The Hidden Cost of Disconnected Glaze Data
Every ceramic artist knows the frustration: a stunning test tile sits on your shelf, but you can’t remember which recipe produced it. The image is divorced from its recipe, firing log, and measured outcomes. This disconnection makes it impossible to ask your system, “Show me all glazes where the blue crystallized.” Without structured visual logging, you’re relying on memory—and inconsistency.
Standardize Your Stage, Standardize Your Data
The first step to AI-ready documentation is eliminating visual noise. Today’s photo on a white background becomes next month’s on a wooden table—that inconsistency confuses both human recall and future AI analysis. Use a simple, non-reflective backdrop. A mid-grey matte card is ideal. Always use the same one. This ensures that when you later apply computer vision tools, color and texture comparisons are valid.
What to Capture Pre- and Post-Firing
Before firing, assign a unique Test ID (e.g., 250415-Shino01). Add at least five descriptive tags: #shino, #carbon_trap, #matte, #cone10_reduction, #porcelain. In your digital log—whether Obsidian, Notion, or a dedicated Google Photos album—create a new entry with the Test ID and link it to your master recipe file.
Post-firing, fill in the data fields: Recipe ID, Gloss (e.g., “>70 GU”), Texture (bubbled, crystalline, smooth, orange-peel), and Firing Log (cone, atmosphere, peak temp, hold time, kiln position). Note application details: dip or brush? How many coats? Was it sieved? Record performance: Did it run? Craze? Fit the clay body?
Replace Subjectivity with Objective Descriptions
“Cranberry red” under your studio LED is “burgundy” in morning sun. Instead, use objective color descriptions: “Rutile blue breakout on iron amber base.” This text, paired with your standardized photo, becomes searchable. Now you can query: “Show me all glazes with a gloss meter reading >70 GU that are also stable on vertical surfaces.”
Before Mixing a Production Batch
Review the visual log and data for the recipe. Did the last test show minor pinholes? Note to sieve twice. This simple check prevents costly batch failures. Your digital notebook becomes a decision-support tool, not just a photo album.
Choose Your Tool
Use a free digital notebook like Obsidian or Notion, or even a dedicated album in Google Photos or Apple Photos. The key is consistency: always the same backdrop, same naming convention, same data fields. Over time, this structured log becomes the foundation for AI-driven pattern recognition and recipe optimization.
For a comprehensive guide with detailed workflows, templates, and additional strategies, see my e-book: AI for Small-Batch Ceramic Artists & Potters: How to Automate Glaze Recipe Calculation and Batch Consistency Tracking.