- From: Jukka K. Korpela <jkorpela@cs.tut.fi>
- Date: Fri, 30 Jul 2004 00:09:27 +0300 (EEST)
- To: www-html@w3.org
On Thu, 29 Jul 2004, Karl Dubost wrote: > In the category of structure, I will put things like: > l, p, h, section, ... The crucial question is what "structure" (and "semantics") really mean. These days, quite often "structure" means just syntax - which _is_ structure of a kind of course. If structure means just grouping, I think we are on thin ice. While <div> and, less convincingly, <section> could be regarded as "pure structure" in the sense of lumping text and elements together, <section> implies some semantics too, in some vague sense. And <p> is even more semantic; it indicates a paragraph, which is a crucial concept, though often misunderstood. > In the category of semantics, I will put things like: > address, quote, code, ... They are semantic indeed, at least in principle, though the semantics is not very well defined. And, in part, they express fairly specific semantic relationships. But is the concept of paragraph less semantic than the concept of quotation? > This is not a simple issue, Indeed. And it will keep haunting until a general solution is agreed upon. If I quote an entire paragraph, then it's probably a block quotation, if we keep the dichotomy between block quotation and inline quotation. If I omit some irrelevant part of the paragraph, does the situation change? What if I quote just two sentences, or one sentence, from a ten-sentence paragraph? Or what if I quote just a few words, like "Mundus vult decipi, ergo decipiatur", but do that in a context where the quotation is presented separately as opposite to being embedded into the flow of text? > "Oh it's a inline citation -> quote. Oh no, wait finally I'll move it > outside of the paragraph... doh now it's a blockquote, I have to change > the tag." Yeah, it gets confusing. But I think a key point that has been missed so far is that "block quotations" really mean quotations that are related to, but not embedded into, normal paragraphs. It's more than just presentation, and it's reflected in the wording of the text. And this requires attention, and better markup. At present, we can have a paragraph that tells that a quotation follws, followed by a blockquote element followed by (e.g.) a div element containing a citation (source indication). At markup level it's just three consecutive elements, p, blockquote, div, with no indication of relationship. Thus the default rendering cannot reflect the semantic structure well. One might say that the quotation, no matter how long and block-like, is logically part of the paragraph that refers to it (and perhaps builds on it), but the HTML syntax prevents us from putting blockquote inside p. So I think the primary question is how quotations should be presented in markup, not whether the distinction between block and inline quotation should be present in HTML markup. -- Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Thursday, 29 July 2004 17:09:51 UTC