Is mass collaboration all it's cracked up to be?

On Friday I spoke at the National Library of Australia’s Innovative Ideas Forum. It is an annual day, free and open to the public, that aims to “highlight new ideas”.

The other speakers were excellent and I will probably post again when the videos/podcasts become available.

The NLA encouraged the audience to take part via Twitter and they certainly did. It’s the first time a talk I’ve given has been so heavily “liveblogged” and I was a bit nervous about it but I needn’t have been.

thanks aldellit! And more tweets. My name proved very hard to spell. Just as well I didn’t tell them my username. ;)

Download this essay as a PDF

(the PDF doesn’t have any links, but it may be easier on the eyes)


Is Wikipedia a one-off?

Is mass collaboration all it’s cracked up to be?

This essay is a slightly modified version of a speech given at the National Library of Australia’s Innovative Ideas Forum 2010. It is licensed under the Creative Commons Attribution ShareAlike license.

What is mass collaboration?

This month I had my fifth Wikipedia birthday. Five years ago, I was 21, in my fourth year of university and probably procrastinating before exams, and I discovered Wikipedia, and created an article for the Victorian Women’s Football League. After a few more edits to the same article, I left a lengthy comment on the talk page of the HSC article. And thus my Wikipedia career began.

Over the years I’ve written articles on famous pubs and theatres, the town where I grew up, bands and authors that I like. I’ve taken and added around 100 photographs of tennis players, and hundreds more from my travels abroad and even just in the streets of Melbourne. I’ve made over 10,000 edits on Wikimedia Commons, which is a sister project to Wikipedia that hosts images and media files. I’ve written over 200 blog posts about Wikipedia related topics. I’ve welcomed newcomers, banned trolls, drafted policies, I’ve volunteered for projects, I’ve mediated disagreements. I’ve answered press requests. I’ve been to three international Wikimania conferences and expect to attend my fourth in Poland this July. And for the past couple of years I’ve concentrated my attention on Wikimedia Australia, which is a local not-for-profit that aims to promote free cultural works, such as Wikipedia.

So it’s safe to say that Wikipedia has changed my life.  It’s easy to forget how Wikipedia has changed the world. It may not impact on your life in a way that you constantly re-evaluate, but Wikipedia is to the internet as the Library of Alexandria was to the ancient world. The thing is amazing in itself, but it is also amazing in what it stands for, and how it opens our eyes and forces us to re-evaluate the possible and impossible. Ten years ago it was obvious that a project that let anyone contribute, with no pre-publish checks or balances, would never have a chance at creating the biggest reference work ever known, or even anything remotely comprehensive. That was obviously impossible. Today, our horizons for the impossible have been pushed back that much further.

Yet the Library of Alexandria was destroyed, and while fire and imperial coups are not a great concern for Wikipedia, there is no guarantee that Wikipedia as we know it will continue. The greater risk at this point is of complacency. Taking Wikipedia and its success for granted has the potential to harm many a fledging mass collaboration project as well as Wikipedia itself. And it would be a sad loss if we were to let that happen.

Wikipedia is just the most well known of hundreds or maybe even thousands of mass collaboration ventures that exist online. Mass collaboration projects involve self-selected participants from anywhere in the world coordinating their work with others towards project goals. Many wikis are mass collaboration projects, even if they currently only have dozens of editors rather than hundreds. Social networks, blogs or microblogs can be used as tools in mass collaboration projects, but they are not mass collaboration projects themselves — there is no single goal directing everyone’s efforts.

I’m interested in the questions around the potential for mass collaboration projects to change society as we know it. We have seen that they can achieve amazing things just in coordinating to write a reference work together, or write an operating system together. Could the goal of a project like this be taken offline? Could a mass collaboration project be as effective offline as they are online, or are such efforts doomed to fail?

What are the promises of mass collaboration?

Mass collaboration promises to change the world. And yet like so many new technologies and technological practices, the reality takes a while to catch up to the dream.

Wikipedia promises comprehensiveness, but somehow we get deletionism, where if it doesn’t look like your brand new article was born among the pages of Encyclopaedia Britannica, it might be deleted before you’ve had time to finish congratulating yourself on hitting the edit button in the first place. The experience is hardly a welcoming one for new editors.

It promises depth by accretion, but we get controversies about biographies made defamatory and left unattended for months.

It promises a destruction of the ivory tower of traditional journalism and academia, but as the community experienced exponential growth and the kind of overnight success that takes years to attain, it also embraced a conservativeness in its belief about the Wikipedia product. Embracing encyclopedia traditionalism was a way of defending itself, a kind of appeal to existing authority.

There is the promise of the potential to overturn European and North American biases in how the academic model of the world is constructed, yet the embrace of traditional encyclopedism and verifiable sources means the existing biases are entrenched rather than overturned.

Most of all, it promised that ‘anyone can edit’ and to let you ‘ignore all rules’, but instead we have the dubious honour of creating what is probably the world’s most arcane bureaucracy that ordinary people are actually expected to interact with. Lawyers speak in Latin, but Wikipedians speak in acronyms, and the result is no less exclusionary.

Now these points represent a cynical view. Most of the time, we get both parts. We get everything. Parts of Wikipedia are bureaucratic messes, and other parts are shining beacons for the wisdom of crowds, and how our shared understanding is improved through argumentation. But the honeymoon for Wikipedia is well and truly over, and as part of the web “wallpaper” we now take for granted, it remains to be seen how it might successfully continue.

What is the real potential of mass collaboration?

The vision statement of the Wikimedia Foundation, the US not-for-profit that keeps the lights on at Wikipedia, is as follows:

Imagine a world in which every single human being can freely share in the sum of all knowledge. That’s our commitment.

The sum of all knowledge. Every single human being. Now that’s a pretty bold claim. It’s also something written with the benefit of hindsight, when Wikipedia was already on its way to making that a potential reality.

Yet as bold as that statement is, to my mind, it doesn’t go far enough. While providing a cost-free, accessible comprehensive educational resource is undoubtedly a force for good, it’s only half the story.

The part of using Wikipedia that really changes how you think, and understand the world and yourself, is the editing part. The contributing, participating part. Understanding that you have the authority, the permission, to contribute to the world’s collective record of itself, that your perspective is valuable and needed, is the game-changing manoeuvre here.

Now if you think that this is nothing special, and it’s always been available to anyone who worked hard enough for it, I invite you to check your privilege. Consider how your nationality, your class, your education, your gender, might contribute to voices like yours being privileged above others. For most people in most of the world, having their voice heard and valued is an unimaginable dream.

Even people in the western world, who generally start out with an advantage, realise how valuable this affirmation is. I have been heartened this week to see criticism of Apple’s iPad for being ‘lean back’ rather than ‘lean forward’. Some media outlets see it as manna from heaven that a popular technology will force people back to being passive consumers, receptacles for advertising and squares of pre-approved “content”. But the presence of the criticism means that we are starting to collectively internalise participation as the default mode for digital interaction. We get a new device, we want to know where the edit tab is. We want to know where the comment box is. We want to know where the feedback forum is.

The owner’s manifesto of MAKE magazine is “if you can’t open it, you don’t own it”. There is no single word yet for the concept of this kind of ownership over digital works. If you own it, it doesn’t mean you created it, but you take some responsibility for participating in it in some way. One day we will be able to say about websites, “if you can’t participate in it you don’t own it”, and the participation right of digital works will be as meaningful and important as the property right of physical objects.

How is Wikipedia typical and how is it atypical?

Now all this talk about participation doesn’t mean that we’re all running around thinking deep thoughts about everything we take part in online. Every community has its die-hards and equally every person has some things which they are passionate about participating in, and others which they are happy to be an observer or spectator of. That’s normal, that’s the power law distribution or long tail that says 90% of Wikipedia readers never edit, 9% have a dabble and 1% become dedicated contributors. All large-enough communities are like that, and that’s what drives these projects, not what marks them as failures. This is what Clay Shirky says in his book Here Comes Everybody, which I highly recommend reading to get a comprehensive understanding of how systems like Wikipedia work.

While Wikipedia serves as a great example of what is possible with mass collaboration, it can also harm potential projects that might take Wikipedia as its template, and expect similar outrageous success.

Taking Wikipedia today as a model for a new project is bound to end in tears. Firstly, the Wikipedia of today has almost ten years of existence baked into it. The Wikipedia of 2001 may be a better comparison, when a guideline that said “ignore all rules” was actually sincere advice and not a sad irony.

Secondly the success of Wikipedia is of course an outlier. The initial promise of Wikipedia was simply that you could add your contribution right away, not that you could help write the world’s biggest encyclopedia. So keep your promise proportional to reality.

Thirdly consider carefully your rhetorical model. What does your ideal contribution look like? It won’t look like Wikipedia’s, unless your project is also to create an encyclopedia. Wikipedia benefited from broad shared understanding of the idea of an encyclopedia. Many new mass collaboration models will have to struggle through creating their own rhetorical model, until such projects become mainstream.

Having said that, there are elements of Wikipedia that any mass collaboration project would do well to consider. The first is that of free content licenses.Wikipedia is licensed under the GNU Free Documentation License and also the Creative Commons Attribution ShareAlike license. These licenses are known as ‘copyleft’ — a “hack on copyright” that turns it against itself, and uses the law to tell the public that certain works can be re-used without seeking permission first. These licenses are a guarantee that the community will always be able to salvage its own work, should everything go haywire. It’s not that uncommon: funding dries up, and a host organisation no longer has the resources to devote to an experimental Web 2.0 project. These licenses also give the community a degree of autonomy, which it will need should a managerial decision not be to their liking. This forces a host organisation to be as honest and transparent as the community requires – the threat of a project “fork” means a host organisation keeps its monopoly because the community wants them to, not because they don’t have any other choice. For the same reasons, projects should use technology that is freely available, uses open file formats, and make managed data sets easily available, such as through “Export” functionality. In this respect I take my hat off to Google for its Data Liberation project, which aims to make it easier for users to move their data in or out of Google applications. If you’re using closed technology and copyright to hold your community hostage, chances are you’re doing it wrong.

The free content license also means I am not as concerned as you might expect about the prospect of Wikipedia imploding. I believe that Wikipedia as we know it, is not currently on a sustainable track. There are too many rules. The community is too unfriendly. It’s even a problem that it’s now too mainstream. While going mainstream is great for the impact of the project, it’s not great for recruiting new editors. Niches, fringes and outsider status are what make people passionate contributors, and Wikipedia as a whole now has none of these.

However, Wikipedia the product, as in what we have all written together, is guaranteed to outlive the current incarnation of Wikipedia the community or Wikipedia the project. Because of the free content license, anyone around the world who wants can download a copy of the entire thing and start a competing fork. Fork is a term from open source software, where anyone who is unhappy with how a software project is progressing, can take an entire copy of it and start working on it independently. While the right to fork is very important in open source development, communities work to avoid doing so where possible, because it divides efforts and loyalties.

Forking the entirety of Wikipedia is likely to be a pretty painful exercise because of its size, and any fork project is likely to be overwhelmed by bad edits before it can make any progress. But I see a missing piece of the puzzle as being the potential for part-forking. Wikipedia in its current incarnation is too reliant on a centralised mechanism for editing. We are missing some technology to let groups easily fork a small set of articles, and just edit those articles, and keep them synchronised. This would let narrowly focused groups take care of a subset of articles, and have them be just as readily available to readers, without the hassle of the centralised bureaucracy. There exists technology for actions like this in the open source software world – we can do it for code, with distributed version control, but we can’t yet do it easily for collections of prose. I think the distributed or decentralised wiki is what will breathe new life into Wikipedia.

Australian mass collaboration projects

Clay Shirky says something which we need to keep in mind when thinking about mass collaboration, which is that we get failure for free, and lots of it. What this means is that the barrier to starting a mass collaboration project is now so low, we can all afford to start dozens each day. We no longer need to evaluate if it will be worth the cost, because if you’re on the internet the cost is next to nothing. This means instead of only seeing projects that have been assessed by an organisation as being likely profitable — including administrative and managerial overheads — we can see nearly every project that’s ever popped into anyone’s head. The overheads have dropped to nearly nil. So what we get looks like an awful lot of failure, and it is. The difference is that previously all this failure never had a chance to get out of the starting blocks. But neither did the rare successes.

It is difficult to convince risk-averse management to take on experimental new projects, and even more so if you have to admit that the most likely outcome is failure. But we will have to find a way of making likely failure acceptable to traditional institutions. If this risk is too great for them to take part, they will need to consider the alternative risk of becoming obsolete by choosing not to engage in new ways at all.

Luckily in Australia we have many bold institutions, or more to the point bold individuals in the cultural and public sectors, who are dipping their toes or even a whole foot, into the 2.0 Web. I will mention just a few, to demonstrate the breadth and inventiveness of our experiments so far.

WikiNorthia is a wiki local history project coordinated by a few community libraries in Melbourne. The City of Melbourne put their 5 year plan into a wiki as part of public consultation, called FutureMelbourne. Founders and Survivors is an Australian Research Council project, tracking Tasmanian convicts and their descendents by combining detailed historical records with publicly contributed family history artefacts. The Powerhouse Museum in Sydney was the first Australian institution to take part in the Flickr Commons project, and engages with its online community in a consistent and considerate way. OpenAustralia is an open source project with no institutional backing that reworks Hansard records to make parliamentary proceedings more accessible and politicians more accountable.

Lastly I could not talk about mass collaboration today without mentioning the National Library’s Australian Newspapers Digitisation Program. After OCRing newspapers spanning some 150 years, Rose Holley and her team put their millions of pages and OCR text up on the web and invited the public to correct the texts. With little publicity, it has found a niche of people who enjoy correcting the text from old Australian newspapers. Thousands of users have collectively corrected millions of lines of text, and told the Library how addictive and interesting they found it. This project has really found a sweet spot for user contributions and I am already looking forward to hearing how it evolves in the future.

The range of projects being undertaken shows there is no shortage of good ideas and interested users in this country. If we keep up the same pace we may soon be showing the rest of the world how it’s done.

Where to now?

Mass collaboration offers a way for groups to organise without having an organisation. But organise to do what? Mass collaboration has been used very effectively to organise around single issues or single events, but it is yet to really cross over to the kind of sustained action that we so far only know how to do through lobby groups, politics, charities and the like.

A major part of the problem lies in crossing back from the online landscape into the political world we are so used to, where groups need to be incorporated to be taken seriously, political responsibilities may stop at an arbitrary border, and the law marches at a speed that could be considered glacial in internet time. These difficulties mean that mass collaboration projects struggle to be taken seriously and often don’t bother, preferring to stay in their sphere of influence — the online sphere.

Wikimedia Australia is such a group — we want to take the interesting, amazing things we’ve learned about mass collaboration offline, and help more sectors and more individuals take part in similar projects. Not only should everyone have equal access to use free cultural works, like Wikipedia, but just as importantly, everyone should have an equal opportunity to participate in creating them. We have certainly struggled at times to meet the expectations of meatspace, and so a challenge I see looming for us as a society, is: how can we better accommodate mass collaborative projects making changes in the real world? What organisational structures might we be able to support? How might they be accommodated in law, especially given their borderless nature? If we can find answers to these questions, then maybe we will see the potential of mass collaboration as a genuinely revolutionary force.

19 April, 2010 • ,

Comment [3]

Bug 23223: Developers (and WMF Staff) need way out of Europe

I thought I saw this on Planet Wikimedia, but it must have only been on Bug 23223 is a good read for those involved/interested in wmcon, the chapters, board and developers meeting just past in Berlin, Germany. As we all know, the Wikiepdias is srs biznis!

18 April, 2010 •


Coming up in April - ACEC & the NLA's Innovative Ideas Forum

I have a couple of interesting gigs coming up in April.

First is a Wikipedia editing workshop at the Australian Computers in Education Conference. I gave a talk called ‘Safe wiki’ at the last such conference, in Canberra in 2008. Now it is rolling around to Melbourne and I am doing something a bit more hands-on.

Registrations for ACEC are now open. You can also buy a ticket just for Thursday (to catch the open source stream!) or even just a half-day if you like.

Second is something quite exciting, the National Library of Australia’s Innovative Ideas Forum. It is a national event they hold, one day long, free, but only about 300 attendees. This year they are encouraging the use of Twitter throughout, and will probably podcast it, if last year is any indication. One of the other speakers is Nicholas Gruen, who was head of the Government 2.0 Taskforce. Esteemed company indeed! My talk is called “Is Wikipedia a one-off?: Is mass collaboration all it’s cracked up to be?

The NLA is a great host for such an event; it doesn’t surprise me that they are encouraging big-thinking and innovation in technology. Their Australian Newspapers site is not only a fine example of making dead-wood products machine readable and comfortably at home amongst the electrons, but it also integrates ‘crowdsourcing’ (OCR correction) in a very natural and, from what I understand, successful manner.

I hear places are filling up quickly so if you are in Canberra, register soon!

10 March, 2010 • ,


Charles Matthews: Backing Limited Perspectives

This is a guest post by Charles Matthews. See also his previous posts. —Brianna

That is not what BLP stands for on Wikipedia, though you might sometimes wonder. This now-notorious three-letter acronym stands for Biography of Living Person. Wikipedia hosts several hundred thousand of them, and the summary deletion of a number of those has recently caused consternation and recrimination, not to speak of admin-on-admin disrespectfulness, of a kind that hasn’t been seen for, oh, all of several years.

BLPs are troublesome because in real world terms they affect lives, in legal terms they may be defamatory, and in Wikipedia terms content policies must be applied very strictly, and still may give poor results. But they predominate among biographies: there is a decay law saying if you go back a decade by birth date the number of biographies for that year of birth drops off by a factor (could be something like 20% or 30%) and this is quite marked as you get back to 1900 and before. Around 1983 is the peak (over 8000), which tells us what? Duh, sport (Finnish speedway stars, anyone?) and popular culture. Editors add but do not necessarily maintain well numerous articles about young stars aged 27 or so who are mentioned in the media.

Back among the grumpy folk known as “old school Wikipedians” the term “MySpace page” may occasionally pass the lips, but surprisingly, perhaps, there is a classic old-style inclusionist argument that works the other way. In a polite form it reads “if you come across work of others on the site that is substandard, your first task is to try to improve it, before cutting it or sending it for deletion”. In the matter of BLPs substandard means just one thing: references absent or low-grade. Wikipedia shouldn’t post things about real people out there that are just made up. We all agree. So, an editor finding a substandard BLP should try to reference it better.

Nice theory. BLPs are speedy-deleted by the thousand as newly-posted pages when unreferenced, sometimes quite wrongly, because as posted they don’t have the references needed to support them (no verifiability and/or no convincing reason to support notability). The drama has come up when the same criteria, or stricter, have been applied to articles dormant on the site for years: unreferenced BLPs that seem not to be going anywhere better.

So what is the “old school” counter-view? ‘Wikipedia has no fixed rules’ is part of the old-time mantra called the ‘five pillars’. Which allows for tectonic shifts in how things are done. Some anti-BLP activism has homed in on the broken nature of incremental change in dealing with the issue. Some BLPs are inherently problematic, functioning only as places of wars between supporters and denigrators of a real person (I have to babysit three of those). As Wikipedia expands, it gets into the area of biographies that are not that easy to reference. And such tenuous biographies may just have to remain, as things stand, because the inclusionist view amounts to saying that you are obliged to nurture them. And indeed better references or a controversy may turn up tomorrow: I started Ruth Padel never dreaming she’d be in the news so prominently. I read a history book by David Gress not knowing he was going to appoint himself to climate change controversy.

So what has happened? The limited perspective that there is no real lower threshold for biography on Wikipedia has created another limited perspective, that only a radical cull and shift to a seriously summary deletionist policy on BLPs can save Wikipedia from a future as a morass of neglected gossip about real people. Some demonstrative admin actions on the site have brought the matter to the top of the agenda. These things get messy and costly in human terms, but the logjam gets broken along with the eggs for the omelette, and the real losers are peaceful editors who detest mixed metaphors. No, this is serious stuff, but the lurching motion is unfamiliar to those who haven’t seen Wikipedia in this mood.

25 January, 2010 • , ,


Information philanthropy

I learned an interesting term today, while reading the draft Government 3.0 report.

From the draft report:

11.5: Gifts of public good – Information Philanthropy

Innovation often occurs well in advance of the regulatory and cultural frameworks needed to support it. Many of the most innovative endeavours have been made by people with an idea, some time to volunteer and the wherewithal to make it happen.

For the many innovations that have social and democratic value but no apparent commercial return there are currently few options. Funding through government grants is unlikely, micro-donations and online advertising will rarely cover any substantial costs and the current philanthropic framework does not support substantial giving to such projects.

In the UK and the US examples such as and demonstrate the potential for social good. In Australia initiatives such as OpenAustralia and the Taskforce’s mashup competition and associated hack day events are clear examples of the potential and appetite to innovate with data and online engagement.

It may be possible for organisations whose purpose is to build online systems for public good to receive Deductible Gift Recipient (DGR) and Tax Concession Charity (TCC) status for organisations but it is far from straightforward. There are no categories that specifically support the provision of public goods online in the Australian Taxation Office (ATO) regulatory definitions. DGR and TCC status provides both tax advantages for the organisation and the capacity to receive grants and donations from philanthropic foundations and other donors.

This is not surprising as Information Philanthropy is new and is not widely understood. Reducing the obstacles to the free flow of philanthropy to projects that use government data for public good, or improve the democratic process will no doubt boost innovation and expand the understanding of the value of such projects.[…]

The consultants to the Taskforce have proposed the establishment of such a Specially Listed Deductible Giving Recipient Foundation to support the initial development of info-philanthropy. For the purposes of establishing this, one might define the foundation’s mission as assisting in projects of properly registered not-for-profit organisations and which, in a way that is not party political or focused primarily on advocacy either:


Recommendation 12 – Encourage info-philanthropy

Because some of the most successful experiments in Government 2.0 have been fuelled by not-for-profits in leading countries such as the UK and the US, Australian policy-makers should minimise obstacles to info-philanthropy being treated as an eligible activity to qualify for deductible gift recipient and other forms of legal status which recognise charitable or philanthropic purposes.

(Emphasis mine.) In my opinion that’s a brilliant idea. I don’t really have a lot to add, I just wanted to point it out to everyone. I hope this recommendation is picked up by government. It would be an easy one to implement and make a start towards useful formal recognition of the good that such projects can do.

I’ve only quickly skimmed the report – it’s kinda huge – but I already found a few other nice nuggets. Like The Three Laws of Open Government Data:

The Three Laws of Open Government Data:

  1. If it can’t be spidered or indexed, it doesn’t exist
  2. If it isn’t available in open and machine readable format, it can’t engage
  3. If a legal framework doesn’t allow it to be repurposed, it doesn’t empower

Summarisable as Find, play, share. I like it!

There are also nice summaries on OpenAustralia, and the Social Benefits of PSI talks about the National Library of Australia newspaper digitisation program.

There’s lots to absorb here, but that’s what caught my eye in a first pass.

07 December, 2009 • , , ,

Older Newer

list of all posts, ever

find articles by tag

monthly archive

most popular articles

  1. [guest] Rethinking the Top Ten
  2. An alternative term for "User-generated content"
  3. How to use Gmail to manage high-traffic mailing lists
  4. NLA Innovative Ideas Forum audio/video now available
  5. Write API enabled on Wikimedia sites!
  6. Free as in Freedom miniconf recap + slides
  7. Top 10 software extensions Wikimedia Commons needs in 2008
  8. Templatology, an essay
  9. Is mass collaboration all it's cracked up to be?
  10. GLAM-WIKI, day one

(from the last 30 days)