• Follow Curiously Persistent on WordPress.com
  • About the blog

    This is the personal blog of Simon Kendrick and covers my interests in media, technology and popular culture. All opinions expressed are my own and may not be representative of past or present employers
  • Subscribe

  • Meta

Can our opinions exist without us?

This article from Jeff Jarvis got me thinking about the evolution of content and opinion over time. Extrapolating past patterns could lead to some bizarre scenarios in future.

(NOTE: the remainder of this blog post is incoherent speculation).

Broadly, the past evolution of storytelling roughly covers four ages

  • Oral stage
  • Hand-written stage
  • Printed stage
  • Multimedia stage

Within this, there have been many trends and patterns in the types of content, the means of production and the methods of consumption.

Many arguments focus on the dumbing down of culture. But instead of rehashing that ground, the article got me thinking about content length

I’m no historian but the following spring to mind:

  • Oral accounts would take pretty long to recount and spread
  • Hand-writing/scribing has similar scalability issues
  • The printing press achieves scalability but also encourages verbosity
  • Newspapers and magazines encourage serialisation and the consumption of articles rather than full-length tracts (I suppose pamphlets come under this heading)
  • The computer age propagates articles, blog posts and shorter-form content
  • Social media reducing creation and consumption time down further – currently at 140 characters

How is this extrapolated further? Two scenarios – one logical progression and one step-change – come to mind


A logical extension would be to reduce opinion down to its underlying sentiment – why use 140 characters when a single word or gesture will do (thumbs ups, retweets etc all fulfil this function, but alongside other forms of opinion)

It is conceivable that a social media service in future could be a single spectrum of opinion, from like to dislike. Links, names, words etc could be placed on that spectrum. Our contacts would take something we like as a recommendation and consume it, and avoid things we dislike.

Would this work? Probably not, since it has no nuance. It would further encourage the balkanisation of online opinion and, even with a potential velocity measure to capture trajectory of opinion, it would make it difficult for new content to rise upwards.


As the shortening of opinion can’t evolve beyond a single word, an obvious revolution would be to move from active to passive.

In other words, once I input parameters or some past behaviour, a service can automatically generate my sentiment to new content that crosses my digital path. With refinement over time, this would become more accurate.

We already have digividuals, based on composites of others. Could we have digi-extensions? Possible, but again unlikely. But it raises some interesting questions about the nature of digital personas. Once my online persona starts acting independently, does it still fully represent my real-world self? If someone died, their digital persona could continue to exist without them although it would cease to evolve.

If you’ve read this far down, then congratulations. This post doesn’t really have a point, or any obvious application, but I wanted to write this down to help formalise my speculation (my thoughts on this were even more jumbled before I started writing). And, on the off-chance that something similar happens around the time of the singularity, then I can go the wayback machine and glow over a small victory.



Promoting quality over quantity

“Findability” (Sidepoint: I’m fascinated by neologisms – my current favourite is “winningest”) regularly crops up in research I conduct, particularly for video services. To put it briefly, it’s crucial.

It appears to me that over the past few months, my blog has become less “findable” – at least on one measure.

This blog hasn’t been as active over the past few months (this will improve from October, when my Diploma and much-needed holiday will be out-of-the-way). As a result of my reduced frequency in posting (and, to be fair, I’ve also spent less time thinking about my posts), I’ve slipped down the AdAge Power 150 rankings a fair bit.

This isn’t a post filled with whining navel-gazing and self-reflection. Please bear with me.

It’s inevitable that a less active blog will drop down the rankings – recency and velocity/momentum are important determinants when considering popularity.

But it did get me thinking about how popularity works.

It is correct that something with high numbers ranks highly (though, arguably, the most popular things tend to be the lowest common denominator consensus choices, rather than things that inspire devotion).

But numbers can be misleading. All readers, viewers or followers are not equal. Anil Dash has made this point very well in terms of Twitter followers.

Reasons for popularity can include

  1. Something is genuinely good – despite my cynicism, the good will out, at least on occasion. I’ve recently started reading (and enjoying) Inspector Insight, while Inception and Arcade Fire provide two cultural examples (note the subjectivity), though Arcade Fire’s willingness to experiment with interactivity helps their PR
  2. Frequency – it’s not a coincidence that the top blogs on the AdAge ranking are updated daily. Volume is a major determinant of popularity – indeed many of the other factors on this list can be considered functions of volume. I believe reduced volume is the reason for my drop in the rankings
  3. Differing motivations behind “link love” – volume of links are a signifier of popularity, but the reasons for linking differ. It could be a genuine desire to share, a reciprocal back-scratching activity, or a ploy to garner the attention of someone. My link updates (which will return later this week) have been guilty of all three, though nowadays it is almost exclusively the former
  4. Intensive distribution/self-promotion – of course, it could be one person linking to themselves across a variety of platforms. Ray Poynter recently started a debate on LinkedIn regarding the multiple linking to blog posts on both that group and several others (there was one particularly egregious offender). The cost of doing this (and the cost of annoying regular visitors) is negligible compared to the benefits. The spam principle.
  5. Gaming visibility – SEO has become something of a dark art, with multiple sources offering tips on improving the volume of traffic. Attracting “Junk” visitors can either be intentional or unintentional. For instance, the total number of visits I get on my blog is heavily affected by tweaks to Google’s algorithm, since by far the most popular post on this blog is my review of a Thinkbox event. Not because of the content, but because of the (presumably copyrighted) picture of the brain that the post contains. “Brain” and derivatives thereof far outstrip other search terms (such as my name)
  6. Gaming views – some tech blogs have taken it upon themselves to auto-refresh, thus grossly distorting the page view count. Similarly, other blogs will spread an article over several pages to inflate numbers
  7. First-mover advantage – Google may not have been the first search engine, and Facebook may not have been the first social network, but it generally helps to get in there early. Robert Scoble has partly cultivated his micro-celebrity around being an early adopter of new services, though he gets supplanted once the macro-celebrities arrive. Frontiersmen and women are able to build up their networks early, and this leads to…
  8. Self-perpetuation – there are power laws where people gravitate towards larger numbers. This Rapleaf study shows the distortion in distribution of Twitter followers – this trend would have been exacerbated by the introduction of “suggested users”. Again, this self-perpetuation can be intentional or unintentional. Initiatives such as SXSW 2011 Panel Picker and Fast Company Influence Project become number chases, where the only goal is to get as many votes as possible, irrespective of their provenance or context. Whereas, a market researcher new to Twitter might gravitate towards Tom H.C. Anderson – there must be a reason he has over 50,000 followers. And while there would have been a good reason for Tom to gain popularity initially, this has been surpassed by the self-perpetuating power law.

These last two factors are particularly concerning. They are completely divorced from the quality of the content, and create barriers.

  • An artificial glass ceiling that makes it hard for others to break through. It is a pyramid structure – something Ben Kay has talked about in relation to advertising agencies.
  • An echo chamber since the community effectively becomes closed off. In an attention economy, time is scarce and popular sources become stickier. I’ve written previously about the perils of balkanisation and echo chambers.

Is there a remedy to this? Can something be done to ensure that the good does break through?

Probably not, since rankings are completely subjective. They depend on relevance and context – two factors that are unique to each individual.

But potential solutions could include

  1. Incorporating popularity per piece of content – rather than overall volume of links or viewers, it could be done on a content by content basis. For instance, Avinash Kaushik only blogs once a fortnight, but his post are of a very high standard and thus I’m sure he has a very high number of links per post
  2. Relying on a curator’s subjectivity – the AdAge Power 150 does include the subjective Todd Points but I have no idea who Todd is, nor what his tastes are. Recommendations are far better when they come from someone trusted
  3. Algorithms – though recommendations can also come from an algorithm. Last.fm looks at taste compatibility between users, while Amazon is able to suggest items based on the patterns of other shoppers. Of course, this requires a centralised body with huge swathes of data to be effective
  4. Self-administered ratings – WordPress has a new feature enabling you to “like” posts, while other blogs allow you to rate on a five point scale – linking these to profiles could provide a framework to produce basic recommendations. The former might be slightly more effective due to the polarisation of opinion when rating online, but it is a challenge to incorporate. It requires a change in behaviour and mindset to be more active in providing feedback. For instance, I’ve never rated a blog and my Google Reader trends say “over the last 30 days you read 3,028 items, clicked 449 items, starred 0 items, shared 0 items, and emailed 0 items”. Perhaps I should start doing so, but again this method is open to gaming.
  5. Awards – I dislike award ceremonies, particularly ones that require payment to enter, since they seem to be more about making money than rewarding success (hence the fight over the Press Gazette awards, but not the magazine itself). Nevertheless, they provide an objective signifier of achievement – witness the recent episode of Mad Men, where Don and Roger talk about how a Clio award would be good for the company profile.

There won’t be a perfect solution, but there must be tools that can enhance the “findability” of relevant information online. Greater diversity in what we read and consider can only enhance the discourse, even if it will require some complex mental calculations regarding what to consume and what not to. Because the attention economy is virtually a zero-sum game – we’ve almost reached our limits and so new consumption sources will replace, rather than complement, existing ones.


NB: Yes, I have committed many of the cardinal sins in artificially inflating popularity within this post. Well spotted.

Image credit: http://www.flickr.com/photos/joost-ijmuiden/4485190116

Enhanced by Zemanta