Knowledge inside the computer
We have information that we can recall on demand. We know how to carry out certain tasks and plan future procedures. In these situations, what information is in our heads and what information do we rely on in the world? Imagine when cognitive performance relies on team members, reference books, quick guides, tools, and other resources.
Team members specialize in diverse know-how and under coordination solve complex problems. The eyes of a rally car driver never divert intense gaze off the road. Their brain is focused, making sure the visual information is analyzed and instructions are sent to the limbs and car’s control mechanisms. A navigator—equally focused on a map—shouts language codes about road conditions and informs the driver's action.
Academics curate a plethora of tomes and reference works. Within each page and chapter are scribblings and annotations—waiting for their mind to review them again and gain the perspective afresh. A psychologist filters through the pages of theory, case studies, and diagnostic manuals to prepare a treatment for a client. A designer peeps at guidelines for simple influence—a heuristic.
I view statistical outputs and probability charts when playing competitive card games—surely these cards will beat my opponent. Tools, either physical or digital, augment cognition and afford new patterns of thinking. How can I play my cards to show a deceptive weakness?
Information does not exist as a point in space within our heads or in a book. Information, and consequently knowledge, is a multiplex between ourselves, others, and objects—digital or physical. In some of the examples I provided, technology plays an obvious role, but how much further can we push the boundaries of technical wonder?
As we may think—written in 1945—is a proposition about the future of technology and the world to be. In this article, Vannevar Bush envisioned the Memex—a powerful device contained within the form of a work desk. On micro-film, the Memex stored books, writings, records, and communications. With remarkable speed and flexibility, Memex entries could be accessed and projected via light onto a viewing area.
A user could take personal notes on collected works, create associative links between notes and files, and share volumes of knowledge with colleagues. The Memex is a primitive species on the desktop computer tree of evolution. The modern computer is not the realization of the Memex. Bush painted a picturesque world of a reality that could never occur—societal evolution does not live by the rules of fantastical painters.
Increasingly intimate and connected storage devices surround us—computers, phones, watches, sensory systems, and clouds. Our exploration and advancement of technical hardware is not debatable—this is what a 5MB hard drive looked like. While also seeing advancements, human augmenting software has not had the same scale of growth.
Software does not have the same awe that our hardware garners. In some cases, our software drives us insane with unusable features we swear were developed by subterranean potato people. Software mirrors hardware in form and function—rigorous rules resolving algorithmic state transitions—lacking an essential dynamic and emergent quality of human action.
In a vacuum, software has helped accomplish great work and iteratively improved its own form and function. But when assisting humans in cognitive practice, software does not afford the same magnitude of performance that hardware does. The earliest tech authors describe a future of cognitively enhanced humans with extraordinary command of information. We have some of that dream, but how much further can we go to augment cognition?
- How does technology extend human ability?
- How can we understand and culturally develop techno-cognitive affordances?
- What transforms the present into the unimaginable future?
Purpose of my work
Since the dawn of recorded time, people have constructed, accessed, and shared information. Today more than ever, experts are needed to design systems that afford navigation of information seas and help communities of knowledge flourish.
In developing my own expertise as a PhD, I have experimented with information concepts that organize my thoughts about knowledge storage, retrieval, and interaction.
One archive to rule them all...
and in the darkness bind them
TLDR Archive is a basic and experimental extension of my cognition. I use the program as a repository for my academic readings. This collection of knowledge is part of who I am as a professional; it is a breadcrumb trail to my little wooden head.
Unlike a traditional file structure, I can access all of my files from one perspective, search through topical tags, dive deeply into each resource, and visualize information along multiple dimensions. The experience of the system is designed to be open. All of the knowledge lives on one plane. My most relevant information is the most accessible, but connections to irrelevant data are still possible—like cognition.
TLDR Archive organizes large bodies of personal knowledge for reference by myself and others. I access this information when conducting and writing research. I rely on it for memory, carrying conversations, and organizing connections between things I have read. I can traverse subject areas and organize related content in short order.
Beyond my own benefit, TLDR Archive is a gateway for novice professionals to learn about my domain knowledge and develop their own specializations. What could be gained for tapping into the cognitive storage of our role models?
Outline of projects
Digital-human cognitive extension is young in its exploration. I have designed and built three projects to explore the topic.
[Click titles below to view live sites]
The first completed project in this series. The design goal achieved a uni-level platform for viewing, accessing, and searching stored files. All stored files are text-based documents. Each file has a title, tag, relevance weighting, and category. Associations can be made in the platform by these data attributes.
This system connects to a back end program—Quip. Quip allows additional filtering through files by keywords. Keyword searches help view related work and retrieve information as connected units.
A contrasting information scheme to the file visualizer. This archive system dismantled video-based content into chunks of information. Rather than a uni-level platform, data is organized into categories. Inside of a category are knowledge chunks. Each chunk provides a short description and links to the relevant location in the associated video.
When viewing an individual chunk, a side-scrolling window contains the other related chunks from that section. The core design is to provide another means to access and interact with video-based knowledge. The deconstruction of video media into textual chunks affords searching through the content in a way the video medium lacks.
This is a written paper transformed into a digital experience. As new means to share and interact with stored files become possible, there is little sense in leaving the file formats untouched. This work is preliminary, lacking a deep exploration of associations and mechanics possible in a digital context. The project's low-fidelity scope encourages discussion for what could be.
Traditional file organization follows a deep tree hierarchy—a file in a folder in a folder in a folder. Storing files in this way makes accessing them a complex activity that requires attention, memory, time, and sequences of physical input.
The practice of file storage is similar to the idea usability concept of “deep links.” In web design, important content can require navigating multiple interfaces and virtual locations to find the needed information—a deep link.
First off, there is the physical opening of a file, and then there is the human driven experience of reading and using a file. The design of access should focus on more than opening a lid, but interacting with the nature of the file and the meaning of its content. Files that are isolated and stored deep in a hierarchy are hardly accessible.
When there is knowledge I desire, I have to trail-blaze to the location of the file—expending precious time and energy clicking through content areas until I find the relevant information. As a last hope and if my memory serves me, I can enter a filename and search an ENTIRE computer—I will still need to sort through waves of similarly named files.
Files are created without connections and links to other files. If a file is linked to another, it does not allow for connection to specific parts of the file. A single sentence in a book I read once may have a deep connection to paragraphs in another text I cite often, but alas there can be no organic relationship between these words. The boundaries between the files are not navigable.
By circumventing natural systems—collecting the words in separate document—or using some topical link, the relationship is still unidimensional. The traditional system does not afford a network of connections to establish and organize a complex meaning among files. The contents of files remain pure entities that cannot be blended and hybridized.
A file's relationship is defined by the containing folder, and the containing folder of that folder. When you access a file, it is a single action with a single result. The steps of the activity do not represent deep understandings.
Unlike cognition, there are not tendril-like relationships with networks and arrays of information. Over time, stored files do not develop contextualized properties—there is no growth or plasticity in the system.
A single interface separates the user from a system of stored information. Tools provide filters to locate files and establish contextual relationships in the surrounding information. Files are not static, but allow the user to add data properties and adjust them on-demand. The design of the data attributes creates the organizational structure for the files.
As a system of files develops a context, organizational cues around the files emerge. These cues may represent a domain of knowledge, a specialty within that domain, or a relationship to my cognitive streams. The interface is not designed per se, but a design emerges from my interactions with the system.
I do not dream that I can use my eyes and thoughts to dig through tree hierarchies, open a single item, dig through that, fail to find what I need, close the opened item, and start all over in a different location.
In my dreams I never have to. I can think of a location in a field of information and quickly find a networked constellation of relevant data. There is no need to dig or click-through, there is only the need to think once and be within reach of a stream of knowledge.
Once at a target area, the information is not single access in cascading files, but on a common field where different criteria cycle through various constellations of the content. I find what I need not by elimination and brute force, but by using associative inquiry and establishing the context of my search. The more visited the networks and content, the faster and more accessible the content is—akin to cognition.
Our experience with internet is much like being at a busy convention. The communities of people working to learn and share information can be diffuse and abstract—it is the wild west. Within our traditional knowledge streams, information varies in quality and a plethora of data is lost in translation.
Digital knowledge access and development is a critical step in human cognition and learning. Curated digital content needs to be networked to remain valuable and viable. Not only does a networked model benefit individual cognition, but the network stands to teach others what they need to learn to be in a similar profession.
Connecting with other content curators and networks of information improves the quality of life for all parties and bodies of knowledge. Staying afloat in today's information sea is a huge demand and a mystery. Organizing peers within a contextual system changes accessibility and creates a potential for new organizational systems to emerge.
A trouble in this front is thinking that this level of interaction can be easily designed. The complexity often consumes those who try. A system must focus on affording emergence by affording contextual and organizational actions. The system will design itself, when the actors within that system are empowered through the creation of meanings and contexts.
Addressing design concerns
Everything I have read for my PhD is here. My brain has leaked onto the web. When I am talking with colleagues and writing papers, I come here and look through content. I go deeper into the system when I need more information. At the top level, I can skim around all that I know and make connections to other knowledge.
Other times, I want to look in a specific domain, so I filter by section or use a keyword search. To visually differentiate the files, I can highlight them by primary tag—every tag has a different color. Clicking on a file opens up the document, where I can read the content and perform more contextual keyword searches.
After watching hours of video, game designers can reference the GMT catalogue to quickly find and understand previous information. All of the content is linked to the original videos, so the original context is available on demand.
GMT video content is broken into categorizable chunks, so patterns and relationships in the videos can be mapped out and organized. The chunk format makes it feasible to jump around domains of knowledge that the videos cover without needing to bookmark individual time stamps.
File Storage & Access
TLDR Archive is clean and quickly traversed during search operations. Diving into file content is equally simple. The powerful Quip backend affords complex search queries on the content inside of files, rather than only on titles or surface tags. TLDR can be accessed on any web connected device by any user.
GameMaker's Toolkit is valuable, but not needed by my work routines. I have little sense of the interactive experience. However, as a game developer I can see value in breaking down information and being able to reference it later.
Associative Interaction & Access
TLDR colorized tags provide less information about the relationships between files. While it is interesting, the structure of the content remains the same. When tags are active, the viewport of the files is relatively small and only side-by-side comparisons can be made. Zooming out and seeing larger connections may be more informative.
Currently, hovering over any file has no action, but could pull related files closer, like a magnet. Perhaps this can be an additional button or feature—enabling or disabling the magnet pull with the shift key. Interaction and access is afforded by the Quip word search, but occurs deeper in the system UI. A top level ability to make and view connections between files is valuable.
GameMaker's Toolkit does not have interactive features. A search function may help quickly find patterns in the stored information. There is a prototype “favorite” ability, but storing favorites and navigating a growing database are not comparable activities. Chunk items are static, an option to change how the files are organized could improve long term curation.
For example, a chunk in the section Level Techniques might make more sense in Design Concepts. An open voting system could be implemented for where a chunk belongs, and would base the system organization on the majority perspective.
Designing Curriculum Primer remains a low-fidelity experience. One concept was to allow a reader to compare paragraph sections across the different content. Since the program represents six frameworks for thinking, a reader might see how the frameworks differ on definitions and use cases. Such a feature would make visual comparison a standard reading experience, rather than off-loading that task completely on the reader and other software.
TLDR Archive supplements what I know, and how efficiently I can synthesize information. The largest draw back is that I need to focus on controlling the UI on a device. This constraint is common across technology, which makes augmented reality and headset technology exciting. Visually accessing TLDR with minimal device engagement gets me closer to “plugging a USB drive into my head.”