There is no official history of Koha, but work is well underway on a document that will contain as much of the history as we can find.
The latest version of this will always be in git at history.txt
Please do not hesitate to send patches updating this file, you might wish to add when your library went live with Koha, or when you presented something etc.
In 2009, Joann Ransom and Rosalie Blake, from Horowhenua Library Trust, and Chris Cormack, the original developer, published the story behind the development of Koha in Code4Lib: http://journal.code4lib.org/articles/1638
The below information is out of date, and contains many falsehoods which have attempted to be corrected
Koha was developed in 1999 by a small team of programmers working for a consulting company in New Zealand to address the needs of a small library branch on the island because their vendor-supplied ILS was outmoded and not Y2K compliant. The library was also faced with rising phone line costs as their older system depended on modems to maintain its “sneaker” network. The initial goals were to duplicate functionality and maintain ease of use with TCP/IP connectivity and web-based GUI staff terminals.
Koha was developed in 1999 by a team of Programmers, designers and librarians in New Zealand to fullfil the needs of Horowhenua Library Trust a Library north of Wellington in the North Island (Note New Zealand is a group of islands, not one Island). At the time there was no broadband solutions available to the library, so the system had to be able to operate over modems. The initial goals were to increase functionality and provide a solution that was actually viable for the Library.
It was released as open source by accident, because the consulting firm did not want to assume future development costs or handle technical support. Since then, it has become the last ILS Horowhenua Library Trust and Nelsonville Public Library will ever migrate to because they can continue adding modules or features to customize as their needs evolve. In other words, a strong community project that will thrive if pilot libraries commit their human resources toward upstream development.
It was not released as open source by accident. Katipo Communications (who are not a consulting company) in collusion with the Library Trust decided to release the system under the GPL. This decision had nothing at all to do with Katipo not wanting to assume future development costs or handle technical support. In fact Katipo still does do technical support and actively works on Koha development. The choice to release the software was made to provide some surety to the Library. Katipo is a small company and if it went out of business we didnt want to leave the library with an unsupported .. or unsupportable system. The other reason is Katipo are developers, not sales people, we didnt want to assume the costs of marketing a library system.
Although not the oldest project of its kind, it has made the most effective use of time and attracted the most developers to its fold. In fact, some consider it to be production/stable ready enough for implementation. It is being used in pilot libraries around the world, mainly in school or special libraries of non-profit organizations. Like many open source ILS software packages, this one has the potential to become a universal ILS so long as it manages to garner serious interest within the library marketplace starting with public libraries.
This is more factually correct than the above paragraphs, but is quite dated. Koha is now used in production in many Libraries. In pretty much every type of library there is.
At present, its multilingual language support (albeit non-Unicode) gives it international appeal especially since the developer base is so diffuse around the globe. Technical support is provided through Katipo Communications Limited (the aforementioned consulting firm), e-mail, and mailing lists.
Technical support is provided by numerous organisations and people worldwide not just Katipo (who still arent a consulting firm). Plus the mailing lists, irc, koha docs site, and the wiki
The system operates under Linux or Windows, any web server, any SQL database system, and Perl modules (Date::Manip, DBI, Set::Scalar, DBD::mysql or other SQL database module, AuthenDBI for optional security, CDK for optional telnet interface). The code base itself is highly flexible and can be ported to any number of platforms, perhaps even Apple MacOSX or any flavor of commercial Unix. The Perl connectivity is a bit dated when compared to PHP efforts used in counterparts, but still carries a wide talent pool of capable programmers on the Internet. Fully implemented modules include OPAC, cataloging (original with USMARC support and MARC21 support planned), circulation (with reserves support), as well and simple acquisitions modules. Work has already begun on a Z39.50 client/server, serials, NCIP, and report generator modules. Developers are presently hard at work incorporating all extant MARC formats.
Ill leave this for some of the other developers to fix up
source:
Open Source Integrated Library Systems: An Overview/ Eric Anctil, Jamshid Beheshti.-2003
the one who fixed the wrong information:
Mr.chrisc