TALAMANA · THE AI LITERACY MAP FOR ARCHITECTURE AND DESIGN · Judgment · AGE 18—21 · POSITIONAL · METHOD
The iteration log
The log is the artefact, and the images are its evidence.
When to use
Any time generated material is on its way into real work: a studio project, a portfolio, a client presentation. If the image will be on a wall, the log starts before the first generation.
The method
Four columns, one row per cycle. Brief version: the exact words and settings you used, numbered. Output: the image or text that came back, the one you kept, with a note on what the batch did. Judgment: what the output showed. What it got right, what it got wrong, what it revealed about the brief. Next move: the one change you will make, and why. Keep it short. A line per column is enough. The log is the artefact. The images are its evidence.
Watch for this
A log written afterwards. A log written from memory is a story about a process, not a record of one, and it will smooth over the cycle where you changed three things at once and got lucky. Write the row before you generate, and finish it before the next one.
The Lab's note
The log is the Lab's method, adopted from the first fifty-images test in 2026 and confirmed in every image workshop since. It is not the only way to keep a process honest. It is the one that turns "show your working" from a request into a habit, and it is the artefact a jury can read.
Try it
Keep a full log across one design exercise: every cycle, all four columns. Then hand the log, and only the log, to a classmate and ask them to rebuild your decision chain from first brief to final frame. Where they get lost is where the log was thin.
Prove it
Rebuild the whole decision chain for a reviewer using the log alone, with no talking and no extra images. If they can follow it, the process can be defended.
How it works
The log works because it makes three invisible things visible: how the brief changed (which shows your thinking), the judgment column (which shows how you read the batch), and the next-move column (which shows the cause of every change). Together they are the trace of thinking that a finished image no longer carries on its own. The Lab's position that output is no longer proof of thinking leads straight here. If the work cannot show its reasoning, the reasoning has to be kept somewhere, and the log is where.
What this idea builds on
What this idea opens up
Sources
- disciplined-iteration-log
- What Counts as Architectural Thinking After AI
- The AI Design Director Protocol
Open this idea on the map · The complete map · Logika · RBDS AI Lab, India · revised every edition.