This is mentioned in the article but treated as a joke:
> Other internet researchers, perhaps less seriously, have used 3D-printed models of the Roman dodecahedrons for knitting experiments, and suggested that the true purpose of the objects was to create differently sized fingers for Roman woolen gloves.
Although IIRC when it was first proposed the people who proposed it seemed very earnest and to see it as obvious.
And it nicely explains a lot of the peculiarities e.g. the nubs make sense for attaching yarn, and different people have different finger sizes (up to layers) but the precise sizing is not super important as long as you have a gradation with steps which aren't too big.
It also looks quite close to spool (or "french") knitting.
Though there's the objection that the dodecahedra are serious overkill for those needs when a plank with a few nails would do the trick. There's also the "OOPArt" factor: knitting spools were invented in the 17th century or so, and known records for circular knitting are not that much older.
> Other internet researchers, perhaps less seriously, have used 3D-printed models of the Roman dodecahedrons for knitting experiments, and suggested that the true purpose of the objects was to create differently sized fingers for Roman woolen gloves.
Although IIRC when it was first proposed the people who proposed it seemed very earnest and to see it as obvious.
And it nicely explains a lot of the peculiarities e.g. the nubs make sense for attaching yarn, and different people have different finger sizes (up to layers) but the precise sizing is not super important as long as you have a gradation with steps which aren't too big.
It also looks quite close to spool (or "french") knitting.
Though there's the objection that the dodecahedra are serious overkill for those needs when a plank with a few nails would do the trick. There's also the "OOPArt" factor: knitting spools were invented in the 17th century or so, and known records for circular knitting are not that much older.