A fault-tolerant toilet seat protocol

Abstract

In this article we propose a fault-tolerant protocol for toilet seat positioning in a hostile environment.

Assumptions

Jay Pil Choi (2002)[1] in his excellent article on toilet seat etiquette makes the assumption that toilet users wish to optimize their toilet behaviour for inconvenience cost. Richard Harter (2005)[2] also tries to find a cost optimum and proposes three cost minimization candidate criteria:

Minimize the joint total cost

Equalize the respective total costs

Equalize the respective incremental costs

Although we shall not refute their findings given these assumptions, we propose an altogether different metric to optimize for:

Given the fact that humans on average, regardless of gender, have very little affection for toilet seats that are covered in feces or urine ^{[citation needed]} we instead propose to optimize for toilet seat cleanliness. This introduces a novel optimization challenge and a new cost/benefit distribution across genders.

Secondly, both Choi and Harter assume that the parties in a toilet seat sharing scheme are cooperative. Thus, they try to optimize for a situation where both parties are receptive to a common solution. For our protocol however will argue that the use of toilets should, in essence, be considered a state machine in a hostile environment. This moves the discussion into a different solution space, where fault tolerance and protection against adverse agents become relevant.

Optimizing for cleanliness

We make the assumption that a toilet seat in the up position is less likely to be smeared than a toilet seat in the down position. Also, we refrain from including the impact of air-borne bacteria due to leaving the toilet seat lid open during flushing, for reasons or protocol simplicity. Research into both of these areas is recommended.

Optimizing for non-collaborative use

We propose the assumption that a generic toilet seat protocol should be equally applicable in cooperative and hostile environments. We define hostile environments as toilets in spaces where toilet users feel little or no shared responsibility for the cleanliness of the toilet or the relationship with those with whom the toilet is shared. In other words, the conflict cost M as defined by Harter nears zero. We define cooperative environments as toilets in spaces where toilet users do feel responsibility for keeping both the toilet clean and the relationship intact. In other words, M nears one.

Inter-agent comparison

Traditionally the literature considers the difference between male and female toilet users. This makes sense since from a cost perspective, especially a Harter incremental cost perspective, there is a significant difference between male and female users.

This makes for a very bad starting point in our hostile, non-collaborative case. We have two agents with diverging desires and differing costs, depending on the chosen strategy. It makes sense that this leads to unresolvable conflict. We take this as our starting point for this protocol, accepting that this inherent difference in cost and desired outcome effectively makes each environment hostile to some degree.

Optimizing for cleanliness changes this dynamic since both parties have a desire for a clean toilet, but generally speaking with different tolerances for dirty toilets, as we shall establish.

Benefit analysis

The study of C. Rose et al. (2015)[7] shows an average defecation frequency of 1.1 movements per 24 hours (with men having a slightly higher frequency than women) and an average urination frequency of six movements per 24 hours. This, on average, makes for an approximate seven toilet visits per person per 24 hours.

Of these seven visits, we expect female users to have an equally high preference for a clean toilet on each visit. Of male users we would however expect that they would display a high preference for a clean toilet one in seven times, with a somewhat lower preference the other six visits.

It intuitively makes sense that each agent will have a stake in a clean toilet relative to their respective requirements of said toilet. So we can expect to see a higher willingness to advocate for and care for a clean toilet in female users than in male users.

In both cases the benefit can be used as an incentive to apply a protocol that is inherently self-serving (albeit not selfish, as per Choi's analysis).

Protocol proposition

Given a state machine with the following states:

[up, down, closed]

for the toilet seat up, toilet seat down and toilet lid closed states, respectively; and the following transitions:

[
  up -> down        (1)
  up -> closed      (2)
  down -> up        (3)
  down -> closed    (4)
  closed -> up      (5)
  closed -> down    (6)
]

we would propose the following protocol:

Upon entering the bathroom we observe any of the three possible states
Given the state, and the user's requirements, the user will either:
1. Observe the up state and apply no transition; or
2. Observe the up state and apply transition (1); or
3. Observe the down state and apply no transition; or
4. Observe the down state and apply transition (3); or
5. Observe the closed state and apply either transition (5) or (6);
The user does their business
After finishing step 3 the user will either:
1. Observe the up state and apply no transition; or
2. Observe the down state and apply transition (3);
  such that in any case the end state of the toilet is the up state.

This protocol, if properly implemented, maximizes the up state of the toilet, making visits to the toilet by lazy users who apply 2.i. less likely to smear the toilet seat, regardless of whether they adopt this protocol or not. This makes the protocol defensive by nature, and thus (as we consider a dirty toilet to be a fault, in our model) the most fault-tolerant solution.

One could argue that as the adoption rate of this protocol reaches the maximum, female users will mostly apply 2.ii. followed by 4.ii., whereas male users will be able to apply 2.i. followed by 4.i. in six out of an average of seven toilet visits, and as such infer less cost in the classical sense. Consider, however, that this distribution of cost perfectly matches the distribution of benefits as outlined in our benefit analysis, making each agent inherently self-serving and yet contributing to the end goal of cleanliness.

Social impact of the proposed toilet seat protocol

Eventhough it might be an unpopular stand point to have to put the toilet seat back into the upright posistion especially from a female perspective as shown by Choi, we argue that this protocol benefits everyone. As shown by UK Bathrooms 4 the arguements around the toilet seat puts a strain on relationships. Hilt et al.5 show that there are a lot of bacteria in urine which pose serious health hazards so reducing the chanche to catch these bacteria will benefit society. The largest problem with the proposed strategy is that it requires a change in behaviour. Changing behaviour is hard and requires time and effort. The most benefitial part of this protocol is that the toilet seat is always in the same state, upwards. Knowing this state and the required action will reduce the current frustration that arises from the suprise effect that women experience from sitting down on the toilet without a seat.

Future research

Gerba et al. 6 show in their research that there are merrits to closing the lid when flushing. Since this step does not influence the end state of the lid it is not part of the protocol eventhough it might be worth to address this topic when thinking about a toilet in a hostile, but even in a collaborative enviroment.

MinThaMie/fault-tolarant-toilet-protocol