Martin Albisetti's blog

13Sep/1014

Why (I think) Ubuntu One exists

One of the questions that took a little while for me to fully understand was a very simple one: why does Ubuntu One exist?

Depending on who you ask, you may get a different answer, but here's my take on it.

Above all, to extend the power of Ubuntu as an environment. Ubuntu One already allows you to many things beyond the basic file sync we started off with, you can keep your contacts from your phone and desktop  (and between other Ubuntu devices) in sync and backed up, notes, bookmarks, all your important files are backed up and synced, you can share them privately or publicly, you can buy music that gets delivered right to your music player, and soon you will be able to stream any of your music to your phone. And this is just today. As the project matures, we are working hard to make it easy for more and more third-party projects to use our platform and out-pace us in ideas and code.
All of this allows Ubuntu to extend its reach into mobile devices and even other operating systems. It feels like integrating into the real world today, not only the world we want to build.

Openness is the next thing on my mind. I know about all the criticisms about the server software not being open, I understand them and I've been through this same process with Launchpad. Right now, Canonical doesn't see a way to fund a 30+ developer team of rockstars, a huge infrastructure and bandwidth usage that is mostly used at no cost and still offer up the code to any competitor who could set up a competing project within minutes. I am sure someday, just like with Launchpad, we will figure it out and I will see all my commits push me up thousands of positions on ohloh. Until then, I'll have to continue working on Wikkid or any of the other 20 projects I use and am interested in, to keep me at a decent ranking.
All that said, the Ubuntu One team releases tons of source code all the time. A lot of the libraries we build are open sourced as soon as we get some time to clean them up and split it out of our source tree. All our desktop clients are open source from the start. On top of that, we work on pieces like desktopcouch, enabling couchdb for the desktop. We even got the chance to work with a closed-source iphone application, iSub, to open source his code so we could base our new streaming client on. We get to pay developers of open source projects on the Android platform as well, to work on improving it so we can deliver a better and more secure experience. We also get a chance to learn to package applications and upload new versions of the libraries we use to the Ubuntu repositories. And hundreds of other small things we do that feel so natural we forget to advertise and be proud of. All of this on Canonical's dime.

Finally, a goal that is dear to my heart. Make Canonical profitable. I have been overwhelmed over and over again by the passion with which Mark personally, and the company as a whole, contributes to making open source be the standard way of developing software in the world. I can understand why it's easy to feel uncomfortable with a privately owned company pursuing a profit while sponsoring an open source project which thousands of people contribute to, but after having sat down in dozens of meetings where everybody there cared about making sure we continue to grow as a community and that open source continues to win over tens of thousands of computers each month, I only worry about Canonical *not* being sustainable and constantly growing.

All these reasons for working on Ubuntu One have been close to my heart for many years now, a long time before I took the final step of investing not only my free time, but my work time, leisure time, and not too seldom, my sleep time,  and started working for Canonical in a very strict sense of the term "full time".

I've spent time working in a few different teams, all of them are interesting, exceptionally skilled and open source is a core part of their lives. Ubuntu One is where I feel I can do the most impact today, and I'm beyond lucky to have given the opportunity to act on it.

Comments (14) Trackbacks (1)
  1. What’s your opinion on the inconsistency inherit in the name of Ubuntu One?

  2. I too care much more about the long term sustainability of Canonical than whether the server side of Ubuntu One is open or not. Viva Ubuntu One.

  3. The real question is, does Ubuntu One make sense as a revenue generator? Is being a profitable web service business in Canonical’s corporate DNA? And if so, can Canonical really compete with Google as a web services integrator?

    http://www.wired.com/epicenter/2010/09/apple-google-clash-christmas/all/1

    Even if you beat Google to the punch in something like online music services and music streaming, once Google enters into a service space can you compete for mindshare? More importantly can you compete for service dollars?

    -jef

  4. Hi Jef. We will see. I think it depends on how much we take advantage of the upper hand we have with desktop integration. It is, at the very minimum, an interesting challenge :)

  5. Hi Flimm.

    I don’t see an inconsistency. I understand the criticism around the server being proprietary and it being called “Ubuntu *”, but I don’t agree with it.

  6. beuno,

    Do you have an upper hand on _desktop_ integration? Hows your support for Windows _desktops_ at the moment compared to Google sync?

    -jef

  7. Well, we’re in the best position to integrate into the Ubuntu desktop. We are going to release a Windows client with Maverick, so we’re starting to extend our footprint there as well.

  8. Yes Canonical is the best position to integrate with Ubuntu desktop. No arguments there. Then again. Google’s best positioned to integrate with Android phones…and Android netbooks…and Android tablets.. and then ChromeOS whatevers after that. Which platform serves as a better starting point for service dollars?

    From the outside looking in, its difficult to see how deeper integration with a niche desktop OS is an “upper hand” when all the leading indicators say the real market growth is with Android for the forseeable future..especially in the new tablet form factor. I know you have a compelling reason to spin it as a strength. But from outside it looks like a really one-sided match up over the next year. I hope Canonical has some OEM tablet announcements up its sleeve to breathe some life into the competition. Or failing that an announcement to tie an advertising revenue business model into UbuntuOne as a reliable revenue generator to offset the costs of its free services.

    -jef

  9. I don’t see anything remotely inconsistent in the name “Ubuntu One”, and I don’t think I really understand the point of view from which one would.

  10. I’m glad it exists and that it helps to generate revenue for Canonical which is why I pay for my account each month. I could pay less and get some more space with google and use that account to store data on, but that doesn’t include the music integration or the contact sync, plus there is some data I would like to keep out of google’s hands.

    The integration is key as well, I’m in the process of swapping out my wife’s laptop, prior to wiping the disk I signed her up for an account, synced her data and then killed the laptop (don’t worry there is another backup as well). But I know when I get the new laptop I just need to enter her details and magically all her files will come back. Does that rock, yes, ever just so slightly :)

  11. I agree with you: I don’t mind it being open or not. Important is that it be another source of revenue to Canonical and it finances it’s development and it’s free-use for individuals. Besides, Ubuntu One is a great tool in adds a lot to the desktop version itself.

  12. You saved the best for last – U1 is around to give Canonical an income stream from consumers rather than corporations. And I’m all for it, but I haven’t found a compelling service yet: I don’t listen to music, I use gmail for my contacts, and I rarely share data between computers.

    I’d happily pony up $50/year for a preferences sync, or an automagic screen sharing service to access my various machines through firewalls. But $120 is too steep for a service I don’t use.

  13. Erigami,
    A preference sync for what exactly?

    I can’t imagine a preference sync that could cover enough applications that did not itself tie directly into DConf instead of being an application by application bolt-on technology like desktopcouch is currently implemented as.

    -jef

  14. Interesting post Martin, thanks for sharing. I’m glad you enjoy working on Ubuntu for Canonical and Ubuntu itself :)


Leave a comment