Top 10 Sources is a great example of an OPML Directory where every item is an OPML URL, in this example the owner has created the OPML Directory and all the OPML URL’s that the directory points to.
Then this OPML Directory can be viewed in an OPML outline, or an OPML Browser/Reader…but most impressive is that the OPML Directory can be presented in a HTML webpage…where it looks like an ultimate newsmastering topic directory stream.
Top 10 Sources (home page)
>Topic Headings (currently 24)
>>Sub-Topic Headings (various amounts for each topic)
>>>10 sources within each sub-topic (these sources are the blog RSS feeds)
At the moment I can only see an OPML URL, and a spliced feed for each Sub-Topic…there doesn’t seem to be an OPML URL or spliced feed for each Topic Heading (where’s the river of news at this level)…and there doesn’t seem to be an OPML URL, and a spliced feed for the whole kit and kaboodle.
To make an OPML Directory out of this page I would have to create an OPML for each of the 24 Topic Headings, and include the Sub-Topic OPML’s as the items…then I would have to make a root OPML, and include each of these newly made 24 OPML’s as items…this would be my dynamic OPML Directory.
Actually since all the items in all these OPML’s are feeds, this is more specifically a Reading List Directory.
[ADDED 13/03/06: Bela has supplied us with an OPML URL for the whole of Top 10 Sources, see it in the outline in OPML Workstation, see it in Optimal OPML Reader].
Just say I went and made my own Reading List Directory, great I could browse/read it at Optimal or my personal RSS Reader (if it can use/subscribe to Reading Lists)…but what about a HTML presentation.
What if I could take this OPML Directory and plug it into a newsmastering service, like Blogdigger Groups, now that would be awesome…I could edit the content remotely (eg. an OPML outliner like OPML Workstation), but it could be presented elsewhere (eg. Blogdigger Groups).
Then I could also search the portal, create search feeds, and all the content would be archived.
Also see Dynamic newsmastering with OPML.
I just noticed that we are offered 2 type of OPML URL’s for each Sub Topic, one for an OPML Browser, and one for an RSS Reader.
The one for the aggregator (RSS Reader) is subscribing to a Reading List, so far BlogBridge and NewsRiver are the only RSS Readers with this feature.
This OPML (Reading List) can also be read with Opod, Bitty, Taskable, OPML2PDF, or Optimal…these are all OPML Browsers that can also read feeds.
The one for OPML Browsers can be browsed at the above 4 OPML Readers (excluding OPML2PDF), plus other OPML Browsers such as OPML Surfer, Voidstar OPML Browser, etc…
This one can’t be used in RSS Readers as the items are not feeds, they are HTML links.
Example - Shoes
OPML - for the OPML Browser
OPod (try it)
Voidstar OPML Browser
I don’t quite understand how this OPML is kept up to date with the latest posts from each source, can anyone shed any light here?
…as the items in this OPML are just links, how do the links for each source appear automatically in this OPML in order for it to be kept up to date?
The only way I could see this happen is if each source blog had an OPML of its own…see here.
Then your would create an OPML and each item would be the OPML of each blog, this is a bit hard when blogs usually don’t have OPML’s.
[ADDED 13/03/06: See comment below, apparently they are updated as they are cached from a database…so now I’m wondering why you don’t see all the items as an archive, maybe you just see the last cache.
This sounds similar to how Utils mirrors the del.icio.us database every 4 hours]
OPML - for the RSS Reader
OPod (try it)
Bitty (notice that you are feed reading and not reading from the native page)
Optimal (notice that you can read from the feed)
OPML2PDF (ignore the page heading)
Notice you can’t read the feed content in Voidstar OPML Browser, or OPML Surfer.
Also see other OPML Browsers/Readers/Editors at these 2 posts.