The open source movement continues to take the software industry by storm. In 2005, corporations such as Oracle and BEA joined IBM in contributing significant source code to open source projects. Smaller companies, such as Mergere and Collabnet, have built viable businesses that directly support open source software through open source code contributions and professional services.
What makes the open source movement so significant that corporations world wide recognize that they must embrace it? The answer is simple: Innovative, quality software that provides real solutions to real problems.
Five products were selected as finalists for the 2006 Open Source Tool of the Year award. Traditionally, nominees in this category have targeted technical audiences—most commonly developers—as their primary user base. Frameworks such as Hibernate, tools such as CVS, and development environments such as NetBeans have long ruled the open source world. In 2006, two nominations broke this mold: Firefox and Open Office.
This shift seems to suggest that the open source community is becoming more mature. Development tools target tech-savvy individuals who are often willing to put up with clumsy interfaces, strange nuances, and little support. End user tools, on the other hand, must provide a polished finish, significant documentation, and reliability. Each of the tools nominated for this year’s 2006 Open Source
Tool of the Year award has displayed this type of maturity.
Second to None!
Functional regression testing has traditionally been difficult to automate. Functional testing tools traditionally use record-and-play–style tests that compare recorded Web page request results to the test responses to determine test failure or success. In this approach, tests often fail upon encountering even the most minor changes to page markup. These failures render traditional tools almost useless for functional regression testing—testing that assumes a certain level of change.
Selenium tests are functional tests that run within the browser. Instead of retrieving and comparing content, the Selenium test runner executes scripts within the browser; they imitate the actions a user may take when using an application. Whereas these tests are not immune to unrelated change-induced failures, they are more reliable than record-and-play–style tests.
Perhaps this year was a little too early for Selenium to take home the product of the year award, but developers and testers familiar with Selenium will attest to the fact that its approach to Web testing is revolutionary. As Selenium matures, look for it to capture a wider audience and become a more viable testing solution. Selenium is truly one of this year’s biggest innovations.
When the stated goal of a project is to “provide a compelling replacement” for software that has a 20 year history and millions of users (CVS), your first reaction may be to question the viability of such lofty goals. Subversion is now considered not only a compelling replacement for CVS, but a significantly better alternative.
Revision Control Systems (RCS) are used to track multiple versions of a document. Typically, RCSs are utilized in software development projects to track source code. They provide features such as automated storage and retrieval of document revisions, logging and identification of changes between revisions, and merging of these differences. Over the past 20 years, CVS has become the defacto standard for Revision Control Systems in the software development community.
Subversion boasts most of the features of CVS, but is valued as much more user and administration friendly. Subversion provides atomic commits, directories and meta-data versioning, HTTP access, offline diff utilities, and many more enhancements. These features have prompted Subversion’s popularity to grow rapidly.
Organizations such as the Apache Software Foundation have migrated millions of lines of code from CVS to Subversion. Successes like these will continue to drive Subversion as the leading Revision Control System.
After winning last year’s Development Tool of the Year award, Eclipse has replaced last year’s winner, NetBeans, as the lone IDE nominated for the Open Source Product of the Year Award. Interestingly, it may not be Eclipse’s progress as an IDE, but rather
its strength as a plug-in oriented development platform that has pushed it into the spotlight.
The Eclipse platform provides a “multi-language, multi-platform, multi-vendor environment” upon which development tools can be quickly created and integrated. Eclipse’s greatest strength is the fact that everything, even core functionality, within Eclipse is developed as a plug-in. This provides a level playing field for all extensions and has forced the plug-in API to mature quickly.
Eclipse IDEs are available for languages ranging from Java and Ruby to C/C++ and COBOL. Hundreds of open source plugins and many more commercial ones have been written for Eclipse. The diversity of these plugins ensures developers the ability to find tools that integrate seamlessly into their development environment.
Top Notch Winners!
Congratulations to Mozilla’s Firefox and Open Office, the two projects that share this year’s Open Source Product of the Year Award. Perhaps even more interesting than the fact both of our winners target end users (and not a more technical audience) is the fact that both applications provide viable alternatives to widely deployed Microsoft products.
Some reports indicate that Firefox, From Mozilla Foundation, has captured almost 10% of the browser market. Considering that over the past several years, Microsoft’s Internet Explorer has held over 90% of the browser market, this jump is substantial.
Mozilla-based browsers have been considered second class citizens in the browser space since 1997. 1997 was the year Internet Explorer 4.0 was released—the event that many consider the turning point at which Internet Explorer became the browser of choice, uprooting Netscape Navigato—a Mozilla-based browser—as the leading browser.
The Mozilla Foundation has overcome this history of playing second fiddle through new innovation. New features such as Tabbed Browsing, integrated search, live bookmarks, and integrated popup blocking have attracted new users. Firefox’s commitment to being a lightweight and fast browser has ensured its success is not temporary. The extension mechanism Firefox provides has allowed developers to supply add ins that users can easily download and install. These extensions provide additional bells and whistles to the core browser, allowing users to customize their experience while maintaining core product’s vision of remaining a lightweight and fast browser.
Open Office 2.0
Breaking the grasp of Microsoft’s hold on the browser market may seem daunting, but OpenOffice.org is in the midst of charting new territory in competing with Microsoft’s Office Suite.
In 2000, Sun Microsystems open sourced StarOffice as under the name Open Office. After five years of development, the suite is beginning to make waves as a viable option to the Microsoft Office Suite. In 2005, OpenOffice 2.0 Beta was released. The suite includes a word processor, spreadsheet application, presentation tool, and more. Advanced tools such as mail merge, the ability to read and write Microsoft office documents, macro recorders, and Macromedia flash integration are just some of the innovations that have users thinking Open Office is ready for prime time.
The recent decision of the Massachusetts government to standardize their office documents on the OpenDocument standard (implemented by OpenOffice) has undoubtedly provided an opportunity for OpenOffice to build on its recent success. Corporations are beginning to follow suit. Although it may come as a surprise to many that OpenOffice shared the title of Developer.com Open Source Product of the Year, OpenOffice is undoubtedly a worthy recipient of the award.
What can we expect from open source communities in 2006? If last year’s successes are any indication, end user applications and productivity tools seem to be the trend of the future. It seems reasonable to expect Firefox and OpenOffice to gain momentum. Will these successes prompt more Open Source developers to target a wider audience for future development efforts? Only time will tell, but if this year is any indication of things to come, the answer is a resounding YES!
About the Author
David DeWolf is a Senior Architect and Agile Coach at Digital Focus, and an active member of many Open Source communities. He speaks and writes on technical topics, open source, and agile software development. David is a member of the Apache Portals Project Management Committee and an active commiter to Apache Pluto.