Shuttleworth on Jaunty, netbooks and more
Exclusive: To celebrate the 10th Ubuntu release - Jaunty Jackalope - we're giving you three in-depth articles about the world's most popular distro. Hopefully by now you've already read our article The Road to Jaunty: a look back at Ubuntu's history and our frankenreview of Ubuntu 9.04, so we're following them up with an interview with Mark Shuttleworth, the founder and Benevolent Dictator for Life of Ubuntu.
We went to visit the Canonical HQ deep in
Mount Doom London, England, and asked Mark about his favourite Jaunty features, the netbook push, cloud computing and more...
TuxRadar: Would you agree that the last release of Ubuntu, codenamed Intrepid, wasn’t all that Intrepid?
Mark Shuttleworth: The folks who said that were probably the folks who got bitten by a few of Intrepid's moves. We took a couple of bold moves in Intrepid which were too complex in the end to try and get everything in.
And that’s challenging because on the one hand we want to keep the releases moving. We have to bring new technology in - we have to make tough decisions.
On the other hand, people are really dependent on their Linux systems. The challenge for us is becoming something that you want to give your mum, but you don’t want your mum to have to discover all the complexities of a new sound system.
So, I don’t know really how to respond. It keeps all of us up at night; it’s a big responsibility that we don’t take unnecessary risks with people’s systems, but at the same time we’re also here to deliver the very best of the open source world and that means bringing in new components and new pieces.
"It keeps all of us up at night; it’s a big responsibility that we don’t take unnecessary risks with people’s systems"
TR: I think the way a lot of folks see it is that after a Long Term Support release, we expect that Canonical will release something amazing and awesome. Will we see more of that in Jaunty?
MS: Well, in that dimension there are two tacts. One is all the amazing stuff that's coming in from up-stream. In that sense we don’t control that flow. If there isn’t a new version of Firefox, or OpenOffice.org, or if it’s a little bit late for the release cycle, then that defines how intrepid we get.
On the flip side, we look at things that are driven within the Ubuntu community - things that we are supporting. And I think you’re right, it turned out that in Intrepid we didn’t have a flood of things we felt were mature enough to be included in that release.
TR: So what’s big in Jaunty. What’s your favourite feature - with the exception of notifications?
MS: Two things. On the server side there’s been an enormous amount of work around the cloud. And we’ve taken a really laser focussed view of the cloud because the cloud is being used as a catch-all term for any kind of internet-oriented stuff at the moment. Specifically, Amazon EC2, and try to make it really slick and easy to flow smoothly, to flow work between an Ubuntu environment and EC2, and then bringing that same experience into your own data centre. Actually creating your own little EC2.
So either being able to throw stuff back to Amazon, or running stuff yourself, but using the same approach, the same cloud framework. On the server side, I think that’s really cool.
On the client side, I think Brasero is a very cool new piece of Gnome 2.26. There is constant change that comes in from things like Brasero, OOo3 (that’s new in Jaunty), and I do kind of like that Notification stuff.
TR: Is that unique to Ubuntu?
MS: Well, it’s open source, so one hopes that it will be more widely picked up.
It was conceived at Canonical, designed and built and then released open source, and moved into the distro. It’s something we hope to contribute to freedesktop.org and to the broader Gnome and KDE communities.
And it’s a lot of fun to do something like that. We thought notifications were a key system level capability that was at version 0.9. It was actually a freedesktop.org specification that was at v0.9 that just needed a bit of work to help it drive towards something definitive and be considered a completion. We’ll be submitting all this work as a proposal for freedesktop 1.0.
TR: Isn’t the cloud computing angle something that’s being pushed for Karmic Koala Eucalyptus (Canonical's open source system for creating EC2-like clouds) ?
MS: Yeah. So, 9.04 places the first pieces of that. Eucalyptus will actually go into Universe for 9.04, so you’ll be able to compile it and get a feel for these pieces in 9.04. And in Karmic, we’ll move those things to main and they’ll become foundational. Hopefully that will set us up for a great server-side LTS in 10.04.
TR: Jaunty had a target of 25 sec boot time. Has that been achieved?
MS: I believe it has. Reference there is Dell Mini 9 with an SSD.
MS: Yeah, although it’s an Atom processor, so it’s not a very hefty CPU, but the SSD does help. And yeah, I think we’re sub 25 there. Maybe pushing 20.
TR: But there’s more work taking place in Koala?
MS: There’s still a fair amount of work that can be done to optimise the boot process. We’ve just kicked off a process of consultations and discussions with Debian because we believe in order to really optimise the boot you have to be willing to make that a priority in the way you do boot system initialization. So that’s totally pervasive. Every single package has to think about that. So we need to do that in consultation with Debian.
But Debian is interested in that too. There’s a sprint going down between folks at Canonical, some of the Ubuntu community guys, and the Debian guys, and hopefully we’ll come out with a clear plan about that. Upstart, which is something we pioneered here at Ubuntu, has now been embraced by Red Hat and a couple of others. Upstart we think is a key piece of that. If we use it to its full potential, we can cut boot time even more.
TR: Looking at wider goals. Jaunty is the 10th release of Ubuntu. Did you expect to get this far?
MS: I did, but I didn’t expect it to be this much fun and I didn’t expect there to be so many of us! We have this extraordinary community of tens of thousands of people, and we build this thing together every six months.
"I didn’t expect it to be this much fun and I didn’t expect there to be so many of us!"
TR: Do you think your goals have been achieved so far?
MS: My goals are complex enough that I have to say yes and no, and a whole bunch of things have been achieved that I never imagined or could have dreamt of. We’ve certainly proved that Linux on the consumer desktop isn’t fantasy. So that’s a check.
We’ve also shown that really producing free software and making it available for everyone to use the same binaries without licensing, and focusing on services, is a really interesting model.
Google is doing that with Android. Microsoft is doing that with pieces of Windows. So, I feel like ‘check’, we’ve got Linux being palatable for consumers, and ‘check’, I think we’re teasing at the right elements of the business model as well.
There’s still a lot of work to do to make it something that anybody would whole heartedly recommend to anybody. We still have rough edges and we still have more than rough edges. We still have some real challenges in figuring out how we deliver something that’s totally consistent and clean in something as chaotic as a vast open community.
TR: Moving on, would you say Ubuntu was a little late to the netbook game?
MS: Sure. But I hope that we’ve helped to make netbooks on Linux really fly. A lot of people are delighted with both Ubuntu on netbooks, and specifically NBR on netbooks. So, I wish I had a crystal ball and could tell you exactly how the PC industry would shape up. But what we’ve tried to do there is move quickly to bring the very best of free software to the netbook environment in a format which is absolutely free and open and gives people really the full power of a netbook as a PC.
"What we’ve tried to do there is move quickly to bring the very best of free software to the netbook environment"
TR: I believe Jim Zemlin (Executive Directory of the Linux Foundation) talked recently about having a more unified front for netbooks. How do you think that’s panning out?
MS: I think he’s right. If you look at the 'Linux on netbook' market, there are maybe 20 or 30 different user experiences. Very few of which are shared. The different manufacturers have typically created their own. An ASUS device looks different to an ACER device, looks different to each of the different devices.
On the one hand that gives us a lot of innovation, on the other it really fragments the user base. To me it would be ideal to have a fantastic project like Gnome for netbooks specifically. I think that Moblin could be that project. There’s a tremendous amount of work that’s going into Moblin. Netbooks are the focus. We can only speculate as to what it will look an feel like. But the team behind it is a dream team.
There are strong commonalities with Gnome, in terms of infrastructure and plumbing, and in time perhaps, even governance and process. While we have UNR (Ubuntu Netbook Remix), and we’re very proud of UNR, and if it remains the best thing out there we’ll continue to make it the focus. But I would like to see consolidation rather than fragmentation, and we’ll be part of that if it looks like there’s a great option there.
TR: There’s massive desktop deployment of Ubuntu already. And you’ve got the server market, and there’s now the netbook market too. Which would you say was the biggest growth opportunity?
MS: Well, they’re very different. To me building Ubuntu for Linux enthusiasts remains a really important job and I don’t want to take anybody off that. But if I think, of the team that we started Ubuntu with, that team continues really to focus and work to that mission - make it great, free software power user’s desktop, every six months that’s stable and maintained. So we continue to do that, and I think that will continue to be important, if only to keep Andrew Tridgell productive, so Samba gets better.
But since the team has grown, we can also take on the mission of delivering the best of the free software server world as a server platform, and as the team has grown even more we have the resources to go and deliver a fantastic netbook experience as well.
Because I care about folks who don’t make computing their life blood, I think the consumer story is a really interesting one. So for that reason, I think netbooks are really fascinating. So I spend a lot of my own time on the desktop experience, and specifically in the case of a colleague who was installing UNR, how we work within those constraints. How we deliver a netbook experience that’s just outstanding - outstanding for the web, outstanding for personal information management, outstanding for doing your homework or whatever you need to do.
We cornered Mark Shuttleworth in his London skyscraper offices, and he kindly agreed to be a guest presenter in our podcast - listen to the TuxRadar podcast with Mark Shuttleworth here.