The issue isn't whether they are bugs in the generator. The question is
whether an aggregator should [for example] try each of the heuristics
below to try to parse an XML document instead of giving Joe User an ugly
error message because the Ziff Davis Microsoft-Watch or National
Geographic RSS feed doesn't specify the correct encoding in the XML
declaration.
--
PITHY WORDS OF WISDOM
Entropy isn't what it used to be.
This posting is provided "AS IS" with no warranties, and confers no
rights.
>-----Original Message-----
>From: Rich Salz [mailto:]
>Sent: Thursday, January 15, 2004 1:09 PM
>To: Dare Obasanjo
>Cc:
>Subject: Re: [xml-dev] Postel's law, exceptions
>
>> Exactly how does this solve the most common errors in RSS (and most
>> likely ATOM) feeds?
>
>I love concrete data. As the original/main author of the
>popular RSSBandit, I assume you're right about these being the
>main problems.
>Let me push back at each one in turn.
>
>> 1.) Unescaped ampersands in link elements
>
>Where do these happen? Is it the links that the RSS/Atom
>generator creates? If so, then those are just bugs, man. I
>don't have anything new to say.
>
>> 2.) documents with incorrect endoding in the XML declaration
>
>This is bogus content generated by the generator again, right?
>
>> 3.) Use of HTML entities in the document such as in the <title>
>
>This is in the content that RSS/ATom is syndicating? If so,
>then carrying it as base64 doesn't break the XML purity of
>Atom and it leaves the "you interpret the content" issue up to
>the client (client's browser). Best of both possible worlds, no?
>
>Sorry for not getting it all. Could you post an example or two?
> /r$
>
>--
>Rich Salz, Chief Security Architect
>DataPower Technology http://www.datapower.com
>XS40 XML Security Gateway http://www.datapower.com/products/xs40.html
>XML Security Overview http://www.datapower.com/xmldev/xmlsecurity.html
>
>
>
