Re: HTML5 encodings (was: Re: BOCU patent)

From: Jeff Senn (
Date: Mon Dec 28 2009 - 12:05:09 CST

  • Next message: Jukka K. Korpela: "Re: Filtering and displaying untrusted UTF-8"

    On Dec 28, 2009, at 12:27 PM, Peter Krefting wrote:

    >> The implications are that (1) the authors of the present draft know better than authors of previous works on character encoding and (2) compatibility with existing, incorrectly or incompletely marked HTML documents is more important than adherence to standards. This is a departure from all other HTML and XHTML specifications I've ever seen from the W3C.
    > Well, considering the current state of the web, these are sound deviations to make. From experience, a significant portion of all pages on the web marked up as "iso-8859-1" are indeed contain windows-1252 content. If you look at current browsers, you will see that all of them make this assumption. HTML5 just documents the current practice in this regard.

    So, following this logic: I notice that many people just speed up and go through a traffic light when they
    see it turn yellow -- so we should change the law to support this behavior!?

    (No, thank you; I'd prefer a world that at least has the *illusion* of converging to sanity...)

    This archive was generated by hypermail 2.1.5 : Mon Dec 28 2009 - 12:07:30 CST