The Copenhagen Project

Citizen journalism and user involvement is just two buzz words from the last five years. Perhaps our media model can help us look through these concepts and see what unexplored possibilities there’s yet to try out - we’re building the model not to make boring analyses, but to help us become creative :)

Two examples of user involvement

OhMyNews, USA Today and The New York Times share the common theme of being oriented toward society and politics. They thereby take on the same role/identity as most newspapers have done throughout their existence. The three sites are structurally highly similar, but the differences are interesting.

All three are run as a business. They all feature original content, are all traditionally prioritized and edited centrally by humans - with USA Today having some mixed features. All three take advantage of the Internet and update their sites continuously. Besides from the continuous update, The New York Times online is more or less a straight copy of the paper version, both in terms of structure and presentation.

The main feature of OhMyNews is of course the user contributed writings, a concept that has received a lot of hype through the years. Our model is open for users “providing” the object while they haven’t got the possibility to also edit and prioritize. In the case of OhMyNews, the users doesn’t edit - it is editors who select and prioritize the content.

Looking at USA Today the situation is opposite - here the editors provide the objects while users can vote (”recommend”) on items. The prioritization is not completely user-based, as an algorithm seems to blend votes with how long time the content has been online. Editors also seem to influence the list, but the concept of opening the site for user editing as opposed to user contribution is new and interesting.

In my personal opinion, user editing should be much more widely considered where appropriate. Currently this feature is mostly seen at sites as social bookmarking services and weblogs, where the users literally edits the Internet. WikiNews is another news site that features both user contributed articles and user editing.

The real value of user contribution might not be content but knowledge about content

Understandably The New York Times really can’t open up for much user contribution of content and editing - their readers expect a high quality product they can trust. But what could be interesting and actually might work very well in terms of user contribution, was if The New York Times opened up for user contributed context in e.g. a Digg-like system connected to each article separately.

It would be both useful and engaging if readers alongside each article could contribute links to background information, different perspectives, maps, blog posts, Wikipedia entries etc. With many million readers a day, each article at The New York Times would probably quickly be “tagged” with a lot of interesting perspectives, and users voting on these links to context would ensure that the most interesting background information was featured first.

Currently context at most news sites, including The New York Times, is limited to earlier articles from the same media. This is probably due to the factor of news sites not wishing to direct readers to other sites and thereby loosing out on advertising revenue, but I guess the biggest factor is the time it takes to find really good context. Journalists simply haven’t got the time to digg through search engines and sort through all the relevant material and subsequently decide what to feature alongside the article itself.

While journalists in most cases are the first in the know and has the best skills to convey information, millions of users together have a much better overview of background information already available than a single journalist does. Giving the users the option to contribute valuable background information would be like giving them the job they are best at, and it would add a whole new explorative dimension to news, especially if it is combined with highly evolved conversation tools, as e.g. Digg’s commenting-system.

Citizens journalism is hard to implement and it’s real value is still doubtful. It would be an interesting experiment to let journalists do what they are best at, while tapping into the real useful resources that users collectively hold: A great overview and a lot of knowledge about what valuable content there’s out there.

Whereas Google News and Techmeme shared the fundamental characteristic of beeing edited by an algorithm, Boing Boing and MetaFilter shares the wide definition of beeing weblogs. But how well does the definition of a weblog hold up when we apply our model to the sites?

As can be seen from the above table, there are two fundamentally differences between the two sites. First of all, Boing Boing is edited by a small number of editors, whereas MetaFilter is edited by anyone who bothers to participate. That’s a pretty huge difference in terms of site dynamics. Secondly, but closely related to the first one, not only does users at MetaFilter edit the site, they also provide the original content.

In that way MetaFilter’s media type is the most open that our model can represent – everything is decided by users, including layout changes, banning of trolls, content etc. Boing Boing’s media type on the other hand shares the common characteristic of medias before the Internet age, as there is an editorial board that produces a final product.

Why, then, does Boing Boing and MetaFilter still feel very much alike? First of all it’s in the presentation of information. The content at both sites isn’t prioritized, whereas content in a traditional media type - e.g. The New York Times – is. And then there’s the social aspects. Both Boing Boing and MetaFilter allows social interaction around their content. So if we look at the aspects of presentation and social interaction, Boing Boing and MetaFilter is identical.

But if we look at the editorial structures, we see that Boing Boing and MetaFilter is fundamentally different. Interestingly we can also see that the editorial structures of Boing Boing and The New York Times are identical. They are both based on a central editorial board that produces the content and edits it into a final product.

Comparing MetaFilter and The New York Times we see that they are very different from each other, whereas Boing Boing is kind of “in the middle”.

If the New York Times decided to omit it’s prioritized frontpage and allow commenting, their media type would be the same as Boing Boing’s - it would be a standard weblog. If it wanted to be as MetaFilter they would have to go much further than that and fire all it’s journalists and editors and give the users the power. I’m not in any way saying that The New York Times should do any of this. I’m just suggesting that editorial changes and differences according to our model might be more fundamental than those regarding presentation.

If we regard Boing Boing as a standard weblog, how much of a weblog is MetaFilter and The New York Times then? In terms of social interaction, MetaFilter is a weblog whereas The New York Times isn’t. But if we look at the editorial structures, we see that The New York Times is more of an weblog than MetaFilter is.

Conclusion: A weblog isn’t just a weblog, and a newspaper site isn’t just a newspaper site. It’s much more complex than that.

Techmeme and Google News shares an interesting common characteristic: They are both entirely edited by algorithms that surveil a predefined list of other medias to determine what’s currently ”hot news”, and it uses this information to construct it’s front-page. The two sites are virtually functioning in the same way with one other important difference: Techmeme isn’t prioritized whereas Google News is, and Techmeme therefore ‘feels’ more blog-like.

Notice that out model holds these two sites as having the property of ”Editors” (Or as I called it originally: Centralized editing). It could maybe seem as if this should be ”voting” or ”decentralized editing” as the sites draws on data from a great number of sites – but the thought behind this property is not to tell on basis of which data the editing is decided, but where ”the power” in the editorial decisions resides. Bloggers that are included in the list of Techmeme’s sites aren’t consciously trying to influence the editing process of Techmeme, neither are the list of medias that Google News use. Users don’t have the power to tweak the algorithms behind the sites and therefore can’t influence what’s on the front-page in the same way users of e.g. Digg can do.

We might have to change the name of this property to make it more obvious what it’s all about – as e.g. voting isn’t the only way that users can participate directly in editing of a site. Structurally Techmeme and Google News are very much alike. What remains is the differences in the topical focus of the sites.

The only fundamental difference that’s left is the possibility for users to customize their Google News front-page – an aspect our model doesn’t yet take into account. We might add the properties ”Customizable content” vs. ”Fixed content” to the model.

Digg and del.icio.us looks and feels somewhat alike and are definitely very different from e.g. The New York Times. Let’s try to compare them, looking at the front-page of each service respectively.

They do indeed have a lot in common, but also differ in at least three crucial points.

First, whereas Digg is very much build and staged as a place for conversation and debate, del.icio.us has no build-in conversation functionality. While Digg is a community with interaction, del.icio.us is much more a product of it’s fundamental staging as a personal tool, where conversation isn’t possible.

Secondly, whereas the basic idea of Digg is to actively engage people in a process of editing a front page, the basic idea of del.icio.us is to enable it’s users to easily file their bookmarks. The del.icio.us-front-page is therefore rather different from Digg - it’s users aren’t consciously thinking in terms of editing. While Digg’s front page is constantly influenced by voters, the del.icio.us front-page is a map of user activity at it’s primary service. This is reflected in the type of material found on the front pages; digg.com’s objects are often characterized by being “news-worthy”, whereas del.icio.us’s objects usually are less timely, and a great deal of the objects are reference material.

Thirdly, Digg utilizes human editors (the users), whereas del.icio.us uses an algorithm to decide the content of it’s front page. It seems to be a rather simple algorithm, but it’s there. It’s impossible to say how it’s tweaked though.

So there you have it: From a media perspective Digg and del.icio.us differs in the areas of social interaction (conversation/no conversation), in the type of editing (editors/voting) and which type of editor that edits (human/algorithm).

The summer is coming to an end here in Copenhagen, and so it’s time for this project to get up to speed. During the last weeks Thomas and I have been discussing what we’re doing here, and we are now making some adjustments to our focus and the process. I will just explain what we have decided.

Our goal here is to push through some creative barriers that seems to exist after media has been moving from the physical world into the Internet. We are gonna do that by breaking down existing medias, offline and online, into the pieces it consists of.

An example of such a ‘piece’ could be ‘prioritized news’. The New York Times and most other newspapers and magazines is a good example of that. Digg.com’s front page on the other hand is just an expression of what stories made it above the front page requirements over time, just as this blog is non-prioritized. Even in conversations people usually decides to prioritize what they are talking about. So here we have two interesting elements that characterize all media: ‘prioritized’ or ‘unprioritized’.

Other examples is ‘participation’ vs. ‘consumption’, but also more descriptive elements such as ‘depth and detail’. By breaking down all media, we’ll end up with a lot of ‘elements’, ‘bricks’, ‘parts’, ‘perspectives’ or whatever we’ll decide to call it.

In the end we’ll be able to make up new media types in minutes by just randomly (and of course also purposely) combining the different bricks. A lot of interesting questions will emerge, e.g. how would digg.com be different if there was a daily, prioritized digest of the last 24 hours front page stories? Which new elements then come into play? Another example: what if The New York Time’s editorial board was extended to consist of all it’s readers, but all other elements was left unchanged?

We’ll use this blog in a wiki-like fashion, blogging about each different element with an explanation of what it is, how we see it relate to other elements etc. We aim for a non-structured, tag-like organization of the ‘bricks’, as many of them are inter-related in ways that makes it difficult to classify them.

The blog posts will be tweaked as we go along, so some of the posts may seem very short in the beginning and then evolve over time. Also, we’ll coin ‘bricks’ that may show to be to general, to vague or irrelevant, or there may be better words to describe what we mean. We therefore invite you to comment and discuss each post and to suggest other wordings, better explanations etc. during the process.

The final product will be a free and open catalog that hopefully will inspire people to experiment with new models, new concepts and new ways of improving this thing we call media.