W3C home > Mailing lists > Public > public-html-xml@w3.org > January 2012

Re: New HTML/XML Task Force Report

From: Robin Berjon <robin@berjon.com>
Date: Mon, 16 Jan 2012 11:55:32 +0100
Cc: public-html-xml@w3.org
Message-Id: <5AADB377-F547-4643-8800-31E7C71681FE@berjon.com>
To: Noah Mendelsohn <nrm@arcanedomain.com>
On Jan 13, 2012, at 20:23 , Noah Mendelsohn wrote:
> On this I'm less convinced. The idea of polyglot is that you have a bass document that can be transmitted on different occasions as text/html or application/xhtml+xml. As far as I know, EXI is usable as a (sort of) encoding with the latter, so it's an option for optimizing in the cases where you choose to transmit as XML.

Technically, it's actually an option for both (assuming no document.write I would think).

> With or without, any given transmission of a polyglot document is either the xml or the text media type, not both; it's the underlying document that remains unchanged. I do understand that using EXI changes the bits on the wire for the Entity-body, and that's a difference; I think there's still some value, so worth mentioning.

It does change the bits on the wire in the entity-body, but just as an encoding. There is nothing preventing you from sending a text/html document (perhaps even not polyglot) with Content-Type: text/html and Content-Encoding: exi in the same way that you can use Content-Encoding: gzip. It's a lossy coding (depending on what you consider matters of course), but "just" a coding nevertheless.

I've pinged the EXI WG to see if they may have some concrete applications to speak of (I know that the application of EXI to HTML is being tested in the lab, but I don't know how public that information is at this point).

-- 
Robin Berjon - http://berjon.com/ - @robinberjon
Received on Monday, 16 January 2012 10:56:52 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 16 January 2012 10:56:52 GMT