The Glass Database isn’t scraped, bulk-imported, or auto-filled. Every entry is added by hand, reviewed with care, and designed to live here for the long haul. This page explains how that happens—and how we think about accuracy, sourcing, and structure.
🫧 Where Entries Come From
Most objects are submitted by registered users who own the piece they’re describing. They upload images, add notes, and sometimes tell a story about how the piece came to them. Our editors review every submission before it goes live.
We’re also adding historically significant items from catalog scans, reference books, and museum collections—but always with proper context and credit. And always with a light hand. This isn’t about flooding the archive. It’s about building something sturdy and shareable.
🔖 What We Do With Sources
When we link an object to a source—like a trade catalog or a specialist book—we log that clearly. We don’t just mention the title; we record the page, the image, and treat the book itself as an object in the database.
(Some catalogs, books, and documents are listed as reference only. If we’ve received permission, we show the full source; otherwise, we just share enough to help you find it elsewhere.)
But: we don’t rely on sources blindly. Even the most respected books get things wrong. Being authoritative isn’t the same as being deferential. If we find strong visual or historical evidence that contradicts a published source, we’ll say so—and note the reasoning.
This is why every attribution carries a quiet confidence level. Sometimes it’s a match. Sometimes it’s a maybe. Sometimes we just don’t know yet.
🗂️ How We Organize the Chaos
Most objects enter the database with more questions than answers. We might not know who made them, where they came from, or when they were produced—and that’s okay. Rather than guessing, we start with what’s visible and verifiable: shape, style, materials, maybe a mark. Each object gets a general label that’s meant to be helpful, not misleading.
If there’s no confident link to a maker or supplier, the object remains unassigned—like an orphan waiting for its story to surface. As more evidence comes in—familiar forms, shared traits, matching documentation—we connect the dots with care. An object only joins a category, network, or maker page when there’s a reason to.
This archive isn’t about jumping to conclusions. It’s about leaving space for things to become clear, when they’re ready.
📚 The Reference Library
We’re building a record of the resources available: books, catalogs, exhibition leaflets, interviews, and more. Each source becomes its own “object” in the database, complete with publication details and links to the items it helps describe.
Over time, this becomes a library you can browse alongside the objects themselves. Want to know where a 1954 catalog page first mentions a shape? You’ll be able to find it, cite it, and cross-check it.
The system also helps us notice patterns—and recognize when a misattribution has quietly echoed for decades, copied from one source to the next without question.
🧾 Change Logs & Transparency
Every submission is versioned internally. We track edits, updates, reclassifications, and corrections. You don’t see all of that yet, but we’re working on making that history visible—so you can follow how an entry evolved over time.
If you suggest a correction or add context to an existing piece and it’s accepted, you’ll be credited. Quietly, but visibly.
🧠 Still Learning
We don’t pretend to have it all figured out. The database is expanding, but so is our thinking. Taxonomies change. Attribution trails get stronger or fall apart. And that’s part of what makes this living record worth maintaining.
Every clear photo, honest guess, or researched challenge strengthens the system.
We put accuracy first—even when there’s not much to go on. The system stays tidy, but that doesn’t mean every object comes with a full story. We show what’s known, make space for what isn’t, and leave the door open for discoveries to catch up.