Archive for November, 2006

#8: Govoritye po Russki?

Wednesday, November 29th, 2006

This is one post in a series, describing challenges we need to overcome to make free software ubiquitous on the desktop.

There are 347 languages with more than a million speakers. But even Ubuntu, which has amazing infrastructure for translation and a great community that actually does the work, is nowhere close to being fully translated in more than 10 or 15 languages.

You can see a great chart of this in the Edgy Translation Status page.

Translation is one of the key opportunities we have to create something radically better than what the Windows hegemony has yet been able to deliver. Something that will give millions – tens or hundreds of millions – of new computer users a reason to stick with free software.

If you want to make a difference, and you are not a native English speaker but nonetheless have good English and are a regular users of Ubuntu or the free software stack, please help! It’s dead easy to contribute with Rosetta in Launchpad, and you can translate many upstreams or add translations for your home language straight to Ubuntu – we ship out a new language pack at least once a month.

Our goal is straightforward – build a huge, well managed community to translate free software into hundreds of languages with a high level of accuracy, then make those translations instantly and freely available. It’s like wikipedia for your desktop.

Binary-only codecs, nyet

Tuesday, November 28th, 2006

Corey, the distinctions between software that enables the hardware to function fully, and software that delivers a specific feature, are manifest.

Ubuntu has included firmware, and used proprietary drivers since its inception. That’s always been a slightly uncomfortable proposition, as Mako observed, but it’s been true since the Warty Warthog. Even Debian goes some way along that road with its inclusion of firmware and other non-free bits, suspending belief in the DFSG in this particular case. We discussed this at length in one of our earliest summits and settled on the hardware enablement vs apps boundary.

It’s worth pointing out that all of the applications that will be enabled by the AcceleratedX decision are free software applications – Compiz/Beryl and related work are all about showing what is possible at the cutting edge of free software. The hardware to run this is part of the very basic standard PC offering today, but the drivers to enable that functionality are tied deeply to hardware that is not publicly documented. The vendors concerned each have their own strategy – some open, such as Intel, and some closed, such as ATI and Nvidia.

We lobby the companies concerned to open up their drivers, and to a certain extent we can apply market pressure, directing users to hardware which DOES have free drivers which I believe is a much more effective approach than “preaching from a distance”. It’s always better to engage and work with someone than to sanction them and isolate oneself. I’m certain that this strategy moves the free software agenda forward more effectively than any other. Our strategy has already put us in a position to influence significant open source strategy with major companies, and we have used that leverage to accelerate their embracing of free licenses. It’s not always possible to claim public credit when that happens, but I’m sufficiently convinced of the merits of the approach that I feel very good about the impact we are having in the world. I think Ubuntu will have a bigger chance of helping to convince Nvidia and ATI to take an open approach if we build a good relationship, apply market pressure, and get them to see the benefits to them of the open source road. We will, I think, get them there, but not by pouting and yelling insults from our high horse. Remember how little power we really have in that discussion, and remember that free software progress has always been made by playing to our strengths. For a very long part of the history of free software it was ONLY possible to run GNU tools on a proprietary OS.

We can’t assume fancy graphics functionality will be on every machine Ubuntu is installed on, we can however allow X and free software apps to take full advantage of it when the hardware is present. I’m very happy that it is possible (and straightforward!) to remove the non-free drivers from Ubuntu, and I don’t believe that the AcceleratedX specification will change that. In fact, I worked quite hard to get Gnubuntu (an ISO of Ubuntu without any restricted elements) off the ground – it has effectively now emerged as gNewSense and I would encourage you to use that if this is a touchstone issue for you.

I hear you when you say “users want proprietary codecs”. That’s why we make sure these items ARE available, at the user’s option, as packages on the network repositories. That allows users who need that functionality, or who choose that functionality over free alternatives, to exercise that choice freely. We don’t make that choice for them, though of course there is huge demand from real users for that. And we will stay firm in that regard. Ubuntu does not, and will never, include proprietary applications.

Why NOT include those items? Because they exist in free forms, for a start. There are free implementations of MP3 and MPEG and other proprietary codes, and in some jurisdictions its perfectly legal to use them. In time, it will be legal to use them everywhere. That’s not true of drivers for your graphics card. Refusing to include the proprietary codecs, and Flash, and until recently, Java, is part of what defines Ubuntu’s core set of values. So is making damn sure the OS enables the hardware you run it on. In the case of modern graphics hardware, which is the particular item that you are talking about, we are getting to the point where the majority of the transistors in your computer are devoted to pixel and vertex shading, and dead unless you enable them properly. So it’s silly to say that this is “unimportant hardware functionality”.

Fresher’s Day

Monday, November 27th, 2006

As part of Ubuntu Open Week, we’ll be hosting a Fresher’s Day on Friday December 1st where any and all questions from new contributors will be welcome. It will happen primarily in #ubuntu-freshers on and I expect there will be analogous conversations in the Forums and other community, erm, forums. So for folks who are interested in the dynamics of the community but not sure where to start, come along on Friday!

An invitation, not a conspiracy

Monday, November 27th, 2006

A number of people have commented on my invitation to OpenSuSE developers to join Ubuntu Open Week, some have expressed dismay that I would risk creating discord in the free software universe by inviting developers to leave one project and join another. There have also been plenty of reasonable comments and suggestions, and I hope the net effect is to leave both communities better informed about the efforts of the other.

I think it may be worth having public “town hall meeting” in the usual Ubuntu style to discuss the invitation and make sure everyone has a fair chance to air their views. Feel free to continue to comment on this blog – I do the spam-moderation thing about once a week usually, will try to get to it more this week. Till then, let me say the following:

  1. No offense was intended to SuSE – it’s a great distribution. This is about Novell’s extraordinary decision to legitimise Microsoft’s IP claims over Linux in general. I have serious concerns about the Novell-Microsoft deal – and so do other people who make huge contributions to the body of free software. Novell and SuSE are of course deeply linked, and so the actions of one do have consequences for the other. I would expect the same sort of consequences in Ubuntu if Canonical made poor decision. In the past two weeks I’ve fielded many mails from SuSE developers in regard to this, so I believe it was reasonable to point out the timely Ubuntu Open Week. I very much hope all of this helps to bring home to Novell executives the folly of their course, and results in the termination of the patent-related aspects of the deal.
  2. Collaboration between SuSE and Ubuntu is welcome, and I would support efforts to make that collaboration happen in practice. Most free software developers want to see the whole free universe succeed, not just one or other distribution, and collaboration is a good step towards that goal.
  3. Ubuntu is not free of controversy, and neither is Debian. I was not suggesting that Ubuntu or Debian are somehow perfect – only that we would have nothing to do with Ballmer’s offer and are deeply conscious of the impact of this sort of deal on the long term future of free software.

Apologies to anybody who was offended by my extension of the invitation to OpenSuSE developers, it was certainly not my intent to upset you. Thanks to cool heads on both sides who have kept the discussion focused on our shared goals of improving the quality of free software and ensuring that it continues to be widely and freely available.

Welcome, OpenSUSE developers!

Friday, November 24th, 2006

Novell’s decision to go to great lengths to circumvent the patent framework clearly articulated in the GPL has sent shockwaves through the community. If you are an OpenSUSE developer who is concerned about the long term consequences of this pact, you may be interested in some of the events happening next week as part of the Ubuntu Open Week:

We are hosting a series of introductory sessions for people who want to join the Ubuntu community – in any capacity, including developers and package maintainers. If you want to find out how Ubuntu works, how to contribute or participate, or how to get specific items addressed, there will be something for you. I’ll also be on IRC on Tuesday 28th to answer any questions you may have of me specifically, such as Luis’ questions about our position on software patents.

There are a couple of sessions that would be particularly interesting for folks familiar with OpenSUSE. The Kubuntu team is hosting some events during the week to look at KDE and Ubuntu and to discuss the roadmap of their project. There are also a few events being hosted by the Ubuntu Desktop team’s, which I think should include some discussion of the ideas that came from the recent Ubuntu Developer Summit in Mountain View. There are a couple of Packaging 101 and Package Maintenance sessions too, specifically for developers.

Ubuntu is structured to empower our community to get things done, and to maximise the opportunity for collaboration between teams that share a common vision (even if it’s not 100% of their vision, such as between the Gnome, KDE and XFCE desktop teams). While we’re always open to new members, we thought it would be a good idea to identify a dedicated week where new members would be the focus for our whole project.

If you have an interest in being part of a vibrant community that cares about keeping free software widely available and protecting the rights of people to get it free of charge, free to modify, free of murky encumbrances and “undisclosed balance sheet liabilities”, then please do join us.

#9: Pervasive support

Wednesday, November 22nd, 2006

This is one post in a series, describing challenges we need to overcome to make free software ubiquitous on the desktop.

I have this weird relationship with the words “it’s not supported”.

Whenever I’m talking to an audience of typical computer users about Linux I’ll hear those words. I also often hear them when I’m meeting with organisations that could well benefit hugely from free software infrastructure or desktops. “I’ve heard about Linux, it sounds great but it’s not supported.”

This is an interesting comment, when Canonical along with many other companies offer 24×7 support for Linux. Red Hat offers support. Novell offer support. HP and IBM and others all offer support. You can get it on commercial terms pretty much anywhere, anytime.

So why do people say “Linux is not supported”?

Because the guy behind the counter at their corner PC-cafe doesn’t support it. Because the guys they deal with every day, who are more than likely a relatively small outfit, don’t support it. And even if they DO support it, they don’t have a big sticker on the front door next to the Windows logo and the Apple logo, saying “Linux”. There are huge amounts of skill in Linux in many economies out there that are effectively invisible, because they are not specifically advertised.

This is why I encourage governments to announce that some portion of their infrastructure will run on Linux – it catalyses the whole ecosystem to make their existing capacity public. It gives IT services companies a reason to put Linux on the door. It gives project managers a reason to learn about Linux deployments and how best to manage them.

There will come a day when Linux shifts from being something behind the scenes to front and center stage. Then, although the actual number of Linux-skilled people won’t have changed, people won’t say “it’s not supported”.

#10: Pervasive presence

Wednesday, November 15th, 2006

This is one post in a series, describing challenges we need to overcome to make free software ubiquitous on the desktop.

We’re increasingly living in an always-connected world, with multiple ways to speak with one another at any given time. I might SMS or call a friend, reach them on IRC or SIP-phone, send them email or – once in a blue moon – a handwritten note. Knowing how best to talk with someone at any given time is the challenge – if you’re like me you try the fast-and-light stuff first – a ping on IRC or Jabber – then bring out the heavy guns like email.

“Presence” is all about turning that haphazard process into a systematic framework – making sure that you (well, more accurately your laptop and your cell phone) know how you should reach out and touch the person you want to communicate with. It’s about an integrated addressbook – no more distinctions between IM and email – and a constant interaction at the system level to keep others aware of your status.
On the reverse side, of course, none of us wants to be SO accessible that our stream of consciousness is perpetually turbulent. Just as spam has come to be a hostile attack on my inbox, I fully expect to have to defend against hostile attacks on my concentration on IRC, SMS and other comms channels. So I imagine we’ll have a nice little arms race in the “presence” department, as the good guys work to make it easier for us to talk to one another at any given instant, and the bad guys try and offer us body modification pills.

This is an opportunity for the free software desktop to outshine the proprietary guys, because it’s going to be an area of enormous innovation. The core pieces are falling into place – look at Galago and Telepathy – what’s needed is innovation in the ways we use that framework.  I believe that free software communities can innovate faster than proprietary companies. This is a good place to prove our mettle.

#11: Simplified, rationalised licensing

Wednesday, November 8th, 2006

This is one post in a series, describing challenges we need to overcome to make free software ubiquitous on the desktop.

Richard Stallman is the man I admire most in the free software world. Nobody else has so clearly articulated, so beautifully argued for the freedom to change your own software and the freedom to share it. I’m absolutely convinced it is free source, not “open” source, which is at the heart of the innovation that will carry free software to ubiquity. Developers are inspired and motivated to climb in and make Gnome or KDE better *because* they have the right to do so, and they know their work will form part of something big and beautiful.

But my voice is only one of many, and I recognise in this world that there are lots of reasonable, rational positions which are different but still, for some people, appropriate. Thus I think it’s normal that we have the BSD-style licences as well as the GPL family of licences, and it’s also normal that there be a few licences in the mix from companies like Microsoft – witness their “Shared Source” initiative.

The result of those reasonable-but-slightly-different positions has been a plethora of licences, most of which fall into a few broad categories, and a few of which are… thpethial. The diversity of them adds nothing to our cause. If anything it makes it harder to build cohesion in our world, and harder to reap the benefits of both collaboration and competition. There are two major desktop environments in Linux partly *because* of these differences.

So what can be done? Well, I turn for inspiration to the work of the Creative Commons. They’ve seen this problem coming a long way off, and realised that it is better to create a clear “licence space” which covers the various permutations and combinations that will come to exist anyway. Hence we have the various options of the CC licences, which allow organisations to specify their INTENT using a few, high level ideas. And as far as I can see we have neatly sidestepped the potential fragmentation or balkanisation of the open content licence space.

By being willing to represent the whole movement, and not just one square on the board, the Creative Commons has saved us from a very messy death by a thousand paper cuts. It is reasonable for me to dream of whole swathes of compatibly licenced educational content, from hundreds of different owners, being combined to create new works of breathtaking scale.

Here’s the hard edge. The OSI needs to decide if it wants to continue to be a driver of fragmentation, or if it wants to lead towards unification and integration. The FSF, much as I am happy to wave their flag, will always remain aligned with a single-minded set of values, and that’s the right place for them to be. They are not an appropriate forum for talk of the whole movement. OSI, on the other hand, while they have much to earn back in the way of credibility with the free software camp,  might just be able to pull this off.

#12: Consistent Packaging

Wednesday, November 1st, 2006

A long, long time ago, packaging was an exciting idea. There were disputes over style and process, there was innovation. There were reasons to prefer .deb over .rpm over emerge and its binary packages…

Today, these differences are just a hindrance. The fact that there are so many divergent packaging systems in the free software world (and I include the various *bsd’s) is a waste of time and energy. We want to focus the collective brainpower of the community on features and bugs, not on packaging. I would like to see the LSB renamed to the FSB – the “Free Software Base”, and get buy-in from the *BSD’s, and then I’d like to see us define distribution-neutral packaging that suits both the source-heads and the distro-heads. Then there’d be sufficient rationale for the relevant upstreams to include that packaging framework in their revision control repositories, and distro patches would become far more exchangeable.

Ubuntu isn’t built on secret sauce in the packaging. We don’t think our patches should be hoarded – we mail them all to Debian daily anyway, and publish them as best we can on the web the instant they are uploaded, often before.

Packaging is also one area where we can definitively improve on the real user experience for most people who treat computers as a job not a passion. It’s a strategic tool in the battle between proprietary and open approaches. I often think that the proprietary software world’s way of distributing software is one of its biggest weaknesses – an Achilles Heel that we should be exploiting to the full extent possible. I’m often asked why Linux can’t make it easy to “write something like Microsoft Installer, or Installshield”. That’s the wrong rabbithole, Alice. Linux can make it so when you dream of databases, PostgreSQL or MySQL are “just there” and “just work”. That’s a much nicer experience – we should make the most of it.