Mark Shuttleworth on the Future of Ubuntu

On LWN.net.

GPM on LWN.net

In the spirit of passing on vaguely useful information, I notice that LWN.net has put together a handy master index of external contributors, including yours truly.

Second Life's Grand Opening

I wonder whether in retrospect Linden Lab's decision to open up the code of Second Life will turn out to be as momentous as when Netscape gave its Navigator code to the new Mozilla project? Interestingly, Linden Lab specifically invoked that precedent when it made the announcement:

In 1993, NCSA released their liberally licensed, but proprietary, Mosaic 2.0 browser with support for inline images arguably heralding the start of the web as we know it today. In an act of either acceptance of the inevitable or simple desperation, Netscape Communications released the bulk of the Netscape Communicator code base to form the foundation of projects as Mozilla, Firefox, and Thunderbird.

We are not desperate, and we welcome the inevitable with open arms.

Stepping up the development of the Second Life Grid to everyone interested, I am proud to announce the availability of the Second Life client source code for you to download, inspect, compile, modify, and use within the guidelines of the GNU GPL version 2.

A year later, it's a good moment to review where we are, and here are two useful contributions, one from Wagner James Au, the other from LWN's Jonathan Corbet. Things seem to be moving on, and it will be interesting to watch how this area develops.

Oh, Tell Me the Truth About...the ODF Bust-Up

The recent decision by the OpenDocument Foundation to shift its energies away from ODF to CDF has naturally provoked a lot of rather exaggerated comment. I wrote a piece for LWN.net (now out from behind the paywall) exploring what exactly was going on, and found out that there are bigger issues than simply document interoperability at play.

It turns out to be all about Microsoft's Sharepoint - software that I am beginning to see as one of the most serious threats to open source today. Read it and be very afraid.

Linux Weather Forecast

Get your umbrellas out for the Linux Weather Forecast:

The need for a Linux Weather Forecast arises out of Linux’s unique development model. With proprietary software, product managers define a “roadmap” they deliver to engineers to implement, based on their assessments of what users want, generally gleaned from interactions with a few customers. While these roadmaps are publicly available, they are frequently not what actually gets technically implemented and are often delivered far later than the optimistic timeframes promised by proprietary companies.

Conversely, in Linux and open source software, users contribute directly to the software, setting the direction with their contributions. These changes can quickly get added to the mainline kernel and other critical packages, depending on quality and usefulness. This quick feedback and development cycle results in fast software iterations and rapid feature innovation. A new kernel version is generally released every three months, new desktop distributions every six months, and new enterprise distributions every 18 months.

While the forecast is not a roadmap or centralized planning tool, the Linux Weather Forecast gives users, ISVs, partners and developers a chance to track major developments in Linux and adjust their business accordingly, without having to comb through mailing lists of the thousands of developers currently contributing to Linux. Through the Linux Weather Forecast, users and ecosystem members can track the amazing innovation occurring in the Linux community. This pace of software innovation is unmatched in the history of operating systems. The Linux Weather Forecast will help disseminate the right information to the ever growing audience of Linux developers and users in the server, desktop and mobile areas of computing, and will complement existing information available from distributions in those areas.

Good to see Jonathan Corbet, editor of LWN.net, for whom I write occasionally, spreading some of his deep kernelly knowledge in this way.

Your (Second) Life Flashes Past You

Although not quite on the level of LWN's comprehensive review of the year, Giff Constable's Second Life timeline is a handy summary of what happened when in undoubtedly crucial year for SL. And for all those getting their knickers in a twist over the issue of SL's user base (you know who you are) he makes an important observation:

There has been a lot of questions and skepticism around the numbers, and retention rates, which Linden Lab estimates around 10% - 15%, but if you just look at concurrent users, a year ago max concurrent users was around 5 thousand. Today it is around 20 thousand, a 300% growth rate, although still a fairly small pond.

I'd underline that this is concurrent users, so the total number in a given day is some multiple of this: if I had to guess, I'd say around six to eight times. This would give an active user base of around 150,000, which is consistent with a 10% retention rate and two million signups.

LWN's 2006 Linux and free software timeline

A lot has happened in the last year in the world of free software. That makes it hard (a) to remember who exactly did what and (b) to get the big picture. One invaluable tool for doing both is LWN's 2006 Linux and free software timeline, which offers all the main events with handy links to the original stories. They've also got other timelines going back to 1998, if you want to see an even bigger picture. Great stuff for a trip down free memory lane.

Update 1: And here's C|net's list of top stories in the same field.

Update 2: Meanwhile, here's Matthew Aslett's open source year in quotations.

Making All the Right Connexions

A couple of months back I had the pleasure of interviewing Richard Baraniuk for an article about open content for LWN.net. As I wrote then:

Just as open source avoids re-inventing the wheel by building on existing code, so open courseware aims to save time, effort and money by making educational material freely available for others to re-use, extend and improve.

The first such project, Connexions, came from Rice University. It was the brainchild of Richard Baraniuk, professor of electrical engineering, who was directly inspired by the example of open source. Connexions uses a content creation platform called Rhaptos, which is released under the GNU GPL.

Connexions is a fascinating exercise in re-inventing university course materials, but Baraniuk told me that they were planning to go even further. Now Rice has announced that it is to use print-on-demand technologies to produce academic textbooks in a completely new way. As the press release explains:

Rice University's innovative Connexions today announced an on-demand printing agreement with QOOP Inc. that will allow students and instructors anywhere in the world to order high-quality, hardbound textbooks from Connexions - in most cases for less than $25.

The deal positions Connexions to take the lead in open-source textbook publishing as soon as it completes software needed to feed each of its titles to QOOP's on-demand publishing platform. Connexions plans to offer more than 100 titles for online purchase by year's end.

"From its inception, Connexions has used the Web to go beyond print," said Connexions founder Richard Baraniuk. "Connexions lets pupils and instructors make cross-disciplinary intellectual leaps with a simple mouse click, following knowledge wherever learning takes them.

"But being Web-based is also about access, and because our materials are freely available to everyone, we needed an easy, low-cost way to let people use a book if that's the medium they are most comfortable learning from," said Baraniuk, the Victor C. Cameron Professor in Electrical and Computer Engineering.

QOOP's on-demand service will allow Connexions users to order customized course guides and a variety of fully developed Connexions textbooks. Standard paperbacks will take just 3-5 days to produce and ship, and traditional hardbacks will take about a week to produce. QOOP ships directly to customers.

No other publisher of open-source educational content can match Connexions offerings. This is partly due to Connexions early adoption of Creative Commons open licenses. Because all content on the site is authored under these licenses, there are no copyright conflicts to negotiate.

Open content, open source - sounds like all the right connexions are being made.

Clueless or What?

You might think this is a typical non-sequitur laden piece of cluelessness about open source. But one of the ever-watchful readers of LWN.net points out the interesting tidbit that the author "previously worked as policy researcher for Microsoft’s Geopolitical Policy and Strategy Group". Now there's a coincidence.

Another Day, Another Open

Open content is an area that I follow quite closely. I've just finished the second of a series of articles for LWN.net that traces the growth of open content and its connections with open source. The first of these is on open access, while the most recent looks at Project Gutenberg and the birth of open content. The next will look at open content in education, including the various open courseware projects.

Here's a report from UNESCO on the area, which it has dubbed open educational resources, defined as

the creation of open source software and development tools, the creation and provision of open course content, and the development of standards and licensing tools.

I'm not quite sure we really needed a new umbrella term for this, but it's good to see the matter being discussed at high levels within the global education community.

Beyond Parallel Universes

One of the themes of this blog is the commonality between the various opens. In a piece I wrote for the excellent online magazine LWN.net, I've tried to make some of the parallels between open source and open access explicit - to the point where I set up something of a mapping between key individuals and key moments (Peter Suber at Open Access News even drew a little diagram to make this clearer).

My article tries to look at the big picture, largely because I was trying to show those in the open source world why they should care about open access. At the end I talk a little about specific open source software that can be used for open access. Another piece on the Outgoing blog (subtitle: "Library metadata techniques and trends"), takes a closer look at a particular kind of such software, that for repositories (where you can stick your open access materials).

This called forth a typically spirited commentary from Stevan Harnad, which contains a link to yet more interesting words from Richard Poynder, a pioneering journalist in the open access field, with a blog - called "Open and Shut" (could there be a theme, here?) - that is always worth taking a look at. For example, he has a fascinating interview on the subject of the role of open access in the humanities.

Poynder rightly points out that there is something a contradiction in much journalistic writing about open access, in that it is often not accessible itself (even my LWN.net piece was subscribers-only for a week). And so he's bravely decided to conduct a little experiment by providing the first section of a long essay, and then asking anyone who reads it - it is freely accessible - and finds it useful to make a modest donation. I wish him well, though I fear it may not bring him quite the income he is hoping for.