Simple ko under area scoring has problems with unbalanced cycles, and superko under territory scoring has problems since in territory passes must lift bans (and a superko variant that simply allows to recreate positions after a pass would revert to the previous problem under area).
Fixing these with additional rules is possible (with care). A prisoner-aware long cycle rule can be such - this is effectively the Chinese solution for the first problem.
Sure, for example, superko proponents tend to confuse textual conciseness with conceptual simplicity. 
But this is exactly why objective principles are useful, Like the principle that the area and territory game must stay coherent.
We know that unbalanced repetition is not enough for a draw, even though in area terms that would not seem worse than a triple ko (which does lead to draw under Chinese). But since this is a loss under territory, it must be prevented under area as well.
Likewise, we know that passes must lift bans under territory, otherwise it could lead to completely different outcomes than area. So a theoretically sound, universal ko rule cannot unconditionally prevent recreating previous positions.
I don’t see the problem. It would be rare, and when it does happen, like suicide, it’s just zugzwang, nothing gamebreaking, and a cool emergent property of the rules to boot
Could you further clarify? We already have situations where area scoring and Japanese rules yield completely different outcomes, such as bent four in the corner with unremovable ko threats. I’ve discussed this in another thread: Life and Death under Chinese Rules
Some differences are inavoidable, like string L/D cannot hope to model partial exchanges for a bent4 for example. But rules are still designed to keep area and territory as close as possible, and violations of this principle would be seen as flaws.
It would be very hard to argue for a common non-crowded endgame situation where kos are frozen open and cannot ever be recaptured just because changing the board cost points. This is simply not how Japanese rules work, partly because this is not how Chinese rules work either (nor even AGA/NZ with their strict superko). In OPs position he should have been able to force B fill by a few passes and resumptions, just like he could have forced the fill under area by playing inside territory.
I’m not trying to argue that area scoring should not use superko, or that territory scoring should use it. It is just a side point that one is not necessarily beholden to tradition or consistency with historical rulings. However, I understand that one could have subjective preferences on such issues.
My ultimate contentions are that territory rules are woefully complex compared to area scoring rules, and that there are no simplified principles that these complexities could be reduced to.
For example, consider how a simple ko rule would handle the double ko seki. With passes to lifts ko bans, should a player in a losing position be able to exploit a double ko seki with endless cycling to force a no result instead? Of course, this is a rhetorical remark, as I believe the Japanese rules handle this by essentially making a special ruling for this case, by specifically explaining in the commentary of the 1989 rules.
Another example to consider is the complexities around life and death determination (under Japanese rules) as arise from the ambiguity of the enablement clause. We touched upon these across several posts in another thread (see Odd Cases 🤔 in the Japanese Rules - #63 by elsantodel90)
My point with these examples are just to contend that the Japanese rules are not at all conceptually simple nor textually concise. I would say that area scoring rules, particularly NZ or AGA rules or the simplified Chinese rules (as implemented by OGS where positional superko is always applied), are both conceptually simpler and more concise.
From what I understand, even the two main territory scoring rule sets, Japanese and Korean, are not completely consistent with each other in all situations. I believe that this is seen in some particular forms of moonshine life, which I think you even posted about previously in this forum. Does this mean that either (or both) of these are flawed? Is one approach better and why?
Unless you’re only talking about popular territory rulesets (only JP and KR afaik), in which case of course those are way worse (if you value elegance) than the popular area rulesets (CHN, AGA, NZD, maybe a few others?), but just like highly elegant area rulesets (like Tromp-Taylor) exist, so too do highly elegant territory rulesets (like Lasker-Maas), so I don’t think it’s fair to dismiss territory scoring as “woefully complex”. Marginally more complex at worst
Yes, I’m talking specifically about JP and KR rules, which are the only that seem to be widely used.
Ah, okay. I’ve been talking too much with people who value elegant rulesets very highly, to the point that they don’t always consider Go elegant, so I’m used to insisting that any refusal to judge Go’s elegance based on Tromp-Taylor rules constitutes a strawman
Conciseness does not imply simplicity, the relation is one-way: Simple rules can always be written in reasonably concise form, but textually concise rules are not necessarily simple. A single English sentence can hide references to very complex logical concepts.
Territory scoring does not equal Japanese rules either. It can be simpler, and oc there are things I would do differently than J89. Double ko seki: its cycle includes periodic stops on two passes, and whatever are the rules, they unlikely allow infinite resumptions. Moonshine life (Korean or not): it changed a lot (was alive just a few decades ago) and may warrant rethinking once more.
In any case, Go evolved to its current (Asian, C/J) rules trying to keep the area and territory game coherently reflect the same ideal/theoretical game (that’s why Chinese allow triple ko draws but not unbalanced cycle draws for example). Both area and territory scoring has advantages, but huge problems as well (L/D for territory, ko and issues with B’s surplus stone for area). Only when considered together do the two systems reveal “the” real game and its cohesive logic (incl. the role of real passes).