Jan-Eirik B. Nævdal schrieb:
Some extra for the pros list:
- Full control over which content to be searchable and not.
- Posibility to make pages searchable almost instant after publication
- Control over when the site is indexed
+1 expecially the last point
you can also add a robot.txt and prohibit spidering of the site to
reduce traffic. google won't index any highly dynamic content, then.
Friendly
Jan-Eirik
On Fri, Nov 13, 2009 at 10:52 AM, Lukáš Vlček <lukas.vl...@gmail.com> wrote:
Hi,
I am looking for good arguments to justify implementation a search for
sites
which are available on the public internet. There are many sites in
"powered
by Solr" section which are indexed by Google and other search engines but
still they decided to invest resources into building and maintenance of
their own search functionality and not to go with [user_query site:
my_site.com] google search. Why?
By no mean I am saying it makes not sense to implement Solr! But I want to
put together list of reasons and possibly with examples. Your help would be
much appreciated!
Let's narrow the scope of this discussion to the following:
- the search should cover several community sites running open source CMSs,
JIRAs, Bugillas ... and the like
- all documents use open formats (no need to parse Word or Excel)
(maybe something close to what LucidImagination does for mailing lists of
Lucene and Solr)
My initial kick off list would be:
pros:
- considering we understand the content (we understand the domain scope) we
can fine tune the search engine to provide more accurate results
- Solr can give us facets
- we have user search logs (valuable for analysis)
- implementing Solr is a fun
cons:
- requires resources (but the cost is relatively low depending on the query
traffic, index size and frequency of updates)
Regards,
Lukas
http://blog.lukas-vlcek.com/
--
Jan Eirik B. Nævdal
Solutions Engineer | +47 982 65 347
Iterate AS | www.iterate.no
The Lean Software Development Consultancy
