My UbuntuLive Talk

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

The slides are up here, here, or here.

I can’t compete with Jeff for laughs, but the audience seemed at least mildly entertained. Adult Swim fans, this deck was for you.

Also, I fleshed out my notes on the flight over and am making the basic talk script available for the first time. Should help with context for the slides.


Those of you who’ve never seen Kung Fu are probably terribly confused at this point, maybe even wondering if you’re at the right conference: my apologies.

This is how Master Kan – a Shaolin master – explains to a young Caine when he’ll be ready to move on.

Master Kan holds out a pebble in his palm. If Caine is quick enough to take the pebble, it will be time for him to leave the temple…and begin wandering more or less aimlessly through the southwest.

This talk is about getting Ubuntu ready to take the pebble, not getting ready to wander through the southwest.

Now most analysts probably wouldn’t use a Kung Fu metaphor in the first place. But if they did, they’d very like tell you that taking the pebbel is all about execution and partnerships and marketing and channels and compliance and a host of other things that are indeed important important.

Obviously, I am not most analysts.


On the face of it, it seems ridiculous to question the maturity of Ubuntu. It is a legitimate phenomenon at this point. Mark can tell you how many CD’s have been shipped out, but last I heard it was well into the 7 figures. For those of you bad at math, like me, that’s a lot.

And then there’s distrowatch. Going back as far as it’s history allows, we find two years where Ubuntu failed to place – the excuse is that it didn’t yet exist – one year where it placed 13th, and then several years where it placed first.

The end all and be all? Hardly. But that’s still a lot of pageviews. By virtually any metric, non-commercial penetration of the distribution is exceptionally high. That’s all good.


But let’s not forget Ubuntu bug #1. Despite Ubuntu’s success, this bug is till very much alive.


We’re not a numbers company at RedMonk, a.) because of the aforementioned math difficulties and b.) because the numbers were problematic before wide adoption of open source. After it? Well, let’s just say I wouldn’t bet on most of them.

That said, there’s little debate that on both the server and desktop levels, Ubuntu and friends have yet to dent the object of bug #1 severely.

Couple that fact with a newly resurgent OS X – count the number of Macs you see at OSCON, then keep that number to yourself – and it’s clear that for all of its success to date, there’s still a lot of work ahead.

A lot of work yes, but it’s not futile work. There are weaknesses to exploit, as the Google’s of the world continue to demonstrate.

So what’s the plan?


Most important is to avoid the technology equivalent of Pickett’s charge. As Sun Tzu reminds us, the easiest target is the one that is undefended.

Grounding that concept, it means that attempting to compete on equal footing with Apple and Microsoft is a difficult proposition at best. The history of the technology business is littered with the corpses of firms and projects that tried to out-Windows Windows or out-Office Office. Microsoft has $50 billion dollars sitting in the bank that says that’s not a great plan.

Instead, listen to your friend Master Tzu – we’s a wise man. Consider first your strengths, and their weaknesses.


I remember the first time I used Gentoo’s apt-equivalent, Portage – and just an FYI, don’t listen to anything the following speaker, Mr. Waugh, says about Gentoo, it’s all jealousy. It was honestly like magic. As someone that had spent years on Windows, hunting down applications, cruising through page after page looking for download links, only to find an installer that might work and that I’d have to maintain, it was a revelation.

As Arthur C Clarke might put it, it was impossible for me to distinguish from magic. Tim Bray obviously had something of the same impression.

But the funny thing is that amongst Linux users, it’s more or less taken for granted. Just part of the background; useful, yes, but nothing to get worked up about.

I don’t agree. I think it’s very much worth getting worked up about.


The fact that I can install – and maintain – a package like Asterisk with a couple clicks, is not just worth getting worked up about, it’s differentiating.

It’s not differentiating enough to have changed by itself the numbers we saw above, true, but it’s differentiating.

And ask yourself this: what’s Microsoft’s alternative?


As impressive as apt is the Ubuntu community. As the de facto sysadmin for RedMonk, I should know. I’m by no mans qualified for that job, but with a little help here and there from folks in the community I’ve been able to get a production server up and running and maintain it over time. No small accomplishment.

The forums are everything CNET says they are and more. It’s rich, it’s active, and perhaps more importantly – it’s friendly. No one’s called me an idiot yet. Not that I know of, anyway.

Without even thinking about it, then, we’re able to cite two real differentiating factors for Ubuntu. Two strengths which have lukewarm at best equivalents amongst its most successful competitors.

Two potentially undefended areas, then – areas of opportunity. Which is good.

What’s much less good is that there is no real linkage whatsoever between then.


Consider the conversation that a prospective user might have with Ubuntu. We’ll choose Asterisk for the sake of argument, but we could just as easily substitute Apache, MySQL or dozens of other projects.

Maybe this customer is fed up with their PBX costs. Maybe they’ve never had one, but are growing and could use the functionality. Whatever the case, they read of Asterisk, or come across it searching through the Ubuntu archives. A couple of clicks later, and it’s installed via the magic of apt.

What now?

In the current model, the expectation is probably that said customer will seek out the forums, use search and post their question if it’s unanswered. Or maybe hit craigslist, Dice, eLance, Monster or one of the other contract and/or job sites available. Or maybe wind up at Digium for official support.

For a portion of Ubuntu’s audience, one of those answers might work. For a much larger portion of Ubuntu’s audience, in my opinion, it will not.

What if, instead, of saying “look, via the magic of apt we got the damn thing installed for you, what more do you want?” Ubuntu could answer as follows?


Wouldn’t it be nice to be able to hold the customers hand through that process?


What if a customer could query a repository of human resources as easily as it could the software library?

That, in my opinion, would begin changing the game. Dramatically. No longer would Ubuntu be competing strictly on software terms – filesystems, drivers, UIs, and so on – but on an operating system experience. At once you’d change the context for operating system discussions and the all important developer economics.

Now you’re competing with an operating system experience. One that competitors would find it difficult to match, I think.

Are there questions? Lots of them. But most of the conceptual moving pieces have been tried before. Successfully.

It’s just that no one’s yet put together those pieces.


Two things are becoming increasingly clear to me over time.

One, that Adam Bosworth is a very smart man.

Two, that he’s right, that the value of the software is being superceded in many contexts by the information and the community built around that software. It’s not that the software is irrelevant, mind you, it’s that it’s not enough.

apt is a magical piece of software. But to really change the tone of the conversation, it needs to be more tightly integrated with the community that surrounds it.


Not that the Ubuntu community is the only one worth focusing on. Consider the reality of today’s computer usage in geographies with at least some access to the network. Absent that access, the computer itself is less useful. Dramatically so, for many of us.

I’m with Havoc – I don’t use my PC for the most part when I’m disconnected, except to write talks like this while on the plane.

What does this mean for Ubuntu?


Opportunity, in my opinion. Who remembers when the first Microsoft Live services were rolled out? Even if you don’t, guess which browser they didn’t support?

Firefox, indeed. The explanation is not terribly complicated: it’s in Microsoft’s best interests, or so it thinks, to advantage its operating system over the competition. That operating system, after all, is one of two licenses it possesses to print money.

Apple finds itself in a similar position. Remember how long it took Apple to bring iTunes to Windows? Why? To protect its hardware revenue stream, of course.

My purpose here is not to debate the ethics, legality or even efficacy of those approaches, but rather to note the opportunity they open up.

Ubuntu, and the upstream window managers it principally relies on, have no such software or hardware revenue streams to protect. So there is no artificial obstacle, then, to going further in embracing the Facebooks and Gmails of the world – better known as the most popular applications in the world – than their competitors can.

Undefended territory, again.


Ubuntu needs, in other words, to impose its will on its competitors, and stop letting competitors impose its will on Ubuntu.


To take the pebble, then, Ubuntu needs to reframe the debate. To do that, it must turn the conversation from basic operating system shootouts to the operating experience. A conversation that, in my opinion, favors Ubuntu.




  1. great deck. havent read your notes yet but the deck was goodness. the pong thing slew me.

  2. First I enjoyed the talk. Second, the link to the odp version of your presentation is broken.


  3. PPT format slides? Are you crazy?!


  4. […] GNU/Linux, Web 2.0 — Mario Diaz Laguardia @ 1:59 am Acabo de terminar la lectura de las notas de la presentaci├│n de Stephen O’Grady en UbuntuLive. Stephen recalc├│ algunos puntos relevantes para Ubuntu […]

  5. I’m unable to be there but read your notes. Well done. Ubuntu must keep pushing forward and lead.

  6. […] tecosystems because technology is just another ecosystem Articles « My UbuntuLive Talk […]

  7. […] de terminar la lectura de las notas de la presentaci├│n de Stephen OÔÇÖGrady en UbuntuLive. Stephen recalc├│ algunos puntos relevantes para Ubuntu ÔÇôla […]

  8. Thank you for sharing even the notes!

  9. I agree that Apt is fantastic, that is why I use Ubuntu, Vector and all the others that base their package management on it’s clever operation.

    Linux has another opportunity battle that hardly anyone, except those bleeding edge game machines, bother about, and that is a public computing grid. Maybe we’d all need to install Inferno, but all those cycles when myself and all the others in my time zone are asleep, could give the other side of the world some super computing powers for their normal tasks.

    I bet such an implementation would give Linux a lot more kudos in the media and how would insecure Windows ever match that.

  10. […] main theme of the keynotes at GUADEC (RedHat’s Online Desktop, PyroDesktop, and seems to be on the agenda at UbuntuLive […]

  11. that’s an astute analysis, but leaves open the rather fundamental question of how those human resources are to be funded. I make a living out of support. If I won the lottery I could happily spend the rest of my life supporting Ubuntu for free: but in the meantime my bread and beer come from commercial software support.

  12. […] g├Âre oynamak, masa├╝st├╝ndeki bedava yaz─▒l─▒mlar─▒ ba┼čar─▒l─▒ k─▒lacak etkendir. Stephen bunu blo─čundaki t├╝m sunumlar─▒nda […]

  13. […] me running 5+ miles 3 or so times a week mid-July through late August. Just before OSCON, I gave a talk at UbuntuLive, staying over for OSCON as usual. And lest the baseball trading deadline go by without reaction, I […]

  14. […] July 23, 2007 My favourite Ubuntu Live slide so far was Stephen O’Grady’s Sun Tzu quote: ÔÇťYou can be sure of succeeding in your […]

Leave a Reply

Your email address will not be published. Required fields are marked *