Re: [syndication] Auto-aggregation of RSS from Blogroll management system

[Julian Bond <julian_bond@voidstar.com>]

Jeff Barr <jeff@vertexdev.com> wrote:
> What exactly are you looking for, Julian? Do you want to pass
> in the URL to some page which might contain links to feeds,
> and get back the list of feeds on the page? Or all you
> looking for the list of feeds which have referenced that
> page in the past?

This is how it works. This week, DW routed weblogs.com pings into an 
aggregator page of Dean campaign RSS[1]. I looked at that and thought 
there must be many more Dean RSS feeds. So I went to the blogforamerica 
page and found a huge Blogroll. Maybe 60% of the sites on the blogroll 
have an RSS feed[2]. Collecting them all involves going to each site in 
turn and scrabbling around through guesswork, auto-discovery, searching 
for an XML gif or the word syndicate and copying the url into a text 
file.

What I wanted out of this is an automated system where I gave it the 
front page URL and it gave me:-
1) A composite RSS feed of RSS found through the blogroll
2) A list of the RSS feeds found for dropping into some other system.

Now imagine going to a typical A-List blogger with their 50-100 blogroll 
entries. The feed in 1) is going to be close to their own reading in 
their own aggregator. Given the automated blogroll tools it may be 
exactly the list of feeds they read. So 1) and 2) are a view into the 
sources of their thinking.

The point here is that the blogroll, especially with blogroll management 
tools, is a human driven classification tool. The blogroll owner is 
making conscious decisions to include or not include entries. And it's 
comparatively trivial to cut and paste a URL out of a browser and type a 
title to add an entry. I want to create the corresponding composite and 
list of RSS streams automatically because finding the RSS is still a 
pain.

I haven't checked it out fully yet but I suspect bloglines does exactly 
this. If you manage your blogroll and RSS lists through them, the things 
I'm after are just a single click away with a link that can be dropped 
into your site. The reverse is where you start with a blogroll and 
generate the lists.

[1]http://deancommunity.scripting.com/
[2]There's an amazing number of Dean supporter sites that are built with 
frames, odd URLs, frontpage, word, blog*spot with no rss and so on. 
Soon, there's going to be one or more campaign aggregators and these 
people will be left out. From a technical POV, this is a great 
opportunity for RSS evangelism.
--
Julian Bond Email&MSM: julian.bond@voidstar.com
Webmaster:              http://www.ecademy.com/
Personal WebLog:       http://www.voidstar.com/
M: +44 (0)77 5907 2173   T: +44 (0)192 0412 433