E
Erik Max Francis
Back in 2000 I made a news aggregation site (REALpolitik,
http://www.realpolitik.com/) since I didn't find anything that fit my
needs. (REALpolitik is unfortunately made in Perl; it was my last
significant project before I started using Python for most of my work.)
At the time, RSS had not reached the near-universality that it has
now, so RP itself uses a combination of custom scraping and whatever
XML-type feeds that were available.
I've checked and all the feeds _I_ care about
are available in RSS
now, so it would make sense to move to an RSS aggregator if it has the
same features. I've looked around at some that are available, both in
Python and not, and haven't found anything that had the feature set I
want. One in Python would obviously be a huge benefit.
I'm not looking for anything all that fancy, but there are a combination
of some seemingly basic features I just can't seem to find in other
aggregators. They are:
- one-page display: It's awkward going back and forth between multiple
feeds in a hierarchical format, and so it's much nicer if they're all
presented on one page available for perusal.
- filtering news items: Preferably for filtering out as well as
highlighting, and also being able to selectively pass on, say, the first
item in an RSS feed, since some popular feeds use that slot as an
advertisement.
- caching news items: I read news sporadically throughout the day, so
one feature I really like is the ability to queue up new items over
time, as distinguished by unique GUID. For example, if an RSS feed only
provided one (unique) item at all in its feed and that was updated once
a day, letting the system run for several days would collect them all,
each stored uniquely and available.
- recent items: When you check for news, it only shows you the news
items in each category that are new since you last caught up (catching
up is the equivalent of a "mark all as read" feature). That way, new
news accumulates, and it's only news you haven't seen before.
Somewhat surprisingly to me, I can't seem to find an aggregator that
supports all these features (using Mozilla). Is it possible it's time
for another Yet Another-type project?
http://www.realpolitik.com/) since I didn't find anything that fit my
needs. (REALpolitik is unfortunately made in Perl; it was my last
significant project before I started using Python for most of my work.)
At the time, RSS had not reached the near-universality that it has
now, so RP itself uses a combination of custom scraping and whatever
XML-type feeds that were available.
I've checked and all the feeds _I_ care about
are available in RSS now, so it would make sense to move to an RSS aggregator if it has the
same features. I've looked around at some that are available, both in
Python and not, and haven't found anything that had the feature set I
want. One in Python would obviously be a huge benefit.
I'm not looking for anything all that fancy, but there are a combination
of some seemingly basic features I just can't seem to find in other
aggregators. They are:
- one-page display: It's awkward going back and forth between multiple
feeds in a hierarchical format, and so it's much nicer if they're all
presented on one page available for perusal.
- filtering news items: Preferably for filtering out as well as
highlighting, and also being able to selectively pass on, say, the first
item in an RSS feed, since some popular feeds use that slot as an
advertisement.
- caching news items: I read news sporadically throughout the day, so
one feature I really like is the ability to queue up new items over
time, as distinguished by unique GUID. For example, if an RSS feed only
provided one (unique) item at all in its feed and that was updated once
a day, letting the system run for several days would collect them all,
each stored uniquely and available.
- recent items: When you check for news, it only shows you the news
items in each category that are new since you last caught up (catching
up is the equivalent of a "mark all as read" feature). That way, new
news accumulates, and it's only news you haven't seen before.
Somewhat surprisingly to me, I can't seem to find an aggregator that
supports all these features (using Mozilla). Is it possible it's time
for another Yet Another-type project?