In about 325 BC, an explorer named Pytheas was charged with the task of finding the source of the tin routes. Although the records are unclear and Pytheas' monumental work describing his achievements called On the Ocean has been lost, it appears that he was funded by the town council and merchants of Massalia. The town council of 300 BC was a bit like the World Wide Web consortium is today in that it provided direction for the activities of the community it represented, even though the merchants of the time didn't always heed its advice. Much like the computer industry’s history with standards, everyone appreciated the need for a somewhat democratic organization to put order on the world but that didn't necessarily mean everyone followed its directions.
A similar though somewhat more modest journey began in the library world in 1998 with the establishment of the oss4lib web site and listserv by Dan Chudnov at Yale1. Initial and continuing discussions about the possibility of an Open Source library system on the oss4lib listserv and the linux-in-libraries list2 led to the creation of the OSDLS (Open Source Digital Library System) site by Jeremy Frumkin at the University of Arizona3. Much like the task that lay before Pytheas, the quest for a truly Open Source library system was not completely without precedent. Pytheas would have been aware of many of the voyages that had ended in failure before him and had first hand knowledge of some of the routes where his journey would begin. For their part, libraries had been major adopters of the "in-house" system trend that had marked the days of the mainframe from the 1960s through to the eighties, and had experienced the backlash against in-house applications when public institutions increasingly turned to the marketplace for technical solutions.
Yet if the waters on the journey to Open Source solutions were not completely unknown, there seemed to be enough differences to make the trip worth attempting one more time. Open Source software is fundamentally based on the availability of ubiquitous networks to foster wide scale collaboration and feedback beyond library walls and institutional borders. Furthermore, the nature of software development has also changed dramatically in the last decade. Object and component-based programming models have led to greater freedom in building applications from a variety of sources and with much greater efficiency since the entire life cycle of developing the component has already been completed by someone else4. There are also a growing number of Open Source success stories in the library world, including jake, MyLibrary, and the Greenstone Digital Library Software5.
One of the most important lessons for embarking on a project on the scale of an Open Source library system or to circumvent a trade blockade is to leverage existing knowledge and tools as much as possible. There is speculation that Pytheas may have been successful because he was willing to tap into solutions that were outside of the Greek sea-faring experience6. He may have utilized the talents of sailors from other countries and gathered information from the inhabitants of each strange land he passed in order to get a better sense of what lay ahead. Similarly, Linux benefited greatly from the building blocks constructed for BSD, and as Eric Raymond points out in his groundbreaking Open Source essay, The Cathedral and The Bazaar, part of the genius of Linus Torvalds, the creator of Linux, was in his ability to find the minimal effort path from Point A to Point B7.
It is critical that libraries not make the mistakes experienced in previous decades when institutions resorted to building everything in-house because so few external building blocks were perceived to be available elsewhere. Pytheas knew better, and in order to follow his example, one of the projects of the OSDLS was named after him in acronym form. Of course, it took some time to determine what exactly the acronym should stand for but the title that emerged was “Powerful Yet Tactfully Helpful Electronic Arranger of Sources”. More than anything, PYTHEAS tries not to get in the way of using existing web and mainstream technologies.