[Vorbis] Can't access www.xiph.org from Japan

Christopher Harrington webkid at webkid.com
Sun Sep 25 22:25:30 PDT 2005


As much as I can agree about frivolous arguments, I personally don't 
think this is "silly". Quite a number of people are quick to assume that 
something is "broken" simply because it doesn't work for them. As much 
as I'm pleased to see that he went to the effort to find a solution, the 
[correct] solution is not in the server.

His suggested approach is a deviation from standards that are designed 
to make the internet a predictable and well-defined place to operate. 
When we deviate from the standards, we solve one problem, and cause two 
more. Now, web browsers that are designed or configured to request only 
a specific language will now receive a different, wrong language.

After that, you could configure the server to ignore HTTP/1.1 requests 
from browsers that don't properly implement the protocol. Oh, hey, we 
already do that.

Well, while we're at it, why not put up websites that generate badly 
formed HTML and CSS to conform to a proprietary browser's shortcomings?

Oh.

Wow.

Look at that, now we've created an entire internet that only works with 
one browser. An entire internet that is as broken as the client side is. 
And now the line between "the website is broken" and "this alternative 
browser is broken" blurs for every uneducated computer user that comes 
along.

The standards exist for a reason. Don't tell me Stallman would accept 
this kind departure from well-formed, well-defined semantics and 
protocols. Ignoring the standards is tantamount to ignoring the browser 
bugs that created this whole problem in the first place.

Is it so much to expect that a user looking at English web pages can 
configure their browser to do so? In Firefox, it's seven clicks:
Edit -> Preferences -> General -> Languages -> (language dropdown) -> 
English -> Add

(That's assuming General isn't already selected)

In IE, it's even simpler.

The original poster pointed out that the server told him his browser 
didn't accept any of the languages the server had to offer. There is no 
"mystery" there; obviously, something about the browser needs to change 
so that the server will respond correctly. It's reasonable to assume 
that the "Language" option in your browser exists for that very reason.

Sure, I'm a zealot. But you can't tell me that you wouldn't react the 
same way if someone suggested using MP3 instead of Vorbis on a site, 
just to cater to one user. ("It wouldn't break any current 
implementations, everyone can play an MP3! Plus, it's easy to implement. 
Just compress using one different command.")

-Chris Harrington

Aaron Whitehouse wrote:
> Let us not be silly about this.
> 
> The original post identified the problem and suggested an 
> easy-to-implement solution which would fix it and not offend the 
> correctly configured browsers. There is so much concern about making the 
> Xiph codecs as accepted as possible and yet people want to make a stand 
> at a user's first point of contact. I would not stop people who tied 
> their shoelaces incorrectly from shopping at my store, regardless of 
> whether I wanted to encourage correct shoe-tieing. I think that the 
> original post was very responsible, it pointed to a guide to remedy the 
> problem and quite rightly pointed out that those who have an incorrectly 
> configured browser would not be able to view the site. The fact that 
> they can fix their browser is irrelevant to the fact that, as there is 
> an easy remedy, it is worth Xiph catering to those who have the browser 
> incorrectly set up as well.
> 
> If you are concerned about people setting up their browsers incorrectly, 
> feel free to set up a website about its evils, but if Richard Stallman 
> can put his ideologies aside (re:licensing) in order to make these 
> codecs widely accepted, I think that you probably could as well.
> 
> Aaron
> 
> Christopher Harrington wrote:
[...]


More information about the Vorbis mailing list