░░░░░░░ WEEK 10 ░░░░░░░
Artefact A Documentation Ideation
This week, my main goal is to complete the documentation for Artefact A, which I've chosen to present in a publication format. I've dedicated the entire week to brainstorming, designing, and printing this book. While it might seem ambitious, I'm confident it's achievable because I've been gathering references and have most of the content prepared.
I've had Roots' publication for a while now, and I'm drawn to its format, colors, and layout. The way the two parts seamlessly come together, with the description greeting you as soon as you open the book, feels intentional and aligns with readers' habits. The pink booklet, consisting of loose sheets with a legend explaining its assembly, adds an intriguing touch. I'm considering adopting this format for Inclusivision's documentation, believing it will do justice to the meticulous thought put into the experiments.
During my internship, my mentor emphasized that content is paramount, dictating the layout. This time, I have the chance to tailor my content to fit the desired layout, an opportunity I can't overlook.
My initial step was categorizing my content into sections:
Overview: This section delves into Inclusivision's purpose, intentional bias, experiment flow, and structure, explaining the artefact's intent. While it encompasses the core of my dissertation, I aim for consistency across my artefacts, possibly excluding detailed dissertation references.
Data Collected: Here, I present the experiment data, including ALEC results, transcripts, and overall sentiment. Given the data-centric nature, I envision creating visually appealing layouts to enhance data visualization and aesthetics.
Biased Outcomes: This section extends Inclusivision's narrative, focusing on the ethical implications of curation algorithms discussed during experiment discussions. It serves as evidence of algorithmic discourse and underscores the project's significance.
While inspired by the reference publication's format, I recognize that my content may not seamlessly fit. Rather than forcing a match, I'm considering three versions of a green book, borrowing design intentions while ensuring a harmonious fit with the content.
A Day To Wake Up To – Ampulets
Working File Setup
Size
One of the things I like most about the Roots publication is the size and how they made use of it. Essentially it’s an A4-sized publication, but folded half into some sort of a DL size. For my size setup, the first thing for me to consider how it’ll fit into the largest possible printable format, which is usually super A3 size. Anything bigger than super A3, most printers wouldn’t do a double-sided print or they’ll prefer to use their in-house paper to not risk damaging their machine. So I’ll have to consider how my publication spreads can be printed on super A3.
The cover flap in the reference publication is a concept I would like to use and it is about 40mm. A super A3 size (from RJ Paper) is usually 330mm x 460mm, which means that the maximum size my publication can go up to is 330mm x 420mm and this does not including printing margins and bleed yet. After trying out multiple sizes within this range, I found that 270mm x 380mm is the most comfortable aspect for me to work with.
Grids & Margins
The next thing I set up was the margins. In this case, I prefer a larger margin at the top, while keeping everything else at a smaller consistent size, accounting for some white space in the composition. As for the grids, it was pretty straight forward, the publication will be folded twice, 1 time to turn it into a book that’s 270mm x 190mm, and another fold to turn it into somewhat of a DL size. The columns in the publication will be decided by these fold lines, giving me 4 columns, and for each column, I gave it a 15mm margin.
I would like my running footer and page numbers to sit directly on the bottom margin line so I created another 15mm margin above the text to preserve some consistency. This margin can be adjusted afterwards to match my body text size just in case it eats into the baseline grids.
Paragraph Styles
Having learned how to properly set up my paragraph styles have saved me a lot of time in making adjustments to text. I’m going to be using the same typeface I used for Inclusivision, Haas Neue Grotesk, which is essentially Helvetica. I think the choice in this typeface saves me a lot of time when it comes to testing readability etc. as I kind of already know how the optimal sizes and the units for tracking, kerning, word-spacing etc.
Text hierarchy can be achieved with either size, position, or weight. I think in the past I’ve made too many mistakes of over-emphasising text, so this time, I’m only going to emphasise using weight. For my headers, I’ll go with bold and the body will remain roman. This applies as well for my running footers and page numbers. Their position, away from the main content area already define their hierarchy in the spread.
It’s a little tricky as to what font size I should go with. The dilemma here is 10-12pt would be an optimal size for the text, however, with 10-12pt, the amount of words I have per line might be too little because of the column size so it’ll look weird. This would also further affect my text ragging later on, creating a lot of empty spaces at the end of my lines. I went ahead with 10pts in the end and amended the issues by tightening up the word-spacing between some lines.
Document Grid
Paragraph styles & custom wordspacing
Overview Section
The overview section of my documentation is akin to the introduction, research objectives, and methodology of my dissertation. However, I aim to avoid overwhelming viewers with excessive information. Instead, I will condense the core content, extracting key points from the literature review and presenting them concisely in a paragraph or two. I won't delve deeply into why algorithm bias is detrimental, opting to provide a few examples of ethical implications. I assume readers have a certain level of general knowledge in this area.
Further breaking down the overview section into subsections helps me to organize the content more effectively:
Introduction: Briefly describes the software's purpose and its role as an artefact in my project.
Narrative: Explores the speculative narrative shared with participants in the thought experiment, highlighting intentional bias, speed versus accuracy in facial detection, and real-world application of Inclusivision.
Prototype Front-End & Back-End: Reflects on feedback received regarding the software's interface complexity and introduces the idea of creating a blueprint to clarify UI elements' functions.
Experiment Methodology: Details the meticulous planning involved in designing the experiment, including the flow, data collection methods(quantitative and qualitative), and the ALEC measuring system.
Summary of Experiment Findings: Provides a concise summary of discussions and insights garnered from the experiment, distinct from the data collected section.
One of the key challenges I face with publications is the unpredictability of the final page count, which significantly impacts the consistency of my content layout. Depending on the page count, I may need to adjust or add/remove sections to ensure a balanced distribution of content throughout the book.
I must admit that documenting images during the experiment wasn't as thorough as I would have liked. The images I've used for Inclusivision seem more effective than those captured during the experiment by my friend. Repeating the documentation process with the subjects isn't feasible, so I'm considering a workaround by treating the images to align with the bitmap/ASCII nuances in Inclusivision's identity. This approach aims to introduce a level of consistency despite the limitations in image quality.
Overview sample spreads
Data Collected Section
This section serves as a platform for showcasing my typesetting skills, targeting a specific audience who will appreciate it during events like open studios or the grad show. As I transition into a design studio career post-graduation, I recognize the value of this skillset and aim to build a portfolio showcasing it. Additionally, having an abundance of long-form text for integration into a publication is a rare opportunity that I'm eager to seize.
Beyond personal motives, this section plays a crucial role in demonstrating Inclusivision's effectiveness. Early in the project, I considered visualizing ALEC chart data to enhance comprehension. My approach involves using a gradient, adjusting the percentage of white to black against the publication's background color (white). While exploring various data visualization references, I found many options too contrived for my project's context. The gradient technique aligns better with Inclusivision's identity.
Regarding the transcript layout, I have an idea inspired by the grid system, which I believe enhances visual appeal and readability. By creating text boxes of varying sizes within the columns, aligned with the grid, I can establish a hierarchical system that improves the overall reading experience.
Before deciding on my subsections, I will always dump a bulk of my content in the spreads first to see how much space they take up, another habit I picked up during my internship. This approach will help me visualise the layout options that I have and if possible, how to separate the sections to achieve a more cohesive and consistent layout.
Data Collected sample spreads
Biased Outcomes Section
I created this section from scratch, which meant revisiting the experiment transcript to extract instances of algorithm bias discussed during the conversations. Many of these implications overlapped, so I carefully selected 6 distinct ones:
1. Biased Interpretation of Representative Diversity
2. Implications on Marginalised Groups
3. Impact on Genuine Social Media Engagement
4. Susceptibility to Misleading Bias
5. Technical Considerations (Lighting & Photography)
6. Vulnerability to Manipulation
The purpose of identifying these bias implications is to showcase examples of Inclusivision's functionality and to emphasize the insights gained from the discussions, proving that the artifact effectively sparked critical thinking about the ethical aspects of biased algorithmic decision-making.
Considering the previous sections' wordiness, I intend to keep this one concise with more visual elements depicting the biased software outcomes. I envision this section as loose sheets for readers to remove and examine each outcome individually, similar to the Roots publication's format.
While I initially felt I spent too much time making the software functional instead of presenting it as a prototype, generating results for this section was straightforward. I ran images through my software to obtain authentic results, ensuring they accurately reflect the mentioned implications.
I devised a new layout following the existing column grid system to display the biased outcomes. Avoiding screen captures from the software, which could be complex for readers to decipher, this new layout simplifies data visualization and includes key components for easier interpretation of the outcomes.
Biased Outcomes sample spreads
Publication Style & Cover Design
Creating the publication's style proved challenging. Inclusivision has its identity, but it wouldn't fit well in the publication context. It also needed to differ from Artefact B while maintaining alignment across Artefacts A, B, and C. This balance was crucial for consistency and coherence within the project.
Back Cover
For Artefact B's cover, I employed a gradient bitmap to suggest algorithms and pixels. I see this as a style element that could transition to Artefact A (Inclusivision)'s design. On the back cover, I used a bitmap image collage of faces that spans all three publications. When aligned side by side, they reveal the complete image, creating an aesthetically pleasing effect.
Front Cover
For the front cover, I kept the main section names on the flaps, as well as styled the index as a fraction to indicate that each publication is part of the set of 3. Inclusivision itself is a long word, so I found it more convenient to just include the logomark in there just to declutter a bit. Considering that the publication will be folded into a DL size, the front cover will be folded in half to have a front and back. There’s no need to overdo it or make the design different as the section names already have different lengths and served as a distinction. This is going to sound confusing but bear with me, I kept the front of the front cover simple and consistent across all 3 sections. While for the back of the front cover, I felt it was needed to credit the people who have helped me with this project, as well as acknowledge the images used just in case I ran into any form of trouble with royalties.
Colours
I've maintained a consistent approach across all my publications, using black and white for text and bitmap images while leaving online images untreated to maintain authenticity. However, upon comparing Artefact A and B's documentation, I noticed that B's neon green color stands out too much. To avoid it overpowering future presentations, I felt the need to add more color to Artefact A's publication.
After using fluorescent color for Artefact B's cover, I searched for similar colored papers and found a pink fluorescent one from the GMUND series. While there was also a light orange option available, it didn't complement the green well. The pink paper had a crosshatched texture similar to the bitmap texture used in my documentation images. I also had intended to go for different coloured covers for each section and selected a white and silver one from the same series. Although the colors weren't fluorescent, they harmonized with the pink really well.
Reflections
While drafting this section, I encountered challenges in articulating my thoughts clearly in writing. I grappled with numerous considerations regarding visual identity due to the structured nature of my project outcomes. Each deliverable is layered with complexity, making it difficult to explain concisely. The crux of the matter revolves around ensuring that Artefacts A, B, and C visually cohere as part of the same project while maintaining distinctiveness that reflects their unique dilemmas and themes.
Cover Design & Colours
Putting it Altogether
File Setup
The file setup for both the Overview and Data Collected section are the same but for the Biased Outcomes which are loose sheets, I needed to export in a different format. I also created another document for the covers (which also contains the index pages) as they were of different sizes as compared to the inner sheets. The process was pretty straightforward, having done this quite a few times I’ve gotten pretty used to it.
Paper
Selecting the right paper for the inner sheets posed a challenge. In the reference publication, they opted for a thin grammage, making folding easier without creasing and allowing the book to flatten smoothly due to less paper resistance. However, for my project, I couldn't use such thin paper because of the double-sided printing with images, which tends to show through on thin paper. After experimenting with various grammages, I settled on 125gsm. While this weight minimizes visible creases near fold lines and prevents seepage from double-sided printing, it compromises the tactile feel of the book and makes flattening it more difficult—a trade-off I had to accept.
For the covers, I had purchased full-sized papers from Artfriend. Normally, I trust myself with simple tasks like trimming the paper to a super A3 size. However, this time I made a mistake and had to make a second trip because of a trimming error. It was quite frustrating, especially after dedicating an entire week to working on this artefact. Originally, I had planned to print on the same day I got the papers, but I was too drained and demotivated by the mistake to oversee the printing process. So, I decided to take a break for the night.
Production
On a fresh Saturday morning, I arrived at the print shop right as it opened, ready to tackle the printing task. Fortunately, the process was straightforward since my print files were well-prepared; I had meticulously reviewed everything to minimize the need for adjustments.
When the printer produced the cover, I was pleasantly surprised by its quality. The rich blacks, achieved through careful tonal conversion, gave the cover a striking appearance. The combination of bitmap prints and texture complemented each other perfectly, marking a small triumph after the challenges of the previous day.
As for binding, my repeated visits to the shop on Saturday mornings had built a strong rapport with the staff. They kindly offered to assist with trimming and saddle stitching my book, an offer I readily accepted. While I typically enjoy handling these tasks myself for practice, the investment in these prints led me to opt for professional help to avoid any potential mistakes.
Reflection
I was genuinely thrilled with the print results; it's incredible to think I produced such a high-quality set of publications in just one week. Perhaps it's because I've been gathering references and subconsciously planning for some time now. My relief mainly stems from having something substantial to showcase during open studios. The intensity of my focus on this documentation even led me to briefly lose track of time, thinking it was already week 11 and scrambling my timeline.
There's no time to take a break just yet; I have a series of tasks to complete before Open Studios. Firstly, I need to conceptualize my final artefact. Once I have a solid concept in place, I can proceed to craft the design statement and descriptions for each artefact. Additionally, I still have to create collaterals for Open Studios, including labels and descriptions. 2 more milestones to go (Open Studios & Viva), this last 2 final push is a measure of my growth throughout the past 3 years. Having missed all the roundtable, I can’t wait to see what people have to say about my work.
Detail shots