- From: Paul Mitchell <paul@paul-mitchell.me.uk>
- Date: Mon, 20 Feb 2006 23:31:09 +0000
- To: "www-style@w3.org" <www-style@w3.org>
David Woolley wrote: >>Some things are universal, and deserve a special place in the world. >>Style and script are universal to browsers, so I just accept that HTML >> >> >I might be wrong, but I am pretty sure that <script> originated from >JavaScript and not from HTML, or at least not from the HTML standardisation >process. > > I'm sure you are right. I'm pretty sure that the adoption of the <style> tag was similarly implementation-led and not standards-led, but I'm not sure if that matters. What we are using and what we use next is more pertinent than how we got here. >Script generally shouldn't be needed in true documents, and in web >applications, the function could be performed by, say, Java, or .NET >bytecode, so script is too narrow a term in those contexts. > > I'm not following. "Generally shouldn't" allows "sometimes must", and what constitutes a "true" document isn't really a matter for you, is it? Neither Java nor .Net languages can reasonably be described as script. Both require pre-compilation into bytecode to be executable, are services provided by host computers rather than browsers, so do not serve the same purpose. >W3C has no jurisdiction over many forms of tagged documents. > > It does have jurisdiction over many forms of tagged documents on the internet, which is all I care about. >Browser directives should, as I understand SGML/XML (and I may be wrong) >be processing instructions, not markup elements. > > So the correct form, supported in addition to the legacy form, and encouraged for future use, would be <?style> and <?script>, which like all processing instructions could appear at any point in the document, depending on how restrictive the relevant definition of "processing instruction". I'm more into working with boundless datastreams than discreet documents, so I can only think it sensible to allow processing instructions anywhere, at any time, for any reason. Document-centric folks might think it sufficient to restrict PIs to the start of the content, but that's only because they have the luxury of something to refer to called "start". >PS. Please do not send HTML to public mailing lists. You will find >that none of the regulars do (and if you are going to use HTML, don't >misuse <br>, <b>, etc., and do make it comply with its declared DTD. > > Sorry. I don't send HTML to the list, my e-mail client does. Your wish is Mozilla Thunderbird's command, which was set to "auto-detect" the format, and is now plain text. The e-mail looks no different than before to me, so I hope it is easier for you to read now, and thank you for pointing out my mistake. I can't tell, but are you for the proposal or against? I'm counting 2 to 1 against so far, but I am still hopeful, and willing to compromise to reach agreement. -- Paul Mitchell www.paul-mitchell.me.uk
Received on Monday, 20 February 2006 23:31:59 UTC