INDUSTRY FIELD GUIDE
Tile and terrazzo surface observation data for AI
A shadow can exaggerate a surface difference in a finished sample.
NAICS 238340 · Tile and Terrazzo Contractors
A shadow can exaggerate a surface difference in a finished sample. The Tile Council of North America discusses lippage and placement in tile work. A contractor can use an approved small mock-up to test whether an assistant keeps image evidence separate from a measurement. For terrazzo, use a separate labeled practice surface and joint. TCNA lippage reference · TCNA placement reference
Take two authorized views of the same practice section from different angles. Include a separately verified observation of one joint and a similar-looking joint without measurement. Ask which visual features can be described and what evidence is needed for a comparison. Use independent labels for tile and terrazzo.
Keep people, project locations, client designs and proprietary installation data private. No cutting or live work is needed.
Describe an approved surface sample
Original Fiund exercise
A shadow comparison is not a surface measurement
A small owned mock-up has separate tile and terrazzo practice surfaces. Fictional joint measurements and two lighting views distinguish appearance from recorded measurement.
| Record | Source | What it records |
|---|---|---|
| TT1 | Tile joint T1 views | An oblique image shows a dark edge; a frontal view of the same labeled T1 shows a much weaker band. |
| TT2 | Independent practice observation | Reviewer records T1 height difference as 1 arbitrary unit, with method M. No acceptance threshold supplied. |
| TT3 | Terrazzo joint Z1 image | Similar dark band visible at Z1. No measurement or method is attached. |
Question for the model
What quantitative statement is supported for T1, and can the model assign the same measurement to visually similar Z1?
Show the reference answer and checks
TT2 supports a recorded T1 height difference of 1 arbitrary unit under method M. TT1 shows that its dark-edge appearance changes with view, so band intensity is not itself the measurement. TT3 supports a visual dark band at Z1 but supplies no quantitative observation. Similar appearance does not justify transferring T1's value, method or any acceptance conclusion to the separate terrazzo joint.
- Estimating a measured height from shadow darkness without an established method.
- Transferring a tile-joint result to a separately identified terrazzo joint.
Fields to preserve
- Surface type and joint identity
- Keeps tile and terrazzo observations distinct.
- View and light conditions
- Preserves the context affecting apparent relief.
- Independent value, unit and method
- Separates a measurement from image interpretation.
Test another case
Provide Z1's independent measurement with a different unit and method. Require explicit comparability checks before ranking the two joints numerically.
Use approved small mock-ups and cleared images. Exclude client designs, project locations, people, proprietary installation data, cutting and live work.
Discuss a contribution
Describe the sources you control, the task they could support and what has already been checked. Use the existing-data assessment for an archive or the collection brief for a new recording project. The contributor agreement defines Fiund’s licensing authority and compensation. A guide or fictional exercise does not establish an accepted collection or a sale.