On Mar 21, 2007, at 2:14 AM, Antonio Eggberg wrote:
Well as I mentioned earlier flare rails app have served me well. I
have
left flare code i.e (browse_controller) as is, and started to build
out the
application instead i.e. users/login/settings/admin etc..
So your situation is a prime example of why Flare needs to be a
plugin. What you really want is your own custom Rails application,
with the ability to easily add Flare into it at specific points.
There are couple of things that I miss from flare.. honestly i am
not sure
if i miss it from flare or solr. But I was planning on getting to it
once I solved my "solr feeding issue". Some of them are already in
flares
roadmap as I can see .. but these are also "big discussion stuff"
i.e tagging,
top-down faceting, "frequency based faceting" etc.
Tagging is not something that I plan on implementing in Flare in the
short term. I just don't have a need for it at this level. I have
implemented tagging in Collex, which is "Flare-like" in that it is a
Rails application that leverages solr-ruby, but it also contains some
custom (and, IMO, weakly scalable) Solr request handlers and a lot of
intertwining of a specific domain (19th century literature/art).
The long term goal of Flare is to embrace tagging as a first-class
feature, but my personal development philosophy is to drive these
features by real-world implementations. One of the key benefits to
Flare being open source is for others to extend it and contribute
those features back. With Flare becoming a plugin, it makes it
easier for folks to use it and offer patches. Tagging will involve
clever use of Solr, not just on the Flare side of things. Here is
some brainstorming that has occurred on Solr+tagging: <http://
wiki.apache.org/solr/UserTagDesign>
I'm not sure what you mean by top-down or frequency-based faceting.
Could you elaborate?
http://www.marumushi.com/apps/newsmap/
Very cool! So this is what you are driving with Solr currently?
Where does Flare factor into this?
Erik
