Can computer software be built to last?
That question has been preoccupying software designer Dan Bricklin, who helped kick-start the personal computer revolution in the 1980s as co-inventor of VisiCalc, the first electronic spreadsheet.
These days, Bricklin is developing software programs, writing a weblog (danbricklin.com/log), speaking to technology groups, and consulting from his perch at Software Garden Inc., which he runs from his Newton home. The focus of his work is open-source software and alternative business models for the companies that make and sell it.
For Bricklin, recently named by Governor Romney to a state information technology advisory panel, open source is key to creating ''Software That Lasts 200 Years," the title of a call to action he wrote and posted on his blog. His essay argues that the structure and culture of most prepackaged software companies -- marked by constant upgrades and replacement -- isn't compatible with society's long-term need for the software that is fast becoming part of our basic infrastructure.
''The world is different now than it was even just a decade or two ago," Bricklin writes. ''In more and more cases, there are no paper records. People expect all information to be available at all times and for new uses, just as they expect to drive the latest vehicle over an old bridge or fill a new high-tech water bottle from an old well's pump."
In addition to keeping records, software today contols and monitors everything from traffic lights to generating plants, Bricklin says. But to produce software as durable as roads, bridges, and other parts of physical infrastructure, he contends we need a new software ecosystem and a new style of software development.
One feature of Bricklin's ecosystem is ''modular" architecture enabling software applications to continue working even as hardware, operating systems, and other platforms change. Another is greater ''transparency" in software code to find needed changes.
Bricklin believes funding for software development should come from users, rather than private sources of capital, and that software projects should be targeted at more than one customer. But he also thinks long-term software is best developed by businesses, using transparency and exploiting other attributes of open-source software.
Some of these ideas are being tested by the new Government Open Code Collaborative, which has created a government software repository.
''We're building a society where we're dependent on software," Bricklin said in an interview. ''We don't want to upgrade everything every two years when a platform changes. You can't go to a Fidelity and say, 'Rewrite all your stuff.' People won't put up with that."
Bricklin acknowledges that open-source software scares many traditional software developers because it has the potential to undermine the business model with which they are most familiar. Ultimately, though, he sees the viability of new business models where companies make their money by maintaining applications, training users, and providing security.
The idea of extending the life of software is gaining traction in academia, said Bala R. Iyer, assistant professor of information systems at Boston University's School of Management. Iyer said he asks students to envision software as a ''stack," with the lower layers -- the operating system, core services, and middleware such as database management systems -- becoming more stable and longer-lasting.
''If you develop in a modular fashion, with well-defined interfaces and good standards, you can get longevity," Iyer said. But he questioned Bricklin's vision of software that can last 200 years. ''That seems a bit of a stretch," Iyer said. ''I think he is challenging us."
Bricklin is clearly doing that as he reimagines the software landscape. ''I don't know where things are going," he admitted. ''I just know it's very fertile. And as an innovator, I look for fertile fields."
Robert Weisman can be reached at weisman@globe.com.![]()