It’s no secret that nowadays we are not only used to, but somewhat dependent on technology, as it has become such an important part of our daily routines. That’s what makes software malfunctioning even more painful—something you once considered your ally has turned against you, likely at the moment when you needed it the most. (I am not sure what law it is, but technology always seems to fail us at the worst possible time.)

While glitches and failures show that there’s still room for improvement—or that it’s time to visit a repair shop—software engineering has been developing in leaps and bounds over the decades (since the 1950s, at least, when the term ‘software’ was coined), and there seems to be no end in sight.

In a recent interview with Bored Panda, professor of computer science and software engineering at University of Ottawa, Dr. Timothy C. Lethbridge seconded the idea that software engineering has been developing at a very rapid pace over the years.

In his opinion, some of the biggest leaps in software engineering include moments when certain new languages and 'frameworks' have appeared, when computing 'in the cloud' became dominant, when mobile software started to work well, when automated testing started to be widely used, and when software engineers could look up a certain answer on the web when they get stuck. “Now, a tool like ChatGPT can help software developers,” he added.

According to Dr. Lethbridge, even the smallest errors in a software can lead to big problems, many of which tend to be safety critical. But depending on the kind of ‘glitch’ the software is experiencing, the reasons for its presence might differ.

In the expert’s eyes, software glitches could be categorized into 5 groups, the first one being freezes, slowness, and failures caused by bad network connections (either on our own device or somewhere else). The key root causes for such an inconvenience, according to him, can be inadequate or nonexistent network connection, excessive data transfer, bits of data arriving in the wrong order, or a cyberattack (overloading the network).

The second type of glitches entail slowness and freezes caused by extreme popularity resulting in overload (for example, a sports app in a major game or opening day of a big sale), which happens because of lack of planning for capacity to process all the requests at peak times.

The third group of glitches include freezes or incomplete results caused by an outage at a 'server' (one of the computers providing the service over the network). It can happen as a result of a crash due to many kinds of bugs, a cyberattack on the server, or a planned maintenance or update, Dr. Lethbridge explained.

Some glitches can also be user experience (UX) defects. “Much software is always too slow due to being built using overly complex components,” Dr. Lethbridge told Bored Panda. “In addition to that, confusing labels, feedback, and instructions leading the user to do the wrong thing, get confused or get stuck.”

Glitches based on UX defects can also be related to excessively complex user interface, making it hard to do the work or leading users to do it wrongly (for instance, too many steps, or steps that are difficult to do without making mistakes).

Then there’s also bad design making it impossible for some users to do something (for example, forcing you to enter a postal code in a country that does not use them) or bad translations when converting software to another country or language, the expert pointed out.

The last group of glitches are software’s wrong answers or outcomes, such as a Maps app sending you round and round in circles, or a product on sale being sold for free or delivered to the wrong address.

This group can also comprise glitches caused by bad data in the system entered by humans or corrupted by earlier glitches, bad algorithms (for example, mixing up imperial vs. metric units, or making excessive rounding errors in computations), the software not considering special cases or large cases or 'edge cases' (for instance, limiting the number of kids in a family to 4), or a cyberattack or hacking designed to get the system to do the wrong thing.

If this list made you laugh instead of crying and you would like to browse some more similar content, we have you covered—continue to our previous edition on instances of software giving up on people and delve even deeper into the wondrous world of software engineering. Happy scrolling!