Friday, August 31, 2007

The problem with knowledge, part deux

In keeping with the theme of maps started by the now extremely famous Miss Teen South Carolina (she should think about a book and movie deal given the amount of press her answer has seen), Isabel (over at God Plays Dice), wrote an interesting post on ways of drawing the US Interstate Highway system.

In particular, she references this map. This is a simplified version of the system.

When you go into a service station and try to find a map in the US, you can't buy this one. Yet, it strikes me as a MORE effective way of displaying interstate routes than a map with exacting coordinates and relationships.

You just don't need to store the level of detail found on a conventional map to understand or navigate your way through this system. This map is perhaps the best internal representation of what most people probably know - the general direction and position of these roads.

When we abstract away layers of knowledge to more efficiently store it for what we USE it for, the details become lost. Testing our knowledge of the details, no matter how simple the test, is ridiculous. When asked to recreate the US highway system, I am sure many people would try and curve roads around landmarks, lakes, mountains, farms fields etc. and scratch their head when they can't quite remember how this or that road curves. Not realizing that a straight grid and a few semi-circles is all they actually know, and need to know.

When you do research on product and brand attributes you see the same effect. While most marketers believe consumers are aware of every differentiating feature of their product/brand, most consumers have already abstracted away these differences into broader 'maps'. 'Maps' anchored by the things they care about the most.

The faster we realize that our heads aren't full of facts waiting to be regurgitated at a moments notice, the fewer stumped Miss Teen South Carolinas there will be.

That's a good thing, right?

Digg this
Sphere: Related Content

Thursday, August 30, 2007

UK House Price Roller Coaster

Ok, so I managed to actually get this thing working (as mentioned in this post). It's definitely not as interesting as the US one - partly because I could only go back to 1953 and partly because the real wild ride in UK house prices seems to have just started.



As a data visualization it's probably not the most effective. A simple graph would tell you all you need to know. Still, by the time you hit the year 2000, you are on a path that is considerably different to anything in the previous 50 years - that at least is very clear.

Be sure to watch it all the way to the end.

Digg this
Sphere: Related Content

Wednesday, August 29, 2007

The problem with knowledge

This has to be one of the funniest things I have seen in a while. Poor Miss South Carolina in the Miss Teen USA competition has a hard time answering the question "Why do you think a fifth of Americans can't locate the US on a map of the world?".



To start with, you can't help but feel sorry for the poor girl. Public speaking is tough. It takes experience to overcome direction-less rambling when stumped.

But what is obvious when you look back at her initial reaction is that she was, well and truly, stumped. But why? The question seems so easy! It's obviously a polite way to ask why Americans are so culturally inept and ignorant of even their own place in the world (which isn't completely true). She could have just rambled on about the need for education, more cultural sensitivity etc, any number of PC buzzwords.

She didn't because she failed to deduce that this outwardly innocuous question had any deeper meaning. She was just trying to figure out why someone wouldn't know where the US was - of course it's due to the lack of maps!

The real problem is that the premise of the question is wrong. Not knowing where the US is on a map of the world is not an indication of ignorance or lack of 'worldliness', it's an indication that the knowledge itself is largely useless.

Why would you ever need to locate your country on a map of the world? Yes, there are exceptions, but for most people (at least a fifth of Americans) it's easy to conceive why this knowledge is redundant. And besides, what sort of test do you use? A blank map of the world with no city names and just coastline? Give anyone a globe fully annotated with landmark and urban center names and I am sure they are going to have no problem picking out NY and LA and guessing the US is in the middle. All the knowledge they need to store is that one is on the west coast and one on the east.

This issue reminded me of another test that seemingly showed a lack of attention to detail. When given 20 different versions of the face of a 1 cent coin (with different positioning of the inscriptions etc.), most Americans couldn't pick the correct one. Why? Because you don't need to know the exact features of the 1 cent coin, the knowledge is, again, useless. You just need to know how it differs from the other coins in your wallet, in size and shape, for the system to work.

And that's the problem with knowledge - it's easy to measure in the wrong way. In the case of Miss Teen South Carolina, her undecipherable ramblings might have made her the poster child for dumb blonds, but in my opinion, the question was as equally stupid.

Digg this
Sphere: Related Content

Tuesday, August 28, 2007

Best4c online flow-chart design site

Just came across this site - Best4c. It's an online flow-chart design tool. It's pretty nifty.

One click on Try For Free brings up the program (inside the browser) and in less than 10 secs you can start to create diagrams.

Easy. Fast. Useful. Everything an online offering like this should be.

For those of you who are still trying to design flow-charts in Excel or Powerpoint, this tool is probably orders of magnitude better!

Digg this
Sphere: Related Content

Thursday, August 23, 2007

Not practical or meaningfull, just... beautiful

A good inspiration piece from our friends at Information Aesthetics.

I was never a huge fan of community art. Just because you invite someone to post something they have created and combine it with other creations, doesn't mean a higher meaning emerges. But maybe that's not the point...

My wife constantly tells me to stop trying to explain the meaning behind movies - I always wondered how you could appreciate something if you didn't understand it as a metaphor? She made me realize that sometimes you just enjoy things and move on.

She's smart like that.

Digg this
Sphere: Related Content

Tuesday, August 21, 2007

Some more data visualizations - UK house prices

Since looking (and screaming) at the Greatest Ever Data Visualization, I was curious to see what UK house prices have looked like over time (spurred on by my sister who lives there, and was curious to see what it would look like):

House Price by Year


This is a pretty nasty looking trend.

A couple of points though. The vertical scale is plotted against the average of the distribution, not the actual house price. This is because I couldn't (easily) figure out how to adjust the scale in Swivel (something I will be trying to find out how to do later). However, the trend when plotting against the average and the actual values trend is about the same - all equally as scary!

The house prices are for the UK as a whole and came from Nationwide's very useful data download facility.

The adjusted figures were from entering the unadjusted house price into this calculator. It uses the Retail Price Index (RPI) to calculate the value of the price in 2006 pounds. This seemed to be a logical thing to do - adjusting for inflation in everyday goods (as that is how you tend to 'value' your money). However, I have seen shorter runs of adjusted house price data that have different price levels, if similar trends. If anyone knows what additional adjustments are used, please let me know. I might have a bit of a poke around some economic sites to try and find out.

Next step, plotting it to a roller coaster.... stay tuned.

Digg this
Sphere: Related Content

Monday, August 20, 2007

Brand in the Machine

Following on from my previous post about Design, Norman divides good design into 4 stages:

  1. Visibility - simply looking at something (with no direct knowledge of how it works) gives the user the state of the device and the alternatives for action.
  2. Conceptual Model - a 'model' designed into the operations of a device to give the user a coherent and consistent image of functionality.
  3. Mappings - making it easy to determine the relationships between actions and results, between the functions and their effects.
  4. Feedback - the user receives full and continuous feedback about the results of their actions.
For building anything with a User Interface (UI), I don't think you could get four better principles.

The only one I would add though, would be 'Tone'. Tone is almost the physical manifestation of each of the above four ideas. It's really the 'aesthetic' delivered in a consistent and coherent manner.

For the innately visible aspects of design (visibility, feedback, even mappings) it's easy to see the importance of Tone - all great products have it. But for certain products, I think you can even extend the idea to the Conceptual Model. It's about having an elegant, or easy, or recognizable, or familiar (a subtle difference) idea running through a product.

Every PC has a basic conceptual model regarding how files are stored - folders. It's a very structured, almost rigid idea that lends itself to very linear thoughts (file size, position in hierarchy, order, etc.). If you use Google Desktop, the conceptual model of 'folders' is replaced by 'search' - files now have relationships to other files based on commonalities of content and meaning, not a position in a hierarchy. 'Search' is less structured, more fluid, less linear.

Conceptual Models thus have 'tone'. Either one may be appropriate. Which one fits best with the design you have?

All of this is Branding - well removed from what many people typically believe Branding to be all about - but Branding nonetheless.

Digg this
Sphere: Related Content