The Ed Techie

Loosely coupled vs integrated again

Continuing our debate around VLEs (which I'm enjoying anyway), Niall Sclater posts that one of the dangers of the small pieces approach is that we have no control over external sites, and he uses (my favourite web 2.0 site) Slideshare being down as an example.

Tony responds rather cheekily (he tags the post 'baiting') showing the OU site was down, to demonstrate that internally hosted systems are not immune from such dangers.

I would suggest that in a loosely coupled system your risk is spread from a user perspective - if Slideshare was down, that's fine I'll do the wiki activity, or look at the YouTube clips, or read the RSS feeds in Google Reader. There is an alternative activity I can be doing, but if your VLE goes down, you are stuffed since everything is in one place.

In general though, Niall is right, this is a concern (and one I mentioned in my original post). But it gets at something deeper I think than just about choice of hosting services - it's about philosophies and attitudes. The almost universal reaction in all higher education institutions is to control things - be it systems, content, dialogue, roles, etc. And there is a good reason for doing this - universities have a duty of care, both pastoral and educational, to their students, and the best way to deliver on this is to control the environment.

Hence we have educational versions of tools, closed systems, selected readings, etc. And then we have web 2.0 which lets anyone do anything and then puts metrics and filters in place to help you find the good stuff. In order to engage in this world there is going to have to be a good deal of letting go of control - not only will that be hard to do, but it may have serious implications for students (and maybe, legally for universities - what if someone has a bad experience on an externally hosted site which you have sent them to - are you responsible?). I'm not arguing that letting go of control is necessarily better - I just think more and more of it will happen anyway, so we need to be able to handle it. I don't think trying to control things more is a valid response though. This tension between control (and thus being able to assure education) and freedom (where bad things may happen) will be one of the key questions higher education faces in the next few years, and it's far more complex than control= bad, freedom = good.

The VLE/LMS is dead

(but we'll probably take five years to realise it). Scott Leslie has coined the term Loosely coupled teaching, for the assembly of a number of different, third party apps to do your teaching with. This differs from a PLE in that it is still the educator who provides the tools, they just bypass the institutional systems.  Scott's post set me thinking and I had something of an epiphany (okay, I know I'm slow) - whereas I have been using the odd application you could completely set up your course outside of the institution.

What might you want to use? For my course I could use the following:

• For content and discussion - wetpaint or wikispaces, upload the content, each page has its own forum, and students get to control the content. Or I could create a blog wherein each post is an activity, this has the advantage that students can subscribe. I'd probably go the wiki route though.
• Dialogue - I'd probably encourage everyone to sign up for Twitter or Jaiku, to create a low-level, non-intrusive dialogue between students. We might also need some ongoing discussion board.
• Virtual meetings - I'd use Flashmeeting and schedule regular sessions.
• News and updates - I'd use a blog and get students to subscribe.
• Quiz/informal assessment tools - there are a bunch of these around e.g. Blogquiz where you can embed them in your own page. There's a lot of commercial stuff too, so I think there is room for better free services in this area.

Apart from that I would use any tool as and when you needed it. The wikispace would be the hub, and hopefully you could hang these apps off it easily enough. An alternative would be a Netvibes/Pageflakes space.

Firstly, what are the issues in doing it this way:

• Authentication - this is quite a big one. Students are authenticated via the University database and this feeds through to the VLE and related systems. Single sign on is obviously a big plus here. For small courses you could manually enrol your students on your wiki (if you didn't want it to be open to all), but for some of our courses we have 1000+ students, so that isn't scaleable. Having said that, this is not a problem that is insurmountable. Authentication isn't really my subject area, but with openid, Shibboleth etc people are moving in this direction. What I want is to be able to apply the OU authentication to any site I want, so if I create a wiki I simply tell the OU authentication system to include that url. Maybe it can do this already? The issue of roles is more complicated, but again if we start on this now, it's not impossible to crack.
• Convenience - there is a degree of convenience for both academic and student in having all the tools packaged in the VLE. However, I think there is also an increasing frustration at being limited to these tools, and also an increased ability to cope with a range of tools.
• Support - if you have one centralised system then you can offer centralised support also. If every academic is using a different collection this becomes more difficult. However, these tools are all pretty easy to use, and one could easily have a collection of supported ones.
• Reliability - if we house the VLE then we can guarantee the server times and service level agreement. If it is housed on an external system you have no control if it goes down. This is true and something that keeps IT people awake at night, but this surrendering control is going to be one of those things we just have to get used to as we use more third party apps simply because they're better.
• Monitoring -  one of the tools that a VLE offers is the ability to monitor a student or cohort's progress. These can be useful tools in identifying problems and offering support. While a loosely coupled system wouldn't offer this at the individual level, there are an increasing number of sophisticated analytical tools available (as Tony Hirst repeatedly tries to get me to realise) which will provide much of this information.

That might seem like a big enough list to make you stick with your VLE, but wait, here come the benefits:

• Better quality tools - because offering each of these loosely coupled elements is what each company does, it is in their interest to make them really good. This means they stay up to date, have better features, and look better than most things produced in higher education.
• Modern look and feel - related to the above, these tools often look better, and also their use makes a course feel more modern to a user who is raised on these tools compared with the rather sterile, dull systems they encounter in higher ed.
• Appropriate tools - because they are loosely coupled the educator can choose whatever ones they want, rather than being restricted to the limited set in the VLE. This is one of the biggest draws I feel - as an academic if I want a particular tool I don't have to put a request in to IT and wait a year to get a reduced quality version, I just go ahead and use it.
• Cost - using a bunch of free tools has got to be cheaper hasn't it?
• Avoids software sedimentation - when you have institutional systems they tend to embody institutional practice which becomes increasingly difficult to break. Having loosely coupled system makes this easier, and also encourages people to think in different ways.
• Disintermediation happens - this isn't really a benefit, just an observation. If a services can be disintermediated then it will be. In this case the central VLE system is disintermediated as academics use a variety of freely available tools.

On balance then, I think this shift to loosely coupled, freely available third party systems will happen. It won't happen overnight, and it will follow a succession model, but I would place money on this being the direction things take. Thank goodness I haven't just published a book on VLEs eh? Doh!

Some quick opensocial thoughts

What with travelling back from Oz and catching up on work I've not had time to really look at Google's opensocial much. Not that there is much to look at currently, except some big announcements. If you don't know what it is, it is an open API approach from Google with a host of partners to allow users to easily create social tools which can be used in more than one location. So, here a week later than everyone else, are my thoughts on it.

Let's get the Facebook thing out of the way first - opensocial is the social networking technology equivalent of a politician waiting to see what their opponents propose and then suggesting the opposite as policy. So, whereas Facebook allows people to develop applications for Facebook, those applications are not easily portable anywhere else, and the data stays inside Facebook. It's classic walled garden stuff. This is the biggest complaint against FB, so naturally it's where Google should attack. Opensocial, in theory anyway, allows you to create tools which will port across to any platform or site that signs up to their API. This is smart - Google doesn't have to go and develop a social networking platform to compete with FB, it just makes the other networks suddenly seem more attractive to developers in particular. And if developers start creating cool opensocial apps, but don't bother with creating a bespoke FB one, then suddenly those other networks look more attractive to users too.

Ironically, FB being closed may be a good thing in the long run, because it forces Google to be open. Google have a decent record here with Google Maps (and the famous don't be evil motto), but without FB on the scene would they have gone for such an open approach? Maybe not.

From an elearning perspective, it has potential. We have been developing FB applications, but if we developed opensocial ones then users could port them wherever they like. The implication is that the PLE is a step nearer - we (the university) could create opensocial tools, for users to assemble in the platform of their choice. As RSS does for content, so might opensocial do for collaboration, and that's no bad thing.

I just need Tony Hirst to develop an app now...

A lexicon for the virtual/real world intersection

It strikes me that there are a number of new social behaviours that are arising when our online/web 2.0/social network selves intersect with our real, physical, everyday lives. So, while awake with jet lag I started compiling some new terms. I have put these up in a Wetpaint wiki, so you can add to them yourselves:

• Twintle - a knowing look someone gives you when you meet them face to face, who is also a Facebook friend or Twitter follower, where they debate whether they should mention your status updates, or whether that would be impolite. e.g. "I met Kathy the other day, and I'd been saying about how much I hated work at the moment on Facebook, and she gave me a twintle, but didn't say anything."
• Unpoke - the disappointment of meeting someone in real life and finding they are not as interesting or amusing as their online personas. E.g. "I met Psychopony the other day at a conference, he was really dull, definite unpoke."
• BackBacn  - when a post generates a lot of real life traffic, for example every second person at work comes up to you and comments on it. E.g. "I did a rant about the RAE the other day, talk about BackBacn!"
• Twetch - when someone reveals rather too much about themselves online. E.g. "Dave's a real twetch, he was twittering about his operation in graphic detail, I almost fetched my breakfast."
• Friendgap - the awkwardness when you meet someone whose invitation to be a friend you declined.

So have a go yourself - okay, it was an excuse to try out Wetpaint, did I mention I'd set up a wiki?

Learning design as aspirin for the web 2.0 headache

In the last couple of posts I talked about some of the ways being online, and web 2.0 in particular, challenges some of our assumptions of higher education. The whole web 2.0 thing represents something of a problem, or headache if you will, for higher ed. On the one hand we can see how enthusiastic people are for it, and how it genuinely creates user participation, community, and quality content. All things we'd like to have in higher ed. And on the other we cherish lots of aspects of higher ed that seem at odds with it, such as the quality assurance of content, careful support and structure, hierarchical structures, etc.

So, one of the things I've been working on at Macquarie is how we can bridge this gap. I've been focusing on learning design (in its broadest, not IMS sense), but I would see that as only one means of attacking the problem. I've written a paper for the LAMS conference on this with James Dalziel, so when I give that I'll post the slides and do slidecast, and I'll also work it up into a longer chapter.

For now though, here are some ways in which learning design could help ease the 2.0 headache:

• Meno’s paradox – or, the need for guidance if you prefer. Learners still often seek guidance and structure. For some subjects they are satisfied with creating this structure themselves, for example by finding resources such as blog postings, tutorials, articles, podcasts and video clips. For other subjects, particularly when the subject is itself complex, or the learner feels less confident in the subject area, then providing a scaffolding structure is essential to help the learner build concepts and skills in a robust manner.
• Granularity of learning – in the post on granularity, I argued that the size of educational unit we commonly recognise has been largely determined by physical factors. If learning designs were created and shared by a community of users, what might be thought of as a Flickr for learning designs, and these could be run by individuals, or by groups of interest, then many of the restrictions on size which derive from a hierarchical, centralised model disappear. I've used the the music industry as an analogy previously, but in education perhaps a more relevant model is that of blogging. Prior to the advent of blogs, the type of academic output was usually limited to books or journal articles. The granularity of these was partly driven by the economics of publishing, as Shirky argues:
  “Analog publishing generates per-unit costs -- each book or magazine requires a certain amount of paper and ink, and creates storage and transportation costs. Digital publishing doesn't. Once you have a computer and internet access, you can post one weblog entry or one hundred, for ten readers or ten thousand, without paying anything per post or per reader. In fact, dividing up front costs by the number of readers means that content gets cheaper as it gets more popular, the opposite of analog regimes."
  With the advent of blogging, academics (as well as many other bloggers) have found the format liberating, so that blog posts can vary in size from small links with comments to full essays.
• Topography of formality – as with granularity, a set of user generated learning designs allows users to bundle their recent experience together into a course which can be formally recorded more frequently. This would be possible not only because the monopoly of formality is removed from universities, but also because a distributed model of learning design production is the best way to attack the long tail of possible learner interests. If a user wants to find small courses to formally accredit their understanding of highland knitting patterns, history of Sydney in the 1960s or anthropology amongst football fans, then most current formal providers will not meet their requirements, but a sufficiently distributed pool of user generated designs might.
• Web 2.0 quality – much of the concern educators have around web 2.0 is of the quality, and how it can be assured. A set of user generated learning designs could go someway to addressing this by providing a pedagogical structure around resources, and those resources are then changeable. Users can see who has created any given learning design, so some designers may be trusted more than others, rather like sellers and buyers on eBay gain reputational status by recommendations from other users. Similarly, users will be able to comment on designs, thus providing information and context for other users. However, by allowing users to create and select learning sequences it is necessary to accept some of the bottom-up metrics mentioned previously, as the ‘filtering on the way in’ approach currently used in education is replaced by filtering on the way out. This is necessary to encourage participation.
• Personalisation – if a learning design pool reached a sufficient critical mass, then users will be able to select designs that are appropriate to them in a number of different ways: subject area, style of learning, level, range of resources, duration, assessment method, etc.

For these to be realisable we would need very easy means of generating and sharing learning designs, which is where tools such as LAMS and also the OU's Compendium (we have adapted a prototype of this for learning design, which I'll release soon).

Last in the series, gratuitous Home and Away daggy tourist photo:

 

The topography of formality

Just as we think of learning being bundled into a convenient course package (see last post) so we think of the formalisation of learning being grouped into large chunks. Informal learning is difficult to recognise and accredit, and is thus often overlooked in favour of formal education. There is an intrinsic paradox with informal learning – in order to reward and recognise it, then it needs to be formalised in some manner. So, even if we accept that informal (or if you prefer non-formal) learning will be brought in to the fold increasingly, it still needs to be formalised in some manner.

If we were to consider the formalisation of learning as a topography then currently it is a flat plain with a few high peaks, rather like skyscrapers in a desert, representing courses. The learner traverses this landscape over their lifetime, most of it spent on the flat plain, with no easy access to formal recognition, and is then requested to climb large peaks of formality, such as a postgraduate course.

This bears little resemblance to how they actually learn, which will have some peaks, but will be more evenly distributed.

In the online world however, this topography could be subject to considerable change. The peaks become shallower, but more frequent, so it is more akin to an archipelago. In this model, the digitisation of content and interaction allows users to gather evidence of informal learning on a daily basis. They may then choose to bundle this into a formally recognised event, for example by having their portfolio assessed, or engaging in a ‘micro-course’ which demonstrates their ability in a given area, or by creating a meta-document of their own, for example a reflective blog post that draws on the different pieces of evidence.

Ironically, this is actually how educators conduct their professional lives. An educator may engage in a research project and they will formalise this learning through conference presentations or journal articles. They will bundle together recent experience into published text books, or project reports. In this respect the academic profession has a number of recognised means of formalising learning. Many other professions and individuals do not have such readily available and acknowledged means of unifying recent learning and experience.

Oh, and why not - gratuituous Bondi tourist photo:

(Photos: Gavinsblog http://www.flickr.com/photos/gavinsblog/91871302/, Stefe http://www.flickr.com/photos/stefe/236802005/)

The granularity of education

(slight apology for repeating elements from previous posts, but worth gathering in one place).

One of the things I've been thinking about here in Sydney is how learning design can help bridge the gap between web 2.0 and higher education. I'm writing a paper on this which sets out the argument in detail, but I'll post elements of it here as I go along.

One issue I think web 2.0 raises for education is about the granularity of education. I've blogged before about the granularity of ideas, so this is extending that a little.

The digitisation of content and its frictionless distribution on a global scale is challenging many of the assumptions we have about the format of content and the underlying business models that support these. Often these are so ingrained in our view of the content and its related businesses that we do not even recognise them as assumptions.

Prior to digitisation, all content was bound up in its physical form. That is, you had to buy or obtain, the physical copy of a book, CD or DVD. The business models followed the demands of this physical form – CDs require production, packaging, storage, distribution and retail. The record company and record stores are thus a logical necessity in a market for music CDs. But with the digitisation of content many of these demands disappear. MP3s can be distributed freely online, they do not require production (in the physical format sense), or a distribution network.

What web 2.0 adds to this process is the removal of the filtering function performed by intermediaries, which were a necessary part of the model. Prior to the internet artists could make tapes or CDs of their own music and try and bypass the record companies, but they lacked the distribution network. With the advent of the internet they had access to a potential global distribution method, but they lacked the promotion and means of people finding their music. With web 2.0 content discovery is facilitated through services such as LastFM and iTunes, whereby users can find recommendations, see similar artists and people with similar tastes. Thus the filtering process of the record companies is also removed.

This leads to some profound implications for content industries, including newspapers, television, music, film, etc. For instance in the music industry, record companies are beginning to be disintermediated by artists, with artists (e.g. Radiohead) offering free downloads of their albums.

It also changes the nature of our relationship to content. Weinberger says of music

"For decades we've been buying albums. We thought it was for artistic reasons, but it was really because the economics of the physical world required it: Bundling songs into long-playing albums lowered the production, marketing, and distribution costs ... As soon as music went digital, we learned that the natural unit of music is the track.”

Education has some similarities with content industries, but also some significant differences. However, it is worth considering whether we hold similar assumptions about the granularity of education as we held about the granularity of music, which would be subject to change with digitisation of content and provision of online services. Higher education, as we normally conceive of it, is typified by the undergraduate degree course. This takes 3-4 years continuous study, comprises a number of modules, has regular exam and assessment sessions, is taught face to face, and students are assessed in terms of the knowledge they demonstrate of the taught modules. There are, of course, variations to each of these elements – study can occur at a distance, it can be part-time, assessment can be portfolio and continuous, there can be breaks in study, etc. But each of these adaptations is usually mapped on to the existing, standard model. They represent modifications to it, not replacements.

However, it may be that many of these assumptions are bound up in the economic models that have their roots in the physical aspects of education. For example, if you are requiring students to come to a physical campus, then it makes sense to bundle all their modules in to a short time span to minimise inconvenience and to manage staff time. If the assessment is then based on an exam, it similarly follows that you package this up into one event. These restrictions have then moulded what we deem to constitute a higher education experience, but like the album, perhaps this packaging is merely a product of the physical format, not a ‘natural’ means of structuring it. Even when courses have moved online, they have usually followed similar conventions in terms of length and assessment.

The digitisation of content, and perhaps more significantly, dialogue, means that the type of learning event we can include and assess now changes. It is not just the standard lecture, but can include video, blogs, podcasts, etc. It can also include discussions between learners, in the form of asynchronous text forums, recorded virtual meetings, instant messaging transcripts, etc. So the type of content we assess the student’s understanding of changes, which will inevitably have consequences for the way we assess it. The formal exam or multiple choice question bears little relationship to the student’s experience when connected to such a range of media.

Just as blogs allow us to have varying sizes in our academic writing, from single links and comments to full essays, so user generated learning designs might facilitate a wide range of 'course' sizes.

Oh, and a gratuitous whale watching picture taken by the boat camera we were on (sometimes a phone camera isn't up to the job):

Slifeshare and my startup idea

(hat tip: Tony Hirst) Rev2 reviews this web 2.0 service, Slifeshare, saying

"Do you really want to know what your friends are doing online right now? Do you fancy a bit of online friend espionage? What article they are reading in the New York Times online or what iTunes track they are listening to in real-time? If so, Slifeshare is the new social network for you."

The idea is that you remove the need to actually contact friends or post things about yourself, you simply track each others activities. So I can see what videos, blogs, music you access. Here's the pitch

"Slifeshare is an activity network for you and your friends where you share your live computer activities, such as videos you watch, music you listen to, web pages you find interesting and much more as a way to stay in touch"

I can see the power in connecting as a byproduct of what you do, but it's a step too far, can you imagine your friends tracking what you're doing all the time? 'Martin you've been looking at that Spurs discussion list for two hours now.' 'Martin, I noticed you only opened that file I sent you for twenty seconds, so don't pretend you've read it.'

That aside, what I need is a meta web 2.0 service, one which finds the web 2.o services I need because I can't keep up with them all. I can't even be bothered to work out which fine niche each service is targeting, and whether that is in fact a service I need. Each web 2.0 startup is so specific that matching it on to your daily life and determining whether it could help, or what the potential might be if you modified your behaviour is a costly overhead. So there's my idea for a startup - one that knows my routine, needs, interests and gathers a set of 2.0 services together for me and then makes recommendations, suggesting how each new one could help. 'We do 2.0 so you don't have to' is the tag line.

Someone will tell me such a thing exists now won't they?

[Update: In the interest of fairness - the slifeshare people contacted me and say that it "doesn't actually expose how much time you spend on web sites or emails. If you want to share a site, you need to do so explicitly. People can see the aggregate of sites you visit on a day, but not the amount of time you spend on these sites. Also, emails are not part of Slifeshare at all. "

So my examples were a bit off the wall. Maybe though, in order to be useful that's the type of data it does need, but that's a different question]

How do I? video search

Tony Hirst has created a neat little search interface using Google's Coop customizable search engine. It searches a number of video archives to find instructional videos - he seeds it with the phrase 'How do I?'  He explains it here, so I won't bother, and thus save myself potential embarrassment by getting it wrong.

I will say it returned some very good results for me. I put in a search for How do I.. create a podcast. And the resources it came back with were great. Having just written a piece on podcasting for a course, this reinforces to me the point that increasingly there is little point in me writing guides or even finding resources - students can do much better themselves and find the ones they want. This doesn't make me totally redundant though (my mortgage lender will be pleased to hear), I think my role then is to place podcasts in an academic context by creating meaningful activities and perhaps producing other material which relates to pedagogy, broadcasting, or whatever - ie to make some links to material they might not get through such means. This is a good thing - educators should concentrate on what they do best, not writing how to guides or acting as a search engine.

Tony says he knocked it up in twenty minutes. This also demonstrates the difference in approach in the web 2.0 world 'out there' and the academic world. You just know if this was a university (any university) project it would have been funded for a year, they'd set up careful consultation, draw up a set of user specifications, run some workshops, produce a report, start coding, get to a demo stage that worked only on a very small subset of video, then disband.

Anyway, give it a go.

The future of content -Pt 1

[Note - this is part of a distributed article, see previous post for explanation]

The Future of Content

In this section I am going to argue that digital content will move toward being free and widely available because of two complementary arguments: the argument from economics and the argument from quality.

The argument from economics

Where content can be digitised, it is having a profound effect on the economics that underlie the business model of that content, and the way society uses and thinks about it. In this opening section I want to look at two examples of how the digitisation of content has led to significant changes in a number of industries.

Newspapers

Many newspapers ignored the online world for too long, assuming their customers would want their news that way. Now that they have been forced to shift all of their content online, they are finding new business models to cope with this. The initial hope was that essentially the same model would apply, that people would pay for content. But the pay per copy, micropayment and subscription model have all failed. Most recently the New York Times closed its TimesSelect subscription business and made its content free. Apart from it not making enough money, the reason seems to be that the subscription model harmed the alternative model of advertising, which is driven by the global audience finding their content.

Vivian Schiller of The Times comments “What wasn’t anticipated was the explosion in how much of our traffic would be generated by Google, by Yahoo … our projections for growth on that paid subscriber base were low, compared to the growth of online advertising”. In other words, it is better to be free to all of the market than paid for by a small section.

Music

Another example of the massive changes wrought in an industry by the digitisation of content is the music industry, which like newspapers was slow, and resistant, to change. The wake-up call for the music industry was Napster, when suddenly millions of users were exchanging songs and albums without paying. This was partly a result of wanting to have something for free, but it was also because Napster facilitated behaviour which was the content discovery and the social function of music. Through Napster users could find other bands that were similar, sample different types of music, find other users with similar tastes and, most importantly, do all this from their laptop.

Eventually, iTunes saw the industry find a model that seemed to suit both parties. Users could download individual tracks at relatively low cost, and also engage in the content discovery through shared playlists.

LastFm and Pandora take this social aspect a stage further, by data mining users actions to build up a network of artists, so that you only need enter an artist’s name and be able to have a personalised radio station playing similar tracks. You can also find users with similar tastes, join groups, see upcoming events, etc. You can’t own music through these sites and you can’t request a specific album, but when the richness of the choice becomes so great, maybe that becomes less significant. And meanwhile the file sharing software is back, with applications such as Limewire allowing users to share files again.

The internet has also changed the relationship of the recording artist to the record companies. Increasingly bands are establishing an online presence, allowing free downloads of their music to build a following, touring, and recording an album, and only then seeking a label. Even this last stage will become redundant once CDs finally disappear. As Chris Anderson puts it in the Long Tail

“At this point, the artists don't need the labels any more. The consumers don't need the labels any more and I think the labels, rather than trying to protect what business they have, need to ask themselves what is their relevance."

Changing our relationship to content

I’ve concentrated just on two of the more obvious examples here, but there are many more. For instance, where audio goes, video will follow once the bandwidth is sufficient. So for traditional television broadcast we are seeing subscription models becoming increasingly difficult to maintain. This has been partly affected by the BitTorrent type sharing services that allow the download of large AV files. At the moment this remains something of a technical skill, but as it becomes more viable and easier, then DVDs will go the way of CDs.

As well as changing the underlying economics of the industry, it alters the way we relate to the original product. To return to music again, in Everything is Miscellaneous, David Weinberger suggests that digitisation of content has altered our perceptions of what we thought was the basic unit of musical output:

"For decades we've been buying albums. We thought it was for artistic reasons, but it was really because the economics of the physical world required it: Bundling songs into long-playing albums lowered the production, marketing, and distribution costs ... As soon as music went digital, we learned that the natural unit of music is the track."

Nick Carr disagrees with Weinberger, stating the artistic structure of the album, using Exile on Main Street as an example but Clay Shirky argues that if this were so, then it would survive digitisation. When you look on iTunes this is not borne out - most people download Tumbling Dice.

Digitisation has made the track the currency, and then users have begun to create their own playlists by mixing tracks together. In addition, attendance at concerts and festivals is on the increase – people will pay for the live event, but less so for the content that supports it. Digitisation has changed our relationship to music, artists and record companies. Forever.

Shirky’s Second Law and the Content Law

In 2003 Clay Shirky wrote an article called Fame vs Fortune: Micropayments and Free Content, in which he argued that micropayments (which he defined as "payments of between a quarter and a fraction of a penny,”) would fail. At the time there was considerable interest, and belief, in micropayments as a model for internet enterprise. He said of the failed attempts of micropayments that “they failed because the trend towards freely offered content is an epochal change, to which micropayments are a pointless response.” Whereas analogue publishing has inherent costs, essentially the cost of the format, storage and transportation, digital publishing doesn’t. The cost involved then is that of the creators, and online the creator can become the publisher. They will then be faced with a dilemma, Shirky argues, of fame or fortune. In an analogue publishing world you could have both, since in order to achieve fame lots of people need to have read your book, or bought your album. If online people are resistant to paying then charging for your content limits your fame potential. As he puts it in a follow up posting in 2007, creators are “in the position of having to decide between going for audience size (fame) or restricting and charging for access (fortune), and that the desire for fame, no longer tempered by reproduction costs, would generally win out.”

We’ll call this Shirky’s second law (the first is generally given as “Diversity plus freedom of choice creates inequality”, a precursor to the long tail), that is given the choice between fame and fortune, fame wins out.

This, combined with what we have seen above led me to propose the content law, which I think embodies what will happen to content in the future:

"Digital content wants to be free, and will seek the path to maximum access."

In their book Blown to Bits, Evans and Wurster argued that the digital marketplace has seen the unbundling of the economics of information and physical product. This is most readily seen in retail, where you have to see the physical object in a shop to know about it, but online the product information is separated out. David Weinberger explores the implications of this unbundling further, in that it allows infinite recategorisation because the information, unlike the physical product, can be in multiple places at once.

Let’s consider a possible future example, that of books. When the web first became popular there were some suggestions that books would disappear since we could download free ones. This did not come to pass for a number of reasons:

1. Transportability – books are easy to carry about, and don’t run out of battery life, can be used on a train and don’t require a special device.

2. Ease of use – books don’t require special help programmes, can be used by most of the population, have good navigation features and are good at presenting text, compared with the small screens of some hand held devices.

3. Cultural value – we cherish books as social artifacts. Few things cause as much outrage amongst civilised people as seeing books being burnt or desecrated. People have a deep affection towards the tactile nature of a book.

For these reasons, and combined with the advanced content discovery facilitated by Amazon and co, book sales have done very well since the arrival of the net. But let us consider what would happen if digital paper really arrived (despite several proclamations digital ink and paper have been stubbornly difficult to realise, but that isn’t the point of my argument). It felt like real paper, you could have it bound in books like real paper, you could write on it like real paper. But crucially the content it displayed could change, you could search it, and it could record all those annotations you made.

Even as something of a bibliophile, this begins to look tempting to me. I like having books as objects on my shelves, but I used to like having vinyl albums and CDs also, but now I only have MP3s (and clear shelves). If digital paper were so good it overcame the 3 benefits of books above, then it would have significant advantages over real paper.

What would the book publishing industry look like then? I would suggest that Shirky’s Second Law and the Content Law would take over. What would be the role of publishers? If you can download a copy of book into your digital paper book, then a good deal of what the publisher does disappears. They don’t need to provide the printing, binding, or distribution. What they can provide is the marketing. But this is where Shirky’s second law becomes relevant – some authors will start to give their content away cheaply (since unless you’re JK Rowling, most of the cost of a book goes to the retailer and publisher to cover all the costs associated with analogue format). As an author you may only get 10% and have to do most of the work, so why not sell it for that and generate interest online? Then another author decides that actually what they’ll do is give their book away free, because that way people link to it, quote from it, mash it up with Google map overlays, or whatever. Being free and open generates a lot more traffic.

For the industry as a whole the content law is now in operation. Books are now digital content that wants to be free and to have the maximum audience. Publishers need to find a new business model and relevance or they disappear.

As George Siemens puts it "Consumers, like learners will in the future, have a dramatically different relationship with content than they have had in the past. Textbook publishers, journals, and other content-centric industries need to take heed of these lessons and adjust before they become the next statistic.”

The argument from quality

The argument I have given above suggests that economics will be the main driving factor in the liberation of content and has focused on individuals or small groups creating content. A second factor is that it will improve the quality of a lot of content through the distribution of the process.

Here we have a powerful analogy in the process of natural selection which shows us that a vastly distributed process can produce things of great complexity. As Daniel Dennett argues in Darwin’s Dangerous Idea, Darwin’s great contribution was to remove the need for top down intervention (or sky hooks in his metaphor) in any explanation of how biological complexity is achieved:

"Cranes can do the lifting work our imaginary skyhooks might do, and they do it in an honest, non-question-begging fashion… Skyhooks are miraculous lifters, unsupported and insupportable. Cranes are no less excellent as lifters, and they have the decided advantage of being real.” (pg 75).

The blind, but distributed process of natural selection was sufficient to do all the ‘lifting’ required in biological design space. Natural selection is distributed over many individuals in a species and over a very long time span. This allows for small, but incremental changes to produce cumulative complexity. The internet allows for similar distribution across individuals, but unlike natural selection, each participant is not dumb, or blind, to the process, thus the overall process is speeded up considerably, and we don’t have to wait millions of years for the results.

What the internet, and web 2.0 in particular, achieves is this massive distribution of the task. Just as prior to Darwin there was no way of conceiving of biological complexity without some designer, without top-down input, so many of the web 2.0 critics fail to understand how you can achieve the relevant sophistication required for, an encyclopaedia say, without a heavily controlled, centralised, top-down process. The democratisation of process that web 2.0 has wrought is key to understanding why they are wrong.

Take photography as an example. There are a lot of people out there who are quite good photographers, some of whom are ignorant of the very specialist skills required, but who have a natural talent, others who have some knowledge, and others still who are specialist in certain types of photography. They were all largely ignorant though of the process required to become professional photographers, of how to share their photographs when the sharing is controlled by the economics of distribution and a top down authority. With the advent of Flickr all of these people can now share their photographs. The result is an explosion of creativity, and real, undeniable high quality photographs.

What the web 2.0 critics would say is that if you compare any photographer at random from Flickr with any random professional photographer then the second will win out, because they have been through the filtering process. But this is to fundamentally misunderstand the nature of the distributed process that is now in place. Sure, if you pick any photographer at random on Flickr then you probably find very average family snaps, but the result of the process as a whole is the production of islands of complexity. And what is more, because the traditional filtering process in the top-down model tends to make professional opinion converge, what you get from the bottom up process is a far greater range of inventiveness and style. In evolutionary terms, the first process is like inter-breeding while the second is akin to broadening the gene pool.

This is what critics such as Andrew Keen fail to appreciate in their criticism of web 2.0 and user generated content. It is not the comparison of any one individual with another, or any one artefact, that is significant, but comparison between the processes. And when it comes to producing complexity, mass distribution wins every time.

So the second reason why content will become free is that only by removing it from behind the confines of payment and strict rights control, can certain types of content be improved. We have seen this in software of course with open source, and with general knowledge in wikipedia, but it also occurs on a smaller basis with blog posts, podcasts, etc where the overall product is improved by making it open, and then incorporating user comments and feedback.

[Over to Ray Corrigan for the second part of this]