James Hurst

Personal opinion adding to the noise of conjecture based on shaky facts.

[taptapio]

Palchinsky Principles

Inertia is driven by many factors, but usually its the fear of getting 'it' wrong driven by as excessive desire for due diligence. It stops people  making a decision, so they deflect it.

Deflect decisions and you'll achieve less. 

Make decisions quicker and you'll expose faults quicker, learn what hasn't worked quicker and find the solutions... quicker.

To help the risk averse, the Soviet engineer Palchinsky bestowed us simple principles that encourage success, by mitigating failure:

Seek out new ideas and entertain different options

When trying something new, do it on a scale where failure is survivable

Seek feedback, learn from mistakes, know when to stop

The first embeds variation, the second ensures survivability and the third  the selection process of what works (and what doesn't). 

He died for those beliefs, and in a world where our Monkey Business renders us blind, lets stop fretting about 'if its right' and adopt a leaner, less invested approach. Lets make decisions and accept (even celebrate) that some will be wrong.

We'll develop a pattern for thinking that moves forwards and finds solutions quicker.

[taptapio]

Saying boo to Tumblr?

I worry, as many do, about Yahoo buying Tumblr.

I worry because I really love my Tumblr blog. I am a fan of the Tumblr user experience, and its frustrating that once again the lazy instrument of 'advertising' is being used to pay for functionality.

I especially worry that a company that relies so heavily on advertising revenue models has purchased a company that only three short years ago described how advertising turned their stomachs. 

I'm not against Tumblr making money, in fact I encourage it, the issues I have are two fold:

The emotional missmatch between the culture of Yahoo and Tumblr (which very nearly killed Flickr)

The exclusion of user participation in discussions and opportunities to grow a user base to 'fund' the platform 

The first issue is a higher order issue. Looking down the barrel of $250 million dollars for personal financial gain would tease the most ardent anarchist to part with a few beliefs.

However, the second issue is still addressable. I really hope that Yahoo listens to the community.

Embrace the platform and its users. We are willing to pay for a premium service. In fact, we'd prefer to pay if it meant we could keep the platform ad free and Yahoo could build something very special, with a truly (in)vested user base.

If that isn't an option, then I can only imagine how happy this acquisition has made John O'Nolan and the team behind Ghost, who have smashed their funding target, and are a proof point (albiet they have some zeros to get to $1.1 billion) that users are willing to pay for  services.

[taptapio]

Data Piracy

Back in 2000 a few interesting things happened. We survived the millenium bug, The State of Vermont passed HB847, legalizing Civil Unions for same-sex couples and The Tate Modern opened. 

Much less reported is that  self anointed Prince Michael of Sealand (aka Michael Bates of Leigh-on-sea) and Ryan Lackey bought a dormant company (Sealand) and  declared the man-made World War 2 defensive rig known as Rough Tower as a sovereign principality. 

The facility was anchored 6 miles off the coast of Suffolk, probably the most exciting thing to have happened in data storage till the caves were dug in Iceland.

As a soverign principality Sealand could live (and die) by its own laws and would store anything you wanted (so long as it wasn't child pornography, or enabled spamming / malicious hacking -  rules they imposed upon themselves). 

Sealand disappeared in 2008 (not the rig, that is still there and well worth visiting if ever you're in a helicopter near the Suffolk coast), but Sealand the data storage service is  making a comeback. 

A timely response as CISPA laws start to unravel how 'safe' our data is from  prying eyes of government agencies and big data marketeers  Lets expect more 'off shore - off regulation' data storage centers as people push to retain their privacy. 

[taptapio]

Whose data is it anyway

I've spent an afternoon at a 'hide & own your data' investment event, followed by a 'what big data personalisation can do for you' launch party (very separate companies), it is an understatement to write data is de-jour. 

But it isn't all celebration a few individuals are asking important questions about a fundamental change in who owns what.

Federico Zannier has a captivating (already 150% funded still 19 days to go from today) pitch on Kickstarter as part of his A Bit(e) of Me project. He is taking back the ability to sell his own data directly, and has bundled up his own personal data to sell direct (and you should too). 

The interface artist Johannes P Osterhoff has an array of amazing projects exploring, questioning, provoking including the brilliant dear Jeff Bezos kindle app, that alerts Amazon CEO Jeff what Johannes is reading. Data amazon usually extracts subtly for commercial gain.

As someone who advises brands how to build compelling experiences, I find myself suggesting all manner of data explorations. They always have 'good' intentions, but does that really make them good?  

I don't know.

What is worrying are that laws surrounding personal data are well established but poorly understood, and as 'big data' mining deepens, so few are following in Federica and Johannes' footsteps and furthering the debate about what is and isn't right. 

[taptapio]

Know your meme

I was asked to correct a typo the other day, concerned I was being self important and chatting all about me. Meme in fact. 

That was not the case. I'm too middle class and British. I was using the accepted shortform to describe a memetic device. A Meme. 

"An idea, behaviour, or style that spreads from person to person within a culture". Wikipedia article here

Meme's occur everywhere (even in the 'real' world), although the ability for lots of people to congregate around ideas allows internet memes to grow at astonishing rates, and achieve astonishing things. 

Some of it is fascile (LOL Cats are an internet meme), but when an idea spreads (or goes "viral") it is astonishging what it can do, ie the moving Embrace Life public service video about wearing a seatbelt with over a million views in two weeks

For a fatter list of 'internet phenomena' there is a fairly well kept wikipedia article here, or (and i've only just been sent this link, so haven't had a great look yet) knowyourmeme.com seems to keep on top of the new and emerging internet meme's.

Now. Enough of that. Let me tell you more about me.

[taptapio]

Mere Exposure Effect

With his experiment Mere Exposure Effect, Robert Zajonc framed how you and I, as animals build associations as a result of exposure to events, sounds, images, smells, stories, dreams and more.

It is (in part) this science that lays the foundation for why clients pay good money not to captivate but saturate the daily grind. 

The arguments around this are covered in detail in too many marketing books, however, when I was lucky enough to be walking around the countryside in India it struck me how logos crudely hand painted onto walls acted as a signifier of everything I know in London.

I was a bit shocked (shit - i'd travelled all the way to India to find something new - right?!) And back country India is an amazing, magical place. You want to feel displaced, uncomfortable, unaware of what is going to happen next.

But those logos keep you (well they kept me) in some sort of home fug. In fact, as they were brighter, more visceral, hand hewn incarnations of the logos we see so polished on city hoardings it felt as if they were the forefathers to the crisp clean advertising running through our cities.

And seeing the farm hands waist deep in their fields, with the looming hand painted signage beaming messages across their fields, I couldn't help thinking how sad it is to have reached an Over Exposure Effect. Brands forcing us to remember them and by association everything our day-to-day associates with them.

[taptapio]

A three pipe problem

Its been a blistering start to 2013. My feet haven't touched the ground for all sorts of good reasons. However, I have been so busy, i've forgotten all the good 'rules' I was going to impose to stop 2013 being as much of a blur as 2012.

Thankfully I spent some time listening to podcasts, including the (as ever) brilliant science weekly show with Alok Jha.

His main guest, Maria Konnikova talked in detail about her book, How to Think like Sherlock Holmes, which I haven't read, but her words resonated. Especially around the theme of Mindfulness  or in in the nomenclature of Holmes "a three pipe problem". 

The appeal runs all the more 'deep' as I had set out, quite deliberately to push my heuristic learning. To sit and write - a shallow form of meditation - on the particular topics that were of interest to me day-to-day. 

However i've been astonished how little time there really is. Or rather, how much time is spent chopped up across so many things that there isn't the mental headspace to do any of that 'thinking' constructively (as the number of half baked drafts for this blog attest).

So. I restate my purpose to have more occasion to enjoy a pipeless, three pipe problem. Listen to the podcast, i'm sure Maria will inspire you the same in you.

[taptapio]

'Direct and well mannered' wins the day

Following in the footsteps of a folding plug and a low energy lightbulb, the British Government has delivered something superb with www.gov.uk and personally i'm thrilled  it has been recognised as Design of the Year 2013, the first website to have done so. 

With crowd pleasing competition in the form of the dramatic Olympic cauldron and the tall skyscraper Shard project dominating London Bridge, i'm sure there will be alot of head scratching around the winning votes being plied on such a simple, functional website. 

That is exactly why i'm in love with this project. Its not grand or wizzy (can we say words like wizzy?) They understand users of gov.uk don't need baubles. They need information.  The quicker they get that information, the longer they can spend gawping at the Shard and re-runs of  Olympic cauldrons being lit. 

This is great design philosophy, that understands human needs and delivers something that makes peoples lives easier every single day.

[alteredclone-deactivated2018051]

[whybray]

[taptapio]

The Simulation Argument

Published in 2003 by Oxford University philosophy professor Nick Bostrom, The Simulation Argument is well worthy of investigation.

The paper argues that:

...at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a “posthuman” stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation.

The compelling (and mind bending argument) proposition is that we are almost certinly living in a computer simulation: 

The possibility expressed by alternative (3) is the conceptually most intriguing one. If we are living in a simulation, then the cosmos that we are observing is just a tiny piece of the totality of physical existence. The physics in the universe where the computer is situated that is running the simulation may or may not resemble the physics of the world that we observe. While the world we see is in some sense “real”, it is not located at the fundamental level of reality.

It may be possible for simulated civilizations to become posthuman. They may then run their own ancestor-simulations on powerful computers they build in their simulated universe. Such computers would be “virtual machines”, a familiar concept in computer science. (Java script web-applets, for instance, run on a virtual machine – a simulated computer – inside your desktop.) Virtual machines can be stacked: it’s possible to simulate a machine simulating another machine, and so on, in arbitrarily many steps of iteration. If we do go on to create our own ancestor-simulations, this would be strong evidence against (1) and (2), and we would therefore have to conclude that we live in a simulation. Moreover, we would have to suspect that the posthumans running our simulation are themselves simulated beings; and their creators, in turn, may also be simulated beings.

Reality may thus contain many levels. Even if it is necessary for the hierarchy to bottom out at some stage – the metaphysical status of this claim is somewhat obscure – there may be room for a large number of levels of reality, and the number could be increasing over time. (One consideration that counts against the multi-level hypothesis is that the computational cost for the basement-level simulators would be very great. Simulating even a single posthuman civilization might be prohibitively expensive. If so, then we should expect our simulation to be terminated when we are about to become posthuman.)

And as professor Bostrom concludes:

Unless we are now living in a simulation, our descendants will almost certainly never run an ancestor-simulation.

A fascinating paper, well worth reading in full here:

http://www.simulation-argument.com/simulation.html

[taptapio]

Publishing digitally

Publishers, usually the smaller, more nimble start ups, are beginning to really embrace and leverage the opportunities that digital affords. 

Its a hodge podge list, but some recent examples: 

In the 'hyperlocal' trend

Spun delivers geotargeted editorial content. It lets you know quickly whats hot near you, with a stunning UI and experience. 

It aggregates thousands of feeds, from sources in 11 cities (all US at the moment) and publishes them as an interactive cube of 'curated' information, which adds additional data based on your location. Part magazine part toy. 

https://itunes.apple.com/gb/app/spun-city-news/id575348873?mt=8

In the same vien, but without the local angle is Flipboard:

http://flipboard.com/

  In the 'Slow Media' trend

Narratively fights the mass march to short bite size information. Its about long form, well written copy. 

It is Local. Organic. In Depth. 

They explore 1 key theme each week, with one story a day told through long form copy, essays, video and photos. They met their funding target on kickstarter and the content is genuinely interesting and well worth taking time to read. 

http://narrative.ly/

Another example of well written long form copy / nice UI and design is the news resource Quartz

http://qz.com/

  New Technology Impacts

Its  worth looking at innovations with augmented reality. One of the most interesting examples is Layar Creator, which doesn't require a target frame: 

https://vimeo.com/43376093

People are also starting to do fun things with Near Field Communication (nfc). A great example is The Impulse Economy book:

http://www.theimpulseeconomy.com/book/

Which is (I think) the first book that features a smart chip on its cover, that allows smart phones with NFC to access additional information. 

Its quite old now, but worth reminding yourself of the great work Berg did with Bonnier on how digital publishing could evolve. Alot of insight about user experience evolving from the paper to  screen:

https://vimeo.com/8217311#at=0

[taptapio]

The wisdom of crowds

Are meetings structured in the most effective way?

In this age of egalitarian social interaction, a work meeting still holds onto traditions that should be challanged. 

Highest paid persons opinion (HIPPO) usually carries the most sway.

Loudest Voice (LV) usually becomes the consensus view.

With a meetings conclusion usually positively confirming a bias preset in the attendee HIPPO and / or LV. 

Daniel Kahneman in Thinking Fast and Slow explains how our system 1 (intuitive) and system 2 (rational / logical) minds react to situations. He introduces an idea based on work in James Surowiecki's book, The Wisdom of Crowds that outlines the benefits of removing singular individual judgement and developing ways of aggregating multiple individual judgements.

A popular example of crowd thinking revolves around the fete 'guess the number of sweets in the jar' competitions. Individuals guess the total number of sweets in the jar. The closest guess wins the sweets.

Individual guesses can be high or low. However studies show that aggregated together, the wisdom of lots of individual guesses divided by the number of guesses made is often surprisingly accurate.

This wouldn't happen if the fete held sessions where individuals worked in groups to solve the challenge. 

Could we, as Kahneman suggests, learn from this as we go into our next meeting? Could we have a series of action points that we, as individuals, respond too on paper before the meeting to undo the potential for others personal biases affecting our views and therefore the outcome? 

[taptapio]

Motive, means and opportunity

Most people have a model (explicit or implicit) to solve problems. The method used will inform how information is evaluated and ultimately shape  action.

The model is usually explicit and we see people sketching quadrants or venn diagrams or pyramids or onions or cones etc full of interesting information dollops building foundations for arguments.

Well this is another one of those. 

Sir Arthur Conan Doyle via Sherlock Holmes introduced a shorthand into popular culture about negotiating our way towards 'who done it'. Sherlock would seek the motive, the means and the opportunity. And employing Occams razor would deduce a truth. 

What if we borrowed this logic to help better interrogate ideas. 

This isn't a magic bullet, the output reflects the input. I'll get some examples of it 'in-use' soon but broadly (and these might change) I'm going to be breaking down the inputs into:

Motive Do they (or you) have the motivation to action the idea

Means Do they (or you) have the capability to execute the idea

Opportunity Is there an audience for a specific opportunity

Some broad examples are:

The client has little motive (shown in orange), but has a great marketing team and therefore the means (magenta) and a huge opportunity (yellow) our diagram would look like:

Or we might determine our client is really motivated for the changes we recommend (orange), but the opportunity within the market is less certain (yellow) and the client doesn't have the right budget or team to execute the idea (shown in magenta). 

[taptapio]

Could we turn a car journey into a premium treat?

Building on the previous post about Bayesian theory and a bit of and experimenting in Google docs, i've built a Bayesian calculator that works in a way that (for me at least) feels intuitive. 

I've used the equation from Nate Silvers book The Signal and the Noise, although i'm not sure where this equation should be attributed. 

You can also see the calculator here (its a static web page)

The example below is fictitious. I've looked at it in two ways. #1 with probabilities based on gut feeling. #2 with probabilities based on hard figures pulled off the office for national statistics about transport here

A fake scenario

We are working for a company that aims to offer an "online car concierge" service. For the sake of clarity lets call it Carcierge. Its a premium offer not that different to Uber but launching in England. The client already has a market presence but as a standard taxi service. Their ambition (and brief to us) is to enter the 'premium' service sector and they want us to build a brand around premium private hire drivers.

Some hard facts we have are:

78,000 licensed taxis in England and Wales with 22,600 in London, 31% of the total (31 March 2011 from this go.uk report)

155,100 licensed Private Hire Vehicles (PHVs), although this includes all of the various minicabs. 

We also know why people travel (from the National Travel Survey 2010 here)

And how they get there:

From this we could infer: 

Our client should focus on short trips.

and an obvious Carcierge customer might be the corporate commuter being hammered by parking premiums, rising public transport costs and congestion charges

However, the market for the corporate commuter is fierce (and obvious), so lets imagine:

Our hypothesis

Carcierge will add value as a premium 'treat' that people can book occasionally when they are doing something special, maybe a night at the theatre or corporate entertainment where a hired 'private driver' feels more discerning and special. We'll define a category code where an evening out starts with the journey. Luxurious service a pleasure to unwind in.

 #1 Developing a probabilistic response but recognising that the data supplied is disconnected with the hypothesis, so basing inputs on gut feel and presenting it as biased.

You can see the calculator on google docs here

So my gut feeling (although I did believe in the hypothesis as I wrote it) when looking at the hypothesis through Bayesian eyes is that actually adding value to 'the journey' probably isn't the most compelling strategy for Carcierge 

#2 developing a probabilistic response using the 'hard data' sources, and trying to infer a pattern or rationale for each input. The data isn't quite 'right' to answer this exact hypothesis so i've knowingly tried to find my own pattern in the data.

You can see the calculator on google docs here

Building a biased analysis by inferring meaning into the data implies that the probability Carcierge could add value as a premium 'treat' is even less. 

In summary

Unless the perfect data set can be found, it seems disingenuous to daisy-chain adhoc data together and present the findings as cod-statistical analysis. The attempt to intellectualise the findings might be read as 'fact' while the factual-ness wont be perfect as the input is convenient rather than designed.

However, in my opinion, having less data, but structuring the calculation as a baseline to be tested and refined on an ongoing basis might be a useful yardstick. In this faux world, it would be encouraging if Carcierge ignored the low probability, rolled out a campaign and examined the hypothesis with actual data.

That ongoing refinement would allow us to build a tool that could be used to better reflect probabilities to questions around how Carcierge ought to behave, structure experience, develop partnership, etc etc 

Although alot of branding and design is awash with patchy 'real data' I'm thinking that as a discipline for examining our thoughts its rigour is exceptionally useful. 

Do get in touch if you have any thoughts or feedback. 

[taptapio]

Could Bayesian theory help with branding?

We've had an interesting run of projects at Figtree where elements of brand strategy have been exposed to a variety of testing situations across various markets. In real words, people have got on planes, flown around the world with research agencies and sat with research groups testing if we've made the right decisions.

But could we better test our thinking before we commit that level of investment? Could we define a set of tools to help us discern viability (or not) in a more nimble way?

I think there is a lesson we can learn from a statistical theory named after Thomas Bayes:

The Bayesian approach is superior to use in decision making when there is a high level of uncertainty or limited information in which to base decisions on and where expert opinion or historical knowledge is available. Bayes is also useful when explaining the findings in a probability- sense to people who are less familiar and comfortable with comprehending statistics. It is in this sense that Bayesian methods are thought of as having created a bridge between business judgments and statistics for the purpose of decision-making.

Bayesian theory in marketing

Bayes' theorem

Its both surprisingly simple and frustratingly complex. An algebraic expressing with three known variables and an unknown one. It helps us understand the probability of something if something else happens.

How it works

Develop a hypothesis

Estimate the probability of the hypothesis being true

Estimate the probability of the hypothesis being false

Estimate a prior probability of something happening without interference from the hypothesis 

With these values, we can determine the posterior possibility, a fancy way of saying how likely something is (or isn't).

There are inherent drawbacks, the model works well with reliable data (a barrier to use in some situations). Its also an algebraic equation which needs a little decoding to understand. But if the data were there it might be an antidote to bias inherent in a profession that is more subjective than objective.

It also encourages probabilistic outcomes which allow us to present (if we find a way of using the theory) our work in degrees of probability, which seems a more modern and elegant solution.

If anyone has experience of working with Bayesian theory, especially outside of a pure data context, i'd love to hear from you.

Some useful links:

A great tool to drop in Bayesian graphs on your own sites. My issue with the tool is a slight missmatch between my understanding (which i'm assuming is over simplified) of the inputs, especially 'the evidence' field. You can view the a demo page on github here. 

As ever, Less Wrong have an interesting visual post about the theory here.

And the brilliant Nate Silver, who explains Bayesian theory and its application (among many other things) in his book Signal & The Noise is well worth a read.

[taptapio]

Should we be more parochial with our data?

Guilty of oversimplification and probably a case of correlation over causality but; should we be more parochial with our data and information architecture?

Mass agriculture has opponents and proponents, both with wide ranging and converging arguments. Although aware of global benefits born from the green revolution, one well documented issue with vast mono-crops is fragility in the face of a powerful virus. Bluntly, a lack of bio-diversity widens potential for wider devastation.

Would we benefit from learning lessens in how we maintain our data (farms)?

As with farming our early online practices look (now) parochial. Most web servers and data houses were built for specific applications, with shared attributes but inherent built in diversity. Some of that diversity was born out of market competition with different companies offering different blends of proprietary and ubiquitous software and hardware. Today the market is saturated with reseller programmes that trickle up to a shrinking number of origin points. Are we are breeding out the diversity that added a layer of protection. 

The counter argument would highlight benefits of enterprise technology protection, versus the (if we take the farm metaphor a little further) rickety gates of the original data cottage industry, and the 'meta' business is a concept baked into the internet with domain names bought through ICANN resellers (most of whom then added additional services like hosting and email).

However with most people re-selling services by established players like Amazon (accessible cloud computing solutions) and enormous innovations like Apache Hadoop  (that leverages distributed computing) acting as an innovation baseline, does our data and the architecture we're using to distribute it resemble enormous exposed fields? If so, would we be better off being more diverse in our thinking to build better resilience?