Re: [tlug] Firefox 3.0.1 doesn't respect <meta http-equiv="content-type">

[Shin MICHIMUKO <smitimko@example.com>]

I think that there was some discussions like follows:

http://padawan.info/2004/07/debugging-chars.html

http://edin.no-ip.com/legacy/?q=apache2_2_adddefaultcharset_default_behavior_changed_debian_etch

> Curt Sampson writes:
>  > Well, I used to think that the priority (server header over meta tag)
>  > ought to be reversed, but now I wonder if it's not correct. After all,
>  > I'm surely not the only one that has written a server that does encoding
>  > conversion on the fly based on the client.

I agree, I think that the spec should be reverted. However, the Debian 
folks seem to set "AddDefaultCharset" as "off" by default, like above 
document, for the moment. It seems like a practical approach to me.

From: "Stephen J. Turnbull" <stephen@example.com>
Sent: 2008年09月09日 19:28:21

> This may be a W3C frain bart, but pretty clearly either the server
> should respect http-equiv or they should fix it. 

Hm, I believe that it is still too difficult for the web severs to 
determine the content enconding on the fly. At least, it is a very big 
overhead for the website in general.

This was the problem I met when I moved to Apache2.0 few years ago, and 
I thought this was a bug, but there seems to be a reason to set the 
charset by default. So, now I may have to use "AddDefaultCharset" set to 
"utf-8" for the general purposes. But in fact, sometimes I am still 
using "Off" for this directive to keep using the old contents...

---Shin.
--
== Money is one of the minimum conditions to do anything, but... =======
  Shin MICHIMUKO <smitimko@example.com> http://www.peanuts.gr.jp/
============================================ Freedom is everything. ====