Skip to content

Month: June 2018

Liquid // Graph Scenario 18 June

In the near future, Joe is sitting down in a comfortable chair with fresh iced coffee, lights down low and comfortable, quiet music playing. It’s raining outside, but not too much. He feels good and is happy to have some time to focus on his work. He looks at his beautiful 27” screen starts tabbing through his favourite news websites to engage his brain this early morning. He then looks at his To Do list and remembers he is due to submit a research paper. 

Since Joe is a professor at a prestigious university, at the forefront of his mind is the need for the research paper to be a good model for his students. He therefore aims to write as clearly and well thought-through as possible and showing his evidence clearly and with citations to source in ways a reader can instantly check.  

This scenario covers basic concept mapping in a dynamic view mode, what Marc Antoine refers to as normal research, instant searches and then enters graph space to follow an argument and returns to the document with the results and adds an element through the hyperGlossary.  

Covered Applications include Liquid | Author a word processor, Liquid | Flow, a text utility for instant searches and more and Liquid | View, a concept map application with the ability to serve nodes and to use nodes from other applications through an open API (related to the graph API I would expect). 

Dynamic View#

Joe doesn’t know everything about the subject but he knows a few things so he opens a new workspace in his Liquid | View application and lays out what he knows as short sentences as Post its called ‘nodes’ (same root as knot, so a very apt term for something which is designed to link) quite randomly. Once he has finished typing what he can think of, he starts to group the nodes and then he gives the groups headings (in View the default node is level 2 and headings level 1). This way of brain storming and seeing structure emerge is very much like the KJ-Ho method from Japan. 

Column View#

His screen now has some structure so he opens his Liquid | Author word processor and chooses to  Live Import from Liquid | View. Here he has a single column word processor view where the nodes now are headings, and he starts writing prose under each heading, some for himself to remind him why the heading is there, some for the final paper.  

He expands and collapses the text to aid his reading of it (to see only sentences with certain keywords, to see headings and names instead of body text or pictures instead of people’s names or company names etc.), looking up terms instantly when he is not sure of their correct use using Liquid | Flow.  

He moves around his text document with extreme ease, giving him the conceptual map of his work as it developed to allow him to really focus on following the nub of the issue he is trying to understand and present.  

Citation Space#

Joe pulls his document to one side and looks at his citation space (no app or name assigned as of yet). Here he searches for articles he has read, by keyword, author, title and so on and stretches the citations out in space across the screen, from old on the left to newer on the right, with less cited on the bottom of the screen and more cited towards the top. He finds a few relevant articles and puts them aside as he follows his intuition to better understand what traditional documents have to offer on the issue. 

He cites books, academic articles, web sites, videos and audio with ease, allowing a reader to follow the citation not just to the cited document, but right inside the document to the exact location, to determine it’s relevance, accuracy and importance.  

Graph Space#

After completing this basic research he delves into the connection space of knowledge graphs to see how others have approached the issue. 

  • Here he follows interactions as primarily specified by Marc Anthony (my assumption is this would entail large scale interactions).
  • He then transfers references across from the connection space into his document as Gyuri scenario (my assumption here is that this would be more detailed or personal interaction.

HyperGlossary Entries#

As he puts the finishing touches on his document, he realises that a few words are used differently by him than would be by a potential reader, so he adds glossary entries for them. As he does so, the graph spaces he has been connected to appears with any relevant other entries, which he can, through simple drag and drop, choose to include or dispute.  

Once done, any glossary term will show up in his document for the reader in the way the reader prefers, with the default being as hard brackets after the keywords [] which can be clicked on to expand and show the short definition right in the text.  

Since he added the glossary entries carefully, including adding explicit relationships to other terms, the reader can choose to see the terms in context in a graph space.  


When he is done, Joe Publishes his document by going through a few Publish modules. His main authoring system, Liquid | Author, first does a spell check, grammar check, reading level and plagiarism check to save him any potential embarrassment. He then gets a summary/abstract automatically generated where he can see if he really did communicate what he thought he communicated. If he finds that the abstract was not what he intended, he can click on any sentence to see what parts of the document contributed to that summary segment and edit it. He can also add text to the summary but then he will need to tag the sections in the document which explicitly expand on those points. The end result will be that the summary will be sufficient for most readers, only the very interested will need to delve deep to understand the full document. 

He then emails a rich PDF exported document to the review body, of whom some only have the ability to read the surface PDF render, others use their own advanced graph-aware systems and a few have the same main authoring package Joe used, Liquid | Author, and thus have access to the original format .liquid document Joe worked on (minus whatever meta he chose to scrub).  

Those reading his document, apart from the PDF only readers, will be able to get a lot of meaning out of the core document and the document will be fully connected to cited sources and graph spaces for full, rich interaction. They will be able to explode the document in a myriad of ways and see relationships with citations and graphs in any granularity they choose–reading the document has become as interactive and almost as fun as playing a video game.  

1 Comment

A dip into the SEA

I wrote this as an excursion into possible interactions into something I believe we are working towards in the group.  


The Name#


To write about this I need to know what to call it. “The augmented thinking space. Jump space. Knowledge Space. Facility (facilitated evolution). Sea” ? I like ’Sea’, since it is also ‘C’, as in Doug Engelbart’s ‘C’ level activity and of course links to my own philosophy of Liquid Information: and  


I’ve even turned it into an acronym (a ‘backronym’): ‘Special Environment for Augmentation’. It’s just a suggestion.  


How we change the evolutionary environment away from simple direct monetary rewards for consumer sales needs to be seriously thought out and I have posted on this as well this morning. We need to facilitate the evolution (in Doug’s terms):  


The Open SEA#


The main difference between a traditional computer work space and this work space is how open it is. By ‘open’ I mean that there is no social media giant where you need to go to interact with your social circle, all the connections are open to be followed anywhere. (This is a key reason why the way evolution needs to be facilitated requires real thought and effort, otherwise we are developing a Dead Sea). 


I open my laptop and simple and nice things are available against a beautiful, dark gradated background: My friends and colleagues arenrepresented with icons docked with status information about them. I can effortlessly see that Bruce has posted something related to AI and that Bjørn has released another drone video. I check them both out and flick into my citation space, it’s time to get some research done. 


In my citation space the research I am doing is not represented by paper-legacy rectangular form, though that is a view I can switch into. I choose to see this space as a basic list to start with, filtering the view to show me only the documents I have read where I have highlighted ‘Engelbart’ or something strongly related to Doug Engelbart. This takes me half a second to specify and the view change is instant. I choose to see more information about all the documents through a quick and familiar keyboard shortcut (I’m not on my tablet where I would use a gesture of some sort) and I choose to hide the documents I have read the most and cited the most–I’m on a quest to learn something new.  


For the sake of this scenario let’s say I come across CoDIAK and I decide to investigate it a little further. I highlight the term and execute a Find command (cmd-f on my Mac) and only sentences which have ‘CoDIAK’ are shown in the document (all others are temporarily hidden), and off to the sides of the document a myriad of connective links appear, pointing to easy-to-read occurrences of the term in hyperGlossaries and the Knowledge Domains my friends have given me access to or which are public as well as public domains. It’s like a beautiful octopus.  


I choose to follow the entries in domains my friends have posted and I navigate through them with the ease of a video game and read and connect as fluidly as possible. Once I see that this is indeed a powerfully useful term I decide to add it to my own hyperGlossary and drag and drop passages of other definitions I agree with and disagree with, being careful to specify the relationships as such. 


During this process Mark Anthony posts an update to a new aspect of the protocols which I can see on my general space when I come back into it. It’s pulsing since he has marked it as important for our community. I have a brief look and fling it as an action item over to Jacob who will know better what to do with it. Jacob reads it and either adds it to a list of actions to do, store it as reference or sends it back to me. If he chooses to take it on as an action my own To Do list will list this and his status, automatically.  


The day grows old and I grow tired and I remember that I have discussed this Christina but I can’t remember when. I decide to speak the command: “Show me occurrences of CoDIAK in conversation with Christina” and my screen morphs into a timeline. I look but find nothing useful, it turns out to be a long list. I then remember that Houria was on the call when we spoke about it so I say: “only show conversations when Houria was there” and here I see a sentence which catches my attention. I click on it to hear the audio and yes, this is what I was thinking off. I can’t remember why we spoke about it so I bring up my calendar for that day and see it was during a planning meeting and that gives me the perspective I need. I now thread my perspective of CoDIAK a bit more carefully and I really like what I have authored and therefore tag it as important. My colleagues who also have worked on this term see my little icon on their systems light up with this new knowledge node, which they can choose to interact with or ignore. If they interact with it we can start a useful, connected dialog or simply list references to each other as not agreeing.  


The CoDIAK notion has sparked my imagination so I message Frank about possible opportunities and Jeff about further clarifications. All of these messages are owned by me, not a service or software provider so I can choose to see status of any of them. For Jeff, since he is so busy, I choose to set a ‘no reply alert’ for one week: If he does not reply I get a notification and can choose to re-send or ignore. When I get a reply from them they can mark the messages as private and I can only read them or leave them public so that I can easily integrate them as citations into my future work on this. 


And thus ends my exploration. I am dealign with information, not applications. I choose what container boundaries and views are. And to end with my old exultation: I dive into the information, I don’t just surf.  


Leave a Comment

Facilitating Evolution

A founding principle of this ecosystem is that it must support what Doug Engelbart called facilitated evolution, which is an interesting and important concept. Evolution in nature happens when the environment the organism is in changes. Random changes making the organism better ‘fit’ the new environment to the degree that the organism can reproduce (the only evolutionary measure of how well it fits the environment–effectively the reward system) or not. 

Planned evolution does not occur in nature and when it does in the human realm we call it innovation and it has proven a powerful force for change. Innovation primarily happens along a designed path though, with specific hoped-for outcomes and this puts it quite firmly in what Doug Engelbart calls the ‘B’ level of activity []. 

In order to facilitate evolution we must design the ecosystem and the reward system. Designing the reward system is designing the investment system and it is crucial since we cannot afford, as a human species, to loose innovation for mind augmentation in a world which is constantly increasing in complexity, we must invest in investing.  

Neither a completely free-market system nor a completely government-run system is sufficient since the rewards would not support enough risk in the free-market (since monopolies would soon appear and stifle innovation) nor enough exploration in a government-run system (who would assign resources?). 

Therefore we must look at multiple rewards to both support a B level of activity but also importantly, C level [], which is riskier but supports more change and therefore can give us more opportunities for fitness in the ever changing information environment we live in. 

I have no real answers in this post, but we need to address this issue, particularly if we develop a more open environment (which I think is a major goal) where no-company can invest in making a walled empire, in the social space like facebook or in the document world like Adobe or Microsoft.  

1 Comment