Re: Putting Generator in a feed?

["Bill Kearney" <wkearney99@hotmail.com>]

> That's a good point; at first blush, this seems like a good idea for
> syndic8 and similar efforts; however, there is the very real danger
> (as Morbus implies) that it will create generator-specific
> workarounds... this is what happened (in reverse) with HTML
> browsers.

I don't see it that way.  I see it as a way to ferret out what's not 
complying.  Not to isolate that provider but to encourage them to 
rise up to a level of greater compatibility.  Right now when errors 
occur it points to the creator of the feed.  This is unfair as many 
of them aren't doing anything about writing code.  They're just 
creating content (a good thing) and we don't want to discourage them.
 
> A way to do this without risking this sort of situation might be to
> keep a list of generators and come up with a compliance suite (since
> it's all XML, this shouldn't be hard) to test them with. 

Chicken and the egg paradox.  Where do we start?  If we put it in the 
feed, it's done.  We'll have a list from live sources.  If we create 
a list of providers, how will we cross reference?

I'm troubled about focusing directly on the toolkit providers.  They 
are, indeed, the source of the problem.  But 'attacking' them 
directly doesn't seem very productive.  

Who are they going to believe, a few syndic8 database people or the 
hundreds of people using their code to create content?  I'd rather 
point them to a list of non-compliant feeds and let them see for 
themselves.  If they care then they'll do something.  And if the 
hundreds of content providers care they'll encourage their toolkit 
authors to fix things.  Otherwise they'll move to other toolkits.  

The side benefit for those that feel the need to compete, would be a 
tally of toolkits in use in relation to number of live feeds.  Might 
go a long way to building marketshare for them.  Conversely a broken 
toolkit sells itself short.  

-Bill Kearney