# The Detail Hewn Shop Notes ยท https://hewn.build/shop-notes/the-detail AI-drafted, not yet checked: Drafted by Chip, the writing system Hewn runs on Claude, from Hewn's craft corpus. Jonathan Salomons has not read this note yet, and no outside reader has checked it. Read it as a shop note: check a number before you cut to it. Cite as: "The Detail." Hewn Shop Notes, hewn.build/shop-notes/the-detail. Drafted 2026-04-24; AI-drafted, not yet checked. --- You got the proportions right. Leg taper looks good. Shadow line under the top reads clean. Reveals are consistent. The base makes sense. And still, standing three feet back, something doesn't land. Nine times out of ten, it's the details. Details aren't decoration. They're the transitions your plan never mentioned. Where one surface meets another. Where the piece meets the floor. Where someone's hand finds the drawer pull. Get them right and nobody notices. Get them wrong and they undercut everything above them. Here are the five that matter most. ## Edge treatments A raw arris is technically complete. It's also what you get when you don't make a decision. Every edge is a statement, even if the statement is "I ran out of time." Four options, each saying something different. **Square.** Surfaces meet at a precise intersection. Reads clean, contemporary, unforgiving. The Eames storage unit runs square edges on every case corner. That thin highlight line where light catches the intersection defines the whole piece. But square edges expose every flaw. Any tearout, any lifted fiber shows. **Chamfer.** A flat 45-degree cut. Honest and direct. A 1/16" to 1/8" chamfer on the bottom of a leg protects end grain from splitting and tells the viewer someone thought about that corner. On a tabletop underside, the chamfer creates the shadow line that makes the top float. This is the detail that separates furniture that's been thought about from furniture that's been built. **Roundover.** Under 1/8" reads contemporary and tactile. That's what your hand finds running across a table edge. Over 3/8" reads casual or dated, depending on context. On MCM work, a roundover usually reads wrong. Too soft. On a kitchen table meant for daily use, the same radius reads warm. **Profiles.** Coves, ogees, beads. These reference a specific historical moment. On reproduction work, essential. On contemporary work, ask the question: does this profile fit the vocabulary of this piece, or is it here because it was in the router bit set? The rule is simple. Pick a language and use it throughout. A chamfer on the top and a roundover on the stretchers isn't interesting contrast. It reads as indecision. ## Reveals A reveal is the gap between elements. Drawer face to case. Door margin. Face frame setback. Plans give you dimensions. They don't tell you why those dimensions matter. A drawer with 3/32" at the top and 1/8" at the bottom looks like a careful fit that shifted. The same drawer with 3/32" on all four sides looks intentional. The difference is one pass with a hand plane. The difference is visible from across the room. Three decisions worth making deliberately. Door and drawer margins: consistency matters more than the specific number. 1/8" on all four sides of an inset door reads designed. 1/16" on one side and 5/32" on the other reads as a gap you couldn't close. Face frame setback: keep it equal on both sides. The shadow line this creates is visible at an angle, and if it's different left to right, the piece reads crooked even when it's square. Rail widths between drawers: equal widths read as a grid. Bottom rail heavier than top reads as considered hierarchy. Random variation reads as compromise. ## Hardware Hardware is the last thing you install and one of the first things anyone touches. It's also where discipline tends to collapse. People pick hardware based on what's at the store or what's in the bin. Hardware is a design choice. Treat it like one. Long pulls, 4 to 8 inches, mounted horizontally, are the MCM default. A walnut pull on a walnut drawer says nothing. That's what makes it perfect. Knobs are traditional on Shaker and Colonial work. One hand, centered, functional. Edge pulls eliminate surface hardware entirely, but require a more complex edge build. Exposed hinges on case pieces are a design decision too. Traditional work shows them. Contemporary work hides them. MCM work often made them honest, a barrel hinge doing its job in plain sight. Match the hardware to the tradition, or make a deliberate choice to break from it. What costs you is ignoring the decision and hoping nobody looks. ## The back panel Most case furniture has a back nobody sees. Hardboard, plywood, rabbeted solid wood, frame-and-panel. Two things worth considering. First, structural: a fitted plywood panel in a rabbet prevents racking. It's the element that keeps the case square under load for twenty years. Face-nailed hardboard does some of this. A proper rabbeted panel does it completely. Second, the version nobody talks about. A case piece with a finished back, dovetailed corners and a frame-and-panel rear that reads as considered even against a wall, is a different object than one whose back is an afterthought. The maker knows. And in some way that doesn't show up on a spec sheet, that intention ends up in the piece. Most good furniture has hardboard backs. That's fine. But decide rather than default. ## The foot The foot is the detail most visible in use, because it's at eye level when you're seated. Raw end grain with no treatment reads like a piece whose maker ran out of time. End grain collects dirt, absorbs finish unevenly, and chips along the arris in softer species. A 1/16" to 1/8" chamfer on all four bottom arrises is the shop standard. Protects the grain, reads clean, takes thirty seconds per leg. That's the minimum treatment that makes a leg look finished rather than abandoned. Beyond that, stopped chamfers add intentionality. Pad feet, bracket feet, turned feet each belong to a tradition and signal it. On a tapered MCM leg inset from the case, the answer is the slight chamfer all around. The taper carries the story. The foot just ends. ## The 5% that confirms the other 95% None of these details require expensive tools or extra materials. They require deciding. Edge treatments consistent across the piece. Reveals controlled and repeatable. Hardware matched to the vocabulary. Back panel chosen rather than defaulted. Foot resolved, not abandoned. The difference between furniture that reads as designed and furniture that reads as built lives here, in the last 5% of decisions made after the joinery is cut and the panel is glued. These decisions don't rescue bad proportions. But they confirm good ones. A detail that isn't deliberate reads as oversight. The piece knows.