Wednesday, May 13, 2009

About UbuntuOne -- The Post That is Sure to Piss Someone Off

I've been waffling for a couple of days over whether to write this post or not. I finally decided to just speak my mind.

UbuntuOne is a new service from Canonical that's currently in Beta testing. It doesn't differ in any important way from similar services like DropBox or, except that it's integrated and designed specifically for Ubuntu (9.04 Jaunty is required). The client is still immature, but it will probably eventually include useful services like integration into Conduit Sync and desktop settings. Once that kind of integrated functionality appears, people will probably line up to pay the USD10 for 10GB of storage.

Unfortunately, Canonical is playing this on-line service the same way they play their others -- closed source. Just like Launchpad, the server software isn't open (though the client is). Why does it choose to do that?

It's not necessary to keep either closed, you know. Launchpad offers a great way for teams to work together, but its value isn't in the code as much as it's in the community behind it. could open its code, too without any worry of someone building a copy. Launchpad links teams and code. Nobody is going to replace it.

UbuntuOne will eventually be integrated into Ubuntu. Users will be able to share files with others on the service. It will be half backup and half social. No matter whether there's an open version of the server or not, the vast majority of users will be on UbuntuOne instead of WannabeOne because that's where everyone else is and getting set up is much more convenient to use the included client than downloading a different client setting it up with the other service.

In both cases, Canonical gains little by playing both sides of the fence, offering closed-source solutions as an open-source business, but it loses a lot of community support. Red Hat, on the other hand, is a pure open-source company, and has no problem:
  1. Offering amazing platforms like the Fedora Directory Server completely openly, and
  2. Making money off of the software service.
Very few FOSS advocates have a problem with dual-licensing projects like MySQL or QT (before the acquisition) selling licenses to closed-source projects, but that's not what we're talking about here. This is more like SUSE's old model of keeping YaST closed (it was GPLed in 2004). That, too, kept a lot of people off balance with regard to SUSE. The open source business isn't one that you can stick one foot into -- you have to jump in all the way.

Most Ubuntu users aren't aware of the FOSS philosophical arguments, but that doesn't change the business arguments. You can't promote open source while holding some of your cards close to your chest. How can I trust you when you say one thing and do another?

Ulitimately, it's Canonical's call and it's their code. They get to license it any way they want to. I still think they're shooting themselves in the foot.


  1. A minor comment I suppose, but I think you are conflating free and open source with this "FOSS" label even though your post illustrates a big difference between free and open source.

    "Most Ubuntu users aren't aware of the FOSS philosophical arguments" -- open source is not much of a philosophy. Its arguments are more practical than philosophical. Open source, as the Wikipedia article you linked to says, is about the strength of the development model. It's not about the freedom of the users.

    Similarly, "Very few FOSS advocates have a problem with dual-licensing projects." FOSS has no meaning here. Open source advocates may well have no problem with dual licensing. Free software advocates would have a problem with it though. The proprietary license restricts the freedom of the users.

  2. @massyset
    Well I think you are wrong. FOSS stand for FREE OPEN SOURCE SOFTWARE and have (for most of the users) more philosophical meaning, although practical meaning is very important. The Liberty to adapt the code to work and act the way you want it, the liberty to take it, combine it with other peace of FOSS software, or modify it - all that perfectly legally for me is like religion. And in 99% of the case FREE (meaning gratis), in this world where people are charged for lousy MS OFFICE update (after which your OFFICE package will probably stop work, but you will pay anyway...) is truly incredible...
    New users are mainly attracted with free (gratis) software, while they understand the philosophy later...

    Never liked Ubuntu anyway... By publicity more easy and simple but pwerfull Linux. That's just not true... Everyone who ever tries Puppy, Mepis or PCLinuxOS knows that...


  3. "It's not necessary to keep either closed, you know. Launchpad offers a great way for teams to work together, but its value isn't in the code as much as it's in the community behind it. Facebook could open its code, too without any worry of someone building a copy. Launchpad links teams and code. Nobody is going to replace it."

    While I'm not entirely happy about the decisions Canonical has made, the above actually fits right in to their statement's about Launchpad. They say that the were waiting until there was enough community buy in before opening the code. They feel that they have reached that point and claim that most of the code will be opened up soon (July 2009 was the claim, we'll see soon I guess).

    Also, I find some of the complaints about UbuntuOne abit strange. I've read so many posts on Planet Ubuntu and Planet GNOME about how great Dropbox is, and if there was ever a mention about its closed code it was always more of a "I wish it was open, but oh well."

    Not only is Dropbox's server closed sourced, but it runs closed source software on your computer (only the Nautilus plugin is open, not the actual client). At least with UbuntuOne, the entire client is open sourced.

  4. First of all, thanks for the comments. I appreciate a civil debate on this subject.

    It's true that very few people even mention Dropbox's closed status: that's the reason that I said in the article "Most Ubuntu users aren't aware of the FOSS philosophical arguments, but that doesn't change the business arguments." (WRT that -- OSS is a business philosophy.) Dropbox isn't an open source company -- Canonical is.

    Another issue is the opening of Launchpad. Canonical has been talking about "sometime" for several years now, and their opening of code in July won't include portions which they view as "business critical." These are the most important aspects of the platform, of course.

    On to Massysett's point about the difference between Free Software and Open Source. I'm a Free Software moderate so I understand the differences between the two camps (and was around when OSS was coined, but there are many Free Software advocates who don't dispute a company's rights to release code under whatever license they prefer: we just don't purchase it. I use Intel graphics, for instance, instead of NVidia, even though the performance isn't on nearly the same level.

    Again, thanks for the debate.

  5. I know I am nitpicking but I am sure you meant to say that you can stick one and not stick on. You also meant you can't promote instead of You can't promoted. I agree with how can you believe some one that says one thing but does some thing else. Kind of reminds me of the President that we have now.

  6. United against,

    Thanks for pointing out the typos. Embarrassingly, I didn't spend any time proofreading the post because it was a fifteen-minute job during a break at work. I always appreciate people pointing out my errors (unlike LifeHacker, which now will ban you for pointing them out. WTF?).

    Oh, and my previous comment doesn't parse due to failure to close parentheses. I'm just on a roll today, aren't I?

  7. Hi! I can tell you that your blog post doesn't piss me off at all, I'm enjoying reading all the different discussion and analysis of the different decisions that we've made so far with Ubuntu One. Mostly I'm really happy that the service is out in the open now and can be discussed, we worked on it for a while and didn't want to tell anyone about it before they could actually try it out.

    I'm interested in understanding what you mean about losing a lot of community support by keeping the server software closed. Currently Dropbox is has more features than Ubuntu One, but are you suggesting that once we get feature complete many people will choose not to subscribe because the server source code is not available? Or are you saying that Canonical the company will lose goodwill and contributions from the Ubuntu community because we are trying to earn some money? As you can imagine, I'm quite interested in finding the right balance for all that we do, and listening to input from various people is part of that.

    At any rate, thanks for the thoughtful article, and please do keep speaking your mind!

  8. Elliot,

    Thanks for visiting my insignificant blog. I appreciate a developer chiming in on this matter (I assume from your post that you are one -- I haven't actually looked you up WRT UbuntuOne).

    I didn't actually expect to piss off Canonical folks. You guys have thick skin, and that's great. Really, the problem is that I took a Free Software position in this debate, and I'll upset:
    a) People who oppose Free software;
    b) Open Source guys who think I went too far, or;
    c) Free software guys who think I didn't go far enough.
    Oh, yeah, and
    d) Quite possibly all of the above.

    Anyway, moving on now in order to clear up my exceptionally murky post. By community support ("losing community support"), I'm talking about the Free and Open Source community, which is exceptionally fickle on these matters, which makes up a smaller proportion of Ubuntu's users than other distros, but which constitute a great part of your developer and evangelical base. As I said in the post, I see no real downside to opening up the server code since UbuntuOne will likely (and hopefully) be integrated into 9.10. Users will choose the default as long as the default is reasonably competitive. Canonical won't lose much, if any, money by opening the code. Some people will install the server at home and use DynDNS and a custom configured client to do the same job, but how many can we really expect to do that? They won't get the "cloud" benefit. They could also use S3, the way you do, but we're talking about an even smaller proportion of users. Probably, 95+% will just sign up for the free service you have integrated and upgrade to the pay service when they need more space (is freemium even likely to generate real money in this case?).

    Speaking more broadly, right now, Launchpad is kind of a thorn in Canonical's side as an open-source company. It is not dissimilar to BitKeeper's role in the Linux kernel just a few years ago. The way that one ended "proved" a lot of the criticism right in the minds of the critics. Canonical has been talking about opening Launchpad soon for long enough that it's kind of become a DNF-style joke. Few believe it will happen, and the partial opening in July will just confirm the nay-sayers' opinions.

    The UbuntuOne server issue will just add to the insecurity of the people who are already unsure about Canonical. I think that's a shame. It appears that you don't know whether you are an open-source company or not. Sun drove itself into the ground fighting that battle with itself, by the way. You can't straddle the fence.

    I raised the example of Red Hat because they are by far the most successful open-source company, and they got that success by opening everything without fear. They aren't worried about CentOS or VARs, either. I do not fault RH for making money -- in fact, I applaud it. Canonical needs to make money soon, too. Shuttleworth's millions won't last forever, and 200 developers means that there's a good-sized payroll. The need for money doesn't mean that you also need to fall back on a business model of keeping everything a trade secret.

    As I said, Canonical gains little by keeping these two projects closed source -- open would work just as well due to the network effects -- and creates doubt in many of the people who would be its strongest supporters. Canonical needs to make money. Creating closed-source software isn't the way to do it. Waffling on your model isn't a good business strategy.

    Does the Mercedes salesman drive a BMW?

    Was that too harsh or over the line? I hope that it wasn't. I'm just expressing my opinion on the matter.

  9. I agree with you completely. But, I must ask: why aren't you using a CMS that is open source?

  10. Touche! Ultimately, it's because I'm not in the open-source business. I also use Google Apps quite a bit, and it's not open, either. I get the domain and hosting for $10 a year, including Google Apps for the domain, which I use for wikis, file storage, and such. I used to host my own blog and other sites, but Korea is a violent part of the Internet and I spent a number of hours a week going over Snort logs and the like. I don't want to do that anymore. Maybe I'll get off my lazy ass one day and find a decent host that lets me use FOSS platforms without giving myself a headache, yeah? We can only hope I get that kind of motivation. ;)

    Don't get me wrong about my post: I'm not demanding that Canonical open anything up -- I just think they need to be consistent with who they are.

    Oh, and, while I'm on the subject of Google Apps ... I'm going to be looking at two open-source alternatives this weekend.

  11. It's quite simple: if you write the code, you get to pick the license. Excuse me for stating the obvious but everything else is just plain BS.

    For every Commercial Open Source Company (COSC) with success (like RedHat, the one you quote) there are probably a bunch of others that didn't succeed. If you think you can decide for a COSC like Canonical what will work, what will not, you're just fooling yourself since I'm sure Canonical doesn't even know exactly.

    What? Is it against the law to make money now? To have a business plan? Did we all become communists over night and nobody told me?

    It's a constant struggle, a constant search for balanced commercial models that work. For better or for worse, you're either part of the Ubuntu community or you're not. Canonical is obviously a huge player in that community, without them, there wouldn't be a community. If you're in that community, why not show at least a little support for them and cut them some slack once in a while. (I mean "you" in the general sense, not specifically the author of the blog post)

    Take care,

  12. Are Canonical actually building something from scratch here with UbuntuOne?
    Reading between the lines I thought it sounded more like they intend re-selling an existing service under the banner of UbuntuOne - possibly partnering with DropBox who have an open client for nautilus but closed server software.
    Sounds a little like what Canonical are offering. Maybe I am wrong.

  13. Matt Casters said...
    "It's quite simple: if you write the code, you get to pick the license. Excuse me for stating the obvious but everything else is just plain BS"

    Did you read to the end of my post where I said "Ultimately, it's Canonical's call and it's their code. They get to license it any way they want to. I still think they're shooting themselves in the foot" or did you just decide to hit "Reply" in the middle somewhere? I support Canonical's right to make money however it sees fit (legally). I think this is not going to do that, though, and will probably hurt the enterprise market, which is where they actually WILL make money. RH makes its money there. So does MS. That's just my opinion, though, based on other businesses I see running the various open-source business models. You're not required to agree.

    See? I told you I'd piss someone off!

    Oh, and I don't believe that UbuntuOne is a rebranded Dropbox. It appears to be original.

  14. Come on buddy, just because you say "sorry", that doesn't give you an excuse to pass an insult right after or before.

    It's the same with this: of-course it's Canonical's business to do whatever they like. I wasn't under any delusion that you had anything to say in the matter. But you still had to take them on over something as silly as this.

    Next time just stop right after "Ultimately, it's Canonical's call and it's their code".


    P.S. As an open source advocate and project lead, I don't get pissed off easily these days. I'm sorry if I gave that impression. ;-) I love a good flame war just like anyone else, but this ain't it.

  15. I think the biggest Error Canonical is making is applying the Ubuntu Label to this Paid Service. Ubuntu the name has come to be the top representative of the Linux community and I find this "Service" to be great but it should be presented as a free component with the option to maybe upgrade to a paid option. I.E. get 5 gigs storage free with this service. then if you want more Say 10 Gig then you can get that for a modest fee. Otherwise At the very least the name should be changed and it should be an add on package to Ubuntu not something that "Comes with it" especially if you are going to charge people. This runs a serious risk of hurting Canonical as a whole by trying to sneak paid services into a free OS like that.

    Mainly I think the service should be named Canonical One thus getting the Company behind the software's name out and providing a little insight that this type of thing is not par for the course because they are basically trying to piggyback the success of Ubuntu's name to try and make money now, which is in their right to do they do own it, but then again this comes down to the good of the community that has stood behind them and help give them this name in the first place.

    I have to admit I was rather upset when setting up my latest server and I was present with an option for launchpad that would be free for a while then I have to pay for it. That is the same kind of petty ripoff scam process used by magazines and other services. I preferre my Free OS without those things.

    Canonical is making a big mistake in the way they are approaching this and it WILL shoot them in the foot. Just like several others that started under the "free" banner then suddenly changed up to "Oh for the good stuff you gotta pay us now" these type of activities can put a serious hurt on how people view your distro and in the end can hurt the supporting community also.

    The most basic advice.

    Change the name and offer a lower storage area free version of this service. otherwise the blow back from the community can be very harsh and end up burying and tarnishing the the good name of Ubuntu as a whole.

  16. Kindreds,

    Canonical offers 2GB free or 10GB for $10 a month. As far as I know, there is no time limit. It is not a "free trial" service.

  17. My Apologies, I was mixed up with another service at the time. I was just on my way to correct myself.

    Yes they do have an outright free version of this which is perfectly fine. Granted I still think the Name change would do more to improve the name of Canonical over all. Most "common" users do not realize the correlation of Canonical to Ubuntu. This would be the perfect way to improve their image by using their name instead of the Ubuntu label.

  18. First, to those that bitch about Launchpad and similar being closed source. If you're going to complain about and boycott that, I expect you to complain about and boycott Google as well. Otherwise you're a hypocrite.

    Also, Red Hat is just as horrible with their vendor lock-in tool they like to call the RHN.

  19. @First, to those that bitch about Launchpad and similar being closed source. If you're going to complain about and boycott that, I expect you to complain about and boycott Google as well. Otherwise you're a hypocrite.

    I was not aware that Google was in any way posing themselves as an "open source company". You? I think you're comparing apples and pitchforks here.

  20. "Also, Red Hat is just as horrible with their vendor lock-in tool they like to call the RHN."

    RHN source code is publicly available under the name of spacewalk

    Unlike Canonical. Red Hat does not depend on proprietary software or services in its business. It is a full fledged Free software company and very successful at that. It actually makes a profit unlike Canonical as well.

  21. This comment has been removed by the author.

  22. This comment has been removed by the author.

  23. This comment has been removed by the author.

  24. Refilwe,

    I enjoyed what you had to say on the subject. It's too bad that you chose to delete it.

  25. [My apologies for grossly messing up the comments section...let me write a much shorter version of what I was going to put.]

    If there is revenue to be had in optional, integrated, online storage why should a company that supports FOSS be forced to abandon that revenue by the same community that has no problem letting a non-FOSS supporting company step in and take that revenue stream?

    I understand the argument of the piece - A FOSS company should only use FOSS - but surely that's a collapsing argument. What if Shuttleworth comes out and says, "Canonical is not a FOSS company, rather we are a Microsoft competitor that uses FOSS where we feel it is appropriate like Novell or IBM", what happens then?

  26. [Although I don't think he or she meant as much by their comment, but here's the long version for the record. I promise not to meddle with it again since frank discussion should trump trying to get the words just right. I feel it's missing some things, but when you shoot from the hip sometimes you miss...]

    I'm not sure how well the argument 'OSS companies should do everything OSS' holds up here.

    I think Red Hat is a poor example for two reasons, one generally and one specific to Canonical's situation.

    In general Red Hat isn't a great example because it seems to be the exception not the rule. The other major corporate backers of FOSS are mixed source and/or sell hardware using FOSS as a low-cost non-MS method of building a software stack on top. Smaller businesses that dual-licence their work tend not to be pushing a complex stack either in the form of an OS or a (near-)complete hardware + software solution where they are the chief architects (or at least curators) of each layer. Further, those smaller outfits' biggest commercial success is often getting bought out by a larger mixed-source company or a closed source company attempting to become mixed-source.

    Specifically to the case of Canonical, they decided not to have a pay-for enterprise edition of Ubuntu long ago. Selling what is effectively an enterprise version is where Red Hat makes its money. It's true one can get CentOS, but Red Hat takes advantage of businesses being risk-adverse when it comes to IT and says "You want day one availability of official updates? Pay. You want official support? Pay. You want one and not the other? Too bad. Pay for both and use one."

    Like the author mentioned, the value of many online services is in the community those services create rather than the code. However, that doesn't really apply to a new service since it doesn't have a large community yet. Until that large community forms, the value of UbuntuOne actually does lie primarily in the code. I would argue that the nature of the service means the value of it may never rest primarily in its community. For an average user UbuntuOne is likely to be first and foremost a place for personal use - a safe place for important files; a place to hold files one needs to move; an internet based USB drive replacement; a way to sync things between computers they own. The potential social uses are secondary, even if Canonical pushes it as a social service - they are not going to make Windows or Mac versions of the client, which limits its social potential even with a lesser web client available. Launching the service as OSS on the server side makes it too easy for competitors to provide the services of UbuntuOne through the integrated client while undercutting Canonical's business model. Such a situation is possible with Day 1 open source code even if it's true that the value is in the community - a competitor could marshall a community toward its free 10GB+ service before Canonical has a chance to build much of anything.

    Unless people are arguing for Canonical to backtrack on its promise for Ubuntu to be free of charge and have free security updates...I'm not sure how they expect Canonical to generate revenue. The companies that use Red Hat or Novell are invested in those distros and change has costs. Combine that with Ubuntu's advantage in consumer-level popularity, it makes sense for Canonical to look for ways to monetise on the consumer-level. For the majority of the community (and potential community going forward) that fall into the 'Free Beer is Awesome!' + 'FOSS 'Freedom' is Meaningless to me' + 'I'm a FOSS Moderate' camps...little has changed: Ubuntu the OS is still free in the ways that matter most to them, now they have 2GB of free online storage that they didn't have before and if they need more then they can pay a small fee. Canonical's hope seems to be that growing storage requirements + a growing userbase + the convenience of integration = a decent revenue stream that doesn't conflict with their prior 'never charge for the OS' commitment.

    Arguing that Canonical needs to make money some other - unspecified - way makes the argument against UbuntuOne less persuasive. I credit the author of this blog post for pointing out Red Hat as an alternative model but RH seems to be an exception. Pointing out what is possible does not mean pointing out what is feasible in a given context. It's like telling a young man to 'swim like a champion' with no other guidance than "well Phelps can do it". The "it's an open source company so different standards apply" argument is a collapsing one, will those that make it drop the issue if Shuttleworth comes out tomorrow and says "Canonical is not an open-source company but a Microsoft competitor that leverages open source like IBM"? Some will, but many others will just shift positions. What I don't understand is why if there is revenue to be had in optional, integrated, online storage why should a company that supports FOSS be forced to abondon that revenue by the same community that has no problem letting a non-FOSS supporting company step in and take that revenue stream? We're not talking about criminal activity, we're talking about closed-source software on a server providing a service you don't have to use...which makes it just like many other optional internet-based services many in the FOSS community already use.

  27. On the difference between "Free Software" and "Open Source Software."

    The term "open source" was coined by Bruce Perens and Eric S. Raymond NOT as a way to set apart two differend software source models, but purely to eliminate the confusion between free as in freedon and free as in beer.

    It was the frequently irritating and opinionated Richard Stallman who decided he never liked the Open Source Initiative for doing what the Free Software Foundation did, but better, who decided to right that article tryiong to pass off free software as something different and better then open source software.

    The reality: They're the same, even if RMS doesn't want to think so. Just like RMS likes to falsely think Linux is GNU, it's all about who outdoes him. Linux outdid GNU and the OSI outdid the FSF for providing a clear definition of FOSS and trying its hardest NOT to be an outright political movement like the FSF.

    RMS doesn't understand one pretty easy to grasp idea about computers and software: Politics and software don't mix. Politics have a nice way of just eating a lot of time by posturing and trying to get people to like you and getting very little done.

    One thing RMS also overlooks is there really IS a divide between the actual GNU and Linux camps. GNU is all about its philosophies and driving forth an agenda. Linux was, from the very beginning, all about being practical. Sure, it's GPL software and strongly built around open source, but Linus Torvalds himself had said that he didn't set out do be some gladiator of freedom and that all e wanted with Linux was for it to be PRACTICAL.

    See what happened, though? GNU gets started in 1983. It produces a fanastic toolchain, but for it to be a system distribution still hasn't hapened ecause the Hurd is incomplete and no one is interested in it. Why? Because Linux is there. It was completed and now people don't have a reason to turn to developing GNU to completion. Linux became more practical and effective. If RMS thought more like a programmer and less like a politician, we'd be using GNU instead of Linux.

    That said, I understand that GNU (The toolchain.) is important to most Linux distributions. They're excellent tools. But I definitely draw and hold a line before saying Linux is actually GNU, because of the actual definitions of operating system and system distribution, which disagree with RMS' own definitions of those terms.

    Now, as for Canonical making a proprietary server for Ubuntu One, it sucks, but I think they want to make sure it's viable before AGPLing the whole system, like Launchpad is scheduled to do in July of this year.

    It makes perfect sense from a business perspective and, to be honest, a lot of great open source projects start out closed an unavailable before becoming open.

    I'm not an open source purist though. There are a few things I use in Linux that are proprietary, like my displa driver or Flash, simply because the FOSS alternative isn't even close to being a viable alternative.

    This is why I avoid distros like gNewSense and Gobuntu.


Other I' Been to Ubuntu Stories