Skip to content

Month: December 2018

Threaded Dialogue

The Thread of Dialogue is too important to leave segmented and only part connected in our digital knowledge environment.

When dealing with digital documents the threads weave a peculiar pattern where the dialogue can take place in different media and remain unconnected, such as in email, chat systems, at the bottom of online articles or floating on top of articles as annotations. For the reader to follow the thread makes a lot of assumptions of the readers knowledge of how the dialogue unfolded.

I suggest that instead of only focusing on supporting neutral knowledge navigation through implicit and explicit links only, that we should focus on surfacing the actively threaded dialogue for a participant of such a dialogue or a new person to be able to follow the movement of the discourse through time.

How can we support the readers ability to follow threads and see them in usefully different ways?

When someone writes a book or a document they are weaving an argument based on their understanding of the world, citing prior work for credibility and context., either explicitly or implicitly, simply by using words and phrases which get their meaning from the wider world.

A knowledge product, whether a book, a document or an article of some sort, carries very little meaning in isolation. The meaning comes from the sources the product refers to and the way the author’s perspective is put together into a linear argument. To gain value from the work the reader must have some means through which to follow the threads.

There is a difference between the threads the author has purposefully woven into the text and incidental connections to the wider web of knowledge. The threads reveal the intention of the author and the longer discussion over time.

The goal is for the author and reader to have a continual relationship where the connections are highlighted and made clear to each other and others. A very basic initial implementation of this would be blogs or hypertext articles which explicitly point to which other articles and sections they refer to and a means through which to view the thread of such a dialogue in flexible ways to filter who and what to see and to further annotate this in ways for yet others to access and add their own threads to.

Too much knowledge gets lost in the marginalia and long, inaccessible articles too long to read or find citation sections in. How can we provide the best environment where connections to be primary elements?

This is worth pursuing since dialogue is perhaps the human activity most in need of attention at the moment.

There are a lot of unknowns but there are also well known issues we can tackle, such as providing for high-resolution addressing, making it possible to link to more divergent media such as digital books and developing powerful and flexible views of the the dialogue record.

Goals will include making documents as findable, viewable, readable and citable as possible in order to provide a rich a view of the thread of dialogue which they embody.

  • By ‘document’ is meant an enclosed unit of information which links to other information implicitly or explicitly and is self-contained in its ability to be shared as an individual item. I think of a document as a framing of human intention and this is why it is a unit of knowledge worth investing in to make more accessible than simply a copy of the paper substrate document which had no inherent interactive or connective capabilities beyond what happened in the readers mind or by the readers pencil.
  • By as ‘findable’ as possible’ is meant that the document should make clear to the world what it contains and how it’s connected as well as possible, so that external systems can, through various means, determine it’s relevance, veracity and credibility without a human reader having to do this evaluation alone.
  • By ‘as ‘viewable’ as possible’ is meant giving the user rich ‘viewspec’ options with which to grasp and reconfigure their information and the dialogue which flows through it.
  • By ‘as ‘readable’ as possible’ is meant augmenting the author’s ability to present their intent and to connect it with the larger environment as clearly as possible in a package where the reader can unpack to extract information and evaluate it with as little time or effort wasted as possible.
  • By as ‘citable’ as possible is meant allowing a reader of the resulting document to cite clearly, with high resolution addressing and types links as well as any other useful means for their reader to determine the credibility and relevance of this document and it’s place in the thread of conversation.



1 Comment

We can do anything

Question, Inspiration

I have been lucky enough to have a few conversations with the magical Alan Kay this year. When we last had to wrap up our lunch he said to me, plainly and clearly: “But you can do anything!” At first this bothered me since it’s obviously not true (of course I was wrong) and not feasible (again, I was wrong).

This morning of the last day of 2018, the last day of the 50th anniversary of Doug Engelbart’s demo, I get some time to put down my thoughts through my laptop and onto your screen, while my beautiful baby boy who has just started skiing (Edgar, 19 months old) is reading for his mum in the early, dark hours of our Norwegian mountain top cabin.

I have thought quite bit about what Alan said and it’s become clear (though I may still be wrong) that the point is not to come up with a neato-design and build it, but to come up with a design direction with clear goals and frames:


Let’s start with the frames. I feel it’s out of scope to look at radical new hardware at this point since we already have high-resolution, fast, networked devices, from large, stationary setups to mobile laptops and tablets, phones and even watches, all of which can contribute as access means to a system.

What I do not feel we need to worry about is legacy file formats or legacy ways of working since file formats can always be translated at the point where they need to meet the wider world and legacy ways of working will quickly and easily be seen as such when they come up against something radically better, which is of course our goal.

The only real frames we have to develop within is the human nervous system and that which can be computed and displayed by a computer system.


The goal must necessarily be quite loose in order to support a long evolution so let’s start with Doug Engelbart’s stated goal in his 1962 paper and add to it based on his further work. I paraphrase:

We must augment our capabilities to
approach complex problem situations,
to gain comprehension and to
derive solutions
And we must keep improving our capacities to do so.

The way it is stated, this could apply to numerous fields and aspects of human life and it’s clear that is what he intended. However, in this context I will apply it to the ‘fire’ of giving us the ability to flexibly interact with symbols. I immediately go to the notion of symbols since I do not think that the most effective route to dealing is with the issues above is to present a ’natural’ scene to the user. From the very earliest moments of human recording onto a substrate we have looked at ways to record that which has the most symbolic meaning, not simply slavishly copying down what is visually there in the world.


Symbols are the ‘stuff’ of recorded thought and deserve our outmost attention. We can further focus our design goal:

How can we design and co-evolve systems to
augment the readers ability to grasp the authors intention,
in a critical and contextual way,
while also citing the original work usefully,
and augment the users ability to produce further clarity in their own work?

Many will criticise my rabbit hole and might say I have simply provided directions to my own work and I will accept such criticism with the comment that there are of course other avenues to pursue, such as persuasive cinema and educational games and open, free-flowing graphs of variously encoded knowledge but the human ability to express oneself, to record and justify ones position and to make this interpretable and understandable by oneself in the future and by others will surely remain one core aspect of our intellectual and moral work.

Directions forward

To simply reproduce the marks on a passive substrate in the digital environment is not enough. We must experiment with various ways of increasing the interactivity of the symbols–as I have written, interactivity is the core aspect of symbols and of the universe.

As we experiment, with free-form non-linear spaces connected to linear reasoning (LiquidView), with colour coding based on velocity, and with the means through which we can integrate the resulting knowledge product into existing knowledge flows, we also need to look at the spaces of symbols themselves and how the symbols can most powerfully relate with each other in a document on a screen and also between spaces–how to connect and link them, how to experience pattens and movements.

We must think freely, with open minds and share our work.

And we must think with the user, not just for the user. We must listen to the user but also encourage a deeper literacy not heir part. If we, the tool people, can demonstrate more powerful ways of working, they will come. History is witness to this, but only when the more powerful ways of working a persuasively presented to a receptive audience, as audience we need to continually foster a hunger within.

We must think fresh thoughts.

We must also read more of what the founders of the field wrote. We we living in their shadows as well as in their light, we need to better understand their un-fogged boy daily use prejudice perspectives. Something I have promised Alan to continue doing.

Personally, I will continue to experiment with Liquid | Author and LiquidView, blog my findings and thoughts here on the blog and to host the Future of Text Symposium. How do you, dear reader, wish to produce and connect? Let’s make 2019 a banner year for breakthroughs in employing powerful computer systems to the aid of our thinking, not only to our distraction.

My dream is to look back within a few years and smile at how naive this all was and that it was obvious that the best solution lay somewhere completely different from where we looked, but it was only by going into the first of possibilities that we found this out.

Happy New Year!

P.S., please have a look at my recent, related blog post
Two New Year Wishes.

1 Comment


The LiquidView is a free-form visual thinking and exploration space in a web browser, where any item (hard/node) can be live data from somewhere on the web or the user’s computer.

Data sources can include:

  • Manually entered text
  • Pasted Web Addresses to multimedia which appears as thumbnails
  • Web Addresses with High Resolution Links with connections straight back to specific sections of documents
  • Wikis
  • Blogs
  • Knowledge Graphs

The user can change the layout arbitrarily to their liking, chunking information and both manually and automatically connecting items to see and show connections.

Unique features include the ability to save the entire layout and code into a ‘Quine’ which can be shared and opened to show the same interactive workspace.


Multiple clusters of different kinds of data can share the same canvas and be visually connected or separate:


Search terms can be entered and appears as their own nodes, with connections to where the terms appear inside other nodes, with thicker lines indicating more occurrences:


The canvases can be small and fleeting or large and permanent:

Leave a Comment