Anyone that knows me will know that I have no shortage of opinions. Below is a selection of a few of my (only-so-slightly) controversial views, to be taken with a pinch of salt.
App subscriptions are great
It appears that the larger tech community has an aversion to app subscriptions, with a vocal minority treating the introduction of such financing methods with the moral fervour that would normally be seen if someone were to kick a baby. The usual argument against app subscriptions sounds something like, “I paid for this once, why should I keep paying £X a month for an app that rarely adds any new features”. There are two flaws with this argument.
The first is the false equivalence between the lack of additional features with lack of any development work. Maintaining an app is relentless work. Even if you somehow manage to get the app to a perfectly working state, that will not last for long. Operating systems add new features, requirements and frameworks that will demand compliance from the app, security and privacy expectations change and vulnerabilities appears, APIs and other dependencies go dead. Even without adding any new features, dealing with the endless stream of issues of this type can take up a lot of time. I’ve been bit by these issues myself, having to migrate from my once favourite but now dormant email app since I required two-factor authentication to access my university email account. If your app has saturated its demographic market, there will be no new app purchase funding this continuing development, leaving the app to fall into a state of disrepair. An app subscription is the most reliable way to ensure that you can still use your favourite services without relying on a virtual machine running a decade old version of Android.
The second is a poor contextualisation of the cost of the subscription. When apps, and in particular utilities, move to a subscription model you’ll often see negative reviews along the lines of “I’m not going to pay £25 a year for just a calendar, that’s daylight robbery”. Sure, £25 may seem like a lot—it could get you a descent meal in a restaurant—but it’s necessary to look at this relative to the cost benefits of the service. Supposing you valued your time as little as the UK minimum wage, you would only need to have saved 3 or 4 hours a year in productivity gains by using the calendar app to have a net financial benefit. For the vast majority of people, that is a low bar to meet.
For this reason, if there is an app that could provide significant time/money savings to me that also has the potential for me to become dependent on, if it doesn’t have a subscription model, I will walk away. The only way to trust that something will stick around and keep meeting your needs is if there is regular funding in the form of a subscription to do just that.
Information boards are better being blank than displaying rubbish
It is commonplace to have large information boards at the entrances to London tube stations. These are meant to be used for displaying critical information such as delays and closures whilst requiring little effort to digest. This is important because at its busiest times, the London underground is fast-moving and chaotic. Easily accessible information is key.
When, however, there is no critical information to display these are sometimes used to show the sort of garbage inspirational quotes your grandmother would email you. The impact of this is two-fold. One, it slows people down and wastes their time as they try to process what the usually-important information board is trying to tell them. Two, it diminishes the expectation that the information board is only for critical updates and so less people pay attention to it in the future
If you are desperate to keep the inspirational quotes, I suggest they go on another board, clearly marked as non-important by writing the text in Comic Sans and adding a fuzzy pink border.
The first “real” prime number is 5
A prime number is defined to be an integer whose only factors are 1 and itself. 2 and 3 satisfy this condition in a vacuous way—since their square roots are less than 2, there is no possible factor other than themself and 1 that could be a divisor. 4 is the first number that has a candidate divisor (namely, 2). For this reason, I feel that the first “real” prime number is 5, with 2 and 3 being “fake” primes in the same way that 1 has sometimes historically been treated as a prime (e.g. by Goldbach).
This claim is not so radical in the sense that theorems can be found containing the phrase “let p be a prime greater than 3”. Clearly there is something fraudulent about the primality of 2 and 3 if exceptions have to be consistently made for them. Even if one cannot be convinced of the fake primality of 3, the case for 2 is stronger with many theorems containing the phrase “let p be an odd prime”, and with a few scholars in the Greek and later Roman tradition refusing to accept 2 as a prime by modifying the definition to only include odd numbers.
The office air-conditioning should be on the colder side
The argument is quite simple: you can keep adding layers to your outfit; if you keep removing layers, eventually you will get into trouble. If you follow this logic, it doesn’t make sense to set the air-conditioning temperature to be the median ideal temperature of all workers because this leaves 50% too hot with nothing to do about it and 50% too cold but with the ability to put on more layers and feel better. That’s up to 50% of workers unhappy with no recourse. Instead, if we set the temperature to the 10th or 20th percentile of desires (too avoid the cryophiles turing the rest of the team into walking balls of wool and down), although we will have people on average further from their desired temperature, the overall total discomfort can be lower.
The kilowatt-hour is a ridiculous unit of energy
What’s the circumference of the Earth? 6.92 miles-per-hour-seconds.
What’s the size of the gravitational force exerted on the moon by the Earth? 49 billion gigapascal-acres.
These units are nonsensical, completely obfuscating the magnitude, and even the dimension of the quantity they represent.
The kilowatt-hour is not much better. It is a unit of energy; we have a unit for energy—the joule. One kilowatt-hour is 3.6 megajoules. Instead of this easily interpretable unit, the kilowatt hour forces us to to conceptualise energy consumption by asking how much energy about 17 standard 60-watt lightbulbs would consume if left running for an hour. The megajoule has an almost identical scale and can easily computed from watts by multiplication by time (easier than dividing by time in the reverse), so there is really little reason as to why it shouldn’t be used.