1. because I prefer to have the excuse of not being able to check my emails when I’m away from my desktop computer;
  2. because I want to enjoy those valuable moments when I can’t figure out the answer to a trivial question (“How old is Obama?”, “How does toothpaste work?”, “Where is Fiji?”), at least until I get back in front of the computer;
  3. because when I meet friends/family I’d like to fully attend to them instead of looking downwards and checking emails/websites every five minutes;
  4. because I’d like to keep my habit of getting lost in books when I travel;
  5. because I like to throw my phone around carelessly;
  6. because my fingertips are too large to type on those tiny touchpad keys;
  7. because if I’m going to spend that amount of money on a piece of technology, I’d rather buy a compound digital microscope and see/photograph cells instead of people.

For the first four reasons, don’t tell me that I have the option not to be doing these things (not checking emails continuously, etc.) while having the smartphone – I know, from observing my own internet behavior and people around me who have smartphones (i.e. people around me), that these habits will become second nature in just a few days. The only way of resistance is to avoid buying it in the first place.

Join the resistance, my friends! (Literally.) Because I don’t want to wait while you check Facebook under the table every five minutes.

Facebook Twitter Email Stumbleupon Tumblr Digg Delicious Reddit Linkedin

My new project The Trouble With Everything is a presumptuous blog by crabby megalomaniacal misanthropic elitists. It also has a Twitter account.

Facebook Twitter Email Stumbleupon Tumblr Digg Delicious Reddit Linkedin

This is an examination of Metallica’s concert history from 1982 to 2012 with a focus on the numbers of songs played live and the albums that they belong to. I and my friends at Çilek Ağacı took the raw data from Setlist.fm (plus Last.fm in the last part), groomed it and visualized it with our own tools.

The colors of the albums – a key feature of the whole visualization – are chosen according to the album cover artworks; Metallica fans can easily understand which color corresponds to which album without reading their names. As a convenient surprise, new albums (Load, Reload and St. Anger) that are stylistically different from the old ones are all tints of orange-yellow and this provides a natural visual grouping in the charts. (Try to see this grouping when you look at the charts.) Death Magnetic, which is their newest album but musically much closer to the old ones, has its brownish gray shade (again, taken from the real album cover), and that separates it visually from the Load – St. Anger period.

In the first chart, the number of concerts given each year is plotted. In the second chart, the bar heights represent the absolute numbers of songs played from each album. The album release dates are marked on the grid columns with the corresponding colors.

The years 1982–91 are a very intense period for Metallica; they write a vast majority of the songs that are played live even today, and are involved in extensive touring. We can see that in the eves of the first four albums, Metallica continues to play gigs. With the self-titled fifth album, the band starts to quit touring and takes more time for songwriting before releasing albums. Maybe this is one of the sources of the famous “problem” (according to the oldschool first-four-albums metalheads) with the new albums; maybe Metallica shouldn’t be left alone before albums. This idea is consistent with the case of Death Magnetic which is not written in absolute isolation according to the chart, and has largely won the hearts of oldschool fans.

The year 1992 sees the climax of touring in Metallica’s carrier where each album reaches their all-time maximum play counts (except Justice), all dominated by the “Black Album”s overwhelming numbers.

Looking at 1996–97 we observe that Load-Reload songs seem to go well with the Ride the Lightning songs in a way that the remaining of the first-four-albums don’t. Interestingly, Reload never reaches the explosive levels of playing of Load after its release but it enjoys a more stable carrier afterwards (thanks to Fuel and The Memory Remains) compared to Load, ultimately not falling too far behind it in total play count (1115, 696).

St. Anger comes after the longest break in Metallica’s touring history (for obvious reasons) and lives the short and brutal life of a monster. (It is seen as their worst album by the oldschool fans.) It’s clear that Metallica aren’t happy with playing St. Anger songs live. It seems like Death Magnetic songs arrive to put St. Anger out of its misery; this is the only case in Metallica’s history where a new album entirely cuts the play count of the previous one. It can also be seen that Death Magnetic songs enjoy the company of the older songs instead of Load-Reload.

In the second part of our examination is a map of the places where Metallica has played live. The opacity of the red outer circles correlates with the number of concerts given in that city. The details about the northern- and the southern-most gig venues where Metallica has played are given in trivia boxes.

In the third part we see the total play counts by albums and songs. As expected, the most played songs are from the old albums, and the least played are from the new ones. But what if there was a way to compare the real “performances” of the songs in concert setlists, independent from their release date? To address this question, in the next part, we introduce the concept of Power.

Song Power is calculated by dividing the total play count of the song by the number of concerts after its first playing date. Album Power is the total play count of the songs from the album divided by the number of all songs played in all concerts after the date when a song from that album was first played (which may be before the release of that album) and then divided by the number of songs on it. This is a normalization to eliminate the advantage that the old songs have (they had more time to be played) in order to create more neutral rankings. You can also say that a song with the Power value 0.945 has a 95% chance of being played in the next concert. The Album Powers are additionally normalized with respect to the numbers of songs on the albums. (The Album Power values on the chart are multiplied by 10 for presentation efficiency.)

We see that the rankings change when we eliminate the time factor. Enter Sandman takes the lead as the most Powerful song and Death Magnetic, their latest album, beats all the old albums in Power; this means that Death Magnetic had a greater share of setlists after its release than the other albums did in their own lifetimes in average. Two songs from Death Magnetic are already in the Top 10 Most Powerful Songs list, prevailing over classics like For Whom the Bell Tolls and Seek & Destroy. Another way of thinking about Power is this: given enough time, Death Magnetic may well be in the top ranks of the Total Count by Albums list whereas Load, Reload and St. Anger do not seem to be moving from where they are. (Note that if we did the same analysis in 1997 we would have been saying a similar thing about Load; time will tell whether Death Magnetic’s Power will subsist.) S&M with its two songs presents an interesting case as it moves on top of the “Orange Albums” thanks to the song number normalization.

This part also contains the complete list of Metallica songs never played live in their entirety; among them are songs that are played in part (for instance, until the second verse part) or have been featured as riffs in a jam. Here Reload sticks out as the album with the most songs that are never played live. Kill ‘Em All and Master of Puppets are albums that had all their songs played live at least once.

In the fifth part we plot the Song Power on the y-axis with song durations on the x-axis. Here the first thing that strikes us is a pattern of orange-yellow squares tiling the lower part of a diagonal line between the upper left and the lower right corners. This tells us that songs from the Orange Albums tend to lose their Power as they get longer in duration; there isn’t one song from Load, Reload or St. Anger among the 19 songs located above that diagonal. This plot also gives insight about the natures of the individual albums thanks to the color-coding; for instance, it is easy to see that the black squares are grouped in the left half of the plot, meaning that the album Metallica consists of shorter songs in average.

Finally we make a comparison between the personal listening statistics taken from Last.fm and the concert setlists (after 2002 when Last.fm was founded). On the left are songs that have a greater share within Last.fm than within setlists in total, and the songs with greater share within setlists are on the right of the axis. This chart suggests that the band may further please the audience by playing songs from the left end (The Unforgiven, The Unforgiven II, etc.) more in concerts. The fact that there are fewer songs on the right side and their bars are much higher than the ones on the left side tells us that the setlist and Last.fm statistics have different distributions: there are songs that Metallica plays on almost every show and there are songs that they have never played live whereas Last.fm listening statistics are much more homogeneous.

[Personal addendum: In addition to all the objective analysis above, I would like to state as a fan that I love the Orange Albums as much as I love the old ones, and I am thrilled to see them played live.]

Praise on Social Media

  • andypotter (Andrew Potter) Lustig and also insane :)
  • jmlacroix (Jean-Michel Lacroix) L’infographique le plus intéressant de 2011.
  • nag_acharya (Nagaraj Acharya) Whoa! This is for those of you who are doing a PhD on Metallica.
  • tmbrntt (Tom Barnett) not exactly a metallica fan but this is an all-time favourite infographic.
  • bquarant (Brian Quaranto) The infographic bar has been raised again.
  • clipperhouse (Matt ☼ Sherman) OK, this is actually an infographic. (Most “inforgraphics” are just lists done in Illustrator.)
  • DannyJWillis (Danny Willis) My new favorite data visualization ever.
  • dkastner (Derek Kastner) Here’s an infographic that’s actually informative, novel, and presents data effectively.
  • Fitoria (Adolfo Fitoria) La mejor infografía de todos los tiempos.
  • Brian Fitzhugh An infographic that is actually interesting.
  • Leah Root Wow! This guy put more research and effort into this than most people do for research papers or companies do for financial statements. He needs to be a CEO of something. A real ‘Master of Puppets’ ;)
Facebook Twitter Email Stumbleupon Tumblr Digg Delicious Reddit Linkedin

“Bugün Ne Giysem?” (“What Shall I Wear Today?”) is a TV show in Turkey where female contestants dress themselves and get critique from judges (fashion icons/designers) who decide whether they go through to the next round or not. The show claims to be in search of “the most stylish woman in Turkey”. The judges also claim that they are educating the viewers while giving critique to the contestants and thus leading Turkish women towards a better-looking future. As an avid fan of the show, however, I observe that their method is flawed. I propose that they may do better in their idealistic task if they take one step further in their educating. Now I don’t know a thing about fashion or clothing, but I’ll make my case as a (visual communication) design professional and you’ll see how it relates.

A typical conversation between the judges and a contestant is like this:

— The color of your shoes is very wrong, considering your dress. Why did you choose that color?
— Because I like that color…
— You could’ve picked a bright green instead.

or

— That dress is a very wrong choice considering your body type. Why did you buy it?
— I don’t know… Because I liked it…
— Look at how it makes your hips look larger. That’s not a good choice.

So what’s the problem here? I see two. The first problem is with the contestant’s almost tautological answer, but it’s natural, given that these are regular people with no design education or high intelligence. The second, and the bigger problem, is with the way judges respond to that. They tend to offer specific and temporary fixes, and ignore the main issue; if only somebody told these women that style (or fashion, or indeed design) decisions are not just about what you personally “like”!

As far as I understand from the show, personal styling is very much like design in general: you start with a problem, and look for solutions. What problem? What is the design brief, indicating the problems to be solved, giving the context against which the solutions are judged, in the case of this show? What is the content at hand? Well, as the judges frequently hint, the “given” content and problems include your height, your weight, your body type, your skin tone, the beauty level of certain parts of your body (face, shoulders, knees, etc.), and where you plan to go in that dress/make-up. Once you have these constraints, the rest is optimization (with multiple solutions, of course). For instance, if you are short, you should avoid flat shoes. If you have a beautiful face, you should make it visible by tying your hair up. Simple as that. Just keep track of the “if”s. [The "if"s multiply in number as soon as you start to make decisions: if you have nice toes, you may pick open-tip shoes. Then this becomes a new "if" in your system: if you wear open-tip shoes, don't ever wear pantyhose. So now you don't have the right to say "I wore pantyhose because I like it."]

It appears that the parallels between styling and other design disciplines go further than this. Consider this dialogue from the show:

— Why did you choose that color for your purse?
— I wanted it to match the color of my shoes.
— But they don’t exactly match.
— They’re close…
— Close isn’t good enough.

This is the famous ambiguity problem in design. Things (colors, distances, typefaces, etc.) should be either exactly the same or distinguishably different from each other; in-between isn’t good since it puts a cognitive load on the viewer – an unconscious monologue like “Are these the same? No, they’re different… Or are they?” Another example:

— You have a huge gold bracelet on your arm, and then you have the huge gold necklace on your neck… One huge gold thing should be enough.

This is the hierarchy problem. “There should be only one boss”, or ”Elements shouldn’t be competing”, as we say in editorial/layout design. So not only styling resembles design in its overall nature, it also shares the basic principles common to other design fields.

As in every design discipline, clothing design (and styling in general) has its long-tried and well-known solutions to common types of problems, though you can always go creative and invent your own solutions. The judges in “Bugün Ne Giysem?” applaud those rare creative contestants, and spell out those well-known solutions for the unsuccessful ones, but they never reveal the true nature of the process: they never explicitly tell that this is a problem-solving process that has very little to do with what you “like”. This may seem like a theoretical issue, but it makes a huge difference in the intended style education of Turkish women. Avoiding it strengthens the mystical belief that picking the right clothes, looking good, being stylish (and getting good critique) all have to do with the so-called “taste”. And this rather metaphysical “taste” thing, you either got it or not. But tastes may also differ. (Yes, it’s not a coherent belief system.) If your look is ugly to me, then you obviously have no taste. If you don’t like my look, then your taste is different from mine. If the judges don’t approve my look, they sure have different tastes, and there’s nothing objectively wrong about my look – now you see how that “theoretical” issue in styling (or design) education ruins everything that the educator tries to accomplish.

[The taste issue is just as problematic in visual communication design: the taste of the CEO of the company – ideally – should have nothing to do with the corporate identity design of that company. He may hate the color blue, but blue may be the objectively right choice given the design brief at hand.]

On “Bugün Ne Giysem?” we too often see this taste defense on the part of the contestants, to which the judges apologetically reply by babbling something like “Well, we’re judges in this show, we’re supposed to give you critique…” Instead, they should be asserting “Forget about your taste, think of problem-solving!” They should, just as design educators do, make it perfectly clear that they won’t take “Because I like it” for an answer.

I am not claiming that taste, or talent, has absolutely no role in styling/design, but rather an immensely overrated one. My point is that making good styling/design is an ability that can be learned within a rational discourse, in a way that the taste metaphysics discourages us from. I propose that the judges in “Bugün Ne Giysem?” should stress in every opportunity that personal taste or “liking” has little to do with the success of styling, that styling is a problem-solving and system-constructing process with predetermined constraints and objective reasons for action, and that one can have rational discussions about those reasons and their results beyond “This is what I like and fuck you if you don’t”. They should do this in order to make a permanent impact on their target group, providing them with the paradigm shift that will illuminate all of their styling decisions in the future. In short, they need to teach design-thinking instead of just giving tips.

Or maybe they need not. This is just a stupid TV show in the end.

[In case you want to watch it, you can do so online here at 15:00 or 02:00 EET.]

Facebook Twitter Email Stumbleupon Tumblr Digg Delicious Reddit Linkedin

Dawkins vs Gould: Survival of the FittestDawkins vs Gould: Survival of the Fittest by Kim Sterelny

I had low expectations for this book because of its (typographically) cheesy cover design and clumsy typesetting. In the end it made me want to read his other books.

The author uses the debate between Dawkins and Gould to walk through many important topics in evolutionary biology, mentioning lots of other scientists and philosophers on the way, and successfully switching between detailed examples and the big picture. He definitely knows what he’s talking about, and his writing is so clear and concise that it somehow resonates with the in-your-face 12-point Baskerville it’s set with. Even so, I would recommend it to those who have read Dawkins/Gould and are somewhat familiar with their ideas. The book ends with an amazing Suggested Reading section where Sterelny makes useful comments on every book he suggests.

In any case, Kim, if you’re reading this: change your publisher. Your book deserves better design.

And this is something I did (based on the original) reflecting where I stand on the debate:

Facebook Twitter Email Stumbleupon Tumblr Digg Delicious Reddit Linkedin