|
ZSearcher provides the two main parts of a search definition within one simple dialogue box.
The database/servers are selected, then the search itself entered. The full set of search
features of Z39.50 (see technical briefing) is overwhelming. ZSearcher offers three search dialogues simple - one field, advanced - one field plus "restrictions" e.g. by date
and "Boolean" - any number of fields combined using AND, OR and NOT - the resulting search statement is displayed as a tree which is easy to read - although as ever with
complex Boolean statements- care is needed to get the desired results..
It i s e a s y t o s e l e c t , b u il d a n d c h a n g e s e a r c h e
s via a "button per field" approach. Other useful features include the query history so that you can easily re-do a previous search if you need to.
ZSearcher supports the scan service - which if it is available on the database you
wish to search is very useful feature. When you click SCAN instead of SEARCH after creating e.g. a name, title or subject query, then your target server will not search for and
send matching bibliographic records. Instead, the server responds with only an alphabetical list of the closest matching terms indexed in the database. If you highlight an entry in the
tree of terms in the Scan window and click the copy button, then the term will be copied into the Query dialogue for you. You then know that you are using the precise, preferred term for
the subject, author etc. that you may be looking for.
Parallel, sequential or cascade searching |
One of the most useful features in ZSearcher is the ability to
select the search mode - parallel, sequential or cascade. Parallel is when your search is sent simultaneously to all your selected servers/databases, and
results are received simultaneously as the servers respond. Sequential searching is when your search goes to the first selected server/database; only after receiving all the results from
that server does your search go to the next selected server; and so on, according to how many servers are selected. Cascade searching is like sequential searching, except that your
search only continues to another server if no results are received from the previous one. The Parallel option is the default, and produces the broadest results. Sequential or Cascade-type
searches are useful if you are connecting to commercial databases such as OCLC or RLIN, and wish to avoid paying for subsequent searches when you have already received results from another database.
Once records have been retrieved, and ZSearcher gives a commentary on how a connection is
proceeding, the search retrieved records can be saved and perused offline. A MARC file (US format only) can be saved and the records can be transmitted via FTP to a server for loading
to a system or other purpose. This will be very useful if a cataloguer / searcher is not on the same local network as the main system. e.g. a home worker or bureau situation.
 ZSearcher comes with a predefined set of
databases which may be grouped for personal purposes - see illustration. New servers can be added but there is no facility for adding local notes about servers or grouping them by country or subject. If a
server fails to respond it is flagged as "dead" automatically - useful to save time when checking multiple servers in sequential search mode.
A very useful addition to available Z39.50 search clients with adequate searching and good
display manipulation features. When coupled with a cataloguing editor, it should be a valuable addition to the professional cataloguer's toolkit. Currently only USMARC is handled
with UKMARC and UNIMARC in the pipline - as this is a single language product then this is reasonable - although I am surprised that a Canadian product does not handle French. If you
want to use UKMARC, UNIMARC or non-MARC formats like SUTRS or GRS-1 then you will need to look at BookWhere for the time being - but support for UKMARC and UNIMARC as well as GRS-1 and SUTRS will be added shortly. In addition to that, ZSearcher will soon
support a variety of output options: text, XML and various MARC formats. |
MARC tags and short or long labels. A key useful point is the ability to view selected, multiple MARC
records from any set of results - useful for comparison - and hopefully leading to record merge facilities in the upcoming MARC editor.