An aggregator should push as well as pull data. There is a new Drupal
sandbox project for this type of operation at
http://drupal.org/sandbox/dasjo/1293622 which made me think of this
topic again.
From their about page http://www.austrofeedr.at/about-us: "Wir stehen
vor einem Paradigmenwechsel von Pull zu Push, was die Grundlage für das
Real‐Time‐Web darstellt."
They are pulling in data from various divers types of sources,
filtering it for relevance to their audience (in this case, Austria),
combining it to create e.g. flood warnings (this part has no obvious
application in Classics!), and pushing it through various channels,
topic based and / or depending on the device or software used by the
subscriber.
The structure, visualised in these slides:
http://www.austrofeedr.at/sites/austrofeedr.drupalgardens.com/files/austrofeedr_presentation_brixen.pdf
Much of what they doing, inlcuding the emphasis on rapid communication,
is not directly applicable to a Classics aggregator, and is needlessly
complicated. However, the central concept of agreggating data from
various channels, arranging it taxonmically, and pushing it out in
relevant channels, does strike me as relevant. It might be fun to play
with the Drupal install profile they have built to do this job, and see
how it might work.
John Birchall
|