[georss] GML georss namespace

[Peter Borissow]

I think Gregor hit it on the head but I'll add another perspective.

My understanding of RSS is that you can extend the XML any way you like provided that you provide a namespace for your non-RSS nodes. 

If you decide to create Java or .NET bindings for the rss schema you'll miss the non-RSS nodes. To bind to the new nodes you'll have to create a new schema that extends the original rss schema with the "custom" extensions (e.g. GeoRSS) and then generate the bindings. This becomes particularly problematic if you want to parse a number of different RSS feeds that each have thier own extensions.

I know in .NET alot of the complexities associated with marshalling/unmarshalling xml are hidden from the developer. In Java, and in JAX-B in particular, the process of generating bindings can be a real pain and can dramatically increase the footprint of your code base. For schemas such as RSS and Atom - it's just not worth it.

Peter



----- Original Message ----
From: Gregor J. Rothfuss <gregor at apache.org>
To: "Knoth, Brian D." <bknoth at mitre.org>
Cc: Peter Borissow <peter.borissow at yahoo.com>; georss at lists.eogeo.org
Sent: Tuesday, March 20, 2007 11:47:35 AM
Subject: Re: [georss] GML georss namespace


Knoth, Brian D. wrote:
> Great, so it seems we agree on the georss application schema - namely,
> that it should probably be updated to reflect correct namespace, etc...
> 
> I did not mention anything about RSS validation and binding, but since
> you've brought it up, I'm curious as to why you think it is a bad idea
> and why it would cost someone down the road. I would think that as more
> machine processing of RSS occurs (such as Yahoo Pipes), that schema
> validation and bindings become more important. Atom has a schema and
> that does not seem to have impact on the extensibility of that
> syndication language.

the reality is that most feeds are not even well-formed xml. if your 
feed stack discards all those, you are left with precious few feeds to 
process. sloppyness in the original rss spec lead to sloppy 
implementations, and eventually a reformulation as atom.

its a long and convoluted story, check wikipedia if you are interested :)

but the takeaway lesson seems to be that a rigorous standard would 
probably not have seen the same levels of adoption, just like sloppy 
html won out over more formal approaches.

in return, we all have to make extra efforts to parse these abominations 
it seems.

-gregor



 
____________________________________________________________________________________
Need Mail bonding?
Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
http://answers.yahoo.com/dir/?link=list&sid=396546091