Re: Will Atom have a "don't process invalid documents" rule?

[Sam Ruby <rubys@xxxxxxxxxxxxxxxx>]

Bill Kearney wrote:
> Ignorance we can help with decent documentation and friendly validators.
> Laziness we can combat with a rigorous validator and, frankly, fear of exposure.
> Should folks find themselves desparate to remain ignorant and lazy, well,
> they're more than welcome to use a spec that better suits them.  It's been my
> experience, however, that by educating people and setting good examples they do
> tend to come around...
>
> -Bill Kearney

I would suggest that people who are participating in this discussion 
spend some quality time with http://validator.w3.org/.  I just tried a 
number of weblogs associated with people in this thread and *each* of 
them failed.

I could congratulate myself by pointing out that my weblog validates. 
But upon further checking, some of my pages with comments don't.  The 
problem is a small leak whereby the URLs associated with comments and 
trackbacks (not the ones inside of content, but the ones associated with 
the author or trackback itself) may have unescaped ampersands.  Oops. 
All it takes is one error to be invalid.  Time to go plug a leak.

Meanwhile, the message here isn't that everybody should go fix their 
problems before they resume their self-righteous indignation, but that 
everybody should go back and fix their problems and SUSPEND their 
self-righteous indignation.  It is causing us to lose supporters [1].

Last month, Mark Pilgrim focused on Atom Aggregator behavior (HTTP 
level) [2].  Perhaps it is time to focus on Atom Aggregator behavior 
(XML level).  Those who feel passionate about this issue should consider 
contributing to the following wiki page:

http://www.intertwingly.net/wiki/pie/AggregatorBehaviorXmlLevel

We *can* provide a list of Aggregator tests.  We *can* provide a 
checklist of things for content producers to be aware of.  For greatest 
impact, lets try to keep this list focused on real-world issues as 
opposed to hypotheticals.

- Sam Ruby

[1] http://journurl.com/support/users/admin/index.cfm?mode=article&entry=633
[2] 
http://diveintomark.org/archives/2003/07/21/atom_aggregator_behavior_http_level