The Timestamp
In 1900, the International Time Recording Company was established in New York. They had improved on a device that was developed about a decade earlier called the timeclock. It was a device made to record the start and end times for hourly employees. We have been watching the clock ever since.
ITRC’s device was called the Time Recorder, and when an employee slipped a preformatted piece of cardboard into the machine, it would punch a hole through it that indicated the exact moment the punch had taken place. (The punchcard as a device to record data was also a recent creation, first widely used in the US Census just a few years earlier.
The breakthrough with this device was to combine that punchcard and the mechanical clock into one seamless mechanism (that also was tamper resistant). The company that purchased the International Time Recording Company and also The Computing-Tabulating-Recording Company (the company responsible for the punchcard data collection during the 1890 US Census) was none other than The International Business Machines Corp (IBM).
Fast forward a few years and the practice of marking time onto documents expands to all kinds of applications. From post office post-dating, to marking time on the execution of legal documents, to recording the minutes of business meetings. Over the decades new gadgets were developed to do these tasks in a more convenient way, most notably with the Time Stamp. A literal rubber stamp with an integrated clock output into the stamping block.
In modern times we tend to balk at the idea of paperwork, but the need for keeping accurate timestamped records has only expanded further. Beyond simply keeping records, most computer systems rely on accurate time keeping to execute critical system functions. The term timestamping is still used to to refer to these actions today. For decades computer systems have synchronized to Unix Time, which started on January 1, 1970. Unix time slices every day into 86,400 seconds and at the time of writing this post 1604891649 seconds have occurred in the Unix Epoch (the amount of seconds since 00:00:00 UTC Jan 1, 1970). When we use a digital timestamp now, the current Unix time (the amount of seconds since the clock started) will be the returned value (reformatted for more conventional translation).
Unix time isn’t actually a time though. At least not in the way GMT and other timekeeping standards are “times”. It’s just another device for recording the passage of time. In this case, it is simply software mechanism for counting the amount of seconds that have passed. Based on that simple task, we can build bigger and more complicated tasks. If the foundational framework of logical systems is consistent, we can expect accurate, predictable, and (most importantly) repeatable results.
It’s so cool to know that almost all of the world’s computer systems all take their cues from the same place. I don’t know about everyone else, but I love the fact that every second has been accounted for. Even if we haven’t always been paying attention to the passage of time, our machines have. Before I’d ever heard of the Time Recorder or the timestamp I’ve done my own version of this same practice. Using a modified camera I can make photographic images of quite long periods of time. Anywhere from 1–5 mins to months, potentially even longer. I’m just another in a long line of people who trying to find new ways to mark the passage of time.
