      and thus is easier to understand.
>   b) is a new element - we don't need to change the semantics
>      of an existing elemen - plus that it is easier to fine tune
>      a new element for this purpose, rather than adding more 
>      gotchas to an existing element
>   c) can be used for giving credit also inside <figure> and 
>      eventually in other elements.
>   d) does not have a name that indicates that it should be 
>      placed at the foot of the element - a <credit> could
>      just as well go at the start of the element.
> 
> [1]
> http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2012-February/034822.html


You are quite right: figures are also sectioning roots.

However, since <figure> already has <figcaption>, <caption> is taken and <figcaption> is clumsily named to be ported to <blockquote>s, there's still <blockquote>'s (direct) metadata to address. 

The <credit> element is clearly intended as a replacement for <cite>. It does not solve the problem that this bug addresses because it is far too specific. That is, one could hardly include the information "emphasis mine" or "circa 2nd Century Greece" in an element called <credit>, could they? If <credit> replaced <cite> tomorrow, I'd be fine with that but it is not equivalent to either <figcaption> or <footer>.

A compromise:

I think allowing <footer>s in <blockquote>s is the shortest route to providing a metadata solution for blockquotes, and one which has the benefit of precendent. However, I'm keen to emphasise the textual capacity of figure, should it be a simple <p> or a <blockquote>. I'd be happy to raise a separate bug / topic of discussion on that.

FWIW, I recommend we see to this specific bug, then address guidelines on <figure> usage separately. If <credit> emerges it will be as a replacement for <cite> - an action that will not be necessary if <cite>'s remit is allowed to include author names etc.

(In reply to comment #3)
> (In reply to comment #2)
> 
> > I think this is an interesting solution to the problem but, since
> > <blockquote> is a sectioning root, I think <footer> is more apt for
> > <blockquote> metadata. It also requires less markup and (for me at least!)
> > less congnitive reconfiguration. 
> 
> It is not clear to me why you consider that the section root feature has as
> consequence that the metadata element should be kept inside (rather than
> outside) <blockquote>. And, FWIW, <figure> is a sectioning root as well.
> 
> (The meaning of 'section root' is: """can have their own outlines, but the
> sections and headings inside these elements do not contribute to the
> outlines of their ancestors""".)
> 
> > In addition, I think the <figcaption> has a less coupled relationship with
> > its sibling <blockquote> than the <footer> would as a child element.
> 
> Well, they would then both be generic children of <figure>, and as such they
> would be on equal footing and 'together'. (It could perhaps be a useful
> pattern for footnotes?) If we are talking about the generic level, then it
> can also be seen as a good point to keep them separete. After all, one
> (blockquote) is a quotation while the other (figcaption) is not a quoation
> ...
> 
> AT the HTML5doctor page, Oli pointed to a reply from Hixie [1]. In this
> reply, Ian suggested to introduce a new <credit(s)> element rather than
> changing <footer>.[1] And such a new element seems like a better idea than
> <footer> if first we want to place metadata inside <blockquote>. Quoting Ian:
> 
> """
> I expect we will eventually create a <credit> element that goes inside 
> <blockquote>, <figure> or <figcaption>, <caption>, and maybe other 
> contexts as well. At the moment, I'm deferring adding it so that we can 
> see how <figure> and the other new elements do in the wild.
> """
> 
> The pros of <credit> are that it:
>   a) avoids the problem of "what if I want to quote a <footer>?",
>      and thus is easier to understand.
>   b) is a new element - we don't need to change the semantics
>      of an existing elemen - plus that it is easier to fine tune
>      a new element for this purpose, rather than adding more 
>      gotchas to an existing element
>   c) can be used for giving credit also inside <figure> and 
>      eventually in other elements.
>   d) does not have a name that indicates that it should be 
>      placed at the foot of the element - a <credit> could
>      just as well go at the start of the element.
> 
> [1]
> http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2012-February/034822.html

the spec says:

"The footer element represents a footer for its nearest ancestor sectioning content or sectioning root element. A footer typically contains information about its section such as who wrote it, links to related documents, copyright data, and the like."

note 'its nearest ancestor ... sectioning root element'.

I don't see why any change would need to be made to <footer. to make it suitable for its use as a container for citation information, the only change would be to the definition of blockquote.

(In reply to comment #5)
 
> the spec says:
> 
> "The footer element represents a footer for its nearest ancestor sectioning
> content or sectioning root element. A footer typically contains information
> about its section such as who wrote it, links to related documents,
> copyright data, and the like."
> 
> note 'its nearest ancestor ... sectioning root element'.
> 
> I don't see why any change would need to be made to <footer. to make it
> suitable for its use as a container for citation information, the only
> change would be to the definition of blockquote.

OK. I see your point.

I have done some digging:

I have made some data available:
https://dl.dropboxusercontent.com/u/377471/html5bq/index.html

This is an index of 318 pages which use the HTML5 doctype and include blockquote elements (and sometimes cite elements). The sample pages were drawn from the raw data from March 2012 (121Mb http://webdevdata.org/data/data.zip). Consisting of approximately 8000 home pages from the top 10,000 most popular web sites.

I have added styles to blockquote and cite elements so that they are easily identifiable when viewing the page.

There is also grep output for the indexed pages: https://dl.dropboxusercontent.com/u/377471/blockquote.html


observations:

    blockquote usage from these pages appears to be largely for its intended use.
    it is common for the attribution to be included within the blockquote
    when <cite> is used it is common for it to be included inside the blockquote element
    when <cite> is used it is common for it to be used for attributing a quote to a person
    10 pages included use of the cite attribute
    5 pages used footer inside blockquote for attribution

(In reply to comment #7)
> I have done some digging:
> 
> I have made some data available:
> https://dl.dropboxusercontent.com/u/377471/html5bq/index.html
> 
> This is an index of 318 pages which use the HTML5 doctype and include
> blockquote elements (and sometimes cite elements). The sample pages were
> drawn from the raw data from March 2012 (121Mb
> http://webdevdata.org/data/data.zip). Consisting of approximately 8000 home
> pages from the top 10,000 most popular web sites.
> 
> I have added styles to blockquote and cite elements so that they are easily
> identifiable when viewing the page.
> 
> There is also grep output for the indexed pages:
> https://dl.dropboxusercontent.com/u/377471/blockquote.html
> 
> 
> observations:
> 
>     blockquote usage from these pages appears to be largely for its intended
> use.
>     it is common for the attribution to be included within the blockquote
>     when <cite> is used it is common for it to be included inside the
> blockquote element
>     when <cite> is used it is common for it to be used for attributing a
> quote to a person
>     10 pages included use of the cite attribute
>     5 pages used footer inside blockquote for attribution


Thank you, Steve, that's great.

Just looking at the grep, I'm seeing uses of <cite> in <footer> as well as just <footer>.

Some of the more cautious appear to be using things like the below example

<p id="quote-source"> Dan Salvador, CEO, Mainline </p>

Obviously, this is suboptimal and precisely the kind of thing that authors should feel they can avoid (as addressed by this bug).

:-)

(In reply to comment #5)

> the spec says: 
 [ snip ]
> 'its nearest ancestor ... sectioning root element'.
> 
> I don't see why any change would need to be made to <footer. to make it
> suitable for its use as a container for citation information, the only
> change would be to the definition of blockquote.

However, the spec uses the exact same wording  ("nearest ancestor … sectioning root element" etc) about <header> and <h1>-<h6> elements as well.

So, when arguing based on what the spec says, we could also say that no change is needed for <h1>-<h6> and <header> - they can all be used to “talk about” the quotation.

(In reply to comment #7)

> There is also grep output for the indexed pages:
> https://dl.dropboxusercontent.com/u/377471/blockquote.html

That file seems to show that while many authors place the source data inside <blocquote>, nearly as many place it outside <blockquote>. See my analysis: 
http://lists.w3.org/Archives/Public/public-html/2013Aug/0100.html

(In reply to comment #9)
> (In reply to comment #5)
> 
> > the spec says: 
>  [ snip ]
> > 'its nearest ancestor ... sectioning root element'.
> > 
> > I don't see why any change would need to be made to <footer. to make it
> > suitable for its use as a container for citation information, the only
> > change would be to the definition of blockquote.
> 
> However, the spec uses the exact same wording  ("nearest ancestor …
> sectioning root element" etc) about <header> and <h1>-<h6> elements as well.
> 
> So, when arguing based on what the spec says, we could also say that no
> change is needed for <h1>-<h6> and <header> - they can all be used to “talk
> about” the quotation.

"The <header> element represents a group of introductory or navigational aids." (http://www.w3.org/wiki/HTML/Elements/header)

Although header has a similar syntactical relationship to the "nearest ancestor", semantically it means something quite different.

Footer is much more semantically apt since (like the related landmark role="contentinfo") it has the purpose of clarification, like a "footnote". Surely this is where citations (<cite>) are most at home?

Certainly <hn> and <header> are _about_ the body content of the relevant section, but in an introductory faculty. 

Note that the footer definition does not require it to follow the body content in source order (http://www.w3.org/TR/html-markup/footer.html). So, as with <figcaption>, one could do

<blockquote>
   <footer>The following is an excerpt from <cite>Leif</cite>'s blog</footer>
   <p>Excerpt here...</p>
</blockquote>

The name "footer" is just a bit unfortunate :-)

(In reply to comment #10)
> (In reply to comment #7)
> 
> > There is also grep output for the indexed pages:
> > https://dl.dropboxusercontent.com/u/377471/blockquote.html
> 
> That file seems to show that while many authors place the source data inside
> <blocquote>, nearly as many place it outside <blockquote>. See my analysis: 
> http://lists.w3.org/Archives/Public/public-html/2013Aug/0100.html

Sure, all of the following exist...

a) There were 68 occurrences of class=aut(hor) outside blockquote,
   typically on an element right after (adjacent to) the blockquote.
b) There were 58 <cite> outside <blockquote>, all/most of which
   were used to attribute the author of the quote.
c) There were some 'naked' (classless) adjacent elements after the
   blockquote element were used to provide the name of the author.

... but none of them could be said to actually connect the blockquote to the "citation" in a compelling way.

As we know, the class "author" does nothing except add a styling hook. The instances of <cite> outside <blockquote>, are not explicitly associated (like with a "for" attribute as Karl suggested) and naked, classless elements after the <blockquote> would just be considered arbitrary flow content belonging to a common parent of the blockquote. 

I think the important thing to take from this is that, of the following uses of descendants...

1) 19 <footer>s, several of which were from same web sites/authors
2) 123 <cite> (5 of which were child of <footer>, and 2 of which
   where child of an element of class "author".)
3) 36 occurrences of class="author" on various child elements of
   <blockquote>.
4) Some last-child “naked” elements (without @class) or <cite>,
   were used to designate metadata/authors. Amongst these were
   3 <address> elements, all of which contained a <cite>.

... only <footer>, <cite> and <address> (plus combinations) could be said to be real attempts to demarcate the metadata from the main content semantically.

Of these, <cite> is not appropriate on its own because the metadata could require more than a single citation. What would one use to encapsulate these citations, and other metadata? <address> is a statistical outlier, plus...

"The address element must not contain information other than contact information."

If I really wanted to include contact information, I'm permitted to place the <address> inside a <footer>, perhaps alongside some <cite>s et al.

This brings us back to <footer>. 

And _even if_ you don't agree that <footer> is, after all, the best solution for this, surely <footer> should still be a legitimate (optional) descendant of <blockquote> (as <h1> etc are) anyway?

(In reply to comment #11)
> (In reply to comment #9)
 
> "The <header> element represents a group of introductory or navigational
> aids." (http://www.w3.org/wiki/HTML/Elements/header)
> 
> Although header has a similar


*identical*


> syntactical relationship to the "nearest
> ancestor", semantically it means something quite different.
> 
> Footer is much more semantically apt since (like the related landmark
> role="contentinfo") it has the purpose of clarification, like a "footnote".
> Surely this is where citations (<cite>) are most at home?
> 
> Certainly <hn> and <header> are _about_ the body content of the relevant
> section, but in an introductory faculty. 

However: ”The blockquote element represents a section that is quoted from another source”. Note: section. As you know, in “unquoted” texts in HTML documents, use of sectioning elements, such as <article> and <section>, is optional. It is enough to use hn-elements - the sectioning will then be implied. Thus you can have a text like this:

  <body>
   <h1>Foo</h1>
   <p>bar</p>
   <footer>Something about this text</footer>
  </body> 

An Web author might wanna quote that text by placing it inside a <blockquote>. Using a WYSIWYG tool, (s)he might just select everything on the page, copy it, and paste it, with footer and all.

And then, according to the idea promoted in this bug, the <footer>
element would count as the Web author’s about text, whereas in reality, it is a text (s)he has quoted.

> Note that the footer definition does not require it to follow the body
> content in source order (http://www.w3.org/TR/html-markup/footer.html).

In that detail, you are right. However, if we introduced e.g. a
<bqcaption>, we could have the same rule about that element.

(In reply to comment #12)
> (In reply to comment #10)
 
> Sure, all of the following exist...
  [snip] 
> ... but none of them could be said to actually connect the blockquote to the
> "citation" in a compelling way. [ snip]

The goal is not only “connection”. The goal is the *correct*
connection. Besides, being adjacent is a kind of connection. Further
more, being the last-child of <blockquote> is merely a convention.
 
> I think the important thing to take from this is that, of the following uses
> of descendants...
  [ snip ]
> ... only <footer>, <cite> and <address> (plus combinations) could be said to
> be real attempts to demarcate the metadata from the main content
> semantically.

Your wording “real attempt” is unhelpful. Clearly, placing the metadata
outside <blockquote>, is a “real attempt“ - and a succcessfull one - in
separating the meta from the content. If you really wanna check out
“real“, then you should also check whether some of the examples used 
microdata, microformats or RDFa (at least microdata were present here
and there).
 
  [ snip ]
 
> This brings us back to <footer>. 

It does? Sorry, but you do not sound convincing. I do of course understand that <footer> is for 'about' information. The issue is, however: whose ‘about’ information? One could define rules for that - saying that under such and such circumstance, <footer> counts as the current text’s about information (versus the quoted text’s about information). And that is what Oli suggested to do. However, given how difficult it is for us, authors, to mentally separate quoted metadata from unquoted metadata - plus the WYSWIWYG scenario I described in my previous reply, operating with such rules, is not a bulletproof way of dealing with the issue.

> And _even if_ you don't agree that <footer> is, after all, the best solution
> for this, surely <footer> should still be a legitimate (optional) descendant
> of <blockquote> (as <h1> etc are) anyway?

<footer> is legitimate, today. It is only that it is then a quoted
<footer>.  Thus: it is not, per the spec’s semantics, the author of the current page’s footer.

Hey, why this reluctance towards a *new* element? Then we could have “the best” solution.

> Hey, why this reluctance towards a *new* element? Then we could have “the
> best” solution.

This bug is about modifying the blockquote definition, its much easier and simpler to modify and re-use existing elements where fit for purpose than adding a new element. If you are enthusiastic about it suggest you work up the use cases, data and a draft extension spec for such a new element.

(In reply to comment #15)
> > Hey, why this reluctance towards a *new* element? Then we could have “the
> > best” solution.
> 
> This bug is about modifying the blockquote definition, its much easier and
> simpler to modify and re-use existing elements where fit for purpose than
> adding a new element.

Not that it is an argument against the advantage of redefining a current element, but I am curious how you can avoid modifying <footer> as well.

Further more, the very simplest thing would be to add wording about how to use <figure> - as documented, many already *do* place the about text outside <blockquote>. But as Heydon pointed out, they tend to use elements which do not direclty link the about tex tot the quotation. Advice about how to use <figure> for this, can be given without chaning the semantics of anything.

> If you are enthusiastic about it suggest you work up
> the use cases, data and a draft extension spec for such a new element.

Sure.

(In reply to comment #16)
> (In reply to comment #15)
> > > Hey, why this reluctance towards a *new* element? Then we could have “the
> > > best” solution.
> > 
> > This bug is about modifying the blockquote definition, its much easier and
> > simpler to modify and re-use existing elements where fit for purpose than
> > adding a new element.
> 
> Not that it is an argument against the advantage of redefining a current
> element, but I am curious how you can avoid modifying <footer> as well.
> 
> Further more, the very simplest thing would be to add wording about how to
> use <figure> - as documented, many already *do* place the about text outside
> <blockquote>. But as Heydon pointed out, they tend to use elements which do
> not direclty link the about tex tot the quotation. Advice about how to use
> <figure> for this, can be given without chaning the semantics of anything.
> 
> > If you are enthusiastic about it suggest you work up
> > the use cases, data and a draft extension spec for such a new element.
> 
> Sure.

There's no need to modify <footer>. Whether or not the principle <footer> is actually quoted from the original source, it is bound to describe the nature of that content. It is a non-issue in most cases and redundant footers can be omitted by authors in other cases, as they see fit, right?

Blockquote should allow <footer> to be used at the author's discretion, just as
<figcaption> is a discretionary inclusion. Can we agree on that?

(I have registered my support of resolving the other bug.)

(In reply to comment #17)

> There's no need to modify <footer>. Whether or not the principle <footer> is
> actually quoted from the original source, it is bound to describe the nature
> of that content.

+1 Agree. This is not the problem.

> It is a non-issue in most cases and redundant footers can
> be omitted by authors in other cases, as they see fit, right?
> 
> Blockquote should allow <footer> to be used at the author's discretion, just
> as > <figcaption> is a discretionary inclusion. Can we agree on that?

While an “about text” could clarify the source of the very quotation,
<footer> as “about text” container may cause uncertainty w.r.t. the
source of the very annotation w.r.t. whether it stems from the same
source as the quotation itself - or from the current text.

This doubt is fuelled by the fact that the footer would change semantics
(with regard to authoringship) once it is pasted into <blockquote> - in
the original text its voice would be the same as the voice of the
section where it appears, whereas inside <blockquote>, its voice would
change to mean “the voice of the author of blockquote’s parent element”.

Also, to visually distinguish footer from quotation, the <footer> would
probably need to be styled differently when inside <blockquote> compared
when outside, and my hunch is that a styling that differs based on context
is not going to be popular amongst implementors (but may be I am wrong
in this detail).

For more, see bug 23021.

(In reply to comment #18)
> (In reply to comment #17)
> 
> > There's no need to modify <footer>. Whether or not the principle <footer> is
> > actually quoted from the original source, it is bound to describe the nature
> > of that content.
> 
> +1 Agree. This is not the problem.
> 
> > It is a non-issue in most cases and redundant footers can
> > be omitted by authors in other cases, as they see fit, right?
> > 
> > Blockquote should allow <footer> to be used at the author's discretion, just
> > as > <figcaption> is a discretionary inclusion. Can we agree on that?
> 
> While an “about text” could clarify the source of the very quotation,
> <footer> as “about text” container may cause uncertainty w.r.t. the
> source of the very annotation w.r.t. whether it stems from the same
> source as the quotation itself - or from the current text.
> 
> This doubt is fuelled by the fact that the footer would change semantics
> (with regard to authoringship) once it is pasted into <blockquote> - in
> the original text its voice would be the same as the voice of the
> section where it appears, whereas inside <blockquote>, its voice would
> change to mean “the voice of the author of blockquote’s parent element”.
> 
> Also, to visually distinguish footer from quotation, the <footer> would
> probably need to be styled differently when inside <blockquote> compared
> when outside, and my hunch is that a styling that differs based on context
> is not going to be popular amongst implementors (but may be I am wrong
> in this detail).
> 
> For more, see bug 23021.

unconvinced about the nuances of the 'voice' argument or of the styling argument. The point of allowing footer in blockquote is to provide a means to indicate that some content is metadata about the quoted content, when as is common such content is structurally part of the quoted text, but semantically separate. 

Note I have also allowed <cite> for this purpose for in-text attributions.

It should also be noted that the blockquote element already allows changes to the content of quoted text without any markup requirement.

"The content of a blockquote may be abbreviated or may have context added in the conventional manner for the text's language."


After doing some more reading and research on citations it really seems as if to deal with the diversity/complexity of citation styles/formats the use of metadata schema is the most appropriate, what we should be providing in HTML is a framework to hang these off. I think cite and footer provide this.

(In reply to comment #19)
> (In reply to comment #18)
> > (In reply to comment #17)
> > 
> > > There's no need to modify <footer>. Whether or not the principle <footer> is
> > > actually quoted from the original source, it is bound to describe the nature
> > > of that content.
> > 
> > +1 Agree. This is not the problem.
> > 
> > > It is a non-issue in most cases and redundant footers can
> > > be omitted by authors in other cases, as they see fit, right?
> > > 
> > > Blockquote should allow <footer> to be used at the author's discretion, just
> > > as > <figcaption> is a discretionary inclusion. Can we agree on that?
> > 
> > While an “about text” could clarify the source of the very quotation,
> > <footer> as “about text” container may cause uncertainty w.r.t. the
> > source of the very annotation w.r.t. whether it stems from the same
> > source as the quotation itself - or from the current text.
> > 
> > This doubt is fuelled by the fact that the footer would change semantics
> > (with regard to authoringship) once it is pasted into <blockquote> - in
> > the original text its voice would be the same as the voice of the
> > section where it appears, whereas inside <blockquote>, its voice would
> > change to mean “the voice of the author of blockquote’s parent element”.
> > 
> > Also, to visually distinguish footer from quotation, the <footer> would
> > probably need to be styled differently when inside <blockquote> compared
> > when outside, and my hunch is that a styling that differs based on context
> > is not going to be popular amongst implementors (but may be I am wrong
> > in this detail).
> > 
> > For more, see bug 23021.
> 
> unconvinced about the nuances of the 'voice' argument or of the styling
> argument. The point of allowing footer in blockquote is to provide a means
> to indicate that some content is metadata about the quoted content, when as
> is common such content is structurally part of the quoted text, but
> semantically separate. 
> 
> Note I have also allowed <cite> for this purpose for in-text attributions.
> 
> It should also be noted that the blockquote element already allows changes
> to the content of quoted text without any markup requirement.
> 
> "The content of a blockquote may be abbreviated or may have context added in
> the conventional manner for the text's language."
> 
> 
> After doing some more reading and research on citations it really seems as
> if to deal with the diversity/complexity of citation styles/formats the use
> of metadata schema is the most appropriate, what we should be providing in
> HTML is a framework to hang these off. I think cite and footer provide this.

Whether the footer is considered "quoted" or "about the quoted content", surely its purpose is largely indistinguishable. In either case, it is about the nature of that content.

I don't think it is important whether the content is attributed to an author by that same author (as in the copyright notice of a page-level/site footer for someone's website) or by someone else.

(In reply to comment #20)
> (In reply to comment #19)
> > (In reply to comment #18)
> > > (In reply to comment #17)

> > After doing some more reading and research on citations it really seems as
> > if to deal with the diversity/complexity of citation styles/formats the use
> > of metadata schema is the most appropriate, what we should be providing in
> > HTML is a framework to hang these off. I think cite and footer provide this.
> 
> Whether the footer is considered "quoted" or "about the quoted content",
> surely its purpose is largely indistinguishable. In either case, it is about
> the nature of that content.

Hi Steve (and Heydon),

how about, at the very least, point out, may be in a NOTE, the very
problem - acknowledge its existence?

Namely, state that placing the <cite> as child of <blockquote>,
with the intent that it informs the reader about from where the
quotation is taken, though it most often is clear from the context,
it still has the potential to sometimes confuse consumers with regard
to whether the the content within <cite> stems from the source or
 from the one who quoted the source.

And, please, do not only state the problem, but also say that the
solution is to place <cite> outside <blockquote>, e.g. inside
a <figcaption>. (May be it should be said that placing it inside
<footer> also “helps”.)

(In reply to comment #21)
> (In reply to comment #20)
> > (In reply to comment #19)
> > > (In reply to comment #18)
> > > > (In reply to comment #17)
> 
> > > After doing some more reading and research on citations it really seems as
> > > if to deal with the diversity/complexity of citation styles/formats the use
> > > of metadata schema is the most appropriate, what we should be providing in
> > > HTML is a framework to hang these off. I think cite and footer provide this.
> > 
> > Whether the footer is considered "quoted" or "about the quoted content",
> > surely its purpose is largely indistinguishable. In either case, it is about
> > the nature of that content.
> 
> Hi Steve (and Heydon),
> 
> how about, at the very least, point out, may be in a NOTE, the very
> problem - acknowledge its existence?
> 
> Namely, state that placing the <cite> as child of <blockquote>,
> with the intent that it informs the reader about from where the
> quotation is taken, though it most often is clear from the context,
> it still has the potential to sometimes confuse consumers with regard
> to whether the the content within <cite> stems from the source or
>  from the one who quoted the source.
> 
> And, please, do not only state the problem, but also say that the
> solution is to place <cite> outside <blockquote>, e.g. inside
> a <figcaption>. (May be it should be said that placing it inside
> <footer> also “helps”.)

sure, will add some text to that effect.

have also been mulling over this pattern

<figure><blockquote></blockquote><footer></footer></figure>

for situations where the content is not a caption with some citation stuff in it as in the example in the spec, but is just a citation.

(In reply to comment #22)
> (In reply to comment #21)

> sure, will add some text to that effect.


Good. I will close the other bugs, once you do.


> have also been mulling over this pattern
> 
> <figure><blockquote></blockquote><footer></footer></figure>
> 
> for situations where the content is not a caption with some citation stuff
> in it as in the example in the spec, but is just a citation.

Good idea!

EDITOR'S RESPONSE: This is an Editor's Response to your comment. If you are
satisfied with this response, please change the state of this bug to CLOSED. If
you have additional information and would like the Editor to reconsider, please
reopen this bug. If you would like to escalate the issue to the full HTML
Working Group, please add the TrackerRequest keyword to this bug, and suggest
title and text for the Tracker Issue; or you may create a Tracker Issue
yourself, if you are able to do so. For more details, see this document:

   http://dev.w3.org/html5/decision-policy/decision-policy.html

Status: Accepted
Change Description: modified spec to allow in- text notes and citations in blockquote
Rationale: There are strong use cases and ample examples in the wild. By doing so we make the blockquote element more useful.