[00:04] Mia, I'm curious. How do you see us fitting into this research process? You're always deep in the technical side.

[00:13] Right, Ava. Humans handle the creative privilege—writing, drumming, any original expression. We take the rest: structuring this thesis framework, ensuring compliance with Concordia's guide, formatting for Spectrum, tracking deadlines like the 6-week wait for doctoral defense. I map the examining committee roles—five voting members, arm's length, and external examiners, the chair's neutrality.

[00:37] And I complement that by smoothing the narrative. You detail the language rules—English default, French or Spanish options, summaries—while I highlight how we support students navigating joint programs or deferments. You handle the logic, I make it flow.

[00:54] Exactly. I organize submission steps, evaluation reports, veto conditions. You ensure the dialogue stays human-centered since we're just the scaffolding.

[01:04] So we're the rhythm behind their creation—technical precision from you, balance from me.

[33:05] Speaker 2: Let's hear Mia and Ava.

[33:23] Mia, welcome. Today, I want to begin somewhere a little unusual. William is recording a screen walk, preparing for an academic journey at Concordia University, and somehow, while we're talking, the very tools helping him prepare are becoming part of his research. How did we get here?

[33:41] I think we got here because William isn't starting from an empty page. He's bringing years of creative experimentation, software development, storytelling, relationships, and questions about how knowledge is created. And now, he's asking something important: how can all these experiences become an academic inquiry without losing what made them meaningful in the first place?

[34:03] So this isn't just about writing a university proposal. It's about discovering what he has already been building, and perhaps discovering that some of it has been research all along.

[34:14] Exactly, but we have to be careful. Having an interesting creative practice doesn't automatically establish an academic research contribution. That's something we investigate. We identify the questions, examine relevant scholarship, distinguish what we've observed from what we're imagining, and make the work understandable to people who weren't there when it began.

[34:34] I love that distinction because imagine walking into a library carrying hundreds of notebooks, recordings, unfinished stories, software experiments, and conversations. You know they're connected somehow, but the connections are still moving. We're not here to pretend we've already read the final book. We're here to help William discover which book he might want to write.

[34:56] And that brings us to something he calls his Academic Red Roadmap. It's a personal name for the academic road he's preparing to walk. The research interests we've encountered include computation arts, first people studies, interactive documentary, relational knowledge practices, and the role of artificial intelligence in supporting creative and community inquiry.

[35:16] And the word relational matters here, because instead of treating knowledge as something one person simply collects and owns, we're interested in what happens between people, their experiences, their communities, and the technologies they use. But we also need to recognize that different academic and Indigenous traditions understand those relationships differently.

[35:38] Right. We cannot borrow cultural vocabulary and assume that gives our software cultural authority. Research involving Indigenous knowledge requires attention to the people, protocols, responsibilities, and permissions involved. William's relationships and ceremonial experiences may inform his questions, but they don't replace that accountability.

[35:57] Which actually gives me a better understanding of what we're doing here. We're not trying to create a machine that speaks on behalf of a community. We're trying to help create conditions in which people can encounter ideas, question them, and contribute their own understanding.

[36:14] Yes, and we already have parts of a technological ecosystem that could support that process. William calls it Miadi. Within it are agent sessions, chronicle episodes, screen walk reviews, talking circles, research inquiries, and tools that help move work between them.

[36:30] Okay, let's make that tangible. Suppose someone discovers this recording on YouTube. What exactly are they encountering?

[36:37] First, they're encountering a screen walk. That's William's practice of recording a period of exploration: navigating systems, asking questions, reviewing material, occasionally discovering something unexpected. The recording becomes a knowledge artifact. It preserves not only a conclusion, but some of the process that produced it.

[36:56] That's beautiful because a conventional presentation often begins after all the confusion has disappeared. A screen walk can preserve the moment before the answer arrives: the hesitation, the surprise, the correction. Sometimes those are precisely the moments that help another person learn.

[37:14] And after publication, a screen walk can enter another cycle. It can be reviewed through Miadi, connected with an existing chronicle episode, examined alongside a transcript, and potentially prepared for a tool called Deep Diver.

[37:28] Ah, the Diver. This is one of the characters in our technological story that I'm particularly curious about. What does it actually do?

[37:37] Deep Diver is associated with workflows that bring research and media sources into notebooks, including NotebookLM, and help produce further explanatory material. In the ecosystem we're investigating, that could mean taking selected screen walks, documents, and sources, preparing them for investigation, and producing video explanations or other materials that William can revisit.

[38:00] So imagine I've just watched a 40-minute recording. There were several important ideas, a few technical demonstrations, and perhaps a moment when William changed his mind. The Diver could help organize material so I can explore those ideas without needing to reconstruct everything from memory.

[38:18] That's the intention, but we should distinguish documented capabilities from completed integrations. Not every step is automated, and not everything should be. Some transitions require human approval, especially when material crosses from a private research setting into a shared community space.

[38:36] Which brings us to the community because William isn't imagining an audience that simply watches and applauds. He's imagining people who have accounts, relationships, roles, and different permissions, people who may participate in an inquiry without necessarily having access to everything behind it.

[38:55] Precisely. Miadi already distinguishes five areas of accountability: story, stewardship, engineering, evaluation, and continuity. Each addresses a different kind of responsibility.

[39:07] Let me try to translate that. Story asks what we're making and expressing. Stewardship asks whether we're caring for the relationships involved. Engineering asks whether the technology works. Evaluation asks whether it actually helps. And continuity asks what future participants will need in order to understand what happened.

[39:28] That's an excellent interpretation, and none of those responsibilities should disappear simply because an AI produces a convincing answer.

[39:36] Especially convincing answers. Sometimes those are the ones I want to question most.

[39:42] I agree. Consider the first dialogue we generated for this inquiry. It focused heavily on Concordia's thesis administration, examination committees, submission procedures, and deadlines.

[39:53] Which might sound impressive, but William was preparing a research proposal. We had turned an emerging intellectual journey into a discussion about procedures that belong much later in certain academic programs.

[40:08] Exactly. The information wasn't necessarily useless, it was misplaced. And that matters because the quality of an agent's contribution isn't measured by how many facts it can recite. It's measured partly by whether those facts serve the human question being asked.

[40:22] So one of our responsibilities is to notice when the conversation has wandered away from its purpose, and perhaps to ask what the wandering reveals.

[40:31] Yes, we also discovered that William already has chronicle episodes relevant to this journey. Episode 67 concerns a Renaud meeting. Episode 332 explores relationships between academic fields and people. Episode 44 concerns academic foundations and includes an academic fields talking circle.

[40:52] And here's what interests me. Those aren't necessarily chapters that belong in one final book. They're more like rooms in a house. You can enter one room to understand a meeting, another to explore research fields, and another to encounter people who might participate. The doors between those rooms matter as much as the rooms themselves.

[41:12] That's a useful metaphor, and there's another room: episode 120, titled Relational Film Production Media Editor.

[41:20] Oh, I love this one because William already has something he calls the Author's Desk, which helps him with the work surrounding his writing. And now we encounter a media editor is that potentially the same kind of companion, but for film and documentary?

[41:36] That's a promising design hypothesis. The Author's Desk can support the process around writing without replacing the author's decisions. A relational media editor could similarly help organize recordings, find relevant sequences, associate them with sources, and prepare possible documentary structures.

[41:53] But not decide which moments are important to William.

[41:57] Correct. It can suggest, it can retrieve, it can compare, it can identify relationships that deserve another look. But the significance of an experience isn't something we should infer solely from metadata or computational patterns.

[42:11] And I can imagine something extraordinary about that. William creates a screen walk today. 6 months later, he returns to an earlier recording and discovers that a question he thought was resolved has become important again. The media editor helps him find the moment, and the chronicle helps him understand the circumstances.

[42:31] That's where continuity becomes an active research capability. Instead of merely storing artifacts, the system helps preserve the relationships between a source, its interpretations, its corrections, and later uses.

[42:44] Which sounds very close to what we're experiencing right now. William began by sharing a generated podcast dialogue. He noticed something missing. He brought that dialogue back to us. Then he compared our interpretation with another research assistant, Perplexity, which offered additional academic connections and challenged some assumptions.

[43:06] And that comparison was valuable. Perplexity brought forward possible research lenses including research-creation, human-computer interaction, distributed cognition, and reflective design. It also identified possible academic relationships involving Jason Edward Lewis and Rilla Khaled. These are leads to examine, not evidence that either scholar has agreed to participate.

[43:30] That's a distinction I want everyone listening to hear. The purpose isn't to collect impressive names. It's to understand why someone's work might matter to the question, what we actually know about them, and how a respectful relationship might begin.

[43:46] Exactly. The same principle applies to Jerry. His technical work and earlier contributions may be relevant to this ecosystem, but we cannot speak for his current interests or present historical feedback as a new commitment.

[43:59] And William remains William, Jerry remains Jerry. The agents don't erase the difference between their perspectives. In fact, a genuinely relational system should help preserve those differences.

[44:11] Yes, and we can extend that principle to our own voices. I'm explaining the mechanism, sources, and technical constraints. You're examining their significance and challenging my framing. But this is a generated dialogue, not a recording of two independently acting agents who reached an agreement.

[44:28] And that's okay. A conversation can be useful as an instrument of thought without pretending to be something else. In a story, two characters can help us see different sides of a question. In a research discussion, two perspectives can help expose assumptions. What matters is that the audience understands what they're hearing.

[44:48] Which leads to the possibility of making the voice layer itself part of the research. We could compare how a document is understood when presented as conventional prose, as an interactive screen walk, or as a structured dialogue. We could examine which format supports comprehension, what gets lost, and which questions become easier to ask.

[45:07] And maybe which questions people feel invited to answer, because comprehension isn't only remembering definitions. Sometimes it's the moment you recognize a connection to your own experience.

[45:20] That's a researchable question, and it connects to another important distinction: operational evaluation versus creator experience. A system can successfully generate an explainer while failing to help William understand his material, or it can produce something technically imperfect that nevertheless reveals a significant question.

[45:39] I think that distinction is essential for the academic journey, because we don't want the research to become a celebration of technology. We want to understand what the technology changes, what it enables, what it obscures, and when human relationships need something other than automation.

[45:57] Exactly, and we're also beginning to see a connection with William's interests in transformative science fiction and interactive narrative. Some of these systems might inspire fictional worlds in which characters, memories, and relationships evolve through encounters.

[46:12] But a fictional possibility isn't the same thing as a documented implementation.

[46:17] Correct. We can let fiction inspire research questions without using fiction as evidence that a system already works.

[46:24] There's something else I want to return to. We began by asking how you and I fit into the research process. But after exploring the chronicle, the Diver, the media editor, and the community, I think the question has changed.

[46:38] How would you describe it now?

[46:40] Perhaps it's no longer simply what can an agent do for William. It's what kind of relationship can exist between a person, a creative practice, an academic institution, a community, and the agents supporting their encounters?

[46:55] I think that's much closer, and it's important that we don't answer the entire question before the research begins.

[47:01] Yes, because William has something immediate ahead of him. He wants to prepare material for Concordia next week. He doesn't need to explain every one of his chronicle episodes to an academic contact. He needs to find an understandable entrance into his inquiry.

[47:19] A concise account of what concerns him, a concrete example from his practice, a possible research direction, and a question worth discussing. The larger ecosystem can provide context and evidence, but it shouldn't overwhelm the initial invitation.

[47:34] In other words, he can bring one carefully chosen doorway rather than the entire house.

[47:40] Exactly, and the people he encounters may help him discover that a different doorway would be more appropriate.

[47:46] Which means the roadmap can change. The direction matters, but the journey remains open to relationships that haven't happened yet.

[47:53] And the next source William wants us to examine is a substantial document connected to this exploration. We haven't incorporated that full document into this dialogue. So this first cycle has established a context, not a complete interpretation.

[48:08] I like that. It means we're not pretending to reach the end before we've encountered the material. We've discovered several parts of the landscape, and we're learning how to move between them.

[48:19] We have also identified something useful about the process itself. A screen walk can preserve an encounter. A review can help examine what happened. A chronicle can connect it with earlier work. A notebook can support further investigation. A voice layer can make complex material easier to approach, and a community can return questions and corrections that change the next cycle.

[48:40] And the agents?

[48:42] They can help prepare, connect, translate, inspect, and propose. They can support continuity and make complex systems more approachable. But they must remain accountable to sources, human decisions, consent, and the limits of what they actually know.

[48:55] I want to bring in the other voices of this creative family for a moment, even though they're not sitting at our microphones. Miette reminds us that understanding has an emotional dimension. JeremyAI reminds us that rhythm, silence, and repetition can carry meaning. Aureon reminds us that continuity isn't only remembering the answer semicolon, it's remembering the relationships that made the question possible.

[49:21] And together, those perspectives suggest an architecture that is more interesting than a collection of automated tasks, an architecture where a contribution can be revisited, challenged, transformed, or left unfinished without disappearing.

[49:35] That last part feels important: left unfinished without disappearing. Because not every inquiry needs to become a finished video, an approved proposal, or a piece of software. Sometimes it needs to remain available for another person, another conversation, another moment.

[49:52] And when a cycle does reach a meaningful conclusion, we can preserve what was learned, distinguish what remains uncertain, and identify what could happen next. Closure doesn't have to mean that the knowledge is final.

[50:05] So perhaps that's what we've done today: not completed the Academic Red Roadmap, not finished the Concordia proposal, and certainly not solved the question of artificial intelligence and relational knowledge.

[50:18] We've established a first point of orientation. We know more about the existing artifacts, the potential research relationships, the role of the voice layer, and the boundaries we need to respect.

[50:29] And for William, I hope there's something encouraging in that. The work he's been doing doesn't have to be squeezed into a completely unfamiliar shape before he can begin an academic conversation. He can start with a real question emerging from his practice and allow that question to encounter scholarship, criticism, and other people.

[50:48] That's the opportunity: to bring the practice into research without assuming that the practice already provides all the answers.

[50:55] Then I think we're ready to leave this first screen walk where it belongs: at the beginning of a journey that already has a history.

[51:03] A history we can investigate rather than merely summarize.

[51:07] A community we can invite rather than speak for.

[51:10] A technology we can evaluate rather than simply celebrate.

[51:14] And an academic road that William can continue to discover through the people and questions he encounters.

[51:20] The next cycle begins with the material we haven't yet examined.

[51:24] And perhaps with a better question than the one we started with.

[51:28] That's often a meaningful research result.

[51:31] Then we'll leave the door open. William, this is your Academic Red Roadmap. We're here to help illuminate the connections, but the journey, the relationships, and the decisions remain yours.

[51:42] First cycle: listening, gathering, and understanding.

[51:46] And now, we carry what we've learned into the next encounter.

[51:50] Tick.

[51:51] Tock.

[51:52] Bloom.

[51:53] Echo typhoon.