Matt's Blog

Content

Sat Nov 18 14:02:33 EST 2006

The important thing about communication is content not style - true or false? In my opinion, both. Communication is the act of constructing a mental model of the world and conveying this model to another person to internalise and integrate into their mental model. This is accomplished through use of some communications channel, such as plain text, speech, demonstration or interactive dialog. Which method is most effective at conveying your meaning to the receiver depends on the receiver, not the transmitter. Using a physical analogy, there needs to be the correct impedence matching for maximum transmission of the signal.

Adding to the problem is the fact that people may be unreceptive to the idea that you are trying to convey. The entire advertising industry is based around the idea of convincing people to buy products by making them want to buy the product ie by altering their mental model to one that includes the product as a necessary requirement for happiness. Similar reasoning applies to political ideologies, religeous belief and education. This is basically Richard Dawkins' memes concept. There is only space for a certain amount of programs to run on the human brain at a time, so the organism builds up filters to prevent useful knowledge from being overtaken by ideas put forth by other agents for their advantage rather than yours. Think of the analogy with spam filters on email systems.

So assume for the moment that your receiver wants to receive your message. They may even actively seek it out, say by going to your website, subscribing to your news feed, having a screen reader verbally read out the content, or by contacting you by email for more details. In all of these cases it is the receiver who decides how they can most effectively make use of the information that you are conveying.

Contrast this to how publishers usually determine their output format. Traditionally the publisher considers what communications methods are available, for example HTML documents, and then coerces their content to fitting this format. This is a model of communication that expects certain capabilites of the consumer, in this case that they have a standards compliant web browser that is able to parse and display HTML documents. This assumption means that the content producer determines to a large extent how their content is used.

  • Sidenote: The drawbacks of this model is that users may want to use your content in ways that you haven't considered. To some extent the content producer is free to ignore the demands for different presentation formats, however this is at best annoying to the consumer and at worst illegal, for example under disabilities leglislation.

A problem occurs when new output formats become popular, as existing content must be reformatted to the new format, for example from Word documents to webpages to RSS feeds to (whatever comes next - chatbots?). XML and XSLT have been developed to give a generic description language to define document formats and how to transform between different formats.

Hack number 15 of "Mind Performance Hacks" by Ron Hale-Evans describes different ways that people learn. The VARK model focusses on the different input channels that a person has access to, while the complementary Honey and Mumford model emphasises how people make use of the information once they receive it.

  • VARK channels

    • Visual stimulus eg diagrams and maps
    • Aural stimuls eg lectures
    • Reading and writing eg plain text
    • Kinesthetic stimulus eg remembering a concept by remembering their physical location when they learned the concept
  • Honey and Mumford model

    • Activists learn by making use of the new information
    • Reflectors tend to take time to examine all the consequences of the new information and making sure that they have all the necessary data
    • Theorists try to make the new information fit into a rational scheme
    • Pragmatists focus on how the new information can help them solve an existing problem

In the context of communication over the internet currently only the visual, aural and reading/writing channels are available (although we're working on it!). So you can communicate with diagrams, words or speech, whatever you as the content producer wish. Or you can produce content in a generic format and have the content consumer process it into a form that they want.

Ten years ago there was the web. Five years ago someone realised that some people prefer to get updates in newsflash form so RSS feeds were developed, podcasts became more popular and so on. What will happen in the next few years? At the moment we are in the "Web 2.0" stage, which apparently means people acting as content producers as well as consumers. Will the next stage be production of content in a generic form, to be processed as the consumer wishes?

A content producer could publish content as a generic document, as well as publishing the description of the machine that they would prefer the content be viewed with. The consumer is however still free to use another machine that they are more comfortable with.

None of these thoughts are new, this was the original goal of SGML. I came to consider it now due to updating my blog software by extending its use of abstract datatypes and at the same time considering adding newsfeed support. Then I thought "Why just newsfeed support? What other ways can I present the content?". Some possibilities:

  • usual HTML document with writing and images
  • newsfeed for more concise description and a link to the full content
  • audio or video presentation (not with a 1GB per month bandwidth limit! Although maybe with torrents...)
  • interactive conversation using a chatbot
  • interactive presentation using an adventure game

I like the chatbot idea in particular - define a knowledge base of the chatbot from the underlying content, then use a standard javascript implementation to run the bot. But focus on newsfeed first (since I roughly know how to do that).

[code]

[permlink]

code (24)

erlang (5)
ideas (19)
lisp (1)
me (11)
notes (4)
ocaml (1)
physics (45)
qo (7)
unix (6)
vim (3)