Skip to content

Category: Notes On…

This category is for writings I consider fuller articles than the very brief glossary terms or other posts.

Liquid Information Environment for Literature Reviews

How the Pieces of the Liquid Information Environment Connects to the larger workflow for Literature Reviews

My single biggest surprise for how literature reviews are done is that there is no best practice or taught system and there is no digital ecosystem to support literature reviews. There are systems for reading and annotating PDFs (the primary substrate for academic knowledge transmission and storage which, in most cases, simply acts as ‘pictures’ or scans of documents, without even the meta-information of the title and authors’ attached to them) and for organising lists of them (citation management software such as Mendeley and EndNote) and nothing widely used to see how all the documents connect.

I have such a hard time working in this piecemeal environment that I feel I need to address this very basic act of research. After all, this is a Clear & Present Opportunity and as an Old Man Navigator I can’t see my soul has a choice. I therefore propose building the following:

Liquid | Author, Reader & View

A word processor which features a rich environment for adding citations as well as other rich interactions for the user and publishing the result as a Rich PDF (with complete meta-data, original document embedded as well as an XML file of information). This is of course what I am already doing with Liquid | Author.

A PDF Reader which will allow the user to read any PDFs and highlight text and then later choose to search based on all text, only highlighted text, annotations etc. (currently Citation Managers cannot search based on highlighted text). The user should also be able to write notes about the whole document and add tags. When it comes time to cite the document the user should be able to simply copy text from the document and paste it into a word processor which will then have the full citation information–the user should not have to go back and forth to manually create a connection–the digital environment should do that for the user. This is what I plan for Liquid | Reader, which will work on macOS and iOS.

A visual thinking space, ideally for large monitors, tablets and VR, where all the users literature review documents are available, visible in any way the user prefers, listing and laying out by keywords, highlighted text, citations, time, author etc. I call this the Liquid | View. I have collaborated with Christopher Gutteridge on a version of this, which he created and called a Webleau.

These components will allow the researcher (academic, business, government or any curios individual) to read and cite documents easily, while having a good overview of how they connect. I further want to support repositories so that the documents are actually findable easily by users searching by various criteria and I agree that we should support PDFs as a reality of the way the world is today but also ePubs and other, more flexible formats.

Open Ecosystem

‘Author’ should produce documents readable by any PDF reader. ‘Reader’ should be able to read and annotate any PDF documents created by any app and save with annotations for any other app to see. ‘View’ should be able to import any documents and save the views in ways accessible from inside apps.

These changes may seem incremental and not very exciting but I believe strongly that by making knowledge more liquid at every stage becomes something more than a sum of features, something it seems uses agree with, since Author is now the 21st most popular productivity app on the macOS App Store.

Who wants to join me in this process and how?

I want to collaborate with anyone who either produce or read any of the components along this workflow. What do you do and how can I integrate with your work?

I am in the process of compiling a list (a literature review of sorts!) of what components are our there and how this can integrate, such as Mendeley and EndNote but also citation visualisers, PDF readers and initiatives to improve or replace PDF. Please do not hesitate to tell me any which you think I may miss.

Leave a Comment

A Clear & Present Opportunity

Time for Better Literature Research

Universities and reading glasses were invented roughly simultaneously (coincidence?), during the 13th century. There were counties for students to have books copied for studying of which I won’t waste our time going through the details of how it wasted their time. It seems however, that not much has changed. Today’s academic is more likely to print out the sources for their literature review than to read it in digital form (Walsh, 2016), even though there are many different types of digital devices to read them on, from reflective to transmissive tablets and from laptops to desktops.

I am not writing about the pros and cons of paper versus digital for general reading but I do mean to highlight that there is a real opportunity to build an ecosystem where digital reading of professional material (someone one reads to learn, not for enjoyment primarily) can benefit from the potentially connected nature of digital text environments.

We will necessarily have to start with PDFs since they are the dominant form of academic knowledge transmission units, though of course support ePubs and other formats as well.


The suggested flow is simply this:


A user reads a document for their research and highlights interesting sections, writes notes in the margin and also generally about the document. The user also also jots down ideas as the reading and research process progresses. Furthermore, the reading applications gives the user a rich set of facilities to view the document as she sees fit, including flexible ways to access keywords, summaries, abstracts, links, references and connections and so on, some of which we have implemented in Liquid Author and Flow.


All the literature review the user has read is accessible through keyword search, including specifying whether to only search highlighted text, annotations, notes or full text. When the user copies text from a document and pastes it into the document their are authoring, all the salient citation information is included and is copied across (since this is not possible by default, they system will need a fast and elegant way for the user to help add this to the document on opening/at leisure and in future automatically use what authoring systems will append as meta-information).

All the literature review the user has yet to read is also accessible through searches and can be bunched based on keywords, citations and other criteria.

All the literature the user has yet to access, which is connected to the users work by being cited in current work is also accessible.

The visual space of connecting literature can be developed by anyone since the connectors will be made clearly available. The weblau is one direction which can be powerfully useful in this regard.

A more Liquid Reader

This could make the literature review process a truly liquid, smooth and rich affair. This is not rocket science or heading far out into the unknown, this can provide clear and immediate benefits.

We have started experimental work around this workflow, with the Liquid | Reader but this will require more investment to continue til completion, particularly since the standards of searching for academic document is so incoherent and therefore many searches engines will need to be strung together.

1 Comment

The Old Man

old man inventor?

As my first major software application reaches 16th most popular Productivity Application on the macOS App Store I reflect again on what Alan Kay said over lunch: “But you CAN do anything!”. I know he was being sincere, it’s the same thing I would say to a curious and passionate student. My concern is that my brain is fully myelinated, all the fatty optimisers have been laid down, my axons and dendrites are not as much in the dating market as before. I have now managed to produce one instance of my vision in Liquid | Author, which includes Liquid | Flow. It’s a different type of word processor where I had a few medium sized ideas (tag for meaning such as for headings, a different Find command and so on) and did a lot of polishing on use, reducing as many button presses and unexpected frustrations as I could see.

So now I have a smidgeon of credibility, which is great, but what I have produced is only a very small incremental step in a different direction. How can I possibly move symbol manipulation along at a usefully large step? How can I invent something amazingly powerful? This is the question any inventor or artist would ask. And my stodgy old brain, is it still able to fluidly create new and useful connections?! I have noticed that my memory has gotten better over the last few years, which worries an artist-type such as myself. I should be able to be flexible and see things afresh. So I worry.

old man navigator

But my worry gives way to perspective. It’s not my ability to invent something which is why I must keep working and why I invite others to join me. It’s not what I build, it’s about where I am going. I am a navigator, not an inventor.

My work is to augment our ability to orient ourselves in our information, to see and make connections. These are all navigation metaphors and they apply to the process of building the tools as well.

I take responsibility of the question, of the goal, not the answer or a particular way of getting there. Sure, my world is that of visual symbol manipulation (text and associated visual communications) but to navigate we must first choose which world to venture forth into.


My future work will be to further improve the navigation and display of text in Author and on the web, including through compressed scrolling, dynamic views and more atomic authoring, including though the user of hyperGlossaries. Much of the work will be to implement small, incremental changes to make the information flow more liquid. The reach of my research and implementation–the twins of progress–will venture as far as my mind will stretch and collages will entertain.

In the distance I see a more liquid information environment where users can become deeply literate because their tools are powerful and they can view and interact with their knowledge in visceral ways, like moulding magical clay.

This is not a solo effort. It is an effort in building dialogue in addition to systems and approaches. This is what I have been doing for almost a decade now with The Future of Text Symposium and look forward to continue.

My premise is simple: Continue to work to employ the occipital lobe to support the prefrontal cortex–to use our eyes to think. And that is my journey.