The Key to the I Ching

[Author’s note: I wrote this article about ten years ago, when I was very interested in the Yi Jing, or I Ching, and how it can be applied to everyday life. While my interest in the Yi Jing has waned, I have received many compliments about this article and its pertinence to understanding this cryptic book. For this reason, I have edited it slightly and posted it here on my blog, instead of in its previous location on my “old” web site.]

The main elements of the Yi Jing, or I Ching, are its hexagrams. These 64 figures, made up only of solid and broken lines, are the foundation of this book which has come to us through more than two millennia, but nowhere in the book is there an explanation of what these hexagrams really represent. It is as if the Chinese of the Han Dynasty did not need a user’s manual to use the book, that the mere words used to describe the situation presented in the hexagram were sufficient. This must be the case, because their diviners knew this system perfectly, and did not need to explain the obvious. Unfortunately, time has gone by, and we do not have this knowledge, this information that they transmitted orally and never put down in black and white. We need to examine this problem if we truly want to understand the Yi Jing.

Many people have written about interpreting the Yi Jing, often by explaining the importance of the lines, trigrams, nuclear hexagrams, and the other permutations that arise when casting a hexagram, but I do not think anyone has clearly explained exactly what a hexagram is, which is the key to understanding any interpretation of the Yi Jing. Different, seemingly unrelated fields, such as linguistics and psychology, can give us new insights into some aspects of this question. This sort of multidisciplinary approach, which has the advantage of examining things from the outside, will allow us to answer the question, with almost total certainty – what is a hexagram?
The Chinese point of view

First of all, what do the Chinese think about this? A hexagram, just like a trigram, is called a gua. This word can be defined as a “pile of divinatory information”. But this word only describes a physical or visual object, it does not explain what a hexagram is, just what it looks like. When the Chinese talk of the idea behind a hexagram they talk of a shi, which is often translated by “moment”. But what is a moment, both for the Chinese and for us westerners? For Westerners it is “a short period of time”, an indivisible, ephemeral unit of time. We tend to think of this as the smallest such unit (at least in common language – it is obvious that some sciences use extremely short units of time to measure events). The Chinese have a totally different concept for the moment. A moment is a situation. It is the son of the past and the father of the future. This word, shi, is used in different expressions to talk about seasons, times zones, chances, opportunities. A situation is far from indivisible, quite the opposite: it is a fence which holds together all of the related moments of an event, which are seen as a whole.

Let us use the word situation to talk about what happens inside a hexagram. This word can clear up a number of points. A situation can be seen in two different ways, and have two totally opposite interpretations. Seen from the outside, a situation seems frozen, not without a relationship to what came before it, but independent of this context, because you can only see the actual moment, and not its evolution. On the other hand, seen from the inside, a situation is quite different. The moment that is seen is lived through, and when you are on the inside you have to distinguish the relationship between the past and the future. It is dynamic, and you can not separate it from what came before it. It is just as difficult to envisage a moment without taking into account its possible evolution, what it may become, whether desired or not, because these evolutions are all present, in the form of possibilities. The relativity of the point of view changes the way the moment is perceived.

Looking at two sides of a coin

The Yi Jing is information, in its rawest form. The sentences in its text are short, concise, and contain no redundancy. It is often this redundancy, however, that helps us understand a text. This naked text is one of the most daunting features of the Yi Jing, since its information does not give much meaning. In fact, the path one must follow to go from information to meaning is a long one, which I will briefly sketch out here.

The question of meaning is central to any discussion of written texts. Meaning is not inherent to a text, it is based on the reader’s interpretation. Since the text is static, there is no direct negotiation of meaning between the writer and the reader. The reader can not ask questions to the writer, but must be responsible for finding all the clues the writer has left so the reader can work out the intended meaning. There is an interactive relationship between the reader and writer, but this relationship is realized through the text, not with the text. This means that the reader can never be certain whether the meaning extracted from a text corresponds to the writer’s goal. Understanding can never be complete: it can only be approximate, and relative to purpose. Not only is comprehension relative to purpose, but it is also relative to the amount of information, both textual and other, that the reader is able to process. “Computing the intended meaning of a speaker/writer depends… on knowledge of many details over and above those to be found in the textual record of the speaker/writer’s linguistic production.” (Discourse Analysis, Brown and Yule, Cambridge University Press, 1983 p. 116)

Meaning is not information; information is not meaning. In this digital society we tend to take for granted that the two are similar. As I write this article on my computer, the words I am typing are converted into the simplest possible form of code so the computer can work with it. This binary code, a code made up of 1s and 0s, is as rudimentary as possible; no code can be less complex. This is paradoxical, because the computer, a machine that can calculate, can do many operations that we, humans, cannot do so quickly, cannot even count to two. This is because the computer is working with information, not meaning.

Even the words you are reading are only information. They are made up of another code, one made of symbols, that we call an alphabet. This alphabet contains roughly 100 such symbols, letters and punctuation marks, that combine to form words, which in turn combine to form sentences, and so on. (I say about 100 symbols in our alphabet, because all the punctuation marks, numbers, and both capital and small letters are different symbols. It is tempting to talk of an alphabet containing 26 letters, but this is an oversimplification.) In order to understand this information you need, first of all, to understand the code. There are two forms of code used here: the letters, and the language.

Many languages, such as Chinese, use a different writing system than we do, and if the reader cannot interpret this system he or she will go no further. In this form of language there is no correspondence between the written symbols and their pronunciation. This makes it difficult to figure out a word one has heard, but cannot read. One must also know the language, of course, for even understanding the first level of code, the alphabet, or the symbols, does not open up the combinatory possibilities of this code.

Let us assume that the reader knows these two codes, he or she must now go up a level to extract the meanings of each word used. Word meaning is a very complex thing. Some words are relatively simple to define, and, therefore, their meaning is not relative to any other information. Words such as tomato, zebra, and chair can be explained by pictures or physical examples. Other words, however, have more complex meanings which require that they be defined in relation to other words, ideas or situations. What is the sky, how do you explain deep, how do you make someone understand the concept of fear? What is more, many words have multiple meanings, and the meaning one must choose in a given situation is relative to the context where the word is found. A word such as pound could mean to hit, an enclosure for stray dogs, a unit of weight, or a unit of British currency. The situation and context of a given sentence will help the reader to decide which meaning is appropriate.

The next level of interpretation is the relationship between word meaning and sentence meaning. As we have seen, the meaning of many words is relative to its context, and, at this level, context includes the surrounding words. The interpretation of a sentence interacts with the interpretation of words in order to create an idea. But even the meaning of a sentence is dependent on its context. A sentence like “It furthers one to cross the great water” has a very precise meaning in the context of the Yi Jing, but elsewhere it would have a slightly different, perhaps less metaphorical meaning. The reader must, therefore, take into account the overall context of the text he or she is reading.

We have so far looked at four aspects of meaning: codes, words, sentences, and context. These four features make up only a part of what is necessary for meaning to emerge from a text. At this point the reader has processed all the information given by the text. Now, the dynamic interrelation between reader and text is shifted over to the reader’s shoulders, and the reader will make out only as much meaning as he or she can, based on knowledge that goes beyond the text.

Take for example hexagram 48, The Well. The idea a Western reader makes of a well is that of, for instance, the well that may be found within the courtyard of a castle, or maybe of a fountain in a small village in Provence. Already these two types of well imply different situations, but neither matches exactly the situation of a well in ancient China. If you look at the character used to write the word well in Chinese, you will see a graphical description of the nine parcels of land that make up the area around a well. There were eight parcels belonging to eight families, and the ninth central parcel contained the well. This parcel was kept up, in turn, by each of the eight surrounding families, and the crops harvested on this land went to the Lord as taxes. One can add to this the social aspect of a well, being a meeting place where information was exchanged among the families, but this background information about the upkeep of the well and its surrounding land is, in effect, vital to the understanding of the situation. Not knowing this means that the meaning extracted from a reading of this hexagram will not correspond exactly to the intended meaning. As I said before, it is impossible to extract meaning that corresponds exactly to what was intended, but the more background information the reader has, the closer he or she will be to that intended meaning.

There is another factor that affects the interpretation of meaning, and this factor, I will argue, has a major role in the interpretation of the Yi Jing. Cognitive science proposes a theory of knowledge called schema theory, which, we will see, can explain why we have difficulty understanding the Yi Jing, and will give us a new outlook on how we may go about understanding what it tries to tell us. Schemata are the key to the Yi Jing.

Schema theory was born in the 1970s as researchers in cognitive science attempted to explain how knowledge is processed in the brain. A number of researchers have proposed alternatives to this theory, such as scripts, frames, etc. While these concepts are not entirely synonymous, they are similar enough that a discussion of one of them will bring forth ideas inherent to all of them.

Schemata are the basic units of knowledge. A schema is an abstract, internal mental representation of an idea, event, an action, or a situation. Meaning is seen as being encoded in different schemata, which also contain information about how such schemata are interrelated. Schemata also contain the default knowledge of a typical, or even a stereotypical member of its class. If you hear the word “dog”, you think of a stereotypical dog, which may be different for you than for me. If you have a better relationship with dogs than I do, you will also have a schema which includes that affective appreciation. There is a relationship between the memory of past situations and current interpretation. If the schema changes, which all schemata do over time, the memory is added to the new information to create a revised schema. So if one day I develop a positive relationship with a dog, my schema for dog will change.

In order to understand how schemata function, I will give a few simple examples. I will then explain how this theory can be used to explain the Yi Jing.

First of all, schemata can act as a visual representation of something. Imagine that you are walking down the street, and you see a person walking toward you on the sidewalk. From far away you can tell it is a person, and, maybe, if it is a child or adult, a man or woman. This is the schema that represents the physical form of a person that is activated. As the person gets closer, you may be able to see about what the persons age is. This is a schema that adds information to your original information, by many possible means. It may be the way the person walks, the type of clothes they are wearing or some other information that helps you come to this conclusion. As the person gets even closer, he or she may look familiar, it may be someone you know. At some point, the number of features you have been able to see on the person converge toward the visual schema you have of a particular person. At this point you know who the person is, and you can not, for the time that you see them, forget who it is, or need more information. Knowing, in this sense, is absolute. It is a question of yes or no. You may however, realize that it is not, after all, the person you thought it was, and at this point the schema which represents the first person will be transformed into that for a different person.

In fact, a great deal of perception is based on hypotheses being confirmed like this. One may look at an object and think it is a certain object, but on a closer look realize it is something else. But let us look again at that person walking down the street. They are getting closer now, and you can see it is indeed the first person you thought of, but they have cut their hair. Now, your mind revises the schema which contains the information about this person to include short hair instead of long. You would be able to recognize the person in another situation with long hair, such as in a photograph, but now the schema for that person includes the possibility of two different hair styles. The previous schema has not been overridden, nor overwritten, just modified. This is always happening to schemata, they are constantly revised according to our interaction with any situation or object. Some things are static, and therefore cannot change their characteristics (rituals, objects), while other things are dynamic and are constantly revised in the mind.

Another analogy which will help understand schemata is that of a play. One could say that a schema is like a play, with actors, props, situations, and a script. In the same way that a play may be performed by different actors, in a different setting, at a different time, even in a different language, it is still, more or less, the same play. Hamlet in Chinese would still be Hamlet.

If, for example, I tell you about a restaurant where I recently had lunch, your mind will activate the appropriate schema, and bring forth the appropriate props and actions to help you predict what I will tell you. In a restaurant there is a table, chairs, a menu, a waiter or waitress, food, and a bill. There are actions such as reading the menu, ordering, eating, and paying. Some of these things may be different, it may be a self-service restaurant, but the overall idea is the same.

Schemata like this are an integral part of our social interaction, but the schema itself is no more than a skeleton around which the salient information is added. It can be seen as a basic model of a situation or action. If the schema we are using to interpret a situation does not correspond to the actions or actors in a situation, we are surprised, sometimes to the point of not understanding. If there are major differences, I will have to explain them to you, since I will know that they do not correspond to the default schema for restaurants. For example, if there was a musician playing in the restaurant, or if the waiters sang ‘Happy Birthday’ to the person I was with, I would have to explain it, these ideas are not part of the default schema for a restaurant. If the waiter tells me I must cook the food myself, I will be not only surprised, but maybe outraged, since this does not correspond to my expectations. I may go to another restaurant, since one of the main reasons for eating out is to not cook.

Schemata and text

Reading is a complex process. It seems simple for us, because we are so used to it. It can seem to be simply a question of deciphering words on a page and making sense of them. But we have already seen how making sense depends on many things. One thing that helps, or hinders making sense of a text is the knowledge the reader has of the inherent schemata.

In order to understand a text the reader must be able to make the connection between the words read on the page and the appropriate schemata in his or her mind. In most cases, this is not a problem. This happens subconsciously so the reader is not at all aware of the work that the brain does. The reader is, however, aware when something does not fit. When the reader does not have the appropriate schema he or she simply cannot understand the information being read. This is the case when someone tries to read a text dealing with a domain that the person is totally unfamiliar with. The words may make sense one by one, but there is no sense at all to the text as a whole.

In other cases, the reader may have the appropriate schema, but may not be able to activate it. This may be because the clues given by the writer are insufficient to help the reader recognize what is being discussed. In this situation, all that is necessary is that the reader find additional clues. One can observe this sometimes when after having read a text and not understanding it, one goes back to read it again, and finds it much easier. This is because the ideas behind the text have become familiar, helping the reader to awaken the schema necessary to understand it.

Sometimes the reader may be able to interpret the text, but not find the interpretation that the writer expected. The appropriate schemata are available, but the reader does not understand the author.

In addition to schemata that describe experiences, events, and actions, there are also what could be called cultural schemata. These are schemata that are firmly rooted in a particular culture, and lead the interpretation of particular information in a culture-specific direction. Since we are talking here about the Yi Jing, I will briefly look at some of the ideas that come from Chinese culture that fit this heading.

We have already seen that the idea of a well is different in China and in the west. The object is the same, but the way it is used and perceived is very different. Hexagram 50 talks about a ritual vessel called a Ding. This is something that dates back very far in Chinese culture, and that we need an explanation for in order to understand its significance. (See the preface to the Wilhelm/Baynes translation of the I Ching by C.G. Jung.) There are also other objects, such as belts and robes, that have no meaning for us without an explanation.

“Crossing the great waters” is an expression that appears many times in the YI Jing. For anyone who has visited China the strength of this phrase is evident: rivers there are often very wide, deep, and dangerous. Crossing a river, in ancient China, was a difficult task.

The idea of the Superior Man is another important idea that needs explanation. It represents the ideal of a man who is acting the correct way in a given situation. The translation used by Wilhelm, superior, does not help to understand this, and, in fact, only makes it more obscure by adding on a concept, that of noble birth, which does not have its place in the Chinese term.
These are just a few ideas that need clarification so the western reader can make sense of Chinese ideas. The YI Jing is full of such ideas, and the biggest problem is when the words used bring forth schemata that correspond to a western idea, such as noble, that is far removed from the concept in Chinese.

As I said at the beginning of this article, a hexagram is a situation. Each situation in our life corresponds to one or more schemata, and each of the hexagrams corresponds to schemata also. Using the idea of schemata for an analysis of the Yi Jing would permit a much simpler approach to the Yi Jing.

All this finally brings us to an examination of schema theory and its relevance to the Yi Jing. We have seen how schemata are necessary to understanding a written text, and how meaning is relative to a number of variables. The Yi Jing has the particularity of coming from ancient China, where both the cultural differences and the time differences are very great. In order to try and understand the Yi Jing it is necessary to find the relationship between the ideas presented in the text and similar ideas that we may be able to understand today.

But the very thing that makes the Yi Jing stand out also makes it very difficult to understand. We do not have the schemata that make up the heart of the Yi Jing. Our culture is so far removed that the best we can do is incorrectly interpret something that seems similar. Without these schemata we are lost, the text seems to make no sense sometimes, and even when it does seem to make sense we cannot be sure that our interpretation is correct.

Today’s hexagrams

If the Yi Jing were written today, it would be necessary to use situations, and schemata, that correspond to our world-view and our understanding of the interrelations of the world. Some hexagrams would talk about politics, and we can imagine one called Cohabitation. This hexagram describes a situation where the emperor is required to rule with a minister who does not think along the same lines as he does. [This refers to the political situation in France, where I live, at the time I was writing the article. As the reader can see, this reference is already obscure, especially to those outside of France.] The Landing would be a hexagram describing how a coalition of foreign armies comes to help liberate a country that is occupied. Or The Old Bridge would describe the symbol of a beautiful centuries-old bridge that is destroyed in a country splitting apart during a bloody civil war, where no other countries come to their aid.

Hexagrams like this are related to situations that we know, that are current. It takes little explanation to understand the situation, and the metaphors that are being presented. The Yi Jing is like that. If we look closely enough at what is being described in the hexagrams, we will find similar information. Once we have discovered the situations described, we can look at them as schemata for other, similar, or metaphorically related situations. When casting the hexagram The Army, it is rarely a question of army, but a metaphorical resemblance to the idea of army. This background of schemata within the hexagrams is present in all 64 of them.

What we need to understand the Yi Jing is to discover the schemata that underlie the 64 hexagrams. No translation currently available can help us do that, because most of them have been made by people who are ignorant of the very concepts that made up ancient Chinese culture. In fact, no translation can translate these concepts. It is necessary to explain them, since they go beyond the words of the text, they are the elements by which the Han Chinese could make sense of their world. Any explanation would include a similar situation related to our world-view, which would enable us to make the connection between the idea in the YI Jing and a similar idea today.

Any translation must respect the text being translated, but a translation that translates only the text and not the ideas within is worthless. Many of the ideas in the Yi Jing are what could be called archetypal ideas, that can resonate even across many centuries, but even those must be discovered. The key to the Yi Jing is simple. We must go back and look at the way the Chinese lived at the time of the Yi Jing, look at their habits and their world-view, and find equivalents in our modern, western world. Only then will we truly be able to understand how the Yi Jing functions. Only then will we be able to use this extraordinary tool that can enable us to discover in ourselves that which we could not find without the aid of this book.

People have asked me to recommend a translation of the Yi Jing. Unfortunately, I cannot recommend any current English translations, but this French translation, Le Yi Jing : Le livre des changements, by my friend and colleague Cyrille Javary, who inspired this article, with Pierre Faure, is probably the closest to what this article pleads for. His other books, Le Yi Jing : Le Grand Livre du Yin et du Yang, Les rouages du Yi Jing, and
Le Discours de la tortue give a great deal of insight into the Yi Jing. Cyrille Javary is one of the most knowledgeable westerners when it comes to the Yi Jing, and especially its historical signification. His work ignores the “new age” interpretations of the Yi Jing, and attempts to reconstruct the mind-set and conceptions of the Han Chinese.

Getting the Most out of Classical Music with iTunes and the iPod

[Update, September 2006. Apple introduced gapless playback to iTunes 7 and to the latest iPods, making the questions of joining tracks, as explained below, moot in many cases. See this article for an explanation of gapless playback.

However, if you have an older iPod (older than the iPod video or nano), you won’t benefit from this feature. In addition, you may still want to join tracks to be able to play music at random, playing entire works, rather than disparate movements. So much of this article remains valid today.]

While Apple is aggressively marketing its iPod to the younger generation, through its ads and commercials featuring black silhouettes dancing to hip-hop and rock music, the iPod is also a valuable device for listening to classical music. However, to get the most out of this type of music, you need to reconsider the way you rip your CDs.

I’ve got eclectic musical tastes. My iPod contains music by the Grateful Dead, The Durutti Column, The Clash, Brian Eno, moe. and Widespread Panic, as well as Bach, Haydn, Handel and Schubert. I’ve long explored all types of music, and the capacity of my iPod lets me carry a diverse selection of tunes with me.

Read more

What is Gapless Playback on the iPod and in iTunes? (Update)

One of the major announcements that Apple made on September 12 was that the latest video iPod (the 5G model) offers gapless playback, and that this feature is available for previous video iPods when updated with the latest firmware. While those who know what gapless playback means embraced this with a loud “Huzzah!”, others are scratching their heads trying to figure out what this means. Here’s an overview of what gapless playback is, how it works, and why you might want it.

Gapless playback is simply the ability for the iPod to play music with no artificial gaps between tracks. For most music, you’ll never notice the difference, but if you listen to operas, dance mixes, or classic progressive rock albums (the standard examples are Pink Floyd’s Dark Side of the Moon or the Beatles’ Sergeant Pepper’s Lonely Hearts Club Band), you’ll hear a jarring half-second space between tracks that are meant to continue seamlessly. This lurch is the audio equivalent of an elevator dropping a floor suddenly, and ruins the listening experience. Fans of live music (The Grateful Dead, and other jambands) especially hate this, since this type of music often has songs that segue from one to another with no break. Finally, Apple answered the call, making the new iPod the only current portable music player that offers gapless playback. (Note: naysayers will point out that the Rio Karma has gapless playback; I said that no current portable music players offer this feature. The Rio Karma is no longer being sold.)

When I wrote about gapless playback here about a year ago, I offered a workaround that would allow classical music fans to rip their music so it could be gapless. The idea was that you simply join tracks when ripping albums, so instead of multiple tracks with numerous hiccoughs, you’d have a single track for an entire work. This is okay, but it’s a workaround; you wouldn’t be able to find which track you were listening to. Now, with gapless playback, operas can have individual tracks, and you won’t heard the difference.

Of course, there’s a down side to this: now classical music fans who did rip their music with joined tracks will probably want to re-rip these discs; I know I’ll be re-ripping all my operas. Sigh.

So how does gapless playback work? You probably noticed that the first time you launched iTunes, the program took a few minutes to analyze your music for gapless playback. iTunes looked at your tracks to determine if there was no silence at their ends to flag them as tracks that would flow smoothly into the following tracks. This works in iTunes, as well as with all 5G iPods and the new iPod nano. However, this does not work with previous iPod models. Curiously, this process works automatically, though there is a “Part of a gapless album” tag that can be set for individual tracks. It’s not clear if this tag exists to turn on gapless playback or to turn it off. Initial reports say it’s not needed for gapless playback (it certainly isn’t needed with iTunes); I’m awaiting a new iPod to test it there and figure out exactly what it’s for. (Yes, I don’t yet own a video iPod.)

Stay tuned for an update to this article in a week or so when I’ve fully figured out how this tag affects playback on the iPod. For now, be happy (if you find gapless playback useful) or yawn if you don’t. I, for one, applaud the fact that Apple listened to its users and provided this needed feature.

Update: Apple has cleared up the question about the gapless tag. They say this tag only matters if you have crossfade playback turned on in iTunes, and it only affects playback from iTunes. All gapless albums are automatically detected and played as such on 5G iPods (video-capable iPods) and 2G iPod nanos (the latest versions of the nano).

iWish: iTunes “Works”

(Click here to view the above image full size: 108 K.)

I’ve written in the past about the complications of using iTunes and the iPod to store and play back classical music. Well, there’s a simple way that iTunes could improve the user experience for classical music fans, and it’s something that the iTunes Music Store already uses.

[Since I first posted this article in late 2004, nothing has changed. I’m re-posting it now just in case anyone at Apple is reading my blog, and happens to notice this. We classical music listeners need this functionality in iTunes and the iPod.]The iTunes Music Store lets you purchase “works” for many classical albums. While some are simply presented in a list of movements, others have works grouped together. As you can see in the above screen shot, Apple groups multi-movement works together to make it easier to purchase single works rather than entire albums; it enters the name of the work in the Grouping tag. But why doesn’t Apple use this same possibility in iTunes after you buy the music? And why can’t you use this same grouping in iTunes for your own music?

While I point out in this article that it’s easy to join tracks when importing them, this has drawbacks: you can’t see the names of individual parts of a work (which is especially annoying when listening to an opera or other long work with many parts) and you can no longer choose to listen to a specific part or movement.

Clearly, iTunes has this ability; at least for the iTunes Music Store. This would be a great feature to provide in a future version of iTunes, not only for classical music fans, but even for those who want to group their albums together in this way. In fact, you can see this in action on the iTunes Music Store listing for the Complete U2 set. Each album is listed at one level with the contents of the album at the next level.

Writing Conversation: An Analysis of Speech Events in E-mail Mailing Lists

Nearly ten years ago, in what was another life, I completed a Master’s degree in applied linguistics with Aston University in Birmingham, UK. My dissertation was about a subject that was, at the time, relatively new: e-mail.

I’ve had this on my web site ever since, and I have received a great deal of feedback about it over the years. While it probably won’t interest many of my readers, I thought it was worth dusting off and mentioning here on Kirkville. So, if sociolinguistics interests you at all, you can read the entire paper here.


Imprecision in Journalism: How the iTunes MiniStore was Reported

With the recent kerfuffle over the iTunes MiniStore and privacy, I have written several articles about the issue, been interviewed by a number of web and print media, been interviewed for podcasts, and invited on a national business channel to discuss the issue. This issue annoyed me from the beginning: the fact that iTunes was both sending personal data to Apple and other companies without warning users, and the fact that iTunes was displaying “recommendations” (that is, ads) when users were not “in” the iTunes Music Store.

But now that Apple has corrected this problem, what really annoys me is the level of journalism I have seen about the problem. From mistakes to clueless writing, from minor technical errors to stupid comments from writers who clearly no nothing about technology (and probably cannot read very well, since I and others have very clearly written about what the iTunes MiniStore does), these errors are legion.

So, here’s an overview of some of the statements I have found that are incorrect, and, in some cases, border on incompetent. Tell me how journalists can get something so simple so wrong…Nick Farrell, writing for The Inquirer, said the following: “There were claims that you had to be a computer expert to know how to switch the data collection facility off, which many of its users are not.” Nick, show me where anyone said that you need to be a computer expert to click a button or select a menu item… Do you run iTunes at all?

Many articles, including this one in the LA Times, an editorial to boot, said things such as, “Besides, Apple said, it didn’t store any of the information it received.” This comment makes me snicker. My friend Rob Griffiths, who wrote an article about the iTunes MiniStore shortly after I published mine. Griffiths received an email from a high-level Apple official to this effect (that Apple was not storing information), and added the following update to his article: “…an Apple official told Macworld that the iTunes MiniStore feature does not collect any information from users.” But at no time did Apple issue an official statement about the iTunes MiniStore. Media comments, such as those mentioned by the LA Times, suggest that this is the case, but they are simply relying on a comment from a journalist about an email he received. Sigh.

Louisa Hearn, writing for The Age in Australia, explains how the MiniStore displays a warning, but goes on to say “Although the MiniStore is turned on by default for new customers, a pointer at the bottom of the playlist page allows them to switch it on or off.” This one is interesting. The article includes a screen shot which shows the Turn on MiniStore button, but this journalist seems to think that the MiniStore is on regardless of this button? The button turns it on, not off, as the button says.

The CBC (Canadian Broadcasting Company), in an article about the iTunes MiniStore, repeats another fallacy that has spread, showing that their journalist didn’t do any fact-checking. “The software scans a user’s existing downloads and recommends new songs to buy,” the article says, which is blatantly false. The iTunes MiniStore only sends information about songs that users click. This suggests that iTunes is sending information about users’ entire libraries. You get an F, unnamed reporter, for this one.

PC Magazine took a different tack. In this article, they tried to sound like investigative network journalists, using advanced forensic techniques to get to the bottom of the question. Journalist Oliver Kaven says such wondrous things as, “Here at PC Magazine, we began dissecting the issue, one IP packet at a time.” Ooh, like CSI but with computers, right? He goes on to say, “We found that this can be prevented by minimizing the MiniStore application or by playing songs from a play list.” Hmm… He didn’t need a packet analyzer to find that no data was sent, but at least he was checking the facts. I reported this, saying, “However, when the MiniStore is hidden, iTunes does not send these requests. You can therefore protect yourself from Apple’s prying eyes by simply hiding the MiniStore,” as did several other web sites. However, “playing songs from a play list”? That’s not entirely correct. What he should have said was that the MiniStore only sends data when you click on a song. If you double-click a song to play it, iTunes sends information about that song, but not about subsequent songs in playlists or albums. Guilty of over-exaggeration, and of a minor error, PC Magazine tried to turn this into a detective story.

What surprises me in all this is that some of these “major media outlets”, such as the CBC or newspapers, have reported this issue with more errors than most bloggers. Not that I believe in the Easter bunny, especially given recent scandals at top-tier newspapers like the New York Times, but I would have thought that these media had better structures in place for fact-checking, and that the journalists are more qualified. This said, the above examples are only a handful of negative ones; there are many more, but a majority of the stories I read about this issue were correct.

The iTunes MiniStore: Apple Does the Right Thing

It wasn’t so hard after all…

As reported here, in an article entitled iTunes: Apple’s New Spyware and Adware Application?, Apple rolled in a new feature to iTunes 6.02, called the iTunes MiniStore. (Read the above linked article for more about this feature and the security issues it raises.) Apple unofficially told a Macworld journalist that it was not collecting data, but I, and many privacy advocates, felt that this was not enough, and wondered why Apple could not simply have iTunes display a warning that explains what this features does.

But today, Apple did the right thing.Well, checking the MiniStore this morning I saw this warning:

And this shows above the button that hides the MiniStore:

I’m pleased that Apple decided to make this change, and reassured that the company has listened to its critics, and that it has reacted so quickly. (A week is not that long for such a large company to react.) Apple, you have restored my faith. Thanks!

The iTunes MiniStore: Fact and Fiction

As more pixels are being spilled about the potential spyware and adware in the latest version of Apple’s iTunes, a great deal of misunderstanding about this issue is prevalent. I’ve written about this issue several times, beginning with this article, which outlined what the iTunes MiniStore does , followed up by this examination of what Apple did wrong, and how apologists seem to want to forgive every mistake that Apple makes, and, finally, a presentation of the actual data that iTunes sends to the iTunes Music Store, including a unique use ID.

In this article, I would like to examine some of the claims that have been made about what the iTunes Mini Store actually does, and explain what is fact and what is fiction. There is a bit of both in some of the articles on the web, especially in the comments on sites like Slashdot. So read on for a reality check.

  • The iTunes MiniStore sends personal information to Apple’s servers. True, in part. It also sends information to a company called Omniture. The since1968 blog has a great article explaining more about Omniture.
  • The iTunes MiniStore sends personal information to Apple’s servers, and other servers, for every song you play, the contents of your entire library, etc. False. The iTunes MiniStore only sends this information when you click a song. If you double-click a song from an album or playlist, for example, the first song’s information is sent to Apple’s servers, but subsequent songs are not. iTunes also sends information for CDs that you insert into your computer (if iTunes is running) to either play or rip. iTunes also does not send the contents of your entire library or anything else to Apple’s servers.
  • The iTunes MiniStore sends a personal ID to these servers. True. As I explain in this article, the iTunes MiniStore sends your Apple ID (or at least its numerical equivalent) with each request for information. It also sends song information (name, artist, and genre) for music you have ripped yourself, or a unique identifier for songs you have purchased from the iTunes Music Store (iTMS). The Apple ID is used for the iTunes Music Store, for .Mac (if you have a subscription), for Apple’s developer program and other Apple services, including purchase you make from the Apple Store. The Apple ID can therefore be linked to your credit card, your address, and your purchasing habits with Apple.
  • The iTunes MiniStore does not send any information to the iTunes Music store or other servers when it is hidden. True. If you want to be sure that your personal information is protected, just hide the iTunes MiniStore by clicking the fourth button from the right at the bottom of the iTunes window, or by selecting Edit > Hide MiniStore.
  • The iTunes MiniStore sends a personal ID to these servers even if you are not signed in to your iTunes Music Store account. False. If you sign out of your iTunes Music Store account, or if you have never created one, no personal ID is sent.
  • The iTunes MiniStore sends other cookie information to these servers. True. And I have no idea what these cookies contain.
  • The information sent to the iTunes Music Store is used for the Just For You feature (a recommendation section on the iTMS main page). False. Just For You seems to only use either your iTMS purchases, or other albums that you have told the recommendation engine that you own.
  • The iTunes MiniStore display is no different from the Just For You recommendations. False. In my case, it displays albums that I have purchased from the iTMS, so, while information is being sent to the iTMS with a personal ID, it is clearly not (yet) being used to check on your purchases.
  • The iTunes MiniStore display is no different from clicking the arrows in iTunes that take you to the iTMS and show you similar music. False. Clicking an arrow is active; the iTunes MiniStore is passive (it requires you to click a song, but you may be doing this simply to play the song). There is a difference, in my opinion.
  • The iTunes license does not mention anything about personal information being shared via iTunes. True. But…
  • The iTunes license refers back to Apple’s generic privacy license. True. It links to the Apple Customer Privacy Policy. However, this document does not seem to cover the type of information that the iTunes MiniStore is sending to the iTMS. It first discusses information obtained during service calls, when you register your computer, and then says, “We also collect information regarding customer activities on our website, .Mac, the iTunes Music Store, and on related websites.” It is hard to imagine that the use of iTunes on your personal computer fits into the definition of “our website, .Mac, the iTunes Music Store, and on related websites.” Perhaps Apple is stretching it by considering that the iTunes MiniStore is part of the iTunes Music Store, but, in most users’ eyes, this is not the case–users “enter” the iTMS when clicking on the Music Store icon, not when they simply click on a song in their library.
  • Apple has said that they are not collecting any information from the iTunes MiniStore. As of now, Apple has made no official statement regarding this. The author of an article on Macworld was contacted by “an Apple official” who “told Macworld that the iTunes MiniStore feature does not collect any information from users.” However, at the time this article was written, it was not known that users’ unique IDs were being sent. While Apple may not be collecting any information now, this does not mean that they will not do so in the future.
  • Apple’s approach to collecting information is illegal. That’s for the courts to decide, should it get to that point. It is interesting to point out that Real Networks was sued in 1999 for a very similar usage of unique identifiers in its music player software. Note that European privacy laws, more stringent than those in the United States, might see things differently. Since iTunes is available around the world, it has to comply with the laws of the country in which it is provided.
  • If Apple can connect song information to unique user IDs, the RIAA might be able to subpeona this information to track down people who have illegally copied music. Um, maybe. This is stretching things a bit, but let’s look at a hypothetical. Before U2 released their last album, a master was stolen then found its way onto file sharing sites on the Internet. Assuming that this were to occur to another band, the iTunes MiniStore could potentially track users who 1) have the MiniStore displayed, 2) have such songs that are not yet officially released, and 3) click them in iTunes. Even if Apple were to collect song information and link it to user IDs, could a court force them to release this information? I’m not a lawyer, and don’t want to speculate. But the technology clearly exists for such tracking to occur.
  • Apple should have realized that there would be a privacy issue surrounding the introduction of the iTunes MiniStore. True. It astonishes me that, given the number of people involved in a product such as iTunes, from programmers to marketing people, that a red flag did not go up at some point. Or, if it did, that it was ignored. Apple should have been proactive and explained this feature from the get-go, rather than wait for users to sniff packets and find out what it is doing.
  • Apple clearly indicates this new feature on its web page. True. The iTunes web page and download page mention this feature. However, Mac users who used Software Update to update iTunes saw no information regarding this feature, but only this: “iTunes 6.0.2 includes stability and performance improvements over iTunes 6.0.1.” So Mac users were not aware of this feature, unless they went to the iTunes download page to get the update.

See other articles about the iTunes MiniStore:

iTunes: Apple’s New Spyware and Adware Application?

The iTunes MiniStore Debacle: What Apple Did Wrong

iSpy: Still More on the iTunes MiniStore and Privacy

The iTunes MiniStore: Fact and Fiction

iSpy: Still More on the iTunes MiniStore and Privacy

Some things just go on getting worse. If it wasn’t enough that iTunes 6.0.2 contains spyware and adware, now it turns out that the program not only sends information about the song you have selected to Apple’s servers, but also sends your Apple ID, or, at least, its numerical equivalent. (If you’ve missed an installment, the story begins with the link just above, then continues here.Michael Griffin first noticed this, as reported on Boing Boing, and I had trouble reproducing it at first. But I quickly found out that he was right, with the exception that his Apple ID is six digits and mine is eight. (See the updates to the Boing Boing story for more on how I discovered this.)

So, after Apple claimed that they were not “collecting” information, it now turns out that the information they send is directly linked to a user’s account identifier, if, of course, the user has an Apple ID. If you have never logged into your iTunes Music Store account, you won’t have this ID, and Apple can’t track you. But if you have, even once, this ID is stored in a preference file on your computer, and sent with each iTunes MiniStore request.

Here is an example of the raw data that is sent, taken from tcpdump output. What is being transmitted is, first of all, song info: the name of the song, the artist and the genre. Then it sends the Apple ID, shown as ######## below. (Note: I’ve inserted link breaks for readability.)

X-Apple-Tz: .3600..X-Apple-Store-Front:.143441..Referer:.http://
(Macintosh;.U;.PPC.Mac.OS.X.10.4.4)..Accept-Language:.en-us,.en ;q=0.50..X-Apple-
Accept-E ncoding:.gzip,.x-aes-cbc..Connection:.close.. .net.... 

You can also see such things as the version of iTunes, the language, and some other cookie stuff (after Apple-Validation).

It then sends this, which is more of the same (without the Apple ID), but with some more stuff from the iTunes cookies files:

Type: .text/xml;.charset=UTF-8..x-apple-lok-response-date:.Thu.Jan.12.04:46:27.PST.200 6..
Vary:.Accept-Encoding..x-webobjects-loadaverage:.0..x-apple-lok-filelastmodif ied-date:.
Tue.Jan.10.21:14:37.PST.2006..x-apple-lok-path:./opt/itms_lokamai/Loka mai/MZSearch/
kin d=song&pn=Another%20Day%20On%20Earth-143441-Ak..x-apple-date-
generated:.Wed,.11. Jan.2006.05:14:36.GMT..x-apple-request-store-front:.
143441..x-apple-max-age:.360 0..x-apple-max-age:.64800..x-apple-application-instance:.
150..x-apple-asset-vers ion:.14571..x-apple-lok-filesize:.1693..x-apple-lok-current-
stor efront:.143441.. Content-Encoding:.gzip..Expires:.Thu,.12.Jan.
2006.12:46:27.GMT..Cache-Control:.m ax-age=0,,.

Here’s more (with my Apple ID hidden again):

3600..Cookie:.asbid=sKUKC49DKFC7T4CHC;.s _vi=
[CS]v4|53C58647-6EC2D2 32|0[CE];.s_vi_jx7Bx7Bgnbx7Ffxxx7Exx=
[CS]v4|53C58647-6EC2D232|0[CE];.s_vi_ox7Ex7 Ebkx7Bx7Dyyygzcx7D=

Most of what is in this part I have found in my iTunes cookies (in the file).

And for a minute, I was thinking that this would all blow over quickly…

See other articles about the iTunes MiniStore:

iTunes: Apple’s New Spyware and Adware Application?

The iTunes MiniStore Debacle: What Apple Did Wrong

iSpy: Still More on the iTunes MiniStore and Privacy

The iTunes MiniStore: Fact and Fiction