Why must every Dorset button cast stitch pass through the ring center hole rather than between adjacent stitches?

The cast stitch in Dorset button making is a specific buttonhole stitch structure, not a blanket stitch. The structural distinction is where the needle passes on each stitch: in a Dorset cast stitch, the needle enters the ring interior through the center hole and returns over the ring edge on every stitch. This means each stitch is anchored directly to the ring as a physical object. In a blanket stitch worked around a cord (the superficially similar technique used in some other ring-wrapping crafts), the needle passes between adjacent stitches rather than through the ring interior — each new stitch locks around the previous stitch rather than around the cord itself. In Dorset button casting, a stitch that misses the center hole and instead loops between two existing cast stitches creates a stitch that is anchored to other stitches rather than to the ring. When warp spoke tension is applied across the ring for pattern filling, this unanchored stitch slides toward the higher tension, distorting the cast stitch spacing at that position. The distortion is small per stitch but cumulative if multiple stitches miss the center hole. More critically, spoke spacing derives from the cast stitch spacing — each spoke exits between adjacent cast stitches at the ring edge. If the cast stitch spacing is uneven due to unanchored stitches, the spoke angles are uneven, and the woven fill pattern shows uneven divisions between the spiral circuits. The fix is to maintain the through-center-hole discipline on every stitch, which is easier to do rhythmically than occasionally — establishing the needle angle toward the center at the beginning of each stitch as a consistent physical motion rather than estimating the correct approach each time.

Why do Dorset button woven wheel patterns require an odd number of spokes?

The woven wheel fill in Dorset buttons — and in all woven wheel embroidery fills — relies on the over-under weaving sequence shifting position after each complete circuit around the spokes. The shift is what produces the spiral appearance: after each circuit, the weaving thread is one position further ahead in the over-under sequence relative to where it started, so each subsequent circuit crosses the same spoke from the opposite side to the previous circuit. This creates the characteristic interlocked spiral appearance of a well-worked woven wheel. The shift only occurs if the number of spokes is odd. With an odd number of spokes (5, 7, 9, 11, and so on), after one complete circuit the weaving needle has passed over or under the spokes one by one and has arrived at a position that is 0.5 circuits out of phase from its starting position — the first stitch of the next circuit will go under a spoke that the first stitch of the previous circuit went over. This phase offset is what produces the spiral. With an even number of spokes (6, 8, 10, 12, and so on), after one complete circuit the weaving needle has arrived back at exactly its starting phase — the first stitch of the next circuit will go over a spoke that the first stitch of the previous circuit also went over. Every circuit is identical to the previous one. The result is not a spiral but a radial pattern: alternating spokes that are always covered and always exposed, producing a striped appearance radiating from center to edge. The striped pattern from even spoke counts is structurally stable and can be intentional in some decorative schemes, but it is a distinct appearance from the spiral woven fill that most Dorset button patterns target. The practical implication: before laying spokes, count out the target spoke number on paper (9, 11, or 13 for most ring sizes) and verify it is odd before anchoring the first spoke.

What is spoke tension equalization and why does a slack spoke produce a visible failure in the woven fill?

Spoke tension equalization is the process of checking and correcting individual spoke tensions after all spokes have been laid across the ring and before the final spoke end is permanently anchored to the cast stitch ring. Each spoke is a structural warp thread — the weaving circuits pass over and under it, and the spoke's tension determines how much the circuit thread deflects toward or away from the ring center at each spoke crossing. If all spokes have equal tension, the circuit thread deflects the same amount at every crossing and the woven surface appears uniform with even spiral spacing. If one spoke has less tension than the others — a common outcome when the thread was held slightly more loosely during that spoke's laying pass — the circuit thread deflects toward the center more at that spoke position because the lower resistance allows it to pull the spoke inward. The result in the finished button is a visible dip or pucker in the woven surface at the position of the slack spoke: the spiral circuits appear closer together at that position, and the spoke itself may be noticeably angled inward toward the center rather than running straight from ring edge to ring edge through the center crossing. This dip is structural, not a visual artifact of lighting or photography, and it becomes more pronounced as more weaving circuits are added because each circuit contributes additional pull at the slack point. The equalization process: after all spokes have been laid in their planned positions, before anchoring the last spoke end, systematically pull each spoke tip at the ring edge very gently in a direction away from the center — approximately one millimeter of incremental pull per spoke — to take up any accumulated slack from the laying pass. Work around all spokes twice. Then anchor the final spoke end. The test: with all spokes equalized and anchored, the button face should feel equally resistant to gentle pressure at any position across the spoke pattern. If pressing at one position feels visibly softer than pressing at an adjacent position, that position's spoke requires additional tension.

How does the center anchor wrap prevent center collapse in a Dorset button woven wheel?

The center collapse failure in woven wheel Dorset buttons occurs at the beginning of the first weaving circuit when the circuit thread's tension pulls the converging spoke crossing off-center. All spokes meet at a single central point — the crossing of all the warp threads at the ring center. In theory this crossing is positioned exactly at the geometric center of the ring. In practice, the crossing is a loose convergence of multiple threads that is held in position only by the tension of the individual spokes pulling outward from it. When the first weaving circuit begins and the weaving thread passes over and under spokes in sequence, each stitch applies lateral tension that tugs the central crossing slightly toward the side being worked. If the central crossing is not secured before weaving begins, it migrates toward the area of the first few stitches of the first circuit, so that by the time the first circuit is half complete, the center crossing has shifted off-center by 1–3mm. All subsequent circuits are then woven around this shifted center, and the finished button face shows an asymmetric spiral pattern that is visibly off-center. The center anchor wrap prevents this: before beginning the first over-under weaving circuit, pass the weaving thread twice around the entire central crossing — wrapping clockwise once, then counterclockwise once, pulling snug after each wrap. These two wraps bind the spoke convergence threads together at the center crossing point and create a small hub structure that resists lateral displacement under the tension of the first circuit's weaving stitches. The hub is small enough to be covered by the first two or three weaving circuits and is invisible in the finished button. After the center anchor wraps are complete, the weaving thread is positioned at the center and can begin the first over-under circuit immediately without any additional anchor. The second and subsequent circuits reinforce the center position automatically because each circuit adds structural constraint to the pattern from the inside outward.

How does Apple Tax affect Dorset button creator Patreon income from November 2026?

Apple's 30% iOS IAP fee on Patreon subscriptions begins November 1, 2026. Dorset button creator audiences are heavily iOS-weighted across the platforms where slow craft textile content performs well. Instagram Dorset button content — photography of finished button collections, in-progress ring preparation, close-up shots of woven wheel surfaces — reaches 70 to 82 percent iOS in UK and US markets where the historical UK craft tradition has its strongest contemporary following. Pinterest boards of historical Dorset button patterns, color scheme explorations, and ring size reference guides reach 66 to 80 percent iOS. YouTube slow-craft Dorset button tutorial videos reach 52 to 68 percent iOS because longer tutorial formats attract desktop viewers on secondary screens more than the short-form photography platforms. Revenue impact from November 2026: at $150 per month with 72% iOS: $150 times 0.72 times 0.30 equals $32.40 per month ($388.80 per year). At $300 per month with 76% iOS: $300 times 0.76 times 0.30 equals $68.40 per month ($820.80 per year). At $500 per month with 78% iOS (a multi-platform Dorset button creator with both Instagram pattern photography and YouTube tutorial content): $500 times 0.78 times 0.30 equals $117 per month ($1,404 per year). A Dorset button creator at $300 per month with a heavily Instagram-based audience loses over $800 annually beginning November 2026 — enough to fund all ring materials, thread stock, and reference materials for a full year of production.

Patreon for Dorset button creators: every cast stitch through the ring center hole, the odd spoke count rule for woven wheel fills, spoke tension equalization before anchoring, the center anchor wrap, why a sharp needle creates an irreversible split spoke failure, button stitch center mechanics, thread ending in cast versus weave layers, and the Apple Tax in 2026

2026-08-15 · ~5,300 words · KeepTier

Dorset buttons are one of the shortest-path crafts in the textile arts: the basic tools are a ring, a needle, and thread, and a beginner can complete a first button in under two hours. This accessibility is the Patreon problem. When the mechanical barrier to entry is low, tutorial content quickly saturates — the first video a subscriber finds shows the finished technique, the second shows a color variation, the third shows the same technique with a different thread weight. After three videos the subscriber has seen everything that video demonstration can convey, and there is no visible reason to pay monthly for more of the same. The solution for Dorset button creators is to produce Patreon content at the specification layer that tutorials cannot reach: the specific mechanical rules that prevent the failures that beginners cannot diagnose from video alone, documented with precision that allows subscribers to apply those rules on ring sizes and thread weights that were not demonstrated in the video. A video showing a woven wheel button being completed correctly cannot convey why the same sequence produces a striped rather than spiral fill if the creator happened to have an even number of spokes on that ring, or why a spoke that was laid two millimeters slack produces a dip in the surface that compounds with every subsequent weaving circuit, or why the needle type matters at all until the subscriber has already completed six circuits over a split spoke and cannot go back. This post covers seven construction mechanics of Dorset buttons that belong in Patreon documentation rather than tutorial video: the cast stitch center-hole discipline; cast stitch count calculation; the odd spoke count rule; spoke tension equalization; the center anchor wrap; why blunt needles are a mechanical requirement; button stitch center mechanics; and thread ending methods that differ between the cast layer and the woven fill.

Cast stitch: why every stitch must pass through the ring center hole, not between adjacent stitches

The cast stitch — covering the ring with buttonhole stitches before any pattern filling is worked — is described in most Dorset button instructions as "buttonhole stitches around the ring, each stitch through the center." The "through the center" specification is treated as description rather than rule, which is why beginners commonly violate it without realizing it. The violation: instead of passing the needle through the ring's center hole on each stitch, the needle passes between two adjacent existing cast stitches on the near side of the ring, loops under the ring, and returns. The visible result is identical to a correctly worked stitch — a neat loop sitting on the outer edge of the ring. The structural difference is critical.

A cast stitch that passes through the center hole is anchored directly to the ring as a rigid object. The ring holds the stitch in place regardless of what tension is applied to adjacent stitches or to the spokes laid across the ring in subsequent steps. A cast stitch that loops between two existing stitches instead of through the center hole is anchored to those two adjacent stitches — it holds them together rather than anchoring itself to the ring. The ring's rigid structure is not involved in anchoring this stitch at all. When warp spoke tension is applied during pattern filling — each spoke crosses the ring and exits between cast stitches at both edges — the tension pulls the cast stitch ring outward on both sides of the spoke's exit points. Every correctly anchored cast stitch (through-center-hole) holds its position under this tension because the ring itself resists the outward pull. An incorrectly anchored cast stitch (between-stitches) slides along the ring's circumference toward the spoke exit points under the same tension, because its anchor is only the adjacent cast stitches and not the ring.

The practical failure: as spokes are laid and tensioned across the ring, any between-stitches cast stitches migrate toward the spoke exit points on each side of the ring. This produces two visible problems in the finished button. First, the cast stitch spacing becomes uneven — gaps appear where the between-stitches stitches have migrated away from, and crowding appears at the spoke exit points where they have accumulated. Second, and more consequential for patterned fills, spoke spacing becomes uneven because spoke exit positions correspond to cast stitch positions. If several adjacent cast stitches have shifted toward one spoke exit, the spokes on either side of that area are no longer equally distributed around the ring, and the woven fill pattern shows unequal sector widths — some sectors are visibly wider than others, and the spiral pattern, if the spoke count is correct, appears to accelerate in the wider sectors and compress in the narrower ones.

The through-center-hole discipline is most easily maintained by establishing the needle angle before each stitch rather than after. As the working thread is drawn through from the previous stitch and positioned for the next, the needle tip is oriented toward the ring center before it approaches the ring edge. Working this way, the needle naturally enters the center hole because the approach angle toward center intersects the hole before the ring edge. If the approach angle is toward the ring edge rather than toward the center, the needle tip tends to catch on the cast stitches already on the near side of the ring, and the beginner's instinctive response — to slip the needle between those existing stitches to release it — produces the between-stitches error. Repositioning the approach angle toward center eliminates this catch.

Cast stitch count: calculating target stitches from ring circumference and thread diameter

The number of cast stitches required to cover a ring without gaps (underpacking) and without stitches overriding each other (overpacking) is determined by the ring circumference and the thread diameter. The target stitch count formula: ring diameter in millimeters × π × 0.75 ÷ thread diameter in millimeters. The 0.75 factor accounts for compression: buttonhole stitches laid around a ring are pulled snug and nest against each other, covering slightly more circumference per stitch than the thread diameter alone would predict. This compression factor varies from 0.70 to 0.80 depending on the thread's softness and the stitch tension used — firmer threads compress less, and cotton perlé (which is plied and relatively firm) sits closer to 0.75 than a soft single-strand cotton.

Example: a 12mm diameter brass ring has circumference of 12 × π = 37.7mm. With DMC perlé #8 thread (approximate diameter 0.90mm), the target stitch count is 37.7 × 0.75 ÷ 0.90 = 31 stitches. In practice, the common instruction "20–24 stitches on a 12mm ring" for perlé #8 represents a deliberately lower stitch count that uses a looser compression factor near 0.55 — this produces a cast stitch layer where the stitches sit side by side but are slightly separated, creating defined ribs rather than a solid wrapped surface. Both targets are correct for different aesthetic outcomes: the tighter count produces a more uniformly covered ring with smoother appearance; the looser count produces visible rib texture that is characteristic of historical English Dorset buttons.

The underpacking failure: if the stitch count is significantly below target, gaps appear between adjacent cast stitches where the ring material is visible through the covering. In patterned fills, these gaps allow spokes to slide sideways off their intended exit positions. The overpacking failure: if the stitch count significantly exceeds target, cast stitches begin to override each other — the later stitches ride up over the earlier ones because there is insufficient space for them to sit flat on the ring. The override creates an uneven surface that causes spokes to exit the ring at different heights, producing unequal visual weight on the two sides of each spoke as it crosses the ring edge. The formula gives the target; starting three to five stitches below target and stopping when the ring is correctly covered is the practical protocol that adjusts for individual thread compression behavior on the specific ring material being used.

The odd spoke count rule: why even spokes produce radial stripes and odd spokes produce the spiral fill

The woven wheel fill — passing the weaving thread over and under the spokes in sequence, circuit by circuit from center outward — is described in tutorials as "weave over-under in circles from the center." What tutorials cannot convey in real time is why the same sequence with eight spokes produces a striped result and with nine spokes produces the spiral fill that is characteristic of correctly worked Dorset buttons and woven wheel embroidery fills. The distinction is mathematical and determines the visual character of the entire button face.

With an odd number of spokes, trace the over-under sequence for one complete circuit. With 9 spokes: the weaving thread goes over spoke 1, under spoke 2, over 3, under 4, over 5, under 6, over 7, under 8, over 9, and arrives back at spoke 1 to begin the next circuit. But the next circuit must start under spoke 1 — because the rule is alternating over-under, and spoke 1 was over in the previous circuit. The phase has shifted by half a step. The second circuit goes under 1, over 2, under 3, over 4, under 5, over 6, under 7, over 8, under 9, and again arrives at spoke 1, which must now be over again. The alternation creates a spiral: each circuit crosses every spoke from the opposite side compared to the previous circuit, and the crossing points march around the pattern, producing the characteristic spiral rotation of the correctly worked woven wheel.

With an even number of spokes, trace the same sequence. With 8 spokes: the weaving thread goes over spoke 1, under 2, over 3, under 4, over 5, under 6, over 7, under 8, and arrives back at spoke 1. The circuit is complete at spoke 1 — and the next stitch must be over spoke 1 again, the same direction as the previous circuit started. Every subsequent circuit starts over spoke 1. This means spokes 1, 3, 5, 7 are always crossed from the top and spokes 2, 4, 6, 8 are always crossed from below. The weaving thread covers the same side of each spoke in every circuit. No spiral occurs. The result is a woven surface that shows alternating thick and thin radial lines from center to edge — the even spokes produce a striped pattern. This can be a deliberate design choice for a crosswheel variant where groups of spokes are treated as paired, but for the standard slick woven wheel fill, even spokes are the mechanical cause of the most common beginner failure: the fill that appears to work correctly but produces stripes instead of the expected circular spiral.

The practical protocol: before laying any spokes, write the intended spoke count on a slip of paper and confirm it is odd. For standard medium Dorset buttons on 12–20mm rings: 9 spokes for rings up to 15mm; 11 spokes for 15–20mm rings; 13 for 20–25mm rings. These counts produce spoke spacings that work with perlé #5 and #8 thread weights without producing spokes so close together that the weaving needle cannot easily pass between them or so far apart that the weaving circuits are loose and wide.

Spoke tension equalization: correcting slack before anchoring the final spoke end

Spoke tension is the most consistently under-documented variable in Dorset button instruction. Each spoke is a structural warp thread — the weaving circuits pass over and under it, and the circuit thread's path is determined by the height of each spoke above the ring face. All spokes at equal tension sit at equal height above the ring face, producing equal circuit-thread deflection at every crossing, which is what makes the woven fill look uniform. A spoke that sits lower than its neighbors — having been laid under slightly less tension during the laying pass — deflects the circuit thread less at that spoke position. The circuit thread, following the path of least resistance, pulls inward toward the lower spoke rather than arcing uniformly from crossing to crossing. The visual result: a slight dip or depression in the woven fill surface at the position of the slack spoke, as if the spiral pattern has been nudged inward at that point. The dip is small for the first few circuits; as more circuits are added, each circuit deposits more thread on the same inward path and the dip becomes more pronounced. After 8–10 circuits, the dip is visible from normal viewing distance without magnification.

Spoke tension equalization protocol: after all target spokes have been laid in their positions across the ring and before the final spoke end is permanently anchored to the cast stitch ring, systematically work around the ring and equalize spoke tension. Method: hold the ring in one hand and with the other hand take each spoke individually between thumb and forefinger approximately 3mm from the ring edge. Pull very gently — approximately one to two millimeters of travel — in a direction away from the ring center, perpendicular to the ring face. The goal is not to tighten the spoke substantially but to take up any accumulated slack from the laying pass. Work around all spokes twice in the same direction. The second pass is necessary because correcting one spoke's tension slightly redistributes tension to its neighbors, and the second pass equalizes this redistribution. After two passes, the button face should feel equally resistant to gentle pressure at any position across the spoke area. Test: press the center of the button face with one fingertip using consistent light pressure, then move the fingertip to a position midway between two spokes at the ring edge and apply the same pressure. The deflection under fingertip pressure should feel equal at both positions. If the center deflects significantly more than the edge positions, the spokes have insufficient tension in aggregate and additional equalization passes are needed. If one edge position deflects significantly more than adjacent edge positions, the spoke at that position requires individual additional equalization.

After equalization, anchor the final spoke end by passing the thread tail twice through consecutive cast stitches at the ring edge (not through the ring center — the spoke end anchors to the cast stitch layer, not to the ring material) and cutting flush with the ring edge. The two-stitch anchoring minimum prevents the spoke end from releasing under weaving tension. One-stitch anchoring is insufficient because the working thread pull during weaving directly loads the final spoke's anchor point, and a single stitch can be pulled through the cast stitch layer by cumulative weaving tension.

The center anchor wrap: preventing spoke convergence migration during the first weaving circuit

The center collapse failure is one of the two most common sources of asymmetric woven Dorset buttons. It occurs specifically during the first weaving circuit, when the weaving thread applies lateral tension to the spoke convergence point at the ring center. All spokes meet at a single crossing at the geometric center of the ring — in a correctly set up button, this crossing is exactly at center. The crossing is a loose convergence of multiple threads held in position only by the outward tension of each individual spoke pulling toward the ring. The crossing threads are not bound to each other at the center; they are free to slide relative to each other under lateral force.

When the first weaving circuit begins and the needle passes over the first spoke and under the second, the working thread arcs between the two spokes and pulls slightly toward the weaving direction. Over the first three to four spoke crossings of the first circuit, these pulls accumulate into a lateral displacement of the crossing point at center — the convergence migrates perhaps one to two millimeters toward the side of the ring where the first circuit's stitches were worked. Because all spokes are attached to this convergence point, they migrate with it. The result: the button center is now slightly off-center, and all subsequent circuits are woven around an off-center hub. In the finished button, the spiral pattern has a visible center that is not at the geometric center of the ring — the woven fill appears to have more coverage on one side of the button than the other.

The center anchor wrap: before beginning the first over-under weaving circuit, position the weaving thread at the center crossing and wrap clockwise once around the entire bundle of spoke threads at the crossing point, pulling snug. Wrap counterclockwise once, pulling snug again. Two wraps, opposite directions. These two wraps bind the spoke threads together at the crossing point, creating a small hub structure that resists lateral displacement under the first circuit's weaving tension. The hub is small — approximately the diameter of one thread — and is completely covered by the second or third circuit of weaving, making it invisible in the finished button. After the two wraps, the weaving thread exits the center crossing in the direction of the first spoke and the first over-under circuit begins. The center remains in position through the first circuit and all subsequent circuits because each circuit's circumferential threading reinforces the hub structure from the inside outward as more circuits are added.

The center anchor wrap is also the correct starting position for a woven fill begun with a fresh thread rather than continued from the cast stitch phase. Anchor the fresh thread to one spoke near the center by making two wraps around that spoke — one wrap clockwise, one counterclockwise — before beginning the center anchor wraps. The spoke-specific anchor prevents the fresh thread end from pulling free as the center wraps are tensioned. After the spoke anchor and the center anchor wraps, the thread is positioned to begin the first circuit without any loose end in the button face.

Why the blunt-tip tapestry needle is a mechanical requirement, not a preference

The instructions for woven Dorset buttons specify a blunt tapestry needle, and this is typically presented as a preference that produces better results or is easier to use. It is not a preference. It is a mechanical requirement because a sharp or semi-sharp needle creates a failure mode that is invisible at the moment it occurs and irreversible after a small number of subsequent circuits have been worked over it.

A spoke is a single thread (or at most a doubled thread) spanning the ring center. Its cross-section, even for a plied thread like perlé, is small relative to the spaces between spokes and small relative to the weaving circuit paths. A sharp needle tip approaching a spoke at a slight angle — which is routine in normal weaving motion, since the hand cannot perfectly control needle angle on every stitch — may penetrate the spoke thread rather than glancing off it and passing over or under it cleanly. When the needle penetrates the spoke thread, one of two outcomes occurs. Outcome one: the needle passes fully through the spoke thread, splitting it into two sub-threads. The stitch appears to have passed over or under the spoke correctly, but the spoke itself is now a split thread held only by the friction of the two split halves against each other. Outcome two: the needle partially penetrates the spoke thread and the weaving motion carries the split partially along the spoke before the needle is pulled free. The spoke appears slightly frayed where the partial penetration occurred but is still intact.

Both split outcomes share the same failure timeline: the split or frayed point is invisible or barely visible at the moment it occurs, because the spoke is still held together by the remaining intact thread structure. As subsequent weaving circuits are worked over the affected spoke, the tension of each circuit pulls laterally on the spoke at the crossing points. The split point, which is weaker than the surrounding intact thread, progressively separates under this repeated lateral tension. After three to five additional circuits have been worked over the split point, the spoke thread separates visibly at the split location: one half of the split spoke lies against the button face on one side of the split, and the other half lies on the other side. The spoke is no longer a single thread running from ring edge to ring edge; it is two unanchored thread ends that have been locked under several circuits of weaving and cannot be extracted without unwinding all circuits above the split point.

The blunt tapestry needle tip cannot penetrate a spoke thread because the rounded tip deforms against the thread surface and redirects itself to pass over or under the spoke rather than through it. This physical property — the tip deforms rather than penetrates on contact — is the mechanical requirement. Any needle with a pointed tip, including craft needles designated as "tapestry" with a mildly blunt point, creates a reduced but non-zero split spoke risk. Verify the needle tip by touch before beginning the woven fill: the correct tip for Dorset button weaving resists being pushed into the thumb fingertip under normal writing pressure. A tip that creates even slight skin indentation under this test is too sharp for weaving over fine spoke threads.

Button stitch center mechanics: the through-center-hole sequence on every stroke

The button stitch center is the alternative to the woven wheel fill for Dorset buttons — it produces a dense, flat, matte center disc of thread rather than the woven spiral that the slick and crosswheel patterns use. The button stitch center can be used alone (a completely button-stitched button), or as the center section of a composite design where spokes extend outward from the button stitch center to the ring edge and are filled with a partial woven wheel or left as open spokes for a cartwheel appearance.

The mechanics of the button stitch: the working thread is anchored at the central crossing of the spokes by two wraps around the crossing bundle, as in the center anchor wrap technique. From this anchor position, the needle enters the center hole from the front (the button face side) going through to the back. On the back side, the working thread is wrapped around a single spoke — passed behind the spoke from one side and brought back to the front through the center hole. The working thread exits through the center hole to the front again, has now wrapped the spoke once, and the stitch is tightened by drawing the working thread gently forward. The stitch sits as a wrapped segment on the spoke between the center hole and the ring edge.

The requirement that the needle enters and exits through the center hole on every stroke is the structural rule that makes the button stitch work. If the needle enters through the center hole but exits between two spokes rather than through the center hole, the working thread wraps partially around a spoke (at the between-spokes exit point) but the return path does not pass through the center — instead it crosses the spoke at a point away from center. These off-center crossings produce wraps that sit unevenly on the spoke at various distances from the center, creating a lumpy rather than flat center disc. Under subsequent strokes, these uneven wraps are pushed inward by new stitches that are correctly placed through the center hole, compressing into ridges that give the button face an irregular texture rather than the smooth flat disc that correctly worked button stitch produces.

The practical discipline for maintaining the center-hole entry and exit: before each stitch, position the needle so that both entry and exit will be through the center hole by feeling for the center hole from both sides before beginning the stroke. On a well-cast ring, the center hole is large enough to accommodate the needle without contact with the cast stitches. If the center hole has become partially obscured by wraps from previous button stitches, the hole can be reopened by passing the needle tip through it from the back before beginning the stitch — this clears any thread that has migrated into the hole during previous stitches. The center hole must remain open throughout the button stitch phase; if it becomes fully closed, subsequent stitches cannot pass through and the button stitch center cannot be completed to the ring edge.

Spoke coverage in the button stitch center: each spoke receives one wrap per stitch row (each circumferential pass of the working thread around all spokes). The first wrap positions near the center hole; each subsequent wrap advances approximately one thread-diameter along the spoke toward the ring edge. The number of stitches required to fill one spoke from center to ring edge equals approximately the ring radius divided by the thread diameter. For a 12mm ring (6mm radius) with perlé #8 (0.90mm thread diameter): 6 ÷ 0.90 = approximately 7 rows. In practice, 6–8 rows fill the spoke correctly for this combination; fewer rows leave visible gaps at the ring edge where the spoke is not covered; more rows cause overcrowding where later rows of stitches have no space to sit flat and begin to ride over each other.

Thread ending: different methods for the cast stitch layer and the woven fill layer

Thread ending in Dorset buttons is performed in one of two ways depending on which structural layer the thread belongs to. The cast stitch layer and the woven fill layer have different structural properties that require different ending methods. Using the cast-layer method in the fill layer produces a visible surface defect; using the fill-layer method in the cast layer produces an inadequately secured end that pulls free under subsequent tension.

Thread ending in the cast stitch layer: when the working thread has insufficient remaining length to complete additional cast stitches — the practical minimum is approximately 12cm of remaining thread, enough to end the current thread and begin anchoring a new one — end the thread by working backward along the ring. Thread the remaining tail through the tapestry needle and work it underneath 5–6 consecutive cast stitches in the direction opposite to the original stitching direction, as if running a thread under existing stitches. The backward-tucked tail is anchored by friction against the ring and against the existing cast stitches; it cannot be pulled free without first pulling each of the 5–6 stitches off the ring. Cut the remaining tail flush with the outermost cast stitch it passes under. Starting a new thread: tail the new thread in the same way under 5–6 consecutive cast stitches at the point where the old thread ended, running in the forward stitching direction, then exit the tail at the point where the new stitching will begin. Work the first 4–5 new cast stitches over the tail — the new stitches cover the tail where it exits the previous cast stitches, completing the anchor. This start method eliminates the need for knots, which would create bulges under the cast stitch layer that distort the ring coverage.

Thread ending in the woven fill layer: the fill layer has a fundamentally different structure from the cast stitch layer. The woven circuits are not fixed to the ring; they are held in position only by the spoke-and-circuit interlocking structure. A backward tuck under existing cast stitches — the cast-layer method — would require the needle to pass through or under the cast stitches and the ring edge at a location that the weaving thread may not reach. More importantly, a tail that exits the fill layer and runs under cast stitches crosses the ring edge at the surface, producing a raised crossing where the tail bridges from the fill area to the cast layer. This crossing is visible on the button face as a slight ridge or lump at the position where the tail crosses.

The correct ending method for threads in the woven fill layer: when the working thread has approximately 10cm remaining — enough to end and begin a new fill thread — work backward through the fill layer itself rather than backward through the cast layer. Thread the tail through the needle and pass it backward (counter to the weaving direction traveled) under 4–5 consecutive circuit rows at the spoke position where the thread runs out. The backward path — going under circuits that the thread went over on the forward journey, and over circuits it went under — creates a locked anchor within the woven structure itself without crossing any ring edge. The tail is then cut flush at the 4th or 5th circuit row it passes under. Starting a new fill thread: anchor the new thread start by passing its tail forward under 4–5 circuit rows from the point where the new weaving will continue, in the weaving direction, beginning one spoke before the continuation point so that the first new stitch at the continuation point also covers part of the tail. The backward-ending and forward-starting methods together produce a fill continuation that is invisible in the finished button face at normal viewing distance, provided the backward and forward passes both stay within the fill layer without crossing the ring edge.

Apple Tax on iOS Patreon subscriptions: what Dorset button creators lose from November 2026

Apple's 30% iOS IAP fee on Patreon subscriptions takes effect November 1, 2026. Dorset button creator audiences are heavily iOS-weighted across the platforms where this niche performs well. The combination of UK historical craft tradition and contemporary textile art revival has concentrated Dorset button content on photography-first platforms where iOS consumption dominates.

Instagram Dorset button and textile craft content — organized flat lays of finished button collections organized by color family, close-up photography of woven wheel surface texture, in-progress photographs showing ring preparation before spoke laying, finished button sets staged on fabric backgrounds — reaches 70 to 82 percent iOS in UK and US markets. The visual appeal of finished buttons as photography objects, the relatively small scale that photographs well on phone screens, and the craft's UK historical association with Pinterest-adjacent audiences all contribute to high iOS share. Pinterest boards for Dorset buttons — historical pattern archives, contemporary color scheme explorations, ring size reference guides, tutorial link collections — reach 66 to 80 percent iOS. YouTube Dorset button tutorial content — slow-craft demonstration videos running 15–30 minutes covering complete button construction from ring wrapping to pattern filling to finishing — reaches 52 to 68 percent iOS, with the lower share reflecting the longer format drawing more desktop viewers than the short-form photography platforms.

Revenue impact from November 2026: at $150/month with 72% iOS: $150 × 0.72 × 0.30 = $32.40/month ($388.80/year). At $300/month with 76% iOS: $300 × 0.76 × 0.30 = $68.40/month ($820.80/year). At $500/month with 78% iOS (a multi-platform creator with both Instagram pattern photography and YouTube tutorial content drawing patrons from both platforms): $500 × 0.78 × 0.30 = $117/month ($1,404/year). A Dorset button creator at $300/month with an Instagram-primary audience loses over $800 annually beginning November 2026 — enough to fully fund all ring materials, thread stock in multiple colorways, ring sizing sets in brass and plastic, and reference books on historical Dorset button patterns for an entire production year.

KeepTier: web checkout that bypasses Apple IAP for Dorset button and textile craft creators

KeepTier provides a branded web checkout page for creator subscriptions that operates outside Apple's In-App Purchase system. Patron subscriptions processed through a KeepTier page are web transactions — Stripe Checkout in the web browser, not the iOS Patreon app — and Apple's 30% fee does not apply. For a Dorset button creator at $300/month with 76% iOS, directing new patron subscriptions to a KeepTier page instead of the Patreon iOS app preserves $68.40/month — $820.80 annually — from the fee that would otherwise apply permanently beginning November 1, 2026. See keeptier.com to calculate your specific Apple Tax exposure and set up a web checkout page before the November deadline.

Tier structures for Dorset button Patreon creators

Dorset button creator Patreons face the same saturation problem as all short-path craft content: the technique is learnable quickly, and free tutorial content covers the appearance of the finished technique thoroughly. The Patreon content tier that retains subscribers past six months is not more technique demonstration but specification content that addresses the failure modes demonstrated above and that allows subscribers to extend the technique to ring sizes, thread weights, and pattern variants that were not demonstrated in the free content.

A Pattern and Specification tier at $10–16/month covers original pattern releases with complete working specifications: the ring diameter and material (brass versus plastic, with the note that parchment rings distort slightly under spoke tension and are not recommended for slick patterns requiring precise spoke spacing), the thread type and weight used with the target cast stitch count and the cast stitch compression factor achieved at that count, the spoke count (confirmed odd) and the equalization protocol used, the center anchor wrap confirmation, the needle type verified to prevent split spoke risk, and the filling sequence with the thread ending method documented for any thread joins in the fill. A subscriber who receives a pattern with these specifications can reproduce the button on a different ring size by applying the target stitch count formula, adjusting spoke count to the nearest odd number for the new diameter, and carrying all other specification values forward. Without the specifications, reproducing the button on a different ring size requires re-deriving all the mechanical variables from scratch on each attempt.

A Technical Depth tier at $22–35/month covers the systematic mechanics in module format: a cast stitch mechanics module documenting the center-hole versus between-stitches failure with comparison photographs showing the migration failure under spoke tension; a spoke count module showing finished button photographs at even and odd spoke counts for the same ring diameter and thread weight, with the mathematical explanation of the phase shift that produces spiral versus striped fills; a spoke tension module with a protocol for equalization and a visual test for detecting a slack spoke before weaving begins, documented as a step-by-step procedure rather than experience-based feel; a center anchor module with photographs of the convergence migration failure (before and after center anchor wraps) and the specific wrap sequence; a needle selection module with the fingertip test for acceptable tip bluntness and photographs of split spoke failures at various circuit counts above the split point; a button stitch center mechanics module documenting the through-center-hole requirement with photographs of correctly and incorrectly executed strokes at the same position; and a thread ending module with comparison photographs of correctly ended and incorrectly ended fill threads showing the raised-tail failure that results from using the cast-layer method in the fill. Each module includes a small worked sample — a single button using only the module's target technique — that the subscriber photographs and submits alongside the reference photographs provided in the module. The comparison exercise trains the subscriber to see the differences between the correct and failure-mode appearances that are documented in the module, producing a visual calibration for their own work in subsequent projects.

The signal for a third tier at $50–65/month for 4–6 patrons: when the Technical Depth tier shows 85% or better retention across five consecutive months, add a Critique and Diagnosis tier. Each patron at this tier submits one photograph per month of a button with a specific technical question — "why does my slick pattern appear slightly striped rather than spiral even though I counted 9 spokes?" or "why does my button stitch center have visible irregularity at one position that I cannot reproduce intentionally?" — and receives a written diagnosis with the specific mechanical cause of the observed failure and the correction sequence. At 4–6 patrons, each diagnosis requires 15–25 minutes of focused analysis per patron per month; above 6 patrons, response quality degrades. The waitlist signal — when the current tier has a waitlist — is the indicator to raise the tier price by $10–15 rather than opening additional patron slots, which maintains the response quality per patron that justifies the tier's value. For additional cross-references in your Patreon posts and pattern documentation, link to the KeepTier explainers for adjacent techniques — the tatting mechanics post on the double stitch flip and the smocking mechanics post on gather-ratio calibration — and to the Apple Tax Calculator for a personalized estimate of your specific Patreon tier structure's November 2026 exposure.