Social Software

The Power Law, iCan and Weblogs…

So the BBC has launched a public pre-beta version of iCan – their attempt to help people re-engage with and self-organise around politics at an local/national/international level. The early stage makes it almost inevitable that there are going to be bugs – and bugs there are – but the idea seems sound and there’s considerable scope for iteration toward something profoundly useful and important. The non-beta version is a way off yet, but I can recommend that anyone who’s interested in starting a campaign or getting engaged around an issue try the site out and feedback into the development cycle. There’s a good few issues starting to populate the site already – including Kevin Marks’ campaign to get the BBC releasing audio as MP3 rather than Real, which is obviously interesting to me (working in R&D at BBC Radio and Music Interactive) but which – also obviously – I can’t really comment upon in the public sphere…

Matt Jones – who has worked on the project for the last couple of years – has written a post (It’s all about the tail) which tries to articulate some of the rationale for the endeavour, based on that old power law chestnut. This time the power law graph has “issues” on one axis and (I suppose) “amount of coverage” or “scale of political engagement” on the other. A tiny amount of national issues get massive amounts of coverage/engagement while a massive amount of smaller local issues get little or none.

The aims of iCan in this space seem to be two-fold – firstly that the tail should get ‘fatter’ ie. that there should be a way to encourage people to engage in the smaller, lower-rent difference-making. Secondly there’s an aspiration towards mobility – that smaller campaigns should be supported in their attempts to get larger, to transition into different scales of activity and to grow.

This latter objective does seem to come with some interesting provisos, however. There are some issues which are by necessity localised, there are some campaigns which will never and should never become national news or motivate hundreds of thousands of people. In a sense, then, iCan is about finding the best place on the power law for a campaign to live. It’s about facilitating the scale and type of engagement that will do the most good for people based upon the kind of issue that they bring to the table.

All of which brings me right back to the issue of weblogs (again), back to Clay Shirky’s article (again) and back to this issue of ‘inequality’. In the case of iCan, there seems to be an acceptance that there’s a difference of type between kinds of campaigns and that certain types will sit at different levels of the power curve. So of course the question arises, is there a difference of type between the weblogs at different points in their analogous curve and what does that mean for weblog inequality.

Any examination of the ‘top-linked’ weblogs brings you to the conclusion fairly quickly that they’re either highly subject-focused or totally subject-focused. Several of them are group weblogs as well. They are almost totally non-conversational. At the bottom end are the small focused, highly conversational clumps of weblogs used almost as mailing-list/group e-mail equivalents for friends, familes and small groups of people. This isn’t a question of quality – the latter type has no aspiration towards massive traffic and web-popularity, while the former model has aspirations towards a publishing model and a larger ‘mass market’.

That’s not to say of course that a conversational weblog will remain conversational or small and downplayed (any more than to say that all human beings are uniformly socially popular) or that the publishing model weblog will necessarily achieve Time magazine levels of success, but simply that the inherent qualities of each type of site make them ideally suited to different points in the power-law – that there are different kinds of interaction which work better at different scales.

In between of course, there are many variants of tone, personality and conversationalism – hybrids designed to operate at different ‘depths’ – some by choice or aspiration, some forced into new forms of interaction by dint of new forms of pressure. Any long-term weblogger is familiar with the changes in tone that come with the arrival into your online social scene of people from your real-world who you didn’t expect to read your writing – and many are equally familiar with the sensation that too much of your life is on display for the benefit of strangers. A trafficked personal site necessarily becomes less personal – more of a publishing style site eventually, as the author is slowly eroded by revealing themselves totally in the public sphere – just as the local campaign becomes less homely and more structured as it extends to the county or national level.

Which leaves us where? My argument would be the fairly obvious one that – in order to create a fair and useful (equal) space within which webloggers can operate, we should be thinking about how to build tools and mechanisms that will encourage movement along the arc of the power-law, helping sites responsively find a level of traffic and engagement that reflects what individuals are trying to achieve, and that we should find new ways (maybe new kinds of weblogs themselves) that help articulate what kind of activity a weblogger is aspiring towards, and help them move in those directions. The level of engagement that has been demonstrated by individual webloggers has clearly been one of many inspirations for iCan – now perhaps it’s our turn to be inspired in turn?

10 replies on “The Power Law, iCan and Weblogs…”

Hi Tom, I’m not sure about your conclusions on ‘A-List’ blogs compared to ‘C-List’ (I know you didn’t use those terms, but to make things easier I’ll use them). You say that the A-List are or aspire to be “highly subject-focused or totally subject-focused”, whereas C-List are “small focused, highly conversational”. These are the same assumptions that Clay Shirky et al make and which prompted this post from me.
What about those weblogs that are highly topic-focused but have only a few readers? Or what about someone like Dave Winer, whose weblog is very conversational in tone but his is an A-List blog? My contention all along has been that the *type* (or “mode” as Clay Shirky calls it) of weblog one writes has nothing to do with quantity of readers. Maybe on average there is the pattern which you describe – A-List = broadcast, non-conversational; C-List = conversational. But I somehow think this kind of compartmentalizing goes against the grain of what the Web stands for – that it’s a big open universe where you can publish what you like and how you like. I don’t identify myself as an “online diarist” (do you?)… I don’t understand the need to typecast weblogs according to size of audience.
Anyway I’m rambling again…not sure why this whole power law thing is winding me up so 🙂 I’ll think some more on it and write another post. But would be interested in yours and others thoughts.

I’ll be honest. There are elements to my argument here that make me uncomfortable too. I think there’s something profoundly troubling about describing disparaties in traffic to something almost ‘natural’. It feels almost like a way of sanctioning prejudice and shooring up the position of one group against another. And there’s no denying that there are a great many good sites that deserve consider traffic that aren’t getting it – and that things like preferential linkage are making it increasingly hard for new webloggers to get seen and read. And you’re not going to get any argument from me that this latter situation needs redress.
Having said all that, and accepting that we have to be careful not to belittle or diminish the actual issues that exist, I still think there are some perceptible trends in the way people use weblogs and how content and audiences interact with one another on weblogs. It seems absolutely clear to me – after four years of experience – that one major trend with webloggers generally is for their personal conversational writing to diminish with time and/or with traffic. People talk towards an audience – even when they expect that audience to be one person – and I think it’s clear that people are prepared to expose themselves (on the whole) much more to small groups of people in perceived privacy than they would be to the web as a whole. Similarly, single issue weblogs are more likely to have focused audiences – for one reason, there aren’t as many of them because they’re harder to maintain. I think for both those reasons you’d expect the low-trafficked end to be more conversational and the high-trafficked end to be more publishing oriented. And that’s precisely what you do find – when you put your eyes slightly out of focus, at least, and look at the larger trend rather than the specific instances…
The specific force of this post has been about trying to develop the mechanisms that can help people get what they want out of weblogging by helping to demonstrate that different people have different objectives that might be fulfilled in different ways. It’s not about A-list to C-list, because that presents the whole thing as a competition, which I don’t think is true of most of those people who don’t sit in the highly connected cluster in the middle of weblogging, and also because I’m very specifically talking about facilitating movement and how weblogs might choose to, or be forced to change and adapt themselves over time.
I don’t know if that’s any clearer or not…

Yes that is clearer now, thanks. The point you make about webloggers becoming less conversational the larger their audience becomes, or when people they know start reading their blogs, is a very interesting point that I hadn’t considered fully before. And I agree that if we look at “the larger trend”, then these theories of ‘broadcast vs conversational’ start to make sense.
Thanks for your feedback – shows the value of “conversations” eh 😉

So what are we doing here; drawing an analogy between the power-law curve of small-campaign news coverage and of small-weblog traffic? And asking the question: how can we ensure greater efficacy for the bottom 90% of weblogs?
One idea: ideological aggregation of some sort? Whereas iCan is an umbrella and facilitor for action, connection and promotion, might a similar umbrella group together – using metadata or textual analysis – posts from hitherto unique individual weblogs that regard a shared subject or interest?
That is, the end result might not be a “group weblog” in the common sense of the term but, by translocating multiple alike posts in one large, popular, open and highly visible space, effectively becomes so – and gives the writers in the invisible bottom 90% a leg-up into visibility, · la iCan.
Or is this method not already employed in existing blog filters etc?
Ultimately, however, I really don’t think iCan is modelled largely on inter-blogger interactions, as you claim, but rather on the other, far-better uses of web technology already used to mobilise issue-led action and connection. In contrast, the blogosphere appears far less effective and far more a vacuum space than, say, FaxYourMP or any of the petrol-price sites from a couple of years back, which achieve real-world agency…
Additionally, I don’t agree that the weblog is a great conversational medium. Rather, it is primarily a publishing medium and, actually, a pretty good one at that. A conversation is a civil interaction between participants, proceeding either on an equal, mutual statement-response basis or according to the most effective conversationalist – an attention economy.
A weblog “conversation” is always started, controlled and ended by the deitified blogger and all conversational content and context are derived from the worship of said host on said page. I prefer conversations in which I have the opportunity to have more fluid interactions, uninterupted by the dictative rhythms of a conversation-keeper.

I disagree about the conversational aspect, and I think if you looked at Livejournal more thoroughly you’d see a much more substantial level of interconnectedness and conversation. As I said – I’m not talking about weblogs at the level of comments on a site, but on intra-site discussion…

iCan, too
Following my previous post on BBCi’s iCan beta service, Tom Coates’ has, as usual, a far better take on it than pretty much anyone else (although out of generosity he links to them anyway)….

Fatten not Flatten
Tom Coates of Plasticbag writes yet another interesting piece, this time on the BBC’s iCan project (I wonder what influenced their choice of names?). iCan is essentially a network building tool with an activist’s bent. It lets users find people,…

Culture of Celebrity and Weblogs
I judge the quality of a weblog by its IDEAS, but it seems some people equate quality with popularity. Is the ‘culture of celebrity’ that afflicts Western movies, television and radio creeping in to weblogs as well? These thoughts were…

Comments are closed.