Logging with a Lens: Using AI to Document Glaze Tests and Results

We need to write a concise educational Dev.to article for professionals about AI automation in AI for small batch ceramic artists/potters how to automate glaze recipe calculation and batch consistency tracking. The title must be engaging, include "ai" or the topic. The topic given: "Logging with a Lens: Using Visual AI to Document Glaze Tests and Results". So title should reflect that, include "ai". Eg: "Logging with a Lens: Using AI to Document Glaze Tests and Results". Must be 400-500 words. Need to count words.

We must include facts from e-book: advanced search, application notes, before mixing a production batch, color description objective, disconnection, firing log, inconsistency, performance, subjectivity, texture, the stage (use simple non-reflective backdrop), the tool (use free digital notebook like Obsidian or Notion, or dedicated album in Google Photos or Apple Photos), unsearchability, post-firing and pre-firing checklist items.

We need to include one specific tool name and its purpose (from facts). Eg: Obsidian as a digital notebook for linking images and data.

We need a mini-scenario: 2 sentences showing principle in action.

Implementation: 3 high-level steps (not exact prompts). Conclusion: summarize key takeaways only.

Tone: professional, conversational, helpful, authoritative.

We must output markdown with # title, ## subheadings, paragraphs.

We must not use placeholders, no thinking process, no explanatory preamble. Just article content.

We must count words between 400-500 inclusive.

Let's draft around 440 words.

We need sections: Title, maybe ## Introduction, ## Core Principle, ## Tool Spotlight, ## Mini-scenario, ## Implementation Steps, ## Conclusion.

We need to ensure each sentence adds value.

Let's draft then count.

Draft:

Introduction

Small‑batch potters often lose valuable glaze data because photos sit isolated from recipes, firing logs, and measured results. This disconnect makes it hard to repeat successes or troubleshoot defects, wasting time and materials. By pairing consistent visual capture with AI‑powered tagging, you turn every test tile into a searchable record.

Core Principle: Structured Visual Logging

The key is to treat each glaze test as a data object that bundles image, recipe ID, application notes, firing details, and objective measurements. When these fields are linked in a searchable notebook, AI can retrieve subsets—like “all glossy, stable glazes”—instantly, eliminating reliance on memory or scattered spreadsheets.

Tool Spotlight: Obsidian

Obsidian is a free, local‑first note‑taking app that lets you create a note for each test, embed the photo, and attach tags and properties. Its graph view shows connections between recipes, firing logs, and test images, turning your visual archive into a navigable knowledge base.

Mini‑Scenario

You photograph a new shino test on a mid‑grey matte card, tag it #shino #carbon_trap #cone10_reduction, and log the recipe ID, gloss reading of 68 GU, and texture “slight orange‑peel”. Two weeks later you ask Obsidian, “Show me all shino tests with gloss >65 GU and no crazing,” and the relevant notes appear instantly.

Implementation Steps

1. Standardize the capture stage – use the same non‑reflective backdrop, consistent lighting, and a fixed camera angle for every tile; save the image with the test ID in the filename.
2. Create a test note – in Obsidian, make a new note named with the Test ID (YYMMDD-Name-Number), embed the photo, and fill properties: Recipe ID, Gloss, Texture, Firing Log (cone, atmosphere, hold), Application Notes (dip/brush, coats, sieving), and at least five descriptive tags.
3. Link and index – connect the test note to your master recipe note via a bidirectional link, then use Obsidian’s search or graph view to run queries like the advanced search example, ensuring future batches can reference verified visual data.

Conclusion

By standardizing photo capture, embedding objective measurements, and linking everything in a searchable notebook, you transform glaze testing from a guesswork hobby into a repeatable, AI‑enhanced workflow. The result is faster recipe refinement, fewer defective batches, and a growing visual library that works as hard as you do.

Now count words.

Let's count manually.

Title line: "Logging with a Lens: Using AI to Document Glaze Tests and Results"

Words: Logging(1) with2 a3 Lens:4 Using5 AI6 to7 Document8 Glaze9 Tests10 and11 Results12

So title 12 words.

Now body.

Introduction

Small‑batch(1) potters2 often3 lose4 valuable5 glaze6 data7 because8 photos9 sit10 isolated11 from12 recipes,13 firing14 logs,15 and16 measured17 results.18 This19 disconnect20 makes21 it22 hard23 to24 repeat25 successes26 or27 troubleshoot28 defects,29 wasting30 time31 and32 materials.33 By34 pairing35 consistent36 visual37 capture38 with39 AI‑powered40 tagging,41 you42 turn43 every44 test45 tile46 into47 a48 searchable49 record50.

So intro paragraph 50 words.

Core Principle: Structured Visual Logging

The1 key2 is3 to4 treat5 each6 glaze7 test8 as9 a10 data11 object12 that13 bundles14 image,15 recipe16 ID,17 application18 notes,19 firing20 details,21 and22 objective23 measurements.24 When25 these26 fields27 are28 linked29 in30 a31 searchable32 notebook,33 AI34 can35 retrieve36 subsets—like37 “all38 glossy,39 stable40 glazes”—instantly,41 eliminating42 reliance43 on44 memory45 or46 scattered47 spreadsheets48.

Count words: Let's count.

The1 key2 is3 to4 treat5 each6 glaze7 test8 as9 a10 data11 object12 that13 bundles14 image,15 recipe16 ID,17 application18 notes,19 firing20 details,21 and22 objective23 measurements.24 When25 these26 fields27 are28 linked29 in30 a31 searchable32 notebook,33 AI34 can35 retrieve36 subsets—like37 “all38 glossy,39 stable40 glazes”—instantly,41 eliminating42 reliance43 on44 memory45 or46 scattered47 spreadsheets48.

48 words.

Tool Spotlight: Obsidian

Obsidian1 is2 a3 free,4 local‑first5 note‑taking6 app7 that8 lets9 you10 create11 a12 note13 for14 each15 test,16 embed17 the18 photo,19 and20 attach21 tags22 and23 properties.24 Its25 graph26 view27 shows28 connections29 between30 recipes,31 firing32 logs,33 and34 test35 images,36 turning37 your38 visual39 archive40 into41 a42 navigable43 knowledge44 base45.

Count: 45 words.

Mini‑Scenario

You1 photograph2 a3 new4 shino5 test6 on7 a8 mid‑grey9 matte10 card,11 tag12 it13 #shino14 #carbon_trap15 #cone10_reduction16,17 and18 log19 the20 recipe21 ID,22 gloss23 reading24 of25 68 GU,26 and27 texture28 “slight29 orange‑peel”.30 Two31 weeks32 later33 you34 ask35 Obsidian,36 “Show37 me38 all39 shino40 tests41 with42 gloss43 >65 GU44 and45 no46 crazing”,47 and48 the49 relevant50 notes51 appear52 instantly53.

Count: 53 words.

Implementation Steps

1. Standardize the capture stage – use1 the2 same3 non‑reflective4 backdrop,5 consistent6 lighting,7 and8 a9 fixed10 camera11 angle12 for13 every14 tile;15 save16 the17 image18 with19 the20 test21 ID22 in23 the24 filename25.
2. Create a test note – in26 Obsidian,27 make28 a29 new30 note31 named32 with33 the34 Test35 ID36 (YYMMDD-Name-Number),37 embed38 the39 photo,40 and41 fill42 properties:43 Recipe44 ID,45 Gloss,46 Texture,47 Firing48 Log49 (cone,50 atmosphere,51 hold),52 Application53 Notes54 (dip/brush,55 coats,56 sieving),57 and58 at59 least60 five61 descriptive62 tags63.
3. Link and index – connect64 the65 test66 note67 to68 your69 master70 recipe71 note72 via73 a74 bidirectional75 link,76 then77 use78 Obsidian’s79 search80 or81 graph82 view83 to84 run85 queries86 like87 the88 advanced89 search90 example,91 ensuring92 future93 batches94 can95 reference96 verified97 visual98 data99.

Let's count words for implementation steps. We'll count each numbered step as part of the paragraph.

Step1 line: "1. Standardize the capture stage – use1 the2 same3 non‑reflective4 backdrop,5 consistent6 lighting,7 and8 a9 fixed10 camera11 angle12 for13 every14 tile;15 save16 the17 image18 with19 the20 test21 ID22 in23 the24 filename25."

Count: 1. (ignore maybe) but we count words after.

Standardize1 the2 capture3 stage4 –5 use6 the7 same8 non‑ref