Product/linkage joint

A ROD OR HOUSING CARRYING ONE OR TWO BALL STUDS IN SOCKETS, JOINING TWO MOVING MEMBERS OF A STEERING OR SUSPENSION LINKAGE SO THEY MAY SWING AND TURN WITHOUT PLAY - a ball joint, a stabiliser link, a tie rod end, a drag link end, a suspension rod. THE QUEUE ASKED WHETHER A STABILISER LINK AND A BALL JOINT EARN A KIND AT ALL, CALLED IT THE HARDEST QUESTION IN THE CAMPAIGN, AND WARNED THAT THEY MIGHT BE SEPARATED BY FITMENT ALONE - which is out of scope - AND SO FOLD AWAY TO NOTHING like b113's plain jug. THE ANSWER IS: THEY ARE ONE CLASS TOGETHER, AND THE TRADE NAMES ARE ITS VALUES. Neither has an exclusive slot over the other, and that is not an inference - it is on febi's and MOOG's own criterion sheets. THE ARTICLES THAT SETTLE IT: **febi 27009, a STABILISER LINK, publishes `Length: 438 mm` AND `Cone Size: 28.6 mm` ON ONE ARTICLE** - the link's own field and the ball joint's, side by side, under `Rod/Strut: Coupling Rod` - and **MOOG BM-BJ-8826, a BALL JOINT, publishes `Length [mm]: 70`** with `Assembly type: Bolted` and `Thread Size: M14x1.5`. Run the differentiating function either way and it returns the wrong answer on a named current part [A13]: the cone size does not tell a joint from a link and neither does the length, because one real link carries both. Every other criterion is shared outright: `External Thread Size` / `Thread Size`, `Fitting Position`, `Material`, `Supplementary Article`, `paired article number` and `Greaseable` are published on ball joints, on links and on tie rod ends alike, out of ONE TecDoc criterion vocabulary, by ONE house, in ONE catalogue. That is the 2026-08-25 amendment biting for the SIXTH proposal running, and it is the answer the queue itself predicted. AND THE MAKERS PUBLISH THE FOLD AS A FIELD, WHICH IS THE b157 ARGUMENT ARRIVING ON A SIXTH TRADE. febi and Lemforder print a criterion called `Rod/Strut` whose values are `Coupling Rod` (the stabiliser link) and `Suspension Rod` - the house distinguishing two members of this family INSIDE ONE FIELD rather than in two catalogues. A field with values is a property. BUT THE CLASS ITSELF IS EARNED, AND THAT IS THE OTHER HALF OF THE HONEST ANSWER [L3]. Folding the two into each other does not fold them into nothing: there is no live class a ball joint could be a value of, and six questions are asked of every member of this family that `Product` cannot answer - `joint_form`, `stud_cone_diameter`, `thread_size` with its role and hand, `lubrication_provision`, `joint_attachment` and `axle_position`. b113's jug folded INTO `Drinkware`, an existing class with the jug's every fact already on it; here there is no such parent, and leaving these unstatable would be a hole in the language [L0]. So: one class, minted, with the trade's five words as values of one required property. THE COMPLETENESS CHECK AT THE MOMENT OF THE FOLD [H2-f16] IS WHAT PUT THE QUALIFIERS IN. A tie rod end publishes an INTERNAL thread beside its external one - Lemforder 33927 01, `External Thread: M12x1,5` AND `Internal Thread: M16x1,5` - and a stabiliser link publishes two external ones - Meyle 30-16 060 0000/HD, `Thread Size 1` and `Thread Size 2`. Without a role qualifier the fold would have made those designations unattributable, so `thread_size` is cardinality many here and carries `thread_role`. And a tie rod is threaded one hand at one end so the sleeve lengthens it, which Lemforder prints as `Thread Type: with right-hand thread` and MOOG as `End 1 Thread Direction: Counterclockwise` - so `thread_direction` was minted too. `stud_cone_diameter` is cardinality many because febi's 32412 publishes `Cone Size 1: 16 mm` and `Cone Size 2: 23 mm` on one article. `axle_position` IS RE-DECLARED HERE, WHICH WITH `Bearing` DISCHARGES b156's BINDING H4-f1 - and the discharge went further than the obligation asked, because the evidence showed the property was NAMED TOO NARROWLY. The trade's field is `Fitting Position`, it is stated MANY TIMES rather than compounded, and its values are not only axles: Lemforder's 31007 01 publishes three separate rows, `Front Axle Right`, `Front Axle Left` and `Lower`. So `AxlePositions` grows by `left`, `right`, `upper` and `lower` and the property's label becomes `fitting position` - see its own definition for the full ruling and for why no lift to a common parent was possible. WHAT IS DELIBERATELY NOT HERE. The dust boot's material, which only Dorman publishes (`Dust Boot Material: Policloropreno` on the BJ90276RD) and which cannot go in the spine's unordered `material [many]` - the b156 backing-plate trap exactly. `Dust Boot Included: No` and `with attachment material` / `with nuts` / `with locknut`, which are box contents and are refused on b156's ruling. The END INDEX - which of a two-ended link's two threads is end one - which no house publishes consistently and which the role qualifier cannot yet carry. And a wear indicator, a cone ANGLE and a load direction, none of which any house read at the mint publishes at all; inventing them would have been a fabrication [A8]. WHAT THE SOURCE MAY AND MAY NOT WITNESS. Habib's two rows here are `Stabilizer Link 48819 52010` and `Ball Joint CBT 56`, whose notes read only *Stabilizer link ya IST* and *Ball joint ya IST* - pure fitment, out of scope, and no specification whatever. They are the DEMAND SIGNAL only [L12]. Every differentia stands on a published article record read at the mint: Lemforder 31007 01, 33927 01, 18003 02, 35004 01 and 2717102; febi 32412, 14426, 183622, 07977, 28090, 21356, 42574, 18134, 14558, 33492 and 27009; MOOG K80026, ES800880, K750147, FD-BJ-7296 and BM-BJ-8826; TRW JBJ779 and JTS7519; Meyle 30-16 060 0000/HD; Delphi TC1907 and TC1939; Monroe L13629; Dorman BJ90276RD.