Sunday, January 16, 2011

Visualize this: Where the public gets its news

I should say, where the US public gets its news.  And US public opinion does seem to affect us all, especially when presidents are elected...
The interesting thing is that it's changing.
The Pew Research Center asks people every year, "How do you get most of your news about national and international issues?"  The latest release shows that for some demographics at least, internet news is at least as important as TV.  Even overall, it's not looking good for traditional media.
The solid line here is a trendline (statistically significant regression) of the actual data points.  Data were collected at uneven intervals, so I've taken that into account in my visualisation.
You'll also notice some coloured crosses - these relate to news sources for specific events.  Pew asked people "How do you get most of your news about the war in Iraq / Hurricane Katrina / the terrorist attacks (9/11)?"  We could argue that people's recall of their main news source is likely to be more accurate when linked to a specific past behaviour, such as following a particular story in the news, rather than news consumption in general.  Interestingly, TV news is relatively more important, and other sources are relatively less so, when asked about these specific events.  It would be interesting to see this kind of measure for the current decade.

The effect you see in the chart of internet news gobbling up newspaper readership and to a lesser degree TV viewing is amplified for younger groups, as can be seen on P3 of the original report.
As you might expect, there's also a socio-economic trend in the data: richer, more educated people are less likely to consider TV news and more likely to consider internet news as a main source.  When we remember that people were prompted to give up to two sources, this is fairly damning for TV news broadcasters.

Figures add to more than 100% because respondents could volunteer up to two main sources.
Last but not least, it's interesting to see the differences by region.  Here I was interested to see if the regions were very similar in their consumption of each medium, or if certain media are more popular in some areas than others.  With that in mind I show the degree above (green) or below (red) average each region is.  You can see the internet taking over from TV effect is most obvious in the west, but it's also interesting to see the preference for print media in the north-east, and to a lesser degree, for radio in the west.
Figures add to more than 100% because respondents could volunteer up to two main sources.
Thanks to Flowing Data for throwing down the challenge to improve on Pew Research's original graphics.

PS Word of warning to anyone who likes coxcomb (polar area) charts as much as I do: sometimes they just don't work so well... Doh!

Saturday, January 15, 2011

How to start in Farmville

In writing this, I know I'm revealing that (a) I waste time playing a two-dimensional, unrealistic and highly commercial game like Farmville and (b) I'm geeky enough to run the numbers to see how to crack it.  Ah well.  I rest safe in the knowledge that if you're reading this, you'll make a mental note of the solutions below in case you ever play yourself.
You can waste a lot of time making coins very slowly in Farmville, so I decided to see which crops would work out best in the long run.
Some assumptions:

  • You've started recently.  Not all crops are available, and you have a 12 x 12 square plot.
  • You don't want to spend real money on Farmville.
  • I've assumed you're playing for coins and to some degree XP, not mastery.  Mastery will come if you focus on specific crops anyhow.
  • I've assumed that you return to the game as soon as your crops are ready and start them going again instantly.  Of course, no one would really get up in the middle of the night to harvest their strawberries, would they?
  • I've looked at the first 8 weeks of your crop - that assumes you don't die of boredom in the meantime. This seemed a good length of time to let economies of scale play out.
  • I've taken into account the fact that space is limited (so a dense crop will make more money than a sparse one, all else being equal).
  • I've taken into account that you have to buy vegetables and pay to plow the land for each harvest; whilst for trees and animals you buy them once and they just keep on giving.
  • I haven't included any extra help (e.g. aborists, tractors, gifts), or limited time items.
We'd like to be in the top right of the chart below - crops that give you high profits and high XP.  This chart shows how much of each we've gained after 8 weeks of planting only that crop in our 12 x 12 plot.  But those crops outlined in blue would require a huge amount of dedication - returning to the game every four hours in the case of blueberries.
The green dots on the other hand, require a visit of once a day or less, and are still fairly profitable.  I recommend these.  Stick to trees if you really don't want to put too much effort in.

Let's look at the trends for profitability.  On the chart below you can see the different types of crop over the first 8 weeks of planting only that crop on your whole plot as often as you can.  Again, the blueberries etc. stand out as being profitable, but we're not interested in dedicating our lives to Farmville so let's look at the best of the rest.  Peppermint is the best of the non-perennial crops.  But by week 8, trees are starting to overtake.  They require a big initial outlay, but you can see it's going to pay off.  Lime has the steepest line - that means it'll be the most profitable in the long run.
What about animals?  Well, they look cute and you can fit four in a square.  But the only one you'll be able to buy when you begin is a sheep (I'm not counting limited time items).  It hasn't even paid for itself by week 8, so I wouldn't recommend it.

This is a beginner's view of the game - and I'm not sure how much longer I'm going to persist with Farmville.  Do let me know in the comments below if I'm missing the point somehow.

Friday, January 7, 2011

A measure of mugg

Pondering the mugginess which has descended upon Auckland like a Fury...

The delightful Bob McDavitt of NZ's Met Service clarifies a few things for me via his blog:
Bob McDavitt's illustration
  • Mugginess has it's upsides.  Pohutakawa trees enjoy it and flower more.
  • Humidity is not a good metric for mugginess.  100% humidity tells us that the air is holding as much water as it can (it's saturated), but the actual amount of water this is depends on factors like temperature and pressure.  In fact, the 100% humidity we're now enjoying in Auckland is also experienced from time to time in Dunedin, at the bottom of the South Island, but doesn't feel half so bad there.
  • "The air feels muggy when we get hot and sticky – and that’s when 1) the air is warm enough to make us perspire, and 2) the air contains sufficient water vapour to interfere with the evaporation of our perspiration and not allow us to cool down much from that.  So mugginess depends on both heat and humidity."  
Mugginess is better expressed by a meteorological concept called the dew point - which is the temperature at which water condenses onto a surface (it only happens at 100% humidity).
All this gave me two ideas.  
  1. Wouldn't it be nice to demonstrate the "mugginess point" via a scatterplot of temperature and humidity rather than Bob's slightly crummy charts (sorry Bob)?
  2. This time series dataset is a perfect opportunity for me to try out Google Docs' motion chart gadget - a freely available bubble chart which is animated over time.  
Lots of these kind of charting tools are available as commercial software, but I neither want to lock myself into one software package (and deal with compatibility issues etc.) or work with something which is completely closed source and not modifiable in all the ways I'd like.
What I came up with is nice.  Not perfect yet (for explaining mugginess at least!) but kinda fun and a good first step on my road to animated charts.

Note: press play at the bottom to see the time series.  The speed gauge is immediately to the right of the play button.  Dates are in US format.  Dew point is colour, and the two series are Auckland (right) and Wellington (left).  Have a play.

Monday, January 3, 2011

Cat adoration society photos

For no better reason than I'm on holiday and can blog my heart out, and that kitty is looking adorable...

New year, new look blog

Out with the old, in with the new!  While Ross spring-cleaned the house, I decided to refresh the blog.  (Stop feeling sorry for him, it was his turn.)
My aim is something more aspirational, which better reflects the things I love and care about.  Or in other words, I want to think about nature more than buildings!
Over the coming months, I hope to share the ongoing delight of harvests and all the yummy things I do with what I've grown, as well as more adventures in the great Nyuh Zilan and beyond.
Let me know what you like and dislike about the refit using the poll and the comment box below.

Sunday, January 2, 2011


I resolve...
To give people the benefit of the doubt,
when my patience is thin and time's running out.
To find a cure for itchy bites,
to keep the cat medicated for fleas and mites.
To climb a mountain, at least just the one,
to indulge myself less and get more done.
To be more efficient in how things are run,
to be ordered and timely and not jump the gun.
To cook fifty new dishes, at least one a week,
using more parsley and cabbage and leek,
To buy new handbags and high heels,
to dress smarter for meetings and restaurant meals.
To enjoy the moment, spend more time with friends,
to make the most of the present before it ends.
In short I'll do everything I ought to do,
And be calm and grateful and tolerant too,
At least until March when the rain starts to fall,
By then I intend to have forgotten it all!

New Year's Day camping under Mount Taranaki