Choosing between Zebra and NoZebra

Koha 3.0 is provided in 2 versions : the zebra one and the No-Zebra one. This page will give you directions to decide which version you set.

Zebra version:

• is highly scalable (can support easily more than 300 000 items on a public library with an OPAC widely accessible on the web : see http://catalog.ccfls.org/cgi-bin/koha/opac-main.pl or http://www.mediathequeouestprovence.fr/)
• provide a public (external) z3950 server
• is complex to configure and administrate

NoZebra version:

• is available immediatly once Koha is installed
• is probably not that speedy for catalogue with more than 100 000 items
• does not provide a public (external) z3950 server

to be written

Koha “NZ” provides the same features as Koha with zebra.

Except for two features that will probably not be considered as harmfull by a lot of libraries : - no 'proximity' search : you can't search “HARRY prox(1) POTTER” to get “HARRY POTTER” and not “HARRY blabla blabla POTTER”. - no z3950 server.

Except those 2, Koha “NZ” is exactly the same as Koha “Z”

Once you have installed Koha and choosen “NoZebra”, just start cataloguing, that's as simple as that !

If you need to import datas from a previous ILS, there is a hint to do the import -much- faster. The biblio indexes in “NZ” are stored in mySQL, in some specific tables. We have developped a small script that rebuild the index for the whole database. It is much faster than add a biblio ⇒ update index ⇒ add a biblio ⇒ update index ⇒ … The way to do that is quite easy :

1. set NoZebra OFF (yes, tell Koha you'll work with Zebra)
2. bulkmarcimport your datas. As you don't have zebra installed, it won't be indexed here
3. run misc/migration_tools/rebuild_nozebra.pl remember : export PERL5LIB and export KOHA_CONF before) That will read all the biblios and reindex them all at once.

It's much faster than importing them manually