Linux expert sees expanded role for Ubuntu on the server

Ubuntu founding member and author Benjamin Mako Hill explains why Ubuntu could have the server chops to make big gains now that version 6.06 is such a success.

After rave reviews as a desktop OS, Ubuntu Linux is finally attracting the support of developers as a server platform.

The expanded sphere of influence for this free operating system is due in part to a renewed effort by developers like Benjamin Mako Hill -- he goes by Mako -- who promote the server side components of the latest Ubuntu release, version 6.06, as a viable alternative to its proprietary counterparts.

More on Ubuntu:

Chapter 7 of "Ubuntu Linux for Non-geeks"

Automatix kicks Ubuntu into gear

Mako is one of the founding members of the Ubuntu project, and today he has returned to graduate school at the MIT Media Lab in Cambridge, Mass., where he works on the One Laptop per Child project -- an ambitious program with a goal to deliver millions of $100 laptops, equipped with Linux, to the developing nations of the world. A member of the Ubuntu Community Council governance board, Mako is also a co-author of The Official Ubuntu Book, available this month from Prentice Hall PTR.

Mako took a break from his work on Ubuntu this week to speak with SearchOpenSource.com about the challenges of generating a community, Ubuntu's server-side additions and the possibility of wide-scale enterprise deployment of this popular OS.

SearchOpenSource.com: Ubuntu has always had a server component, but the discussion around this area of the operating system really ramped up with 6.06. Why now?

Benjamin Mako Hill: Ubuntu has always been a perfectly good server OS. Part of the problem though is that, historically, Debian has been really good for servers, and Ubuntu was really good for the desktop. People assumed that because there was so much effort given to the desktop by Ubuntu developers, the server side would not work at all. But, the truth is we [at Ubuntu] have been running the OS on our servers ever since the Warty Warthog release. The success of the desktop distribution really distracted people from that good effort on the server side.

That said, there have been a number of cosmetic changes. In the first release [it] was called 'custom,' but for the second release we renamed it 'server.' But there were also more meaningful changes. Server administrator stuff is more consistent. There is longer-term support now, as a five-year lifecycle for servers is not unusual anymore. Some telecoms can get up to 10 to 15 years, but we're not ready for that level yet.

So, this renewed focus on the server capabilities could result in an expanded sphere of influence for Ubuntu in the future?

Mako: I have every reason to believe that's the case. We're already seeing more server installs, and over the next six to 12 months, we plan to continue to grow the server team, like with the SPARC version. We've been talking about this since the beginning, but it just made much more sense right now. SPARC support was another area that just made sense, and other kinds of server architectures will arrive depending on what the community wants.

What are your thoughts on Ubuntu becoming a third player in the current battle between Red Hat and Novell?

Mako: Well, it's interesting the way the debate has been framed. People are always describing Ubuntu as if it were in opposition to one of these distributions -- either Debian or Red Hat or Novell, but I think that our goal, honestly, as a project is much more ambitious than that. Our major competitor is proprietary software -- companies like Microsoft. Bug number one in the Ubuntu tracking system is that a majority of people in the world use an OS that is propriety. That bug will only be closed when a majority of people are using free software.

Some big names were thrown around with Ubuntu in the past since months, like Oracle, which was looking to start its own Linux for use in its own stack. Could that be a possible future for this OS?

Mako: Sure. One of the goals of the Ubuntu project was to become a platform on which other projects are developed. With Debian, the primary motto early on was that it should become a universal OS. It is an interesting idea, but I became increasingly convinced that the idea of a single OS that everybody used was a little bit unrealistic.

The reality of the situation is that hundreds of OSes have been based off of Debian, and there have been 90 or so on Red Hat. One size does not fit all. Ubuntu is a derivative distro of Debian. We want Ubuntu to become a platform from which other people build operating systems.

What have been some of the biggest challenges for Ubuntu, even with its popularity?

Mako: The growing community has been a challenge. Today, we are getting more bugs than we can manage. Also, I think part of the biggest challenge was just creating a project as unique as Ubuntu, creating a project that builds a thriving community-based model, and then balancing that with the fact that a paid number of people will be working on it full time. Most other projects that use paid labor, like Red Hat and Novell, tend to have all the code contributed from people paid to work on the software full time. By introducing paid labor, you run the risk of crowding out the volunteer stuff.

I'm quite sure this has happened with Ubuntu, but we tried to make it a compelling volunteer community. People should feel empowered to modify it. Balancing the corporate relationship with Canonical Ltd. and the community has been one of the most difficult things.

The community response to the latest release has been positive, to say the least.

Mako: The reason Ubuntu was able to get all of this support is because of the way the project worked: Dozens of the most active Debian projects got together, and they were asked the question: If you had several million dollars to make Debian rock, what would you do? Partly it was also that Linux distros tend to fall into two places. You have your Red Hat and [Novell] SUSE and Mandrake and other projects whose goal seems to be to build some sort of enterprise-class system. They tend to be relatively smaller in scope, with several hundred packages of software, but they tend to be well integrated. As long as you are using a piece of software that they support, then the package tends to do very well.

Ubuntu went through the stack and got a number of people who were good at touching certain parts of the stack. There weren't a lot of software developers, and there was not a lot of new software, but it allowed all the existing pieces of software to shine. With Ubuntu, we provided a platform on which to build.

Dig deeper on Open source Web and application servers

0 comments

Oldest 

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to:

-ADS BY GOOGLE

SearchDataCenter

SearchServerVirtualization

SearchCloudComputing

SearchEnterpriseDesktop

Close