When specifying a time, some positions may have either one or two digits (e.g. “1 PM” vs “11 PM”), but other positions always have a leading zero, no matter what (e.g. “1:01” and “1:11” both have three digits).
The inconsistency in digits is unnecessary and leads to weird sorting behavior. For example, if we sort the times from 11 AM to 2 PM alphabetically, we end up with this: “1 PM, 11 AM, 12 PM, 2 PM.” Outrageous!
Let’s fix it by using new symbols to give a unique single-character glyph to 10 o’clock, 11 o’clock, an 12 o’clock. If we use existing keyboard characters like “A,” “U,” and “=,” we retain both the ability to easily type these characters, and they will sort in ascending “alphabetical” order (the ASCII default sort order is numbers → letters → symbols), and they can be represented in a traditional 7-segment LED display (Figure 1).
If we want to get really comprehensive in fixing date representations, we might want to also replace every day of the month with its own symbol. Figure 2 proposes using the long-defunct Linear B script, which has no living defenders and is thus vulnerable to our repurposing.
So in the end, we’ll have a comprehensive time-and-date rework where months, days, and hours can always be represented by a single digit (Figure 3). This will replace the current unpredictable mix of 2-digit and 1-digit values.
This should save ink and make life easier for programmers, who can now always rely on times fitting into the format YYYY–M–D H:MM:SS (at least for years between 9999 B.C. and 9999 A.D.).
PROS: Should make alarm clocks cheaper to manufacture, since they’ll only need three digits for HOUR:MINUTES rather than four.
CONS: The Linear B symbols aren’t supported by most fonts. Sometimes, they’ll be replaced by a “missing character” symbol, so we might end up with things like “The graduation is on /9!” or “Please report for your court appearance on /!”