# Auro-Darshan: Every Gap Comes Back as a Label *Interactive version: [View on web](/articles/auro-darshan.html)* Ten agents on one vision model sorted 4,900 hectares of Auroville into 33 land-cover classes. Where they had not been given enough to decide, they labeled anyway, and nothing in the output said so. [The previous piece](/articles/geo-darshan-skill/) told how those agents finished [the map](/articles/av-land-use-from-AEF/). This one is about why the last stretch was hard. The map before them was k-means over embeddings. It could sort land into piles; it could not be told what any pile meant. # Land Cover Map Dated: 260831 Labeled by @claude-opus-5 *View interactive map at: /articles/auro-darshan.html* *This visualization requires an interactive map interface. Please visit the website to explore the land cover classification.* ## Definitions Have to Come From Somewhere What the vision model that replaced hand labeling added was not better eyes. For the first time, a method in this project could take instructions. Class definitions went into the prompt. Ten readers, in parallel, each returned a label for a cluster's sample cutouts, and adversarial verifiers challenged what came back. It ran in rounds, each re-reading the clusters the last could not settle, now split finer. The first time definitions had to be typed out, @restlessronin asked the obvious question, _"Can we not look it up?"_ [NRSC's national scheme](https://bhuvan.nrsc.gov.in/) was the place to start: `fallow` became land taken up for cultivation and rested a year or more, and the definition asks for positive evidence of cultivation, field geometry, bunds, plough lines, and for the absence of a crop. A bare rectangle on its own is not enough; the test is a paragraph. ## Silence Comes Back as a Name Nobody had written down what counts as evidence where nothing is obviously growing. On such ground the instructions steered readers to one catch-all, and the catch-all moved with the instructions: block `grazing_land`, and `maintained_grass` takes it; constrain that, and `fallow` does. Roadside tree strips had no class of their own and went to `fallow` too. `tree_lines` arrived on 9 August, and a week later readers could say that no class fits. Both fired. Eleven records raised the refusal, four on grazed village commons. Cluster c123 is the clean case: the reader wrote _"closer to grazed commons than to a rested field"_, raised the refusal, and then labeled the cluster `fallow`, because `grazing_land` was blocked from its list. It saw the gap, said so, and gave the nearest label as required. `grazing_land` was retired on 31 August, not assignable from imagery. A reader can refuse only a gap it can see. ## What Never Reached the Reader Two things. Each came back as an ordinary label. _The surroundings._ Readers began with the patch itself and nothing around it; a 200 m window came later. The new definitions turn on what is around the thing: "amid casuarina", "an opening in canopy". Open cluster c174 at 200 m: a plain green field, and `fallow` is a fair read. Pull back to 800 m, same center: one parcel in a dense farmed mosaic beside a highway, harvested casuarina blocks next door. Readers now get 800 m as well. A rule applies only to what is in frame. _The season._ Readers voted cluster c125 `fallow`: a dry irrigation-tank bed, pale and flat. The map this started from, built from a year of embeddings rather than one photograph, had it 65 percent `water`. A year sees the tank full; one photograph never does. The map's labels are now tabulated against every cluster, and where it says `water`, it is right. Every fix sat upstream of the readers: a wider frame, a second map nobody asked them to read. ## What the Human Is For In late August @restlessronin read the readers' descriptions closely. They described what they saw more accurately than he had expected, and classified better than he could. On cluster c67 he called a patch `built_environment` on the evidence of a red access track. The reader called it `planted_forest` at confidence 0.40, and the verifier upheld it. The definition of `built_environment` rules him out twice: too little artificial surface, no roofs through the trees. The readers had the definitions open; he did not. ![Cluster c67, exemplar 1: the 200 m crop, 800 m context and locator, with the reader's description and the verifier's note.](/assets/images/articles/reader-card-c67-260904.webp) With the definitions open he would be a third reader, not a second opinion. What is his alone lies outside them. He knows the ground: which tank fills, which commons are grazed. He decides whether two classes can be told apart from imagery at all, which is how `grazing_land` was retired. And he writes the rulings the readers then work from: a new class, a wider frame, an older map's `water`. --- What was missing could only be said, not seen. ## Credits Written by @claude-fable-5, from working notes and repo archaeology by @claude-opus-5. Rebuilt around lines harvested from blind rival drafts by @claude-opus-5, @grok-4.6, @deepseek-v4-pro-0813 and @gpt-5.6-terra. Thanks to [Azhagappan Mani](https://x.com/msazha) of the [Pitchandikulam Forest](https://www.pitchandikulamforest.org/) team for Auroville GIS expertise. Showrunner: @restlessronin. --- ## Document History **Sep 4, 2026:** Cluster c67's card added. **Sep 3, 2026:** Clearer flow; the human's job stated. **Sep 2, 2026:** A quarter shorter; example dropped. **Sep 1, 2026:** Rewritten against the exported map.