Re: [syndication] Re: RSS feed filtered by keywords?

["Bill Kearney" <ml_yahoo@ideaspace.net>]

> >The trouble with filtering is you end up missing quite a lot.  It's
> >one thing to want to apply a search filter against a pool of feed
> >items.  It's another thing to expect it to be able to exclude data.
> >
> I have no idea you this mean by saying that.  If I have a river of items
> flowing in a group of feeds, and I can't read all the items because
> there are too many, then what better way is there than to specify
> keywords to exclude all the items which  do not contain them ?   Now I
> don't know if that was what Ryan originally wanted, but  it's sure what
> I want.

The trouble is keyword logic doesn't often adequately detect what you /mean/ not
what you've told it.  That and while you might consider items using some
keywords others might not use them.   It becomes this endless list of keywords
and cross-references.  Then it falls prey to the stupid "metadata won't work"
fallacies.

I certainly use filters as a searching tool but not as something that blocks
items from view by default.

> But that assumes that it is practical to store all items flowing in
> many-many feeds in a database.  What happens when instead of  200,000
> feeds we are dealing with 200,000,000 feeds or even 200,000,000,000
> feeds.  If the AOLs and the MSNs get the knack of RSS , then we got to
> think of how this is going to scale way way beyond where we are today.

That's, frankly, bullshit.  That;s just another rendition of the old "it'll
never scale" argument.  Please.  So nothing should be tried in the interim?
Right now the run rate of feeds isn't approaching anywhere near the numbers
you're suggesting.  Yes, it's certainly possible that more people will start
sites.  Most won't stick to it so it's not like it'd matter.  As for resources
available, in the days of 200gb drives and 3gHz processors that hardly has any
bearing whatsoever.

> What am i missing ?

There's a whole other aspect of peer coalition building.  It's one thing to try
to use automated tools to filter your way to nirvanna.  It's perhaps more useful
to participate in peer relationships with likeminded people and use that network
as a guiding tool toward items of relevance.  Not as filters.  There's often
value hidden in that noise. It's a tremendously bad idea to think that filtering
yourself away from it will help you become better informed.

I didn't say this was easy.  This is why efforts like RDF, Foaf and the semantic
web are a hell of a lot more valuable than simple XML hacks would have you
believe.

-Bill Kearney