1 <chapter id="introduction">
2 <!-- $Id: introduction.xml,v 1.25 2003-06-26 08:46:31 mike Exp $ -->
3 <title>Introduction</title>
6 <title>Overview</title>
9 <ulink url="http://indexdata.dk/zebra/">Zebra</ulink>
10 is a high-performance, general-purpose structured text
11 indexing and retrieval engine. It reads records in a
12 variety of input formats (eg. email, XML, MARC) and provides access
13 to them through a powerful combination of boolean search
14 expressions and relevance-ranked free-text queries.
18 Zebra supports large databases (tens of millions of records,
19 tens of gigabytes of data). It allows safe, incremental
20 database updates on live systems. Because Zebra supports
21 the industry-standard information retrieval protocol, Z39.50,
22 you can search Zebra databases using an enormous variety of
23 programs and toolkits, both commercial and free, which understand
24 this protocol. Application libraries are available to allow
25 bespoke clients to be written in Perl, C, C++, Java, Tcl, Visual
26 Basic, Python, PHP and more - see
27 <ulink url="http://zoom.z3950.org/">the ZOOM web site</ulink>
28 for more information on some of these client toolkits.
32 This document is an introduction to the Zebra system. It explains
33 how to compile the software, how to prepare your first database,
34 and how to configure the server to give you the
35 functionality that you need.
40 <title>Features</title>
43 This is an overview of some of Zebra's most important features:
51 Very large databases: logical files can be
52 automatically partitioned over multiple disks.
58 Arbitrarily complex records. The internal data format
59 is a structured format conceptually similar to XML or GRS-1,
60 which allows lists, nested structured data elements and
61 variant forms of data.
67 Robust updating - records can be added and deleted ``on the fly''
68 without rebuilding the index from scratch.
69 Records can be safely updated even while users are accessing
71 The update procedure is tolerant to crashes or hard interrupts
72 during database updating - data can be reconstructed following
79 Configurable to understand many input formats.
80 A system of input filters driven by
81 regular expressions allows most ASCII-based
82 data formats to be easily processed.
83 SGML, XML, ISO2709 (MARC), and raw text are also
90 Searching supports a powerful combination of boolean queries as
91 well as relevance-ranking (free-text) queries. Truncation,
92 masking, full regular expression matching and "approximate
93 matching" (eg. spelling mistakes) are all handled.
99 Index-only databases: data can be, and usually is, imported
100 into Zebra's own storage, but Zebra can also refer to
101 external files, building and maintaining indexes of "live"
108 Zebra is written in portable C, so it runs on most Unix-like systems
109 as well as Windows NT. A binary distribution for Windows NT is
111 <ulink url="http://ftp.indexdata.dk/pub/zebra/win32/"/>,
112 and pre-built packages are available for some Linux
115 <ulink url="http://ftp.indexdata.dk/pub/zebra/RedHat7.X/"/>
116 and Debian packages at
117 <ulink url="http://ftp.indexdata.dk/pub/zebra/debian/"/>
126 Z39.50 protocol support:
133 Protocol facilities: Init, Search, Present (retrieval),
134 Segmentation (support for very large records), Delete, Scan
135 (index browsing), Sort, Close and support for the ``update''
136 Extended Service to add or replace an existing XML record.
139 You can insert/delete/replace an XML record given an
140 "external" ID. Actually this way of doing ES Update was
141 meant for an OAI application that Ian Ibbotson had in
142 mind to implement. The "update" command in YAZ client
143 implements this on the client side. My plan is to make
144 this available in ZOOM "extended" soon..
151 Piggy-backed presents are honored in the search request - that
152 is, a subset of the found records can be returned directly with
153 a search response, enabling search and retrieval to happen in a
160 Named result sets are supported.
166 Easily configured to support different application profiles, with
167 tables for attribute sets, tag sets, and abstract syntaxes.
168 Additional tables control facilities such as element mappings to
169 different schema (eg., GILS-to-USMARC).
175 Complex composition specifications using Espec-1 (partial support).
176 Element sets are defined using the Espec-1 capability,
177 and are specified in configuration files as simple element
178 requests (and, optionally, variant requests).
184 Multiple record syntaxes
185 for data retrieval: GRS-1, SUTRS,
186 XML, ISO2709 (MARC), etc. Records can be mapped between record syntaxes
187 and schemas on the fly.
198 <title>Applications</title>
200 Zebra has been deployed in numerous applications, in both the
201 academic and commercial worlds, in application domains as diverse
202 as bibliographic catalogues, geospatial information, structured
203 vocabulary browsing, government information locators, civic
204 information systems, environmental observations, museum information
208 Notable applications include the following:
212 <title>DADS - the DTV Article Database Service</title>
214 DADS is a huge database of more than ten million records, totalling
215 over ten gigabytes of data. The records are metadata about academic
216 journal articles, primarily scientific; about 10% of these
217 metadata records link to the full text of the articles they
218 describe, a body of about a terabyte of information (although the
219 full text is not indexed.)
222 It allows students and researchers at DTU (Danmarks Tekniske
223 Universitet, the Technical College of Denmark) to find and order
224 articles from multiple databases in a single query. The database
225 contains literature on all engineering subjects. It's available
226 on-line through a web gateway, though currently only to registered
230 More information can be found at
231 <ulink url="http://www.dtv.dk/help/dads/index_e.htm"/>
236 <title>NLI-Z39.50 - a Natural Language Interface for Libraries</title>
238 Fernuniversität Hagen in Germany have developed a natural
239 language interface for access to library databases.
240 <ulink url="http://ki212.fernuni-hagen.de/nli/NLIintro.html"/>
241 In order to evaluate this interface for recall and precision, they
242 chose Zebra as the basis for retrieval effectiveness. The Zebra
243 server contains a copy of the GIRT database, consisting of more
244 than 76000 records in SGML format (bibliographic records from
245 social science), which are mapped to MARC for presentation.
248 (GIRT is the German Indexing and Retrieval Testdatabase. It is a
249 standard German-language test database for intelligent indexing
250 and retrieval systems. See
251 <ulink url="http://www.gesis.org/forschung/informationstechnologie/clef-delos.htm"/>)
254 Evaluation will take place as part of the TREC/CLEF campaign 2003
255 <ulink url="http://clef.iei.pi.cnr.it or http://www4.eurospider.ch/CLEF/"/>
258 For more information, contact Johannes Leveling
259 <email>Johannes.Leveling@FernUni-Hagen.De</email>
264 <title>ULS (Union List of Serials)</title>
267 has created a union catalogue for the periodicals of the
268 twenty-one constituent libraries of the University of London and
269 the University of Westminster
270 (<ulink url="http://www.m25lib.ac.uk/ULS/"/>).
271 They have achieved this using an
272 unusual architecture, which they describe as a
273 ``non-distributed virtual union catalogue''.
276 The member libraries send in data files representing their
277 periodicals, including both brief bibliographic data and summary
278 holdings. Then 21 individual Z39.50 targets are created, each
279 using Zebra, and all mounted on the single hardware server.
280 The live service provides a web gateway allowing Z39.50 searching
281 of all of the targets or a selection of them. Zebra's small
282 footprint allows a relatively modest system to comfortably host
286 More information can be found at
287 <ulink url="http://www.m25lib.ac.uk/ULS/"/>
292 <title>Various web indexes</title>
294 Zebra has been used by a variety of institutions to construct
295 indexes of large web sites, typically in the region of tens of
296 millions of pages. In this role, it functions somewhat similarly
297 to the engine of google or altavista, but for a selected intranet
298 or a subset of the whole Web.
301 For example, Liverpool University's web-search facility (see on
303 <ulink url="http://www.liv.ac.uk/"/>
304 and many sub-pages) works by relevance-searching a Zebra database
305 which is populated by the Harvest-NG web-crawling software.
308 For more information on Liverpool university's intranet search
309 architecture, contact John Gilbertson
310 <email>jgilbert@liverpool.ac.uk</email>
314 <email>lee@arco.de</email>,
315 has recently modified the Harvest web indexer to use Zebra as
316 its native repository engine. His comments on the switch over
317 from the old engine are revealing:
320 The first results after some testing with Zebra are very
321 promising. The tests were done with around 220,000 SOIF files,
322 which occupies 1.6GB of disk space.
325 Building the index from scratch takes around one hour with Zebra
326 where [old-engine] needs around five hours. While [old-engine]
327 blocks search requests when updating its index, Zebra can still
328 answer search requests.
330 Zebra supports incremental indexing which will speed up indexing
334 While the search time of [old-engine] varies from some seconds
335 to some minutes depending how expensive the query is, Zebra
336 usually takes around one to three seconds, even for expensive
339 Zebra can search more than 100 times faster than [old-engine]
340 and can process multiple search requests simultaneously
343 I am very happy to see such nice software available under GPL.
352 <title>Support</title>
354 You can get support for Zebra from at least three sources.
357 First, there's the Zebra web site at
358 <ulink url="http://indexdata.dk/zebra/"/>,
359 which always has the most recent version available for download.
360 If you have a problem with Zebra, the first thing to do is see
361 whether it's fixed in the current release.
364 Second, there's the Zebra mailing list. Its home page at
365 <ulink url="http://indexdata.dk/mailman/listinfo/zebralist"/>
366 includes a complete archive of all messages that have ever been
367 posted on the list. The Zebra mailing list is used both for
368 announcements from the authors (new
369 releases, bug fixes, etc.) and general discussion. You are welcome
370 to seek support there. Join by sending email to
371 <email>zebra-request@indexdata.dk</email> with the word
372 <literal>subscribe</literal> in the body of the message.
375 Third, it's possible to buy a commercial support contract, with
376 well defined service levels and response times, from Index Data.
378 <ulink url="http://indexdata.dk/support2/"/>
385 <title>Future Directions</title>
388 These are some of the plans that we have for the software in the near
389 and far future, ordered approximately as we expect to work on them.
397 Improved support for XML in search and retrieval. Eventually,
398 the goal is for Zebra to pull double duty as a flexible
399 information retrieval engine and high-performance XML
400 repository. The recent addition of XPath searching is one
401 example of the kind of enhancement we're working on.
407 Access to the search engine through SOAP/RPC API to allow the
408 construction of applications without requiring Z39.50 tools.
409 This will shortly be available by means of Index Data's
410 SRW-to-Z39.50 gateway, currently in beta test.
416 Finalisation and documentation of Zebra's C programming
417 API, allowing updates, database management and other functions
418 not readily expressed in Z39.50. We will also consider
419 exposing the API through SOAP.
425 Support for the use of Perl both for access to the Zebra API
426 and for building extension ``plug-ins'' such as input filters.
427 The code for this has been contributed to the source tree by
429 <email>pop@technomat.hu</email>,
430 and is in the process of being integrated and tested.
436 Improved free-text searching. We're first and foremost octet jockeys and
437 we're actively looking for organisations or people who'd like
438 to contribute experience in relevance ranking and text
447 Programmers thrive on user feedback. If you are interested in a
448 facility that you don't see mentioned here, or if there's something
449 you think we could do better, please drop us a mail. Better still,
450 implement it and send us the patches.
453 If you think it's all really neat, you're welcome to drop us a line
454 saying that, too. You can email us on
455 <email>info@indexdata.dk</email>
456 or check the contact info at the end of this manual.
461 <!-- Keep this comment at the end of the file
466 sgml-minimize-attributes:nil
467 sgml-always-quote-attributes:t
470 sgml-parent-document: "zebra.xml"
471 sgml-local-catalogs: nil
472 sgml-namecase-general:t