- From: Martynas Jusevičius <martynas@graphity.org>
- Date: Wed, 15 Oct 2014 02:35:43 +0200
- To: Anne van Kesteren <annevk@annevk.nl>
- Cc: David Sheets <sheets@alum.mit.edu>, Yehuda Katz <wycats@gmail.com>, Sam Ruby <rubys@intertwingly.net>, "www-tag@w3.org" <www-tag@w3.org>
By "addressed" I guess you mean loosening the syntax enough to encorporate most of the tag soup, as in HTML5? Currently the only plausible vision of the next-generation machine-readable Web is Semantic Web/Linked Data. To implement it, we need stronger semantics. For semantics, we need strict syntax. How do we get to semantics if we can't even get the syntax right? How can we get developers to publish and consume Linked Data if they cannot even manage this with HTML? Lowering the standards to the lowest common denominator is not a long-term solution that we need. As a side-efect, HTML5 syntax made writing parsers harder and broke XML pipeline processing model. P.S. Please don't say JSON is a solution to semantic problems. Martynas On Fri, Oct 10, 2014 at 2:54 PM, Anne van Kesteren <annevk@annevk.nl> wrote: > On Fri, Oct 10, 2014 at 2:47 PM, Martynas Jusevičius > <martynas@graphity.org> wrote: >> Last time browser vendors applied this kind of reasoning we got >> <blink> elements, table layouts, and tag soup. >> I guess history is meant to repeat itself. > > I'm not sure how that follows. <blink> was an easter egg that got out > of hand. Table layouts were done because CSS sucked for non-trivial > layout. And tag soup existed because the people in charge of defining > HTML parsing at the time deferred to SGML for that, a system so > complex that there has never really been a complete implementation. > All of these have meanwhile been largely addressed. > > Not having tracked how URLs are actually used and implemented has > certainly lead to "URL soup". That's what I've been working on > addressing for a while now. > > > -- > https://annevankesteren.nl/
Received on Wednesday, 15 October 2014 00:36:10 UTC