The failure of the wisdom of crowds

James Surowiecki’s The Wisdom of Crowds argues that across a large and diverse group, the average response will be better and smarter than individual experts. He illustrates this point with the jellybean answer. In a large room of people, few will get close to guessing the correct number of jellybeans in a jar. But the average of their collective responses will be remarkably close to the true number.

A photography exhibition at The Brooklyn Museum entitled Click experimented with curating through crowd-sourcing.

The exhibition was a critical failure. Crowdsourcing works when there is a quantifiable number. When it is subjective opinion on what makes a good photo, opinion congeals into lowest common denominator crowd-pleasers. (Does Hollywood operate via crowdsourcing?)

To quote the Slate article:

Ultimately, “Click!” demonstrates that people—whether they’re experts or laymen—like pictures that remind them of things they’ve seen before.

Curators need to look forward – to know what’s been done before; to recognize exhausted styles and idioms; and to select art that confounds, surprises, and provokes

Which makes sense. In my experience, design by committee descends into a bureaucratic nightmare – appeasing everyone by pleasing no-one.

Long may the power of the auteur continue.

Hat tip to Mintel Alerts for bringing this to my attention.

sk

Photo credit: http://www.flickr.com/photos/vividbreeze/

Advertisement

Ask questions

Simplifying somewhat, there are two main ways to respond to a research brief

• Answer all issues raised point by point in a methodical and thorough way
• Question the brief, outline the options and problems and offer a starting point for discussion

The former will be successful where the client knows the answers and outputs he or she wants to receive. The research needs to be moulded to ensure that. Advertising effectiveness research falls into this category.

The latter will be successful when the client recognises that there is no right or wrong answer, and that no methodology is perfect. But through iterative progress, greater accuracy can emerge. Usage and attitude research falls into this category.

Both kinds of brief can bring in revenues and contribute to the business. But given the option, I know where most of my efforts would be concentrated…

sk

Picture credit: http://www.flickr.com/photos/elprofe/

Linkbaiting is a tactic, not a strategy

Will blogging eat itself?

While taking into the account the existential question of what a blog actually is, and the gamut of prose that it encapsulates, the trend for ever-increasing noise does seem apparent (this blog being but one example). From microblogging to reblogging via splogs and linkrolls, are we reducing ourselves to inanity repeated endlessly? And does this degrade the wider media environment? Two excellent posts have brought these questions to my attention

Warren Ellis argues that we have come through to the end of the age of blogging he calls “The patchwork years”. Does this mean original content will make a comeback?

Possibly, but Jason Calacanis points to a wider, potentially damaging, effect of this era. Jason may have quit blogging for private mail-outs, but his presence is still felt (well, perhaps reblogging isn’t always so bad). I recommend that you read the entire entry, but this quote grabbed my attention (spotted via A VC).

The life of a startup CEO dealing with the rabid but sometime naive blogosphere is one of extremes. You’re killing or you’re killed, you’re the shinny new object or yesterday’s news. You can couple the link-bait based blogosphere with main-stream media journalists who, instead of acting like the voice of reason and “sticking to what got them there,” have taken the link-baiting bait. The MSM has had to incorporate the flame warring, rumor mongering and link-baiting ethos in order to keep up in the page-view cold war.

This is either the shot in the arm MSM needs to compete, or they’re chasing the blogosphere Thelma and Louise-style off a cliff. Time will tell I suppose

This harks back to my earlier post on the problems with auditing online metrics. Page views and unique users are not the complete answer and we risk cheap stunts overpowering quality content. Trivia may be hugely popular on the Internet, but MSM risks damaging their brands if they try to compete.

Perhaps I am being Utopian but if a work of genius like The Wire can survive on 38,000 viewers then surely websites can survive on a commercially orientated but BBC-inspired mindset. By providing us with a useful service. Content should remain King.

sk

Photo credit: http://www.flickr.com/photos/leecullivan/

Links automatically opening in new windows

Whenever I link to other pages or sites, I’ve specified that the link should open in a new tab or window. Ever since the advent of tabbed browsing, this is the way I have surfed the Internet and so it made sense to me.

However, my opinion has changed after reading the posts and comments over at Problogger, Useit and David Airey. The choice should be with the user.

This is a policy I completely agree with. The web is democratic and people should browse as they wish. This is one of the reasons why I allow the full text of my posts to appear in RSS feeds. If RSS users want to click through to my site they can, but it is their choice.

I will continue to specify that direct links to files should open in new windows, but from hereon in the majority of links will open in this tab. If you wish to open the link in a new window or tab you can

• Click the link with the middle mouse button (the scroll wheel)
• Hold control when you left-click on the link
• Right click on the link and choose the option to open it in a new tab or window

(I realise that these are PC instructions, but this is what I have and know. Plus, I figure that if you have a Mac you probably know how to navigate already – SIDENOTE: Is it fallacious to assume Mac users are more web savvy?)

This policy should become more apparent in my next, overdue series of links update – which should be up over the next couple of days.

I’d be interested to hear other bloggers thoughts on this – how do you create your links? As I always control-click it had never occurred to me that I might be in the minority. How do you prefer to navigate?

sk

Photo credit: http://www.flickr.com/photos/pinkmoose/

Person David Airey

Right click for SmartMenu shortcuts

Mygazines and online magazine sharing

Everything is Miscellaneous points to Mygazines, a new website where people can upload and share their magazines.

My previous post was on piracy; would this venture come under the banner? Perhaps, though I’m not sure whether content owners would be as keen to pursue the owners in court (not yet, anyway)

• Unlike music or films, magazines have a built in obsolescence – whether weekly, monthly or longer
• It is not just the content being lifted, but the advertising as well
• The popularity and mass appeal is unproven

A nice feature of Mygazines is the ability to tag individual articles as well as magazines. This means people can search for specific content – whether it is jokes, recipes or technology essays – without having to guess which magazines to trawl through.

But the site is almost a no-win situation as if it proves popular, magazine owners may go after it. I’m not convinced it will get to this stage as

• I see the site as informing users of new magazines and driving them to those destinations – on or offline. In the first instance, I would find a useful article through Mygazines. In the second instance I would go straight to the website of the magazine I had previously read. This would make Mygazines transitory.
• Assuming the content is online, why would users want to scan through pdfs when there are fully functional web articles out there
• And as David points out; the site is slick but the process of uploading magazines certainly isn’t

I’ll be keeping an eye on the types of magazines uploaded and the frequency of uploads as an indication of whether the venture is taking off.

sk

BigChampagne and measuring piracy

Photo by http://www.flickr.com/photos/sharynmorrow/

Through this Economist article on Internet piracy, I came across the company BigChampagne. Among the data they compile are statistics on p2p downloads.

I can’t fathom from their website how exactly they measure this activity (I presume they crunch IP addresses of seeders and leechers), but it is certainly an area work monitoring. I expect that some very useful findings can be accrued.

So far the talk seems to be about music, but I see their being potentially more scope for TV producers and networks. Music, like radio, is try before you buy. You may hear a track by a band that you like on a compilation, and want to check out some more before you decide. As such, you cannot guarantee that a downloader is a fan of that artist and it brings into doubt the insights that can be leveraged from the location of the IP address, or the other simultaneous downloads.

As a sidenote, the Internet should be thanked for minimising the record label’s ability to con the public into buying the albums of one hit wonders. Though a bit too late for the people that have Babylon Zoo, Eagle Eye Cherry or OMC albums gathering dust in their attics.

However, if someone is downloading episode 6 of a show it is safe to assume that they are investing in the show as fans. While the geographic and cross-taste analysis is interesting, the key is to dig into the reasons why people are watching the programmes this way. Is it because of the length of time it takes to broadcast in their territory? Are legal video players sub-standard? Do people prefer to episode stack, and cannot wait for the DVD?

BigChampagne offers a potential aid to explore these issues. And these learnings can then go on to help companies improve their content offerings to the benefit of all involved – producers, distributors and consumers.

sk

Public accepting of advertising supporting free online video

New data from Ipsos MediaCT in the United States shows that the public are largely accepting of the ad-supported online video model.

Over four in five online video users think that it is reasonable to include advertising on full length TV programmes. To me, this is expected but what I did find surprising is that almost two thirds would accept advertising on short-form content.

The only format where a majority feel that this model is unreasonable is user generated content – an interesting result as Youtube consider various methods to shore up its revenue model.

As this data indicates, the question over whether ad-funded models can work online is largely settled. But while people may be in favour of the principle, the practice is still a huge cause for debate.

If someone knows the perfect method, volume and tone to distribute video advertising online, could you please let me know?

sk

Observations from New York

1. With the notable exception of the staff at Century 21, everyone (whom I met in Manhattan, anyway) is really nice

2. With the ability to run air conditioned subway trains 24 hours a day, you’d think it wouldn’t be so difficult to have electronic boards telling you how long it would be before the next train will arrive

3. Speaking of air con, most stores I went in were freezing

4. Portions, even at non-American cuisine restaurants, are huge

5. Cars toot their horns an awful lot

6. Although it applies in London to a small extent, it must be a nightmare having the main office blocks right in the middle of tourist central

7. Having 3 ad breaks during a 30 minute TV programme is really annoying

8. It is a fantastic city

sk

Vacation time

Photo by http://www.flickr.com/photos/stuckincustoms/

Macy’s 4th July Fireworks; Mets vs. Giants; shopping til I drop; Circle Line Tour; Guggenheim; Rockefeller; Lady Liberty; Hot dogs; bagels; cocktails; jet-lag.

I’ll be back in a few weeks.

sk

Links – 3rd July 2008

This will be the last link update for a couple of weeks

Blog-related

Happiness as your business model – presentation and analysis (Experience Curve)

Top 10 TED talks (Boing Boing)

Noah Brier’s Brand Tags has been adapted for the UK

The Marketing Society’s 50 golden brands microsite

Excellent business quotes (Junta42)

Appraising Forrester (Technobabble 2.0)

ESPN’s 360 model (Mediapost) and the move to interactive (Brand Republic) – the transmedia interconnected offering can really work for some genres

Collection of 41 new business ideas (Trendwatching/Springwise)

Military blogger forced to delete his musings (Wired)

Long tail debate using Rhapsody data (Harvard Business Leader)

The UK government have set up a competition to find the best use of its data in a mash-up – I’m guessing most will be using Google Maps, but I’ll be interested to see what comes out of this

Corporate blogs are largely rubbish (Wall Street Journal)

Has the Internet failed as a storytelling medium? (Advertising Age)

10 sites/applications account for 30% of web time (Brand Republic) – the equivalent of the commercial/network broadcasters?

Digg vs Jakob Nielsen (Mashable) – nice argument in favour of the Digg model

New Seth MacFarlane cartoon to be distributed by Google (New York Times) – There has been lots of chat about this. I’ll be surprised if it works, but it is an interesting move

Random

Website collecting one post wonder blogs

Worst videogames based on films ever (Wired)

The Lego secret vault (Gizmodo)

Profile of the inventor Buckminster Fuller (New Yorker)

News and comment on the Large Hadron Collider at CERN (Guardian)

7 people with superpowers (Cracked)

Extreme life forms (New Scientist)

Great moments in history – rendered via etch-a-sketch

One woman crusade against urban development (Seattle PI)

Recycled urban architecture (Web Urbanist)

Autocorrect doesn’t work when reporting on athletics (Guardian) – so much for free speech…

Particular recommendations are

Blog-related: Happiness as your business model, Military blogger forced to delete his musings and Digg vs Jakob Nielsen

Random: Profile of the inventor Buckminster Fuller, News and comment on the Large Hadron Collider at CERN and 7 people with superpowers

sk