Skip to content

‘the ultimate knowledge environment’

So here I am, working in the ultimate knowledge environment, or at least something quite far beyond what exists in 2015.

The first thing to note is how deeply connected and richly interactive everything is. I can instantly look any text up in a myriad of sources, change the colour of the text based on grammar, meaning, contents – anything I choose. Numbers can be automatically converted to whatever system I prefer, whatever currency I need.

I’m reading a document where I come across an assertion which has a citation attached to it. A click and I can see all I need in order to decide whether the citation really does back up the assertion; who is behind it and what credibility the wider community has attached to the citation. I can access the original document where the citation was made, straight through to the paragraph it was made.

In this case I don’t agree with the way the author has used the citation and I make a note of this in the document, easily tagged to make the author aware of my position, with my own citation backing up my point – in a way the author can easily interact with – even though 3 million other people are also annotating this particular document.

I come across a dense, listy section and choose to view it as an exploded mind-map type view to better see how pieces relate. I can do this with any parts of the text and easily dial in whatever variables I choose to decide what view is most effective for me to understand and evaluate the document.

Sections of note come up and I easily stretch these into my own work, almost like pulling bits of text along a rubber band – they stay connected but flexible. All the contextual meta-information of the original text is kept, as is the further contextual information I add by taking it as a note.

While reading, something puzzles me so I decide to go through earlier versions of the document since I cannot understand how a particular conclusion was reached. I can thread through dialog about the text as easily as I can ski down the hill in sports video game. I see who was involved in the discussion and I send a message based on the text, and get a reply later, answering my question. This act has now also been woven into a layer of this document’s texture.

I change the structure of the document a bit, delete some sections highlighting others and it automatically gets wrapped in a new frame. This new version I share with my students, including a few in-section questions for them and I track their reading speed, perceived attention and see what sections they struggle with. I also reply to questions they have and some of their insights I feel I should share with the original author, whom I have never met. The original author likes the insights and attaches them to the next ‘official’ version, threading back to the original student comments.

I keep reading, choosing to see some of the text visualised through using tiny pictures of the relevant actors (people, companies, products etc.) in the text since I simply don’t have time to read it all in slow sequence right now.

Since all the words in the document have full data behind them, approved by the authors (extracted from Wikipedia or whatever other site is useful) I can search the document and change the view based on simple queries such as ‘who in the document were in NY in 1989’ without any of the people in the document explicitly having this written about them.

So then I start writing my own document, which I invite collaborators to edit at will, add to at will, and I also assign professional fact checkers and editors to work on their own layers of the document.

I can easily find information I have come across before, with human queries like ‘show me the articles I read last week about robotic cars in the evening’ and so on. Anything I put into my document retains a link to where it came from.

It’s not a simply document so I pin different sections to different parts of my monitor, pulling them together as needed, back out as required.

And so on and much more.

There is a lot to do; we need to develop real requirements, develop a technical infrastructure to support this and then develop, develop and redevelop.

Published inThoughts