<chapter id="examples">
- <!-- $Id: examples.xml,v 1.8 2002-10-10 14:27:18 heikki Exp $ -->
+ <!-- $Id: examples.xml,v 1.13 2002-10-18 16:16:59 mike Exp $ -->
<title>Example Configurations</title>
<sect1>
<screen>
$ yaz-client tcp:@:9999
Connecting...Ok.
- Z> find @attr 1=/GENUS/MEANING @and lizard earthquakes
+ Z> find @attr 1=/GENUS/SPECIES/AUTHOR/@name Wedel
Number of hits: 1
Z> format xml
Z> show 1
</para>
</sect1>
+
<sect1 id="example2">
- <title>Example 2: Supporting Z39.50 Searches</title>
+ <title>Example 2: Supporting Interoperable Searches</title>
<para>
- You may have noticed as <literal>zebraidx</literal> was building
- the database that it issued a warning, which we ignored at the
- time:
- <screen>
- $ zebraidx update records
- 00:45:46-08/10: ../../index/zebraidx(5016) [warn] records/genera.xml:0 Couldn't open GENUS.abs [No such file or directory]
- </screen>
- <!-- FIXME ### This needs more text -->
+ The problem with the previous example is that you need to know the
+ structure of the documents in order to find them. For example,
+ when we wanted to know the genera for which Matt Wedel is an
+ author
+ (<foreignphrase role="taxon">Sauroposeidon proteles</foreignphrase>),
+ we had to formulate a complex XPath
+ <literal>1=/GENUS/SPECIES/AUTHOR/@name</literal>
+ which embodies the knowledge that author names are specified in the
+ <literal>name</literal> attribute of the
+ <literal><AUTHOR></literal> element,
+ which is inside the
+ <literal><SPECIES></literal> element,
+ which in turn is inside the top-level
+ <literal><GENUS></literal> element.
</para>
- </sect1>
-</chapter>
-
-<!--
-
+ <para>
+ This is bad not just because it requires a lot of typing, but more
+ significantly because it ties searching semantics to the physical
+ structure of the searched records. You can't use the same search
+ specification to search two databases if their internal
+ representations are different. Consider an alternative dinosaur
+ database in which the records have author names specified
+ inside an <literal><authorName></literal> element directly
+ inside a top-level <literal><taxon></literal> element: then
+ you'd need to search for them using
+ <literal>1=/taxon/authorName</literal>
+ </para>
+ <para>
+ How, then, can we build broadcasting Information Retrieval
+ applications that look for records in many different databases?
+ The Z39.50 protocol offers a powerful and general solution to this:
+ abstract ``access points''. In the Z39.50 model, an access point
+ is simply a point at which searches can be directed. Nothing is
+ said about implementation: in a given database, an access point
+ might be implemented as an index, a path into physical records, an
+ algorithm for interrogating relational tables or whatever works.
+ The key point is that the semantics of an access point are fixed
+ and well defined.
+ </para>
+ <para>
+ For convenience, access points are gathered into <firstterm>attribute
+ sets</firstterm>. For example, the BIB-1 attribute set is supposed to
+ contain bibliographic access points such as author, title, subject
+ and ISBN; the GEO attribute set contains access points pertaining
+ to geospatial information (bounding box, ###, etc.); the CIMI
+ attribute set contains access points to do with museum collections
+ (provenance, inscriptions, etc.)
+ </para>
+ <para>
+ In practice, the BIB-1 attribute set has tended to be a dumping
+ ground for all sorts of access points, so that, for example, it
+ includes some geospatial access points as well as strictly
+ bibliographic ones. Nevertheless, the key point is that this model
+ allows a layer of abstraction over the physical representation of
+ records in databases.
+ </para>
+ <para>
+ In the BIB-1 attribute set, an author search is represented by
+ access point 1003. (See
+ <ulink url="###bib1-semantics"/>)
+ So we need to configure our dinosaur database so that searches for
+ BIB-1 access point 1003 look the
+ <literal>name</literal> attribute of the
+ <literal><AUTHOR></literal> element,
+ inside the
+ <literal><SPECIES></literal> element,
+ inside the top-level
+ <literal><GENUS></literal> element.
+ </para>
+ <para>
+ This is a two-step process. First, we need to tell Zebra that we
+ want to support the BIB-1 attribute set. Then we need to tell it
+ which elements of its record pertain to access point 1003.
+ </para>
+ <para>
+ We need to create an <link linkend="abs-file">Abstract Syntax
+ file</link> named after the document element of the records we're
+ working with, plus a <literal>.abs</literal> suffix - in this case,
+ <literal>GENUS.abs</literal> - as follows:
+ </para>
+ <itemizedlist>
<listitem>
<para>
- The master configuration file, <literal>zebra.cfg</literal>,
- which is as short and simple as it can be:
- <screen>
- # $Header: /home/cvsroot/idis/doc/examples.xml,v 1.8 2002-10-10 14:27:18 heikki Exp $
- # Bare-bones master configuration file for Zebra
- profilePath: .:../../tab:../../../yaz/tab
- </screen>
- Apart from the comments, which are ignored, all this specifies is
- that the server should recognise the attribute set described in
- the file called
- <literal>bib1.att</literal>.
- ### What is an attribute set?
+
</para>
</listitem>
-
<listitem>
<para>
- The BIB-1 attribute set configuration file,
- <literal>bib1.att</literal>, which is also as short as possible:
- <screen>
- # $Header: /home/cvsroot/idis/doc/examples.xml,v 1.8 2002-10-10 14:27:18 heikki Exp $
- # Bare-bones BIB-1 attribute set file for Zebra
- reference Bib-1
- </screen>
- Apart from the comments, all this specifies is that reference of
- the attribute set described by this file is
- <literal>Bib-1</literal>, a name recognised by the system as
- referring to a well-known opaque identifier that is transmitted
- by clients as part of their searches.
- ### Yeuch! Surely we can say that better!
- </para>
- <para>
- ### Can't we somehow say this trivial thing in the main
- configuration file?
</para>
</listitem>
--->
+ </itemizedlist>
+ </sect1>
+</chapter>
+
<!--
The simplest hello-world example could go like this:
</caption>
</mediaobject>
-Whene the three <*object> thingies inside the top-level <mediaobject>
+Where the three <*object> thingies inside the top-level <mediaobject>
are decreasingly preferred version to include depending on what the
rendering engine can handle. I generated the EPS version of the image
by exporting a line-drawing done in TGIF, then converted that to the