Re: Re: The Profile and the Feed Validator

Sam Ruby wrote:
>> It depends what you're trying to represent. For most titles, there are
>> several forms of encoding that work equally well. However there are some
>> rare titles (namely those discussing markup) that are best represented in
>> one particular way (assuming you want your feed to be valid).
>
> I think that's backwards.  Here's a title:
>
> <title>I bought some AT&amp;T; it went up.</title>
>
> With some aggregator/browser combinations, there will be an attempt to
> look up an entity named "T".  Fewer would do so if &#x26; was
> substituted in place of &amp;.

I don't follow. In both cases, the feed first has to pass through an XML 
parser which will decode the "&amp;" or "&#x26;" and return the string "I 
bought some AT&T; it went up". If the aggregator treats titles as HTML, it's 
going to end up trying to interpret that "&T;" as an entity regardless of 
how you encoded the ampersand at the XML level.

It generally doesn't matter, though, since T isn't a valid entity and the 
default fallback for an unknown entity is to display it undecoded. The only 
time this is typically a problem is when the the resulting text (of the XML 
level decoding) does contain a valid entity, i.e. you're talking about 
markup.

> Again, I think that's backwards; but we don't need to theorize.  I've
> deployed a test version of the feedvalidator at beta.feedvalidator.org;
> I've looked at the output of a number of feeds, and don't see the
> problem of which you speak.  Can you provide a few examples of
> non-artificial feeds for which the beta feedvalidator presents advice
> that isn't warranted?

Here are just a few examples that I've collected in the past:

Gizmodo:        <title>Other Bits &amp; Bytes</title>
Engadget:       <title>AT&amp;T U-verse set top box "hacked"</title>
NYTimes:        <title>Hackers Gain Data on AT&#38;T Shoppers</title>
Techdirt:       <title>Stolen AT&amp;T Credit-Card Info Used To Launch 
Phishing Scam</title>
gigaom:         <title>Now Even AT&#038;T Loves MuniFi</title>
geek.com:       <title>MAIN: AT&amp;T getting into the Wireless 
Cities</title>
BBC:            <title>M&amp;S to expand electrical goods</title>
talkingpoints:  <title>Rove &amp; Company...</title>
ubergizmo:      <title>Audio &amp; Illuminated Comforter for kids</title>
youtube:        <title>Pam &amp; Jim</title>

I'm assuming all of the above would be given warnings in the beta validator. 
However, there are very few aggregators that would have a problem displaying 
such titles; and those that can't handle those titles, wouldn't cope any 
better with hexadecimal entities.

Regards
James



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/rss-public/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/rss-public/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:[email protected] 
    mailto:[email protected]

<*> To unsubscribe from this group, send an email to:
    [email protected]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/