Re: Sticks, carrots, real XML, ultra-liberal parsing

[Gary Burd <gary@xxxxxxxxxxxxxxxx>]

On Jan 10, 2004, at 10:07 AM, Isofarro wrote:

> Now today I reckon it would be impossible for any single person to 
> author a browser that can be used as a replacement for Internet 
> Explorer. Not because IE is good, but because it has a huge number of 
> error-correcting features to deal with invalid and tag-souped markup. 
> Basing a browser on implementing the HTML specification is no longer 
> good enough for a browser people can use.
>
> Isn't RSS in the same muddle

RSS may not be in the same muddle as HTML. There are differences 
between what's happening with RSS and what happened with HTML.

Netscape Navigator's dominate market share during the early growth of 
the web allowed Navigator to define a de facto HTML standard. This 
standard was not thought out. It was defined by the implementation 
quirks of Navigator. Page authors relied on this de facto standard.

The RSS client market is more diverse. An RSS client can not define a 
de facto standard the way that Navigator did for HTML. This might 
change in the future, but it's true for now. Client diversity 
encourages RSS authors to follow the written specification.

When it comes to generating content, diversity is flipped around.

A large percentage of early HTML content was created by individuals 
typing away in Notepad and other text editors.  It was not practical to 
ask these individuals to fix their HTML.  There were too many of them 
and there was nothing in it for them.

A large percentage of RSS content on the web is generated by a small 
number of applications. If you can get these applications to generate 
valid RSS, then a large percentage of RSS on the web will be valid. The 
developers of these applications should be motivated to produce 
standards compliant RSS because it improves interoperability for their 
customers.