[tweetmeme]Reading the interactive design news I discovered W3C research, released for public view and feedback on October 29th, 2009. W3C created the first draft of the Emotion Markup Language 1.0, conceived as a “plug-in” language for use in three different areas: manual annotation of visual and audio material involving emotionality; automatic recognition of user behavior from sensors; and generation of emotion-related system responses. According to this document, EmotionML can be applied to numerous aspects of modern technology including sentiment analysis, monitoring, control for virtual worlds, social robots, expressive speech synthesis, emotion recognition, and support for people with disabilities. For example, in ambient assisted living systems, this language would measure simple satisfaction levels at mealtime for people with autism, as well as proactive response when things go wrong with elderly individuals, allowing for a more patient-centered rather than institutionally-centered system. Under normal circumstances – and prior to critical conditions – there is a sensation in the human body that, if recognized in time, allows the prevention of a fatal situation. Multimodal Interaction Working Group proposes a general emotional language for various applications, as well as providing default options.
Here is an example of a how simple emotion annotation might look like:
<emotion>
<dimensions set=”myFriendlinessDimension”>
<friendliness value=”-0.7″/><!– a pretty unfriendly person –>
</dimensions>
</emotion>
New language provides new ways to describe emotional dimensions by valence, potency, arousal, intensity and unpredictability. Most language specifics require a deeper understanding of the scale values, but by just examining the code, I found it amusing, and bewildering, to see numbers assigned to something that consists of more than just physical aspects. Is it then possible to capture the subtlety of human emotion? There are primary emotions that are undoubtedly recognized during typical interactions, such as fear, joy, and happiness, but there are also a wide range of complex experiences that consist of moods, feelings, stances towards objects and people, etc., that are so multilayered, momentary, and fleeting, that it seems virtually impossible to recognize them – let alone put numbers on them.
We possess a well-developed sense of emotional encounters and we experience emotions consistently. Our emotions become an inseparable part of our communication, and of our relationship to the environment. So it only makes sense for current technology to create an environment where interactions are emotionally rewarding. The recent Milo demo from the Natal project demonstrates amazing character development; making it clear that the modeling of emotion is the next step in the advancement of human-machine communication.
I find it exciting imagining an emotion markup language 10 years from now. How much of what we experience as humans could be translated into a digital language? How would our emotions be affected when they are encountered with a virtual replica? Will we experience positive emotions with digital characters more so than with real people? And what would we prefer? I asked my friends about the meaning and expression of emotion in their daily lives. The passion I heard could only come from another human and I hope this language will never be forgotten: “Emotion is a flood of feelings – taken outside of a person’s body or heart; expressed from hands and breath as music, in words as poetry, in paintings as color, and in dance as movement. Emotions are like colors of blood in your veins that carry memories, both happy and sad, and when they flow past your heart, feelings are triggered. Emotions make us feel sick, happy, full of life, or longing for life. Without emotions the earth is colorless, and there is no one to see its variance and beauty.”