Brian Behlendorf unwired: `Learn the science'
|
`Those who learn software engineering as a science will always have an advantage.'
|
WHEN he was an undergraduate student at the University of California, Brian Behlendorf set up wired.com, one of the earliest non-academic web sites.
In 1994, he led the team that built hotwired.com, the first ad-supported site.
Brian's big contribution came in 1995 when he founded the Apache Web Server project.
Apache is a freely available, Unix-based web server programme that is now host to more than 65 per cent of world's web sites; it is also seen as the most popular open source software programme ever. In an e-interview with Karthik Subramanian, Mr. Behlendorf talks on how open source software will live up to its billing as being the "future."
What do you think are the key points that are going to influence major companies shift to open source software?
Software cost and licence conformance. It's no longer acceptable to simply use commercial software without paying for the licence. Not only are the technical mechanisms (licence keys, online registries, etc) for using commercial software becoming more sophisticated at battling piracy, but also international treaties and organisations like the WTO are starting to enforce a stronger stance on IP laws within the Indian Government and large companies. And as the outsourcing industry matures, foreign customers will look more closely at an outsourcer's policies around management of IP. Because of this, companies will need to make their existing use of Microsoft Windows and other commercial applications more legal, and this will be costly; so pursuing Linux, Apache, My SQL, and other alternatives will become much more attractive, as it will be much less expensive even if the Open Source software is less capable (which it usually is not, anyway).
How is open source the right technology when people are still apprehensive about how to make money from it?
There are some who teach software engineering as a skill: here is how to manage a Windows box; there is how to write Visual Basic code. There are others who teach computer engineering as a science: here's how to write an operating system, a hash table, a GUI application, a web service.
Those who learn it as a science will always have an advantage over those who learn it as a skill, since technologies and platforms are under constant change when looked at with an eye for the long term, and the ability to learn about new tech is the most important thing.
From my perspective, Open Source software, by providing both code and community, is a much more valuable collection of software to examine for the purposes of learning software engineering as a science, because you can peer inside and really get to understand how things work, watch how real engineers write code in a distributed way, you can debug a problem with much greater depth, and you can borrow code to use in your own application (within certain constraints). There should be no apprehension about "how to make money from Open Source." The question to ask is "Given that I need to acquire or build some software to solve a problem, is the open source software usable (or improvable) to solve my problem, or is the off-the-shelf commercial software better?" Open Source will not always be the right answer for every technical problem; we will always have commercial software. But it is becoming a viable answer for more and more of them.
Printer friendly
page
Send this article to Friends by
E-Mail
Education Plus
Chennai
Hyderabad
