Skip to content

the current model of researcher-advisor document discourse

As sent to my advisors:

Les, Wendy,

As requested by Les yesterday, here is a brief overview of issues with the current model of researcher-advisor document discourse as I see them at this point. The current model is a bastardisation of paper-imitation (through PDF) and Apple over-simplified ease-of-use computer sales-focused products, swirled around decades of the monopolistic Microsoft Office featuritis with an every sprawling set of features and buttons developed for everyone and no-one. 

Apologies for length and inevitable typos.

• Why Documents 

Documents are framings of human perspectives and they are important devices to aid thinking in both the reading and the writing process. 

There is much that needs to become richly interactive and fluid, but there is also a core need for the freezing of the ‘frame’ of a perspective in time, particularly for the research student, but also for all knowledge workers in general (the future expansion potential).

• Danger

The danger of not realising the full potential of digital interaction is profound; perpetuating the ‘fact-less’ politics we are seeing in current elections and for both allowing for dubious science to be taken as fact and brilliant insights to be ignored. The danger is increased ignorance in an ever increasing knowledge world and the erection of further barriers between those who do not relate, rather than building bridges.

The ‘document’ gives authored ‘authority’ to the material presented and as such must be the clearest, cleanest and most connected to the wider worlds media as possible. 

• Opportunity

Much work done in academia and by early pioneers to augment our ability to craft clear and more credible documents have not had the opportunity to augment the knowledge worker of today, for commercial and market reasons. However, we are now at a position where there are sufficient infrastructures developed (internet, web, document formats and so on) and API libraries which are available that it can be viable at this point to try to effect real change.

We have a tremendous opportunity to afford rich new digital information literacies to a new, 21st century, digital native, game-playing,, social-media savvy, smart-phone experts who are eager for more control and have the mindset to learn and implement new tools and system, much like the ‘I’m not a computer person’ who spends ages on Facebook tagging pictures. 

To foster this potential Deep Literacy for new kinds of deep thinking requires developing ever evolving deep literacy tools and environments. 

• Assertion

I assert that the current dialogue between student and teacher is not fit for the 21st century requirements of speed, accuracy and fluidity as we expect them in other professional arenas. The hugely growing mass of academic papers are released in document formats whose aims are to be printed (WYSIWYG), with very limited digital support. 

At a time where computer games are of almost photo-realistic quality and are moving into Virtual Reality displays, the fact that there has been no real innovation in research tools over the last few decades is thrown into sharp relief, sunlight glinting of the edges of the pioneers dreams.

What is required is simply a laser focused investment in time, research, thought, experiment and evaluations to provide ever richer mental freedom of movement through documents in the research document lifecycle from research material interaction to creation of new knowledge documents which go into the world as new knowledge documents. 

We can build the tools for new possibilities of thought. 

• Contribution Intention

I intend to research, test and create new models where rich interaction gives rise to a much greater freedom of intellectual movement for both researcher and advisor.

These will be models augmenting the researcher’s ability carry out research through active reading (rich interactivity), visually augmented thinking (concept map integration), collaboration (annotation systems) and then to communicate the work cleanly with clear references to the rest of the world (advanced citations) where the resulting document will support the ability of future researchers who will use this new paper for their own literature review. 

I further aim to augment the advisor’s ability to interrogate the researchers thesis while reducing the need to spend time and mental energy on clerical tasks. 

The aim is to help the researcher have clearer thoughts and clearer communication, both through HCI contributions and through new models of interactions through documents. 

Ideally the final result will be an open source product which the academic community can benefit from.


• Issues in Research

A most basic issue is that if you annotate a PDF (highlight, doodle or type a comment) the researcher cannot search for that annotation easily later and cannot tag what they find in a way which will benefit the researcher later. Yes, that’s right, PDF annotations are not searchable, neither in-document nor in bulk. (improvement need: annotation system for PDF or PDF import)

The ability to ’tear up’ the research documents and lay them out in a spacial configuration and connect relevant elements (either manually and through some sort of automatic weights) limits the digital interaction more than if the researcher used paper and a dining rom table. With digital tools and environment we have the potential to create ‘magical’ mind environments but in this crucial case we are lacking behind paper.  (improvement need: concept map integration)

There is very little, if any, meta-data connected to the text in the document such as would allow for instant re-formatting of the document to show timeline views for example.  (improvement need: rich tagging)

Text documents can be very long and uniformly laid out, making skimming difficult. (improvement need: automatic highlighting of keywords, summarisations and more)

Access to information resources such as Liquid | Flow which is modelled on Doug Engelbart’s NLS toolbar, is not in wide use, meaning that when terms come up that the researcher is not familiar with the effort required to look the terms up are above the ’no-cost’ mental threshold and thus are slightly less likely to be looked up. (improvement need: powerful tools)

• Issues in Authoring

Authoring an academic paper is an involved and complicated process which entails thinking, organising, searching for connections, checking intuitions, and working this into a coherent structure which conforms to the structures of the field, all the while keeping citations clearly attached to any sources. (improvement need: concept map integration and other views)

Questions as to how to store cuttings of thoughts or found material come down to ‘a different document’ or ‘at the end of the document’ but not in an integrated way.  (improvement need: document structure)

The researcher lacks efficient ways to zoom in our out of their work, such as an integrated outliner could do.  (improvement need: flexible views)

• Issues With Citation Handling

One of the tragedies of the current model of researcher-supervisor interaction via documents is how citations cling to a paper-past; instead of using the links and interactions as we do in other fields, the citations are ‘dumb’ numbers or Names & Years in parenthesis with further information provided at the end of the document (only). There is no opportunity for the advisor to immediately interrogate the veracity or relevance of the citation. There is also no means through which the researcher can see how citations and other information line up along different dimensions, something which could bring further insight.  (improvement need: new models of citation uses, views, storage and interactions)

Citations to recorded media is at arm’s-length; a researcher cannot instantly look up a specific part of a video or audio at a real-world time-location, only the relative time inside the document, making it harder to gather multiple perspectives of events. (improvement need: rich media citation management)

• Issues with Lack of Integration with available Tools

There have been advancements in text analysis over the decades but it is expensive for individual coders to add these to their word processor projects. This excludes serious competition among indie software vendors for high-performance, advanced interaction, such as text summaries.  (improvement need: open API high-performance libraries)

• Issues with Lack of Increased Literacy Potential 

The literacy of the academic user, outside of specialised needs, have stagnated because the tools which can provide more control are not available. This is why this project is also a call to develop ever deeper literacies, both in terms of tool use and tool & environment development. (improvement need: more powerful tools)

• Issues with ’Dumb’ Publishing

When publishing the document to a supervisor or as a final copy to the world, much work needs to be done manually to get the work to a high standard, something modern systems can easily plug-in to do, such as plagiarism and citation check, auto-summary for the author and more, saving the writer and reader considerable time. (improvement need: publishing needs analysis and modular testing)

• Considerations & Questions

Considerations arising out of looking for new, coherent and efficient models which will integrate into current infrastructures and which will serve real user needs, includes the question of what the most effective knowledge representations and connection representations will be. Questions will include:

– What can an effective tag infrastructure be, including basic grammar for the tags, building on what knowledge we have of tag systems and annotation systems? 

– How about glossary management?

– Can we design a connection architecture or should we focus on the essential principles which would let it evolve?

– What user roles are appropriate?

– What will be the most powerful visualisations possible and how will boundaries of ‘documents’ be considered in relation to the researchers larger collection of work and the world in total?

– How can we build dynamic, stable and open systems for incredibly rich interaction, where everything can be a spark point and any connection can be brought to bear?

In other words, how can we make the information, the tools and the users – the researcher, peers and advisors – more liquid, how can we unleash the power of symbol manipulation? 

• Path

I propose a path forward of thorough literature review, interviews with pioneers, academics and prototyping and improving, building the theoretical framework one step ahead of the implementation and modifying both as user needs dictate. 

• Models which will need to be Developed

New models will need to be developed to investigate how external information is added to the document (through inspiration, citation, media etc.), how the information will be interacted with, annotated and shared – and more. These are issues which call for further literature review, concept maps and large paper discussions, testing and refining. 

* * * Change is expensive so I aim to leverage as much of current infrastructure and practice as possible, only effecting change where it will have a high ‘lever’ potential * * *

• Conclusion


There are rafts of known issues which are sure to uncover a larger amount of unknown issues, which will only come to light as we realise the potential we already understand. 

I ask that I may proceed along the route of augmenting researcher-advisor interaction through documents, building a new model along with doing smaller HCI smoothing, all the while creating a new, more powerful – more flexible and robust – model of this most critical interaction. 

I also thank you for this opportunity, WAIS is a wonderful place. 


Recording of meeting with Les today:

Published inThoughts

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.