- From: Robert Brodrecht <whatwg@robertdot.org>
- Date: Thu, 22 Mar 2007 19:27:06 -0700
On Mar 22, 2007, at 5:08 PM, Nicholas Shanks wrote: > ? Bullet 7: I think people marking up computer code in HTML are > completely wasting their time. Most sample code I have seen doesn't > bother. e.g. some random OpenGL sample code: > http://developer.apple.com/samplecode/Red_Rocket/listing4.html > http://nehe.gamedev.net/data/lessons/lesson.asp?lesson=Mac_OS_X > The usage case for this vs. usage of HTML for the rest of the > internet is insufficient to earn the right to be in HTML. Just because "most ... doesn't bother" doesn't mean it ought to be removed. Most people just use <i> instead of <em>, <cite>, or whatever. By your logic, we don't need anything other than <i>, but you already argued for using things other than <i> to counter the articles points. The problem is that to write semantically valid markup, there will be tags that don't get used very often. So let's not ignore elements because "no one uses them." Ignore them because they are useless. <code> is not useless. Code needs to be monospaced, and, sometimes formated like <pre>. We could use <pre> but <pre> is just as bad as <i>. It only says "this is preformatted" and it has no semantic meaning. <code> means "This is computer code." Using some CSS, I can make <code> act like <pre> AND it will mean something (e.g. a search engine that helps people find computer code can index stuff in <code> elements). And, frankly, you are wrong. Lots of places I go markup code with <code>. [1][2][3][4][5][6][7][8][9] The examples you give markup their code. The problem is that there is no "blockcode" element (<blockcode> would be to <code> as <blockquote> is to <q>). They use <pre> because it is all they have. <code> isn't powerful enough as it is, in my opinion. > I fear that in 100 years we'll be downloading free shampoo to our > molecular synthesizers that will come wrapped in HTML <samp> tags. Well, only if the shampoo sample is output from a computer program. We do have to care about the semantics... Joe Clarke isn't calling for the removal of computer science elements. [10] He's asking for other elements be added that will help, e.g., newspaper publishers. As far as I know, HTML 5 accounts for many of these. [1] http://www.456bereastreet.com/archive/200702/worst_code_ever/ [2] http://alistapart.com/articles/flashembedcagematch [3] http://www.cherny.com/webdev/70/javascript-event-delegation-and- event-hanlders [4] http://robertdot.org/2006/07/18/json-an-alternative-to-xml-in- ajax.html [5] http://bitesizestandards.com/bites/automatic-coloured-rows [6] http://www.json.org/js.html [7] http://www.php.net/manual/en/function.echo.php [8] http://developer.mozilla.org/en/docs/DOM:document.getElementById [9] http://joeclark.org/book/sashay/serialization/Chapter08.html [10] http://blog.fawny.org/2006/10/28/tbl-html/#TBL-HTML-not-the- problem-2006.10.28 ---------------------------------------------------------- Robert <http://robertdot.org> -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.whatwg.org/pipermail/whatwg-whatwg.org/attachments/20070322/009bde66/attachment.htm>
Received on Thursday, 22 March 2007 19:27:06 UTC