1 <chapter id="introduction">
2 <!-- $Id: introduction.xml,v 1.45 2007-02-02 14:42:44 marc Exp $ -->
3 <title>Introduction</title>
5 <section id="overview">
6 <title>Overview</title>
9 &zebra; is a free, fast, friendly information management system. It can
10 index records in &xml;/&sgml;, &marc;, e-mail archives and many other
11 formats, and quickly find them using a combination of boolean
12 searching and relevance ranking. Search-and-retrieve applications can
13 be written using &api;s in a wide variety of languages, communicating
14 with the &zebra; server using industry-standard information-retrieval
15 protocols or web services.
18 &zebra; is licensed Open Source, and can be
19 deployed by anyone for any purpose without license fees. The C source
20 code is open to anybody to read and change under the GPL license.
23 &zebra; is a networked component which acts as a reliable &z3950; server
24 for both record/document search, presentation, insert, update and
25 delete operations. In addition, it understands the &sru; family of
26 webservices, which exist in &rest; &get;/&post; and truly &soap; flavors.
29 &zebra; is available as MS Windows 2003 Server (32 bit) self-extracting
30 package as well as GNU/Debian Linux (32 bit and 64 bit) precompiled
31 packages. It has been deployed successfully on other Unix systems,
32 including Sun Sparc, HP Unix, and many variants of Linux and BSD
36 <ulink url="http://www.indexdata.com/zebra/">http://www.indexdata.com/zebra/</ulink>
37 <ulink url="http://ftp.indexdata.dk/pub/zebra/win32/">http://ftp.indexdata.dk/pub/zebra/win32/</ulink>
38 <ulink url="http://ftp.indexdata.dk/pub/zebra/debian/">http://ftp.indexdata.dk/pub/zebra/debian/</ulink>
42 <ulink url="http://indexdata.dk/zebra/">&zebra;</ulink>
43 is a high-performance, general-purpose structured text
44 indexing and retrieval engine. It reads records in a
45 variety of input formats (eg. email, &xml;, &marc;) and provides access
46 to them through a powerful combination of boolean search
47 expressions and relevance-ranked free-text queries.
51 &zebra; supports large databases (tens of millions of records,
52 tens of gigabytes of data). It allows safe, incremental
53 database updates on live systems. Because &zebra; supports
54 the industry-standard information retrieval protocol, &z3950;,
55 you can search &zebra; databases using an enormous variety of
56 programs and toolkits, both commercial and free, which understand
57 this protocol. Application libraries are available to allow
58 bespoke clients to be written in Perl, C, C++, Java, Tcl, Visual
59 Basic, Python, &php; and more - see the
60 <ulink url="&url.zoom;">&zoom; web site</ulink>
61 for more information on some of these client toolkits.
65 This document is an introduction to the &zebra; system. It explains
66 how to compile the software, how to prepare your first database,
67 and how to configure the server to give you the
68 functionality that you need.
72 <section id="features">
73 <title>&zebra; Features Overview</title>
81 <entry><xref linkend=""/></entry>
87 <entry><xref linkend=""/></entry>
93 <entry><xref linkend=""/></entry>
98 <table id="table-features-protocol" frame="top">
99 <title>&zebra; networked protocols</title>
103 <entry>Feature</entry>
104 <entry>Availability</entry>
106 <entry>Reference</entry>
111 <entry>Fundamental operation types</entry>
112 <entry>&z3950;/&sru; explain, search, and scan</entry>
114 <entry><xref linkend="querymodel-operation-types"/></entry>
117 <entry>&z3950; protocol support</entry>
119 <entry> Protocol facilities supported are:
120 Init, Search, Present (retrieval),
121 Segmentation (support for very large records), Delete, Scan
122 (index browsing), Sort, Close and support for the ``update''
123 Extended Service to add or replace an existing &xml;
124 record. Piggy-backed presents are honored in the search
125 request. Named result sets are supported.</entry>
126 <entry><xref linkend=""/></entry>
129 <entry>Web Service support</entry>
130 <entry>&sru_gps;</entry>
131 <entry> The protocol operations <literal>explain</literal>,
132 <literal>searchRetrieve</literal> and <literal>scan</literal>
133 are supported. <ulink url="&url.cql;">&cql;</ulink> to internal
135 conversion is supported. Extended RPN queries
136 for search/retrieve and scan are supported.</entry>
137 <entry><xref linkend=""/></entry>
144 <table id="table-features-search" frame="top">
145 <title>&zebra; search functionality</title>
149 <entry>Feature</entry>
150 <entry>Availability</entry>
152 <entry>Reference</entry>
157 <entry>Query languages</entry>
158 <entry>&cql; and &rpn;/&pqf;</entry>
159 <entry>The type-1 Reverse Polish Notation (&rpn;)
160 and it's textual representation Prefix Query Format (&pqf;) are
161 supported. The Common Query Language (&cql;) can be configured as
162 a mapping from &cql; to &rpn;/&pqf;</entry>
163 <entry><xref linkend="querymodel-query-languages-pqf"/>
164 <xref linkend="querymodel-cql-to-pqf"/></entry>
167 <entry>Complex boolean query tree</entry>
168 <entry>&cql; and &rpn;/&pqf;</entry>
169 <entry>Both &cql; and &rpn;/&pqf; allow atomic query parts (&apt;) to
170 be combined into complex boolean query trees</entry>
171 <entry><xref linkend="querymodel-rpn-tree"/></entry>
174 <entry>Field search</entry>
175 <entry>user defined</entry>
176 <entry>Atomic query parts (&apt;) are either general, or
177 directed at user-specified document fields
179 <entry><xref linkend=""/></entry>
182 <entry>Data normalization</entry>
184 <entry>Data normalization, text tokenization and character mappings can be
185 applied during indexing and searching</entry>
186 <entry><xref linkend=""/></entry>
189 <entry>Predefined field types</entry>
191 <entry>Data fields can be indexed as phrase, as into word tokenized text,
192 as numeric values, url's, dates, and raw binary data.</entry>
193 <entry><xref linkend=""/></entry>
196 <entry>Regular expression matching</entry>
197 <entry>Regexp </entry>
198 <entry>Full regular expression matching and "approximate
199 matching" (eg. spelling mistake corrections) are handled.</entry>
200 <entry><xref linkend=""/></entry>
203 <entry>Search truncation</entry>
206 <entry><xref linkend=""/></entry>
209 <entry>Fuzzy searches</entry>
211 <entry>In addition, fuzzy searches are implemented, where one
212 spelling mistake in search terms is matched</entry>
213 <entry><xref linkend=""/></entry>
220 <table id="table-features-scan" frame="top">
221 <title>&zebra; index scanning</title>
225 <entry>Feature</entry>
226 <entry>Availability</entry>
228 <entry>Reference</entry>
235 <entry><literal>Scan</literal> on a given named index returns all the
236 indexed terms in lexicographical order near the given start term.</entry>
237 <entry><xref linkend=""/></entry>
240 <entry>Facetted browsing</entry>
241 <entry>partial</entry>
242 <entry>&zebra; supports <literal>scan inside a hit
243 set</literal> from a previous search, thus reducing the listed
245 subset of terms found in the documents/records of the hit set.</entry>
246 <entry><xref linkend=""/></entry>
249 <entry>Drill-down or refine-search</entry>
250 <entry>partially</entry>
251 <entry>scanning in result sets can be used to implement
252 drill-down in search clients</entry>
253 <entry><xref linkend=""/></entry>
260 <table id="table-features-presentation" frame="top">
261 <title>&zebra; document presentation</title>
265 <entry>Feature</entry>
266 <entry>Availability</entry>
268 <entry>Reference</entry>
273 <entry>Hit count</entry>
275 <entry>Search results include at any time the total hit count of a given
276 query, either exact computed, or approximative, in case that the
277 hit count exceeds a possible pre-defined hit set truncation
280 <entry><xref linkend=""/></entry>
283 <entry>Paged result sets</entry>
285 <entry>Paging of search requests and present/display request can return any
286 successive number of records from any start position in the hit set,
287 i.e. it is trivial to provide search results in successive pages of
289 <entry><xref linkend=""/></entry>
292 <entry>&xml;ocument transformations</entry>
293 <entry>&xslt; based</entry>
294 <entry> Record presentation can be performed in many pre-defined &xml; data
295 formats, where the original &xml; records are on-the-fly transformed
296 through any preconfigured &xslt; transformation. It is therefore
297 trivial to present records in short/full &xml; views, transforming to
298 RSS, Dublin Core, or other &xml; based data formats, or transform
299 records to XHTML snippets ready for inserting in XHTML pages.</entry>
300 <entry><xref linkend=""/></entry>
303 <entry>Binary record transformations</entry>
304 <entry>&marc;, &usmarc;, &marc21; and &marcxml;</entry>
306 <entry><xref linkend=""/></entry>
309 <entry>Record Syntaxes</entry>
311 <entry> Multiple record syntaxes
312 for data retrieval: &grs1;, &sutrs;,
313 &xml;, ISO2709 (&marc;), etc. Records can be mapped between record syntaxes
314 and schemas on the fly.</entry>
315 <entry><xref linkend=""/></entry>
322 <table id="table-features-sort-rank" frame="top">
323 <title>&zebra; sorting and ranking</title>
327 <entry>Feature</entry>
328 <entry>Availability</entry>
330 <entry>Reference</entry>
336 <entry>numeric, lexicographic</entry>
337 <entry>Sorting on the basis of alpha-numeric and numeric data
338 is supported. Alphanumeric sorts can be configured for different data encodings
339 and locales for European languages. </entry>
340 <entry><xref linkend=""/></entry>
343 <entry>Combined sorting</entry>
345 <entry>Sorting on the basis of combined sorts  e.g. combinations of
346 ascending/descending sorts of lexicographical/numeric/date field data
348 <entry><xref linkend=""/></entry>
351 <entry>Relevance ranking</entry>
352 <entry>TF-IDF like</entry>
353 <entry>Relevance-ranking of free-text queries is supported
354 using a TF-IDF like algorithm.</entry>
355 <entry><xref linkend=""/></entry>
358 <entry>Relevence ranking</entry>
359 <entry>TDF-IDF like</entry>
361 <entry><xref linkend=""/></entry>
369 <table id="table-features-document" frame="top">
370 <title>&zebra; document model</title>
374 <entry>Feature</entry>
375 <entry>Availability</entry>
377 <entry>Reference</entry>
382 <entry>Complex semi-structured Documents</entry>
383 <entry>&xml; and &grs1; Documents</entry>
384 <entry>Both &xml; and &grs1; documents exhibit a &dom; like internal
385 representation allowing for complex indexing and display rules</entry>
386 <entry><xref linkend=""/></entry>
389 <entry>Input document formats</entry>
390 <entry>&xml;, &sgml;, Text, ISO2709 (&marc;)</entry>
392 A system of input filters driven by
393 regular expressions allows most ASCII-based
394 data formats to be easily processed.
395 &sgml;, &xml;, ISO2709 (&marc;), and raw text are also
397 <entry><xref linkend=""/></entry>
400 <entry>Document storage</entry>
401 <entry>Index-only, Key storage, Document storage</entry>
402 <entry>Data can be, and usually is, imported
403 into &zebra;'s own storage, but &zebra; can also refer to
404 external files, building and maintaining indexes of "live"
406 <entry><xref linkend=""/></entry>
415 <table id="table-features-scalability" frame="top">
416 <title>&zebra; data size and scalability</title>
420 <entry>Feature</entry>
421 <entry>Availability</entry>
423 <entry>Reference</entry>
428 <entry>No of records</entry>
429 <entry>40-60 million</entry>
431 <entry><xref linkend=""/></entry>
434 <entry>Data size</entry>
435 <entry>100 GB of record data</entry>
437 <entry><xref linkend=""/></entry>
440 <entry>File pointers</entry>
441 <entry>64 bit</entry>
443 <entry><xref linkend=""/></entry>
446 <entry>Scale out</entry>
447 <entry>multiple discs</entry>
449 <entry><xref linkend=""/></entry>
452 <entry>Performance</entry>
453 <entry><literal>O(n * log N)</literal></entry>
454 <entry> &zebra; query speed and performance is affected roughly by
455 <literal>O(log N)</literal>,
456 where <literal>N</literal> is the total database size, and by
457 <literal>O(n)</literal>, where <literal>n</literal> is the
458 specific query hit set size.</entry>
459 <entry><xref linkend=""/></entry>
462 <entry>Average search times</entry>
464 <entry> Even on very large size databases hit rates of 20 queries per
465 seconds with average query answering time of 1 second are possible,
466 provided that the boolean queries are constructed sufficiently
467 precise to result in hit sets of the order of 1000 to 5.000
469 <entry><xref linkend=""/></entry>
472 <entry>Large databases</entry>
473 <entry>64 file pointers assure that register files can extend
474 the 2 GB limit. Logical files can be
475 automatically partitioned over multiple disks, thus allowing for
476 large databases.</entry>
478 <entry><xref linkend=""/></entry>
485 <table id="table-features-updates" frame="top">
486 <title>&zebra; live updates</title>
490 <entry>Feature</entry>
491 <entry>Availability</entry>
493 <entry>Reference</entry>
498 <entry>Batch updates</entry>
500 <entry>It is possible to schedule record inserts/updates/deletes in any
501 quantity, from single individual handled records to batch updates
502 in strikes of any size, as well as total re-indexing of all records
503 from file system. </entry>
504 <entry><xref linkend=""/></entry>
507 <entry>Incremental updates</entry>
510 <entry><xref linkend=""/></entry>
513 <entry>Remote updates</entry>
514 <entry>&z3950; extended services</entry>
516 <entry><xref linkend=""/></entry>
519 <entry>Live updates</entry>
521 <entry> Data updates are transaction based and can be performed on running
522 &zebra; systems. Full searchability is preserved during life data update due to use
523 of shadow disk areas for update operations. Multiple update transactions at the same time are lined up, to be
524 performed one after each other. Data integrity is preserved.</entry>
525 <entry><xref linkend=""/></entry>
528 <entry>Database updates</entry>
529 <entry>live, incremental updates</entry>
530 <entry>Robust updating - records can be added and deleted ``on the fly''
531 without rebuilding the index from scratch.
532 Records can be safely updated even while users are accessing
534 The update procedure is tolerant to crashes or hard interrupts
535 during database updating - data can be reconstructed following
537 <entry><xref linkend=""/></entry>
543 <table id="table-features-platforms" frame="top">
544 <title>&zebra; supported platforms</title>
548 <entry>Feature</entry>
549 <entry>Availability</entry>
551 <entry>Reference</entry>
558 <entry>GNU Linux (32 and 64bit), journaling Reiser or (better) JFS filesystem
559 on disks. GNU/Debian Linux packages are available</entry>
560 <entry><xref linkend=""/></entry>
564 <entry>tarball</entry>
565 <entry>Usual tarball install possible on many major Unix systems</entry>
566 <entry><xref linkend=""/></entry>
569 <entry>Windows</entry>
571 <entry>Windows installer packages available</entry>
572 <entry><xref linkend=""/></entry>
575 <entry>Supported Platforms</entry>
576 <entry>UNIX, Linux, Windows (NT/2000/2003/XP)</entry>
577 <entry>&zebra; is written in portable C, so it runs on most
578 Unix-like systems as well as Windows (NT/2000/2003/XP). Binary
580 available for GNU/Debian Linux and Windows</entry>
581 <entry><xref linkend=""/></entry>
591 <section id="introduction-apps">
592 <title>References and &zebra; based Applications</title>
594 &zebra; has been deployed in numerous applications, in both the
595 academic and commercial worlds, in application domains as diverse
596 as bibliographic catalogues, geospatial information, structured
597 vocabulary browsing, government information locators, civic
598 information systems, environmental observations, museum information
602 Notable applications include the following:
606 <section id="koha-ils">
607 <title>Koha free open-source ILS</title>
609 <ulink url="http://www.koha.org/">Koha</ulink> is a full-featured
610 open-source ILS, initially developed in
611 New Zealand by Katipo Communications Ltd, and first deployed in
612 January of 2000 for Horowhenua Library Trust. It is currently
613 maintained by a team of software providers and library technology
614 staff from around the globe.
617 <ulink url="http://liblime.com/">LibLime</ulink>,
618 a company that is marketing and supporting Koha, adds in
619 the new release of Koha 3.0 the &zebra;
620 database server to drive its bibliographic database.
623 In early 2005, the Koha project development team began looking at
624 ways to improve &marc; support and overcome scalability limitations
625 in the Koha 2.x series. After extensive evaluations of the best
626 of the Open Source textual database engines - including MySQL
627 full-text searching, PostgreSQL, Lucene and Plucene - the team
631 "&zebra; completely eliminates scalability limitations, because it
632 can support tens of millions of records." explained Joshua
633 Ferraro, LibLime's Technology President and Koha's Project
634 Release Manager. "Our performance tests showed search results in
635 under a second for databases with over 5 million records on a
636 modest i386 900Mhz test server."
639 "&zebra; also includes support for true boolean search expressions
640 and relevance-ranked free-text queries, both of which the Koha
641 2.x series lack. &zebra; also supports incremental and safe
642 database updates, which allow on-the-fly record
643 management. Finally, since &zebra; has at its heart the &z3950;
644 protocol, it greatly improves Koha's support for that critical
648 Although the bibliographic database will be moved to &zebra;, Koha
649 3.0 will continue to use a relational SQL-based database design
650 for the 'factual' database. "Relational database managers have
651 their strengths, in spite of their inability to handle large
652 numbers of bibliographic records efficiently," summed up Ferraro,
653 "We're taking the best from both worlds in our redesigned Koha
657 See also LibLime's newsletter article
658 <ulink url="http://www.liblime.com/newsletter/2006/01/features/koha-earns-its-stripes/">
659 Koha Earns its Stripes</ulink>.
663 <section id="emilda-ils">
664 <title>Emilda open source ILS</title>
666 <ulink url="http://www.emilda.org/">Emilda</ulink>
667 is a complete Integrated Library System, released under the
668 GNU General Public License. It has a
669 full featured Web-OPAC, allowing comprehensive system management
670 from virtually any computer with an Internet connection, has
671 template based layout allowing anyone to alter the visual
672 appearance of Emilda, and is
673 &xml; based language for fast and easy portability to virtually any
675 Currently, Emilda is used at three schools in Espoo, Finland.
678 As a surplus, 100% &marc; compatibility has been achieved using the
679 &zebra; Server from Index Data as backend server.
683 <section id="reindex-ils">
684 <title>ReIndex.Net web based ILS</title>
686 <ulink url="http://www.reindex.net/index.php?lang=en">Reindex.net</ulink>
687 is a netbased library service offering all
688 traditional functions on a very high level plus many new
689 services. Reindex.net is a comprehensive and powerful WEB system
690 based on standards such as &xml; and &z3950;.
691 updates. Reindex supports &marc21;, dan&marc; eller Dublin Core with
695 Reindex.net runs on GNU/Debian Linux with &zebra; and Simpleserver
697 Data for bibliographic data. The relational database system
698 Sybase 9 &xml; is used for
700 Internally &marcxml; is used for bibliographical records. Update
701 utilizes &z3950; extended services.
705 <section id="dads-article-database">
706 <title>DADS - the DTV Article Database
709 DADS is a huge database of more than ten million records, totalling
710 over ten gigabytes of data. The records are metadata about academic
711 journal articles, primarily scientific; about 10% of these
712 metadata records link to the full text of the articles they
713 describe, a body of about a terabyte of information (although the
714 full text is not indexed.)
717 It allows students and researchers at DTU (Danmarks Tekniske
718 Universitet, the Technical College of Denmark) to find and order
719 articles from multiple databases in a single query. The database
720 contains literature on all engineering subjects. It's available
721 on-line through a web gateway, though currently only to registered
725 More information can be found at
726 <ulink url="http://www.dtv.dk/"/> and
727 <ulink url="http://dads.dtv.dk"/>
731 <section id="infonet-eprints">
732 <title>Infonet Eprints</title>
734 The InfoNet Eprints service from the
735 <ulink url="http://www.dtv.dk/">
736 Technical Knowledge Center of Denmark</ulink>
737 provides access to documents stored in
738 eprint/preprint servers and institutional research archives around
739 the world. The service is based on Open Archives Initiative metadata
740 harvesting of selected scientific archives around the world. These
741 open archives offer free and unrestricted access to their contents.
744 Infonet Eprints currently holds 1.4 million records from 16 archives.
745 The online search facility is found at
746 <ulink url="http://preprints.cvt.dk"/>.
750 <section id="alvis-project">
753 The <ulink url="http://www.alvis.info/alvis/">Alvis</ulink> EU
754 project run under the 6th Framework (IST-1-002068-STP)
755 is building a semantic-based peer-to-peer search engine. A
756 consortium of eleven partners from six different European
757 Community countries plus Switzerland and China contribute
758 with expertise in a broad range of specialties including network
759 topologies, routing algorithms, linguistic analysis and
763 The &zebra; information retrieval indexing machine is used inside
764 the Alvis framework to
765 manage huge collections of natural language processed and
766 enhanced &xml; data, coming from a topic relevant web crawl.
767 In this application, &zebra; swallows and manages 37GB of &xml; data
768 in about 4 hours, resulting in search times of fractions of
775 <title>ULS (Union List of Serials)</title>
778 has created a union catalogue for the periodicals of the
779 twenty-one constituent libraries of the University of London and
780 the University of Westminster
781 (<ulink url="http://www.m25lib.ac.uk/ULS/"/>).
782 They have achieved this using an
783 unusual architecture, which they describe as a
784 ``non-distributed virtual union catalogue''.
787 The member libraries send in data files representing their
788 periodicals, including both brief bibliographic data and summary
789 holdings. Then 21 individual &z3950; targets are created, each
790 using &zebra;, and all mounted on the single hardware server.
791 The live service provides a web gateway allowing &z3950; searching
792 of all of the targets or a selection of them. &zebra;'s small
793 footprint allows a relatively modest system to comfortably host
797 More information can be found at
798 <ulink url="http://www.m25lib.ac.uk/ULS/"/>
803 <title>NLI-&z3950; - a Natural Language Interface for Libraries</title>
805 Fernuniversität Hagen in Germany have developed a natural
806 language interface for access to library databases.
808 url="http://ki212.fernuni-hagen.de/nli/NLIintro.html"/> -->
809 In order to evaluate this interface for recall and precision, they
810 chose &zebra; as the basis for retrieval effectiveness. The &zebra;
811 server contains a copy of the GIRT database, consisting of more
812 than 76000 records in &sgml; format (bibliographic records from
813 social science), which are mapped to &marc; for presentation.
816 (GIRT is the German Indexing and Retrieval Testdatabase. It is a
817 standard German-language test database for intelligent indexing
818 and retrieval systems. See
819 <ulink url="http://www.gesis.org/forschung/informationstechnologie/clef-delos.htm"/>)
822 Evaluation will take place as part of the TREC/CLEF campaign 2003
823 <ulink url="http://clef.iei.pi.cnr.it"/>.
824 <!-- or <ulink url="http://www4.eurospider.ch/CLEF/"/> -->
827 For more information, contact Johannes Leveling
828 <email>Johannes.Leveling@FernUni-Hagen.De</email>
832 <section id="various-web-indexes">
833 <title>Various web indexes</title>
835 &zebra; has been used by a variety of institutions to construct
836 indexes of large web sites, typically in the region of tens of
837 millions of pages. In this role, it functions somewhat similarly
838 to the engine of google or altavista, but for a selected intranet
839 or a subset of the whole Web.
842 For example, Liverpool University's web-search facility (see on
844 <ulink url="http://www.liv.ac.uk/"/>
845 and many sub-pages) works by relevance-searching a &zebra; database
846 which is populated by the Harvest-NG web-crawling software.
849 For more information on Liverpool university's intranet search
850 architecture, contact John Gilbertson
851 <email>jgilbert@liverpool.ac.uk</email>
855 has recently modified the Harvest web indexer to use &zebra; as
856 its native repository engine. His comments on the switch over
857 from the old engine are revealing:
860 The first results after some testing with &zebra; are very
861 promising. The tests were done with around 220,000 SOIF files,
862 which occupies 1.6GB of disk space.
865 Building the index from scratch takes around one hour with &zebra;
866 where [old-engine] needs around five hours. While [old-engine]
867 blocks search requests when updating its index, &zebra; can still
868 answer search requests.
870 &zebra; supports incremental indexing which will speed up indexing
874 While the search time of [old-engine] varies from some seconds
875 to some minutes depending how expensive the query is, &zebra;
876 usually takes around one to three seconds, even for expensive
879 &zebra; can search more than 100 times faster than [old-engine]
880 and can process multiple search requests simultaneously
883 I am very happy to see such nice software available under GPL.
891 <section id="introduction-support">
892 <title>Support</title>
894 You can get support for &zebra; from at least three sources.
897 First, there's the &zebra; web site at
898 <ulink url="&url.idzebra;"/>,
899 which always has the most recent version available for download.
900 If you have a problem with &zebra;, the first thing to do is see
901 whether it's fixed in the current release.
904 Second, there's the &zebra; mailing list. Its home page at
905 <ulink url="&url.idzebra.mailinglist;"/>
906 includes a complete archive of all messages that have ever been
907 posted on the list. The &zebra; mailing list is used both for
908 announcements from the authors (new
909 releases, bug fixes, etc.) and general discussion. You are welcome
910 to seek support there. Join by filling the form on the list home page.
913 Third, it's possible to buy a commercial support contract, with
914 well defined service levels and response times, from Index Data.
916 <ulink url="&url.indexdata.support;"/>
922 <section id="future">
923 <title>Future Directions</title>
926 These are some of the plans that we have for the software in the near
927 and far future, ordered approximately as we expect to work on them.
935 Improved support for &xml; in search and retrieval. Eventually,
936 the goal is for &zebra; to pull double duty as a flexible
937 information retrieval engine and high-performance &xml;
938 repository. The recent addition of XPath searching is one
939 example of the kind of enhancement we're working on.
942 There is also the experimental <literal>ALVIS &xslt;</literal>
943 &xml; input filter, which unleashes the full power of &dom; based
944 &xslt; transformations during indexing and record retrieval. Work
945 on this filter has been sponsored by the ALVIS EU project
946 <ulink url="http://www.alvis.info/alvis/"/>. We expect this filter to
947 mature soon, as it is planned to be included in the version 2.0
954 Finalisation and documentation of &zebra;'s C programming
955 &api;, allowing updates, database management and other functions
956 not readily expressed in &z3950;. We will also consider
957 exposing the &api; through &soap;.
963 Improved free-text searching. We're first and foremost octet jockeys and
964 we're actively looking for organisations or people who'd like
965 to contribute experience in relevance ranking and text
974 Programmers thrive on user feedback. If you are interested in a
975 facility that you don't see mentioned here, or if there's something
976 you think we could do better, please drop us a mail. Better still,
977 implement it and send us the patches.
980 If you think it's all really neat, you're welcome to drop us a line
981 saying that, too. You can email us on
982 <email>info@indexdata.dk</email>
983 or check the contact info at the end of this manual.
988 <!-- Keep this comment at the end of the file
993 sgml-minimize-attributes:nil
994 sgml-always-quote-attributes:t
997 sgml-parent-document: "zebra.xml"
998 sgml-local-catalogs: nil
999 sgml-namecase-general:t