Matt Asay Answers Your Questions About Ubuntu and Canonical 244
A couple of weeks ago you posed some questions for Matt Asay, who recently moved into the COO role at Canonical. Click below to read his answers.
Adoption stories and influences
by eldavojohn (898314)"Every so often I see an adoption story about so-and-so taking up some open source solution and sometimes I think 'Wow, French government? Now it's really going to take off. This is it. It's time.' And then I wait. And wait. Are these stories at all positive for the project? I mean, you would think with states and governments using Ubuntu or Red Hat that it would catch on like wildfire if the savings are there so why isn't that happening? I know Microsoft sends out a lot of Wormtongues to stick in the ears of important people. Do you plan on targeting governments in a similar manner? Does/will Canonical work on making a presence in things like the EU Commissions where we've seen corporations collecting members in their pockets?"
Matt: No, we have no plans to turn Wormtongue. We do, however, have aspirations to play Frodo. :-)
Ultimately, governments (good ones, anyway) are established to reflect the voice of their citizens. At Canonical, we believe that real, lasting change happens from the bottom up, as citizens within government and IT and those served by it clamor for change. We try to help this along by working with government organizations, including open source-friendly lobbying groups, to promote free markets and expanded choice through free and open-source software, but I personally believe that individuals will make the difference.
Change can be expensive, whether in terms of cost or bother, and so as individuals or organizations we generally try to avoid it. But people are now starting to feel enough pain - be it software costs, inefficient use of hardware, viruses and other malware, etc. - that Linux and open-source software, generally, are getting plenty of attention. The cure, in other words, now outweighs the effort of applying it. Yes, Microsoft will do its part to thwart this progress,but even so I've seen broad and ever-increasing government adoption of open source. It's just that most of it doesn't get reported.
Don't lose heart and, in particular, don't lose "voice." We're being heard. The worst thing we could do is to slacken our pace now.
Revenue
by Enderandrew (866215)"Shuttleworth is still funding Canonical. At some point however, this needs to turn into a profitable venture to endure. How does Canonical create lasting revenue streams, and will those decisions come at the cost of usability and freedom in the distro, such as the recent decision to use Yahoo search (powered by Bing) as the default)?"
Matt: First off, it's critical to understand that Canonical doesn't make decisions at the cost of usability. Ever. Usability is our cardinal virtue.
The Yahoo! deal is not at the cost of usability. Yahoo! is an excellent and wildly popular search engine with many many millions of users. We are very pleased to have reached an agreement that will pump additional revenue into the community compared to the existing default. For those worried about Microsoft's involvement with Yahoo!, it is trivially easy to switch to Google or other alternatives.
We will make more commercial for-pay services available to our users, but we will never make then a requirement to have a full experience of the Ubuntu desktop. If you don't like them don't buy them and nothing will make you need to.
We have very healthy revenue coming from our various businesses, the most visible of which is providing support for our OEM partners like Dell as they roll Ubuntu-based devices globally. Less visible, but also fast growing, include our enterprise business (providing support and other services for Ubuntu in cloud and traditional server deployments) and our Ubuntu One services for Ubuntu client users.
I like to think of our guiding principle as "make money because of the Ubuntu community, not from it." At the scale where we operate, all sorts of financial opportunities become possible, opportunities that don't require us to hold back Ubuntu bits to goad people into purchasing. As we roll new services out, I hope you'll let us know how we're doing, and ensure we never sacrifice usability for financial gain.
Freedom, second?
by TheModelEskimo (968202) "Matt, you were intensely criticized by members of the Free Software community for your critical stance facing 'vague concepts' like software freedom and 'no vendor lock-in.' Reading your blog, it seems to me like you are still a fan of focusing on 'high quality software at a compelling price' rather than these other concepts. How will this position affect your work with Canonical and more specifically, its relationship with freedom-first software advocates?"
Matt: I've never considered myself at odds with the goals of freedom-first software advocates, though I sometimes disagree with the means and the timing. Some, for example, have criticized Canonical in the past for including non-free bits (codecs and such).
I'm not among that number, because I believe that if we ever want to see mainstream adoption of Linux, we need to provide solutions, preferably short-term, that map to users' requirements. How likely is it that the mainstream could adopt a Linux desktop, for example, that doesn't offer support for Flash so that people can watch YouTube videos, as just one example?
It's easy to demand that everyone be like us, right now. But that, to me, is the antithesis of freedom. I'm not interesting in forcing people to make a choice. That's no choice at all. I believe the best way is to consistently offer a better experience, and invite prospective users to try it.
Here's a personal example. In my new role, I have switched from using Mac OS X to Ubuntu Linux. I've been using a Mac since 2002 when I switched off Windows. This switch would have been painful but for the fact that Firefox runs so well on Linux, and gives me access to a range of online services (like Google Calendar) that I was using before on the Mac. It would have been doubly so if I couldn't keep using Tweetdeck and other software to which I'd grown accustomed on my Mac.
Over time, I'm sure I'll migrate to open-source alternatives, for the same reason I used Adium, not iChat, on my Mac: the open-source alternatives are often the best available.
But to force-feed "freedom" on me or anyone else is a foolish, losing proposition. Especially in the short term.
I believe that Canonical and the Ubuntu community are creating software that people will want to use, not that they have to use. In the three weeks I've been with Canonical, I've used my Mac exactly once (still moving music out of old, DRM-encrusted iTunes songs). I haven't missed it.
Your version of their vision
by eldavojohn (898314) "Late last year, you heralded some moves by Shuttleworth and you said:
This, I believe, is an opportunity for Canonical to tighten its focus. While Shuttleworth suggests that Silber's appointment 'doesn't mark a change of direction,' perhaps it should. With over 300 employees and products that span mobile, Netbooks and other personal computers, cloud computing, enterprise servers, and more, Canonical has its fingers in a lot of pots.
As COO, what are you going to do to improve the products you highlighted above? I'm not looking for a soft answer like 'I'm going to promote Ubuntu on netbooks' but more so an itemized list of measurable goals, with milestones, dates and areas of focus (for instance, power minded ARM distributions). Is there anything about their vision you intend to change or influence the most?"
Matt: I don't want to offend you with a "soft-ball" answer, but it would be inappropriate for me to provide the level of detail you request, in part because much of this information is confidential to Canonical and our partners, but also because a big part of our strategy is to undergird and rely upon the community to take Ubuntu into devices that we as a company cannot or choose not to cover.
That said, two things have impressed me in my three weeks with the company. First, there are, as you point out, a lot of things going on with the very real potential for inefficiency and lack of focus.
But two, the company is remarkably consistent in what it does choose to go after. In particular, we are relentlessly focused on improving the Linux user experience. Canonical, in conjunction with the Ubuntu community, builds the industry's best Linux distribution, one that even a (former) Mac user like myself can easily digest.
We intend to take this emphasis and expertise in user design into a wide array of devices, but importantly will continue to focus on those that require a general purpose operating system. The good news is that even despite the increasing diversity of devices, the world is actually converging on fewer platforms, not more.
For areas that require expertise or focus beyond ours, we encourage our community to take Ubuntu into such opportunities, and they have. You might be surprised to learn just how many of the devices out there are powered by Ubuntu, often without Canonical involved. I see this as very healthy. It's the only way to compete with much bigger competitors like Microsoft: beat them with a bigger community like Ubuntu.
This isn't to say that we couldn't focus more. But that was already underway through Mark's and Jane's guidance. My job is to accentuate it and ensure that we stay on track.
Gaming and drivers
by HungryHobo (1314109) "I like Linux, I like programming on a Linux machine, I like learning on a Linux machine but I can't really game on a Linux machine and that's a big thing in the home PC market. What are the plans to induce game makers to port their games to Linux? What moves are being made to try to encourage graphics chip companies to create good drivers for Linux?"
Matt: You're asking the wrong man! My favorite game is Rogue, originally developed for Unix and still going strong in the guise of Qt NetHack and other variants. I'm easy to please, I guess.
As for the general gaming market, yes, gaming is a weakness on Linux, but addressing that is not a priority for Canonical. Games developers will make their decisions based on their market dynamics and those dynamics are pointing more and more towards dedicated consoles rather than the general PC market.
We work very hard with the Linux Foundation and others in the Linux community to encourage component manufacturers to either open source their drivers or make them available for Linux and with considerable success. This is not to enable gaming per so but to make Linux a peer experience on all machines.
Proprietary products
Enderandrew (866215) "You often praise proprietary, closed-source products on your blog (especially products from Apple and IBM). What is your stance on mixing proprietary and open products?"
Matt: Ubuntu is about choice. While we believe that an operating system is best developed with the source code openly available, that does not mean that the applications running on it need to be restricted to only those using the same development method. Our own users tell us, in large numbers, that they would like to see apps become available from the likes of Adobe and the games developers. On server the case is even more apparent where there are excellent proprietary applications that we would love to make available to Ubuntu users and we work to do that.
We can't boil the ocean. We want people to adopt Ubuntu Linux, and part of that requires us to support the applications that the mass market requires. Our focus is to continue to provide the industry's best Linux experience, and to make that Linux experience superior to any other platform. This process is well underway, and will encourage more and more application developers to port their software to Ubuntu.
Along the way, we hope that others will follow our lead and open source their software, but we intend to lead by example, not force-feed the industry. Google, for example, is arguably putting more pressure on Microsoft's closed-source approach than any amount of lobbying ever has or will. You can argue that Google is only doing this out of self-interest, to which I reply, "Exactly." Once the industry recognizes its self-interest in open sourcing software, we'll have even more from which to choose.
I love great software, whatever its license. But I joined Canonical because I believe the open-source development model can create better software than closed alternatives, and I'm determined to prove that.
Enterprise versus desktop emphasis
by eldavojohn (898314) "You used to write a lot about desktop Linux distributions but now that you're COO of Canonical, the revenue comes most from enterprise support. Do you plan on trying to change that or maintain any value in pleasing the at home Ubuntu user? Your blog post talks about your kids achieving basic tasks with Ubuntu, will you still keep them in mind despite the fact your new employer doesn't see a dime from them? Any plans to make it more user friendly or make it more mainstream and less server room?"
Matt: Actually, the majority of Canonical's revenue does not derive from providing support to enterprises, though I of course hope and expect us to continue to grow that area of our business. Our revenue will be a mix of making Ubuntu available to everyone on a wide range of hardware, from selling services direct to users (e.g., Ubuntu One), enabling hardware manufacturers to deliver a solid, supported Ubuntu experience on a wide range of devices, and from selling support and other services to enterprise IT.
Our market opportunity derives from Ubuntu's global user community, but it's a matter of making money with or around that community, not from it. All sorts of business opportunities are possible once a platform becomes ubiquitous, which business opportunities don't depend on charging users for the right to use that platform. That's a 20th-century model that we eschew.
So, yes, you'll see Canonical putting a great deal of effort into making the Linux experience even more user friendly: the more users, the better our revenue opportunities from ancillary services. It's in our interest to have millions upon millions of people happily running Ubuntu, and our unwavering focus is on improving the usability and design of Ubuntu to ensure that they do just that.
Ubuntu and KDE and GNOME
by Enderandrew (866215) "I loathe Gnome personally but don't begrudge people the freedom of choice. However, with Ubuntu becoming almost synonymous with Linux, do they have a responsibility to try and put out a quality KDE desktop along with a quality Gnome desktop?"
Matt: I'm new to the Ubuntu party, but I believe we already do this with Kubuntu. No?
Ubuntu and KDE and GNOME (cont.)
by Anonymous Coward "More importantly, we see GNOME falling further and further behind KDE. We need to know exactly when Matt will be pushing for GNOME to be deprecated in favor of KDE (or even XFCE). He really doesn't have a choice; GNOME needs to go, and it needs to go very soon. We're seeing the GNOME community fragmenting, and quite badly. Some people still advocate using C, others are saying that Mono is the way to go. And yet others are pushing for Vala. Frankly, the internal strife will tear the GNOME project apart, much like happened to XFree86. I, for one, sure hope that Ubuntu has moved away from GNOME far before then."
by Enderandrew (866215) "I think Ubuntu is actively hurting the KDE community by giving it a bad name. When Canonical works on new features for each Ubuntu release, they work independently of the Kubuntu team. Kubuntu is constantly trying to play catch-up on base issues. Even worse, they [Ubuntu] put out unstable, buggy, and sometimes flat-out broken KDE packages. Almost every I've talked to that has had really bad experiences with KDE complain about bugs and constant crashes they had when testing KDE packages from Ubuntu. Read KDE forums, mailing lists, etc. You'll see some serious hate and vitriol from users who blame KDE devs... They don't realize it is their distro that is causing their problems. I've seen several KDE devs walk away and stop contributing because of all the hate they're getting. If Ubuntu wasn't putting out broken packages, it would remove a lot of this backlash. That is not to say that 100% of KDE backlash is Ubuntu-created. ... But Ubuntu certainly hasn't done KDE any favors the past two years with the packages they've put out."
Matt: I remember my first taste of the KDE/Gnome divide when I was involved in the Linux Business Office at Novell. It was fractious then and, judging from your "question," it remains so. I don't want to add to this rancorous debate, but do hope you'll continue to talk actively and openly with Canonical and the Ubuntu community to ensure your views are heard and the Ubuntu distribution remains one that you will enjoy using.
Quality control
by davidm2005 (1453017) "I have been using Ubuntu as a software developer for the past several years. I have been extremely disappointed with the most recent release of Ubuntu, 9.10, as it has been extremely buggy and seems like a step backwards. The conclusion of this review also expresses a lot of my thoughts about Ubuntu 9.10. I had so many problems in using 9.10, that did NOT exist in 9.04, that I switched one of the two computers I use at work to Windows 7, for stability (yes, these are crazy days). Do you have any plans to increase quality control in Ubuntu, even if it comes at the cost of delaying the every-six-month release schedule?"
Matt: We are not complacent about bugs or quality. Far from it. In fact, I've been surprised by the level of attention it gets within the company.
You can criticize Canonical and Ubuntu for many things, but the work of the engineers and community in making an incredible operating system for servers and desktops on a huge array of hardware available for free to all is not one of them.
Every release of Ubuntu gets more users and is used on a wider choice of hardware. This creates complexity. Making an operating system entirely independent of the hardware that it is run on is hard and it's harder again when you are trying to push the performance of that product with each release.
As for Ubuntu 9.10, I've heard people call it a buggy release but that has not been my personal experience, and it's an accusation that the data do not support. Yes, we're constantly trying to improve, as Canonical CTO Matt Zimmerman calls out. But I look at this as a very good problem to have.
Why?
Because it's a symptom of a very positive thing: growth. There are more users using Ubuntu on more hardware than any previous release. Millions upon millions of users. Importantly, with our hardware partners we are providing certified, pre-installed, and supported Ubuntu on an ever-widening array of hardware. Dell's XPS 13 is just one awesome example.
For those who prefer to go off the beaten track and install Ubuntu on alternative hardware, as I did recently with a ThinkPad X200s, there may be some manual labor involved, just as there would be if you were running Windows or Mac OS X on unsupported hardware. In my personal experience, however, everything "just works." I've yet to have a single problem. Coming from a former Mac user (motto: two buttons are too hard - just give me one button on my mouse! :-), that's high praise.
Quality control (cont.)
by bcrowell (177657) I've been using ubuntu since edgy eft, and I'm really dismayed by the quality of jaunty and (especially) karmic. The biggest issue is that sound, which worked for me in edgy through intrepid, started working poorly in jaunty, and is now essentially completely broken for me in karmic. I've spent a lot of time surfing ubuntuforms.org, collecting information, trying to write useful and well documented bug reports, etc. But the upshot is that there have been major, major regressions in sound for me.
Matt: I'm sorry to hear that (no pun intended). But see my response above.
Is there a time to fork?
by nine-times (778537) "I've been thinking about the relative lack of success of Linux on the desktop lately. By 'relative lack of success' I don't mean to bash the quality of Linux, but only that it doesn't seem to be very widely used in spite of being pretty good for a lot of purposes. So first, to what do you attribute the relative lack of success, and what plans do you have, if any, to do something about it. It seems to me that a fair amount of the problem isn't the OS itself, but the associate applications. For example, lots of people have complained about GIMP for reasons ranging from lack of specific functionality to an unconventional UI, and even to the awkward connotations of the name 'GIMP.' Even having personally gotten some graphic designers to try the GIMP, I have yet to know any professional designers who find it adequate. I'd like to use Linux, but don't find I can come close replicating an equivalent workflow to what I have available using tools like Microsoft Office, Adobe Creative Suite, and Sound Forge. (those are the applications I'm personally stuck with, though I'm sure other people have other applications on their personal lists.)"
Matt: As to the relative lack of market share, it comes down to inertia. I didn't give the Linux desktop much attention until I joined Canonical. I had used it off and on over the years, but there was never a compelling reason to change.
Now that I've switched, I'm surprised by how much my ignorance of desktop Linux was coloring my opinion of it. I've been using it as my dedicated OS for three weeks, and have had only one (minor) reason to revisit my old Mac machine. I simply haven't missed it, and I thought that I would struggle.
Until someone has a compelling reason to shift, however, they're unlikely to discover this. For those picking up new machines, for example, a low-cost netbook, they won't have to overcome this inertia. Email, Internet, IM, etc. all work just as well on Linux as they do on Windows or the Mac. These are the applications we spend 99.9999% of our days in (most of us, anyway). As a result, I think we'll start to see barriers come down.
The irony in this is that these application incompatibility concerns are the exact same ones I had when I started using a Mac in 2002. Years later, application support on the Mac is much better, though still not at the same level as Windows. And yet 99 percent of the time it doesn't matter, just as it doesn't on Linux. As more applications move to the Web and as application developers improve their support for Linux (a trend I've noticed happening), it will matter even less.
In the interim, if you are happy to pay for and need these specific Windows-only applications then Windows is probably the right OS for you. Microsoft Office, however, is not a compelling reason to keep paying the Windows tax for many people. It's one of those applications that we think we use more than we actually do, and which OpenOffice (or Google Docs, if you wish) more than adequately handles.
We would love Creative Suite to be available for Linux but the open source or web-based alternatives are satisfactory for many users.
Mobile platform plans
by abhikhurana (325468) "What are Canonical's plans for mobile platforms? With Maemo, another Debian based distro, now available for smartphones, would Canonical also get involved with either that or maybe develop a completely new Distro? With the desktop Linux market being extremely small and server markets being dominated by Red Hat and Novell, mobiles probably are the sweet spot for Canonical, with its strong focus on usability. Additionally, the lack of standardization means that users are more willing to experiment with interfaces. So what is the relative priority of Mobile, Netbook, Desktop and Server platform in Canonical's roadmap?"
Matt: Mobile is a top priority for Canonical, especially as it looks less and less like the traditional embedded market and more and more like a general-purpose OS market. That's our sweet spot, and given our concern for and expertise in user interface design, we will be leaders in this market.
We will do a lot of work on ARM and Intel platforms this year that will see Ubuntu popping up all over the the computing landscape. Ubuntu is a platform: it is not a desktop product or a server product or a mobile product. So where there is a requirement for an OS you will find Ubuntu. You'd be surprised by the kinds of devices you already own that run Ubuntu today.
Re:Ubuntu needs two things added. (Score:5, Informative)
I don't think you've used Ubuntu in a long time. The hardware manager pops up on first boot and gives you the option to install proprietary drivers for devices it's found on your system (like Nvidia/AMD cards). Also, the first time you try to use a media player you get the option to install proprietary codecs. This has worked for at least the last couple of years.
He'd say mass (Score:5, Informative)
From this interview: I've never considered myself at odds with the goals of freedom-first software advocates
And just last September he wrote that "Free software is dead. Long live open source [cnet.com]", where he writes:
Free software makes for great headlines ("Miguel de Icaza is basically a traitor to the Free Software community"), but it is far too demanding, and of largely the wrong things, to capture mainstream interest. ... The path forward is open source, not free software. Sometimes that openness will mean embracing Microsoft in order to meet a customer's needs. ...Free software has lost.
Dude, that's you being at odds with the goals of the freedom-first advocates. You can embrace Microsoft all you like. Meanwhile, I'm busy working to minimise the patent and FUD harm they're doing to us. Canonical [swpat.org] have been helpful so far in campaigns against software patents. I hope there won't be a new "don't offend Microsoft" vibe that changes this.
Re:Fix Sound! (Score:5, Informative)
Larger/Better Graph [adobe.com]
Original Article it came from [adobe.com]
And it looks like hardware decoded video on Linux is going the same way [adobe.com]
KDE (Score:5, Informative)
I was disappointed with his response to the questions regarding Gnome & KDE. What I read in his response was 'We have Kubuntu. Please keep using it!'.
Ubuntu and KDE and GNOME by Enderandrew (866215) "I loathe Gnome personally but don't begrudge people the freedom of choice. However, with Ubuntu becoming almost synonymous with Linux, do they have a responsibility to try and put out a quality KDE desktop along with a quality Gnome desktop?" Matt: I'm new to the Ubuntu party, but I believe we already do this with Kubuntu. No?
Have you any idea what's going on in Kubuntu with Operation Timelord [kubuntu.org]? That's as close as you can get to saying 'We're tired of Ubuntu is fucking us, so we're blowing this popstand and doing it right.'
Re:Fix Sound! (Score:3, Informative)
OSS is pretty much gone already, so that need not confuse anyone.
Nope, OSS is back from the grave in the form of OSSv4. I believe it's currently being used in FreeBSD, and claims to clean up a lot of the problems with ALSA.
Re:Fix Sound! (Score:3, Informative)
Yet, gnash has working hardware acceleration. Do not listen to the whiner adobe has hired to make linux flash.
Re:This was good (Score:4, Informative)
The last graphics card I bought was two and a half years ago, and it sill plays every new game at a solid 60fps with everything turned on high. This is one advantage to having consoles being powerful. It doesn't make pc gaming a constant horse race as it was in the early 2000's.
Re:9.10 is buggy imo (Score:4, Informative)
I've had the same hardware since 8.04 (now using 9.10) and I've had almost no problems of any sort, with one exception. Since most of my work in web based I tend to have firefox open continually, and after a while (ranging from a few hours to a day or so) I will lose sound in flash (outside of flash it is fine) until I restart FF, which actually requires me to kill firefox since just closing it doesn't work at that point. Everything else seems to bee exceptionally stable for me (better than 9.04 actually). I wonder if these complaints come from a particular chipset? I used to have problems with sound pretty frequently years ago (with my older system) but I haven't had any (any that can't be traced back to flash anyway) sound based problems in a few years at least with Linux.
Re:9.10 is buggy imo (Score:3, Informative)
If you are using the same computer since 6.06, that might be an issue unto itself, in part. Off brand and "special deal" low end pre-built computers have all kinds of goofy cheap hardware with enough soft drivers to make one wonder how the computer even turns on. I can hardly imagine anyone wanting to hack together a soft driver for a crappy piece of hardware. It would be so much more work for that developer than to just buy better hardware. It is one thing to write a driver for something like audigy pro, or other real sound card, and another thing to write a driver for a wire, 2 magnets, and a plastic cone.
The problem that linux (and windows and freebsd) has with specific hardware is not the cheap nature of it, as the os can emulate it in software, the problem is that some low, and high end lie about what they can and cannot do so you have to know that if hardware id = X know that it lies about being able to perform Y.
Linux seems to do better about blacklisting devices do to it's larger number of bug testers. But one computer that windows was crashing on every five hours or so I installed linux on it and the boot screen showed lots of messages about hardware features being disabled because the hardware was known to perform out of spec, so linux was more stable than windows on that machine. I will leave others figure out who's fault it is that windows would not work reliably on the machine.
Re:9.10 is buggy imo (Score:3, Informative)