[cvs] [Wiki] changed: Project/Thomas

[Chuck Hagenbuch]

chuck  Sun, 01 Jun 2008 21:52:15 -0400

Modified page: http://wiki.horde.org/Project/Thomas
New Revision:  1.4
Change log:  add notes on search, more details for hippo, etc.

@@ -23,9 +23,9 @@

 * Jonah and Thomas merge into one application that creates feeds, one or
more per user
 * Hippo is created as a content aggregation application
 * IMP gets feed reading capabilities in a virtual folder
-* Horde_Feed gets the ability to composite feeds together natively, and the
ability to address feeds by universal names
+* Horde_Feed gets the ability to compose feeds together natively, and the
ability to address feeds by universal names (normalization)
 * The feed-reading portal block is retained and improved
 * A generic Horde linkback service is created
(http://en.wikipedia.org/wiki/Linkback)

 +++ Jonah
@@ -56,12 +56,19 @@

 Hippo will be a new Horde 4, PHP 5 application that aggregates content. It
will initially be written to aggregate feeds, both internal from Jonah
natively, and external (through Horde_Feed). Hippo will work with
Horde_Content types and have Input and Output classes. Input classes will
allow reading a content type, such as Horde_Feed feeds, Jonah internal
feeds, comics, etc. Output classes will allow outputting a "remixed" feed
from one or more inputs (such as combining an RSS feed with a few comics
into a daily feed).

 * There will be no direct storage of inputs or outputs, but Horde_Cache
will be used.
-* Some content types such as comics might include code to fetch and store
themselves locally
+* Some content types such as comics might include code to fetch and store
themselves locally; we might decide that Horde_Feed external feeds should be
fetched into a local content table also
 * This application should be able to provide the backend for Planet Horde
(http://planet.horde.org/)
+* Fields to aggregate for feeds: title, url, uid, summary, content if any,
content-type for content, date
+* During Hippo runs, Horde_Log will be used for logging
+* There will be the ability to run filters on Hippo_Input plugins. For
instance, if a certain feed has the same junk in every entry, a filter could
be attached to the Hippo_Input for that feed that would strip it out.
Generic (regex-type) filters should be configurable via the UI; it should be
possible to configure any kind of filter in a config file or separate input
plugin class.
 * This application should eventually be able to replace Klutz
-* Additional possible content types include podcasts, content from other
Horde applications (iCalendar data from Kronolith), etc.
+* Additional possible content types include podcasts, content from other
Horde applications (iCalendar data from Kronolith), "friend information
(http://hasin.wordpress.com/2008/05/31/building-friendfeed-using-php-part-1/), etc.
+
++++ Search
+
+Once we have Horde_Content, we should be able to build Horde_Content_Index
on top of it that builds a full text index of anything in the
horde_content_master table (using the content type objects to fetch each
piece of content for indexing), with either mysql full-text as a backend, or
something like Solr (http://lucene.apache.org/solr/)

 +++ CMS front end?

 I'm just going to preserve a quote from Duck here since I'm sure there's
something useful to be distilled from it but I'm not capable of doing that
distilling right now:
@@ -75,8 +82,10 @@
 > companies uses Horde groupware will really like to see and module for
such
 > publishing and linking to their data that are already in their groupware
 > intranet.
 </code>
+
+Horde_Content might be the backend that we can build this frontend on.

 --ChuckHagenbuch