I try for the simpliest approach that addresses the main problems of physics. The first principles are:
The ideal arm is such that on a flat LP, stylus deflection associated with normal tracking produces no vertical movement of the cartridge. (Horizontal movement is a necessity.) In this regard there is an optimal effective vertical arm mass (as well as a unique compression and rebound damping factor) associated with each unique cartridge compliance. This is because cartridge compliance is integral to the tonearm system that in aggregate controls tracking. From a practical viewpoint, continuously variable effective mass is the first order of business in matching an arm to a specific cartridge. Otherwise we are relegated to trial and error and the approximation of trying arms of generally "light", "medium", and "heavy" mass. This is expensive fun, but is unnecessary in view of the reasonable prospect of engineering an arm of continuously variable effective mass. Moveover, while reviewers like to say that a particular cartridge is fine with a particular arm, who really knows how good a particular cartridge can sound without an *exact* match to tonearm?
Assuming availability of an external mechanism to match tonearm effective mass to cartridge compliance, the ideal arm tube is infinitely rigid and weightless: a perfect conductor or absorber of residual cartridge vibration that does not reflect vibration back into the cartridge or itself add movement to the system. In this sense the optimal arm is no arm at all-- the shortest possible arm as available only in a linear tracker.
The rest of it can be dealt with in conventional terms. If the above problems are addressed, in the final analysis the only problem with a linear tracker is fluctuating geometry over warps. In a world of record clamps, vacuum hold-down, and periphery rings, this is an ancillary matter.