|InfoVis.net>Magazine>message n║ 64||Published 2001-10-29|
|TambiÚn disponible en Espa˝ol|
The digital magazine of InfoVis.net
We have talked about the semantic web and its possibilities regarding information visualisation in other issues (number 26┬á┬á and number 62). We have seen how the inclusion of information regarding the semantics of contents can help in allowing the machines to manipulate and interpret the information present in the Net without human intervention.┬á
But XML and the semantic web techniques can also help to achieve better communication between humans. At least that's what the members of OASIS think, the consortium for XML based interoperability, that announced last August the creation of a committee to promote the specification of a language allowing you to include human communication features in XML.
A quite rudimentary form of this already exists in the widespread emoticons like ;-) or :-o among many others, that appeared with the use of Internet, especially in e-mail and chats.
Nevertheless the purpose of this committee chaired by Ranjeet Kumar Thunga is much broader (see the announcement of its creation). According to Thunga┬á┬á this specification will include contextual human characteristics like cultural, social, kinesic (body language), psychological and intentional features within information
The goal is to reduce the possibilities of misinterpretation and allow humans to communicate over Internet in a deeper way, eliminating to some extent the cultural barriers and the lack of information about the emotions, intentions and even body movements and gestures.
Applications of Human Mark-up Language are multiple. Maybe one of the most evident is virtual reality. An avatar could help to visualise the contents of a web site including gestures, interpreting the intentions end emotions coded in HumanML, as well as subtle aspects of human behaviour like non verbal communication.
Other applications mentioned by HumanML promoters are artificial intelligence, art, advertising, agents, conflict resolution, cultural dialogue or psychology. In this last section there could be the possibility of creating containers able to include cultural background, personality, contextual information that could be relevant for the diagnosis of the psychological disorder. Another option is to create specifications to express analysis based on the culture, neurology, beliefs, etc.┬á
Maybe sociologists could benefit from the possibility of visualising emotional information and its evolution over time of more or less extended social groups.┬á
The work of the committee is being transformed into several drafts that describe classes, elements and attributes of HumanML and some sample schemes related (for the moment) to emotions, gestures, intentions and, perhaps the most bizarre, that of politics (!).
Emotional information has been slippery for machines and even for humans up until now. The inclusion of this type of information in the web opens new visualisation possibilities, in its widest sense, that of forming a mental image.┬á
Mental images that we humans form have visual and non visual components, like olfactory tactile and, in general, of perceptual type. There's no doubt that taking all these aspects into account can help to better understand information (and also to produce more powerful disinformation).
A related article written by Thor Olavsrud appeared in Internetnews. There's another article┬á by Kendall Grant in XML.com, written in a more critical mood, about XML but mentioning HumanML. The HumanML home can be found at www.humanmarkup.org┬á
Links of this issue:
Subscribe to the free newsletter