This evening I spent some time analysing results from a survey and took the opportunity to buid these representations in R.

The exact topic of the survey doesn’t matter here. Let just say it was a survey about opinion and recommendations on some people. The two questions were:

1. How do you think these persons were, last year? Possible answers were: very bad, bad, average, good or very good.
2. Would you recommend these persons for next year? Possible answers were just yes or no.

For the first question, the data was collected in a text file according to these three fields: Person, Opinion, Count. Data was similar to this:

Person,Opinion,Count
Person 1,Very bad,0
Person 1,Bad,0
Person 1,Average,4
Person 1,Good,9
Person 1,Very good,3
Person 2,Very bad,3
Person 2,Bad,4
Person 2,Average,4
Person 2,Good,5
Person 2,Very good,0

The trick to represent this is to use  geom_tiles (from ggplot2) to display each count. There is an additional work to be done in order to have the Opinion categories in the right order. The code is the following:

library(ggplot2)
data1 <- read.table("resultsQ1.txt", header=T, sep=",")
scale_count <- c("Very bad", "Bad", "Average", "Good", "Very good")
scale_rep <- c("1", "2", "3", "4", "5")
names(scale_count) <- scale_rep
ggplot(data1, aes(x=Opinion, y=Person)) +
geom_tile(aes(fill=Count)) +
xlim(scale_count) +
scale_fill_gradient(low="white", high="blue")+theme_bw() +
opts(title = "Opinion on persons")

And the graph looks like this:

For the second question, the data was collected in a text file according to these three fields too: Person, Reco, Count. Data was similar to this:

Person,Reco,Count
Person 1,Recommend,16
Person 1,Do not recommend,0
Person 2,Recommend,5
Person 2,Do not recommend,11

And we use approximately the same code:

library(ggplot2)
data2 <- read.table("resultsQ2.txt", header=T, sep=",")
ggplot(data2, aes(x=Reco, y=Person)) +
geom_tile(aes(fill=Count)) +
scale_fill_gradient(low="white", high="darkblue")+theme_bw() +
opts(title = "Recommendations")

And the graph for the second question looks like this:

Easy isn’t it? Do you have other types of visualization for this kind of data?

I was in New York recently and I was happily surprised to see the number of people still reading something (real books but also magazines, e-books, and of course their e-mails). E-book readers are more and more common, especially in planes and other public transport. Every big book store on 5th Avenue has its own e-reader, even web-based book stores promote their own e-readers. Most if not all of those e-readers promote its own closed, freedom-depriving file format.

We’ve all in mind pictures of the Rose Main Reading Room of the NYPL Schwarzman Building, where the beautiful room is packed with people reading and their laptops. I was also surprised to see the amount of computers available in other rooms (like the Bill Blass Public Catalog Room above). Books – real ones, made of paper and cardboard – are consigned to the walls, very few people are actually browsing through them. We now look for books in electronic catalog ; these books are stored in more functional room for librarians to find them. We now expose our collection of books in “social bookshelves” like librarything.

It’s not a reaction of an old man (well, maybe …). It’s more that we should think about what we want to create and what we want to leave for the next generations. It’s good to have a business plan to allow customers to have the best possible experience while reading your e-books for the next 2 years ; it’s even better if you would allow the same customers to keep their library open, transparent and “sharable” with others. Gutenberg’s invention allowed books and knowledge to be widely shared ; don’t let the “digital revolution” take that freedom back!

If early humans had been vegans we might all still be living in caves.

It says nothing and everything at the same time … Not eating meat would have stopped our “evolution” from early humans? Not eating meat would make us dumber? Or does it have something else to do? It does.

The original article on PLoS ONE is in fact a study about the impact of carnivory on human development and evolution. And the method used is a model of weaning in Mammals (thus no intelligence test per se). Psouni and her colleagues showed that:

• Brain mass is a better fundamental predictor of time to weaning than is female body mass (figure 2);
• Limb biomechanics is a predictor of time to weaning (figure 3);
• Dietary profile is a predictor of time to weaning (figure 4) and that
• Time to weaning in humans is quantitatively predictable from a carnivorous diet (figure 5).

So eating meat made human women wean more rapidly than if we stayed vegetarian. Their model suggests that “the contribution of carnivory was to shorten the duration of lactation and suckling despite the overall prolongation of development associated with increased adult brain mass”. Nothing about intelligence thus.

However this paper came to my knowledge after OAD published a (quite long) infographics about the dangers of red meat (not meat in general). On the presentation-side, I’m not sure such long vertical banner is powerful enough: after some time you are tired to scroll down. On the content-side, it’s a well-known fact that red meat comes with a lot of healthy risks. On top of that, the infographics focuses a lot on the USA, one of the countries where the consumption of red meat is high. This reminded me a TV programme from Jamie Oliver where he showed how meat is processed in the USA …

It’s “good” to be reminded of all this only once you’ve come back from there …

Key messages? Always read the original paper (even diagonally it’s better than general press) and know what you eat!

I know it has little meaning: very often the string of numbers in the pi value was not there for pi itself but for other reasons. I thought it was fun, that’s all.

Now compare how a movie showed Holi in 1981 (“Silsila“):

With a movie showing a Holi celebration in 2010 (“Action_Replayy“):

In Belgium, at least two celebrations are planned:

• one in Leuven, organised by ISAL, on March 10th, 2012;
• the other one in Brussels, organised by InBAG, on March 24th, 2012.

Now I had to delete this structure. And, to my surprise, it was impossible. When you try to just press the “Delete” key with the root directory selected in the File Explorer, you get a Path Too Long exception. The reason is that the maximum length of a path according to the Windows API (MAX_PATH variable) is defined as 260 characters. I tried some other methods but all of them failed:

• write a small Java program that tried to delete the whole path: Netbeans (Java) was able to create this mess, why shouldn’t another Java program be able to delete it? Impossible.
• write a small C++ program that tried to delete the whole path: as long as you stick with the Windows API, it’s impossible (I read that it could be possible using the boost::filesystem library but didn’t try).
• try some Portable Apps utilities for file management: impossible (even when the software was using another framework like Qt).

Finally, I just ran a Cygwin terminal, went to the ad hoc location and did a simple “rm -rf libtest“. And voilà. So, next time Windows forbids me from doing something, it might be a good idea to directly rely on a true terminal from a Unix-like environment. I didn’t try a liveCD (I didn’t have such CD to hand) but it might be also possible.

The chart shows the incidence of disease X in various age ranges. That incidence is split by 8 severity levels. The chart shows that the disease especially affects age ranges 4 and 5, at different severity levels. However I didn’t feel comfortable …

• what are the different levels of severity in age ranges 1, 2 and 3?
• how can we compare levels C, D and E in age ranges 4 and 5?
• is there anywhere some severity A?
• (it’s even worst when some age ranges don’t have any incidence at all: what is happening?)
• etc.

I looked on the web but couldn’t find much information apart from the fact “The Economist says they’re so bad at conveying information, that they’re a great way to hide a bad number amongst good ones” (but are still using them in their graphic detail section) or “a stacked column chart with percentages should always extend to 100%” (this doesn’t really apply here). Then in a post on Junk Charts, someone mentioned Steven Few who would have said “not to use stacked bar charts because you cannot compare individual values very easily and as a rule [he] avoid[s] stacked bars with more than six or seven divisions”. And Steven Few also participated in his forum here.

This reminded me I read a book written by Steven Few, a few years ago: Information Dashboard Design (O’Reilly Media, 2006). Inside, on pages 135-136, one can read stacked bar graphs are the right choice only when you must display multiple instances of a whole and its parts, with emphasis primarily on the whole. And that this type of graph shouldn’t be used if the distribution changes must be shown more precisely.

If one wants to clearly display both the whole and its parts, Steven Few recommends to either use two graphs next to each other or a combination bar and line graph (with two quantitative scales).

As I’m not really interested in the whole but mainly in the parts and their relative distribution, I suggest another way to present the data. This isn’t really new. Actually everything was already in the table. You just format the table nicely and add some colour gradient. And voilà:

You still see where the incidence is the highest (in age ranges 4 and 5), what levels of severity are the most important (C, with lower but approximately similar levels of D, E and H). In addition to the graph above, one can notice there isn’t any severity levels A, B, F and G represented and we can quickly grasp the proportions between the different incidences.

Of course, if your criteria for “sexiness” is that there shouldn’t be any digit on your chart, then this chart is not sexy. But I find this presentation really more appealing and meaningful than the stacked bar graph. Isn’t it?

 FBReader is a very nice e-book reader for Android. It supports a lot of e-book formats like epub, fb2, (partially) mobipocket, html, RTF and plain text. It works very well with Calibre (a free software e-book reader / manager / converter) – or is it the opposite? I also really like the fact FBReader can browse and download some free e-books directly from the internet. Of course, reading an e-book on a small 3.2 inch screen isn’t the perfect user experience. However, the night mode (white text on black background) is very handy to read something when it’s late and you don’t want the harsh white background. You can find FBReader on the Android Market for free (it’s a free software, under the GPL).

Google Reader is the Android version of the web-based Google Reader. If you follow RSS/Atom feeds with the latter, you will be interested by its Android version. A very nice feature is that all your feeds will be synchronized between the different versions of your Reader: read a post on your phone and the web version knows you already read it (and vice-versa). This time the small screen isn’t really a handicap since posts are usually quite short (compared to books and short stories you could read on FBReader). Up to a certain version, I thought the Gooogle Reader for Android was not really using all the capabilities of a touch screen. But in a version recently introduced you just have to swipe a post to read the next one (a bit like turning a page with only one finger). You can find Google Reader on the Android Market for free.

Mustard is a Twitter client for Android. It is also a StatusNet client. Being both on Twitter (@jepoirrier) and Identi.ca (@jepoirrier too), it’s quite interesting to be able to read and post on both platform quickly one after the other. I may have missed something but Mustard lacks the possibility to post on all registered platforms at the same time. And it’s impossible to have all timelines merged in a common one with duplicates removed. But apart from this small annoyance, it’s a very good and fast microblogging client. You can find Mustard on the Android Market for free (it’s even a free software – under the GPL).

According to its own introduction, Shuffle is “a personal organizational tool, styled around the Getting Things Done methodology”. In layman’s words, it’s a very easy app to use to remember things you have to do. Beside just adding a note, you can also give them a deadline, a location and a context. I know I didn’t use everything (for instance, I didn’t use the synchronisation feature since I don’t have any Tracks installation) but it’s a nice tool to remember small ideas on the way and help prioritize them (note: a Moleskine is also good for that purpose). You can find Shuffle on the Android Market for free (and it’s even free software - under the Apache license).

Finally, I really liked the last versions of WordPress (the one that introduced the big white panels/buttons in a “dashboard”). First, these last versions crashed much less often than before (in fact I didn’t see them crash anymore). Then, its developers made it easy to quickly approve comment, add pictures/videos directly from the phone, read the stats and edit some previously posted messages (of course, it’s not very handy to edit a lengthy post on a small screen but it’s not WordPress fault here). If you have a blog on WordPress (or using the WordPress engine at your own website), it’s a must. You can find WordPress on the Android Market for free.

One last word about apps for kids (the “sixth” application). The free versions of Whiteboard and Kids ABC Letters are quite interesting for a 3 year-old boy.

As you can read, I mainly use free software (as in free speech) and apps to read things on the phone. Another use is for small, quick, tasks (like posting a tweet or adding something to remember). I think I will be able to live without that

For those interested there will be a celebration at Théâtre Marni in Brussels on January 28th afternoon. And BelVietnam is mentioning three celebrations in Brussels on January 21st, 29th and February 12th.

Photo credit: Dragon boats resting on Sông Hương river, Huế, Vietnam (from my photos on Flickr, licence CC-by-sa)

I was also wondering: after China and Belgium, how will the population age in the USA?

The US Census Bureau maintains national projections of the population and its latest data is from 2009. Among other things, it takes into account the resident population and demographic components of change (births, deaths, and net international migration). For international migration (in and out of the USA), there are four alternative assumptions (described in the method statement):

• in the High Net International Migration scenario, they increase the previously projected net international migration by a fixed ratio ;
• in the Low Net International Migration scenario, they decrease the same previously projected net international migration by the same fixed ratio ;
• the Constant Net International Migration scenario illustrates the effect a level trend in international migration would have if maintained over the projection period ;
• finally, in the Zero Net International Migration, the number of immigrants and emigrants is held constant at a value of zero for the entire projection period, thus assuming a closed population and no movement of individuals into or out of the United States.

By proceeding in this way, the overall number of migrants projected to enter or leave the population is (optionally) modified while maintaining the assumptions about the distributions of demographic characteristics.

Now, as expected, the American population is indeed aging. From approximately 12% in 2000 the population above 64 years old will increase to more than 21% in 2050 (in the constant scenario, see below). We also see an acceleration of the increasing number of elderly in the USA between 2010 and 2030. This US estimation is a bit lower than the estimations for China (>23%) and Belgium (>25%).

In the graph above, I took the Constant Net International Migration scenario as I consider it as the most conservative. When one plots all the scenarii, we can see the difference is not so big: the US population above 64 years old in 2050 will be between 19% (High scenario) and 23% (Zero scenario) of the total US population (see below).

The main issue remains to maintain older people as much as possible the same levels of health and independence as they enjoyed during their active lives. As highlighted by GOOD, America’s workforce will need to work well past age 65 to save enough money for retirement.