On a Monday in September 2023, police broke into a dark storage room in a small apartment in Brøndby.

They found an emaciated young man lying on a thin mattress with nothing but a jug of water and a bucket in the corner.

The young man had been locked up here for 35 days. This is stated in the indictment in the case, which is expected to end on Friday.

According to the indictment, four siblings, three women and a man, kept their brother locked in a small locked room with no windows or light for over a month because he told them he was gay.

He was denied food or drink at times and forced to sleep on a thin, scabies-infested mattress and defecate in a bucket.

The young man only escaped after 35 days when a neighbor heard him crying for help and immediately called the police.

By then he had lost 14 kilos, had symptoms of scabies and several bruises from blows.

Using "literally" to mean "figuratively"

Obligatory relevant xkcd. While I understand the argument of validating this use as a natural extension of using exaggeration for emphasis (and it's intuitive enough that I sometimes catch myself doing it), I do think the words that are supposed to mean "I am not exaggerating, using a metaphor or joking, I mean this in exactly the way I worded it" should be exempt from this. Language history is no stranger to words shifting their meaning until they're the opposite of their former meaning, and there are plenty of situations where words simultaneously have opposite meanings (in fact, enough of them that a term exists for this which has its own Wikipedia article) where I don't think it's much of an issue, but I do think this matters here, especially since this is happening to many similar words used for that purpose (such as "actually" and "genuinely"), not just "literally". Unambiguously clarifying that the meaning of your statement isn't figurative is something important enough that the words for it shouldn't have their meaning diluted IMHO.

bytes vs bits

More of a matter of technical standardization than pure linguistics, but two separate albeit related issues are happening here. First, a "bit" here is a unit of digital data, being either 0 or 1, whereas a "byte" is another unit usually made of eight bits. While you can define bytes following a different amount as some older and specialized machines do, in practice there is no ambiguity with 8 being the accepted norm and other words (including the word "word" itself, funnily enough) being available should the distinction matter. The bit/byte distinction is commonly understood and usually not a matter of confusion... until you start bringing up shortened unit names and disingenuous marketing. Despite the unit that the average user is most familiar with being the byte, shortened to an uppercase B as a unit symbol, while bits are in turn shortened to a lowercase b, unscrupulous advertisers will often take advantage of the fact bits are a lesser known unit while using almost the exact same symbol to refer to, say, network speeds for a broadband plan they're trying to sell using bits, not bytes, allowing them to sneak in numbers (in bits) eight times bigger than the values (in bytes) the customer would expect in the unit they are more accustomed to.

Using SI prefixes in place of equivalent binary prefixes

The second part (and with it another relevant xkcd) is the distinction between decimal SI prefixes and the IEC binary prefixes, or lack thereof in common usage. For context, a convenient coincidence for computer science is that the value of 1000^x and 2^10x (equal to 1024^x) are similar enough for a low enough value of x to map binary prefixes according to the usual SI metric prefixes (so while 1kg is a kilogram equaling 1000 grams, 1KiB is one kibibyte equaling 1024 bytes, 1MiB is a mebibyte equaling 1024 kibibytes, and so on) and using them to refer to data sizes, which is a lot more convenient to deal with when everything related to computing is binary rather than decimal. This also led before the adoption of those binary prefixes to the practice of "incorrectly" using the SI decimal prefixes' names and symbols when referring to binary data sizes. I'm perfectly fine with this in informal contexts precisely because it's a convenient shortcut and the inaccuracy usually doesn't matter, but this also means a company can pretend, say, a storage device they're selling has a higher capacity than it really does by mixing up the units to their advantage. Worse, a company that does want to convey the capacity of their devices to the user in good faith has another issue: MacOS defaults to computing sizes displayed to the user using the decimal prefixes (1MB = 1000kB = 1 million bytes), while Linux generally defaults to the binary sizes (1MiB = 1024KiB = 1024² = 1048576 bytes).

In which I manage to blame Microsoft for a grammar error

Not much of an issue if you stick to the correct units, but given which OS I pointedly didn't mention yet, you probably realized where this is going. Microsoft, in their infinite wisdom, elected to show data sizes on Windows that, while computed according to binary sizes, are displayed using the decimal prefixes, leading to 1"MB" = 1024 "KB" = 1048576 bytes, but displayed in units that imply it should be 1000000 bytes. This is a holdover from the older practice of using the metric SI prefixes' names as binary prefixes when specifically referring to bytes I mentioned above, which is nowadays discouraged in favor of the international standard for binary prefixes established back in 1999... but clearly Microsoft didn't get the memo. Is it a minor problem in the grand scheme of things? Absolutely, but I consider this negligent handling of a pretty fundamental question where a clear consensus has been established given that this is coming from the company that publishes the consumer OS running on the overwhelming majority of personal computers. As someone who is familiar with computing, I understand why the mental shortcut makes sense. As a consumer, if I buy a kilogram of something, I expect to receive as close to a thousand grams as the manufacturing process reasonably allows, not consistently end up with 976.5625 grams instead of the advertised kilogram. In any other context, "it's more convenient to pretend we count in base 10 but we're actually counting in base 2 and not properly converting the numbers back, usually to the detriment of the customer" would be seen as absurd, but the IT industry apparently got away with it. By not following the internationally standardized terms in their own OS, Microsoft is perpetuating this issue which is doing us a disservice... and I'll move on to the next example before this becomes yet another computing rant in what's supposed to be part of a thread about language, and not even the programming kind.

I could/n't care less

I'm starting to see a pattern here. Another case of "saying something but actually meaning the opposite" which I think is important to be mindful of. Granted, "I couldn't care less" is a common enough stock phrase that omitting the negation usually is recognized as such and not interpreted to mean the opposite, and there are other (and probably more intuitive) ways to convey the literal meaning of "I could care less", but given that there are generally a whole lot fewer things people care about (and therefore occasions to state it) than the alternative, I think it matters more to keep a way to mean that something does actually matter to you intact than expanding the way people can say that they don't care about something by including the exact opposite. I've also seen this used in yet another way to refer something they care about to at least some degree, but still little (with the reasoning that feeling the need to state explicitly that you are able to care less implicitly states that you cared very little in the first place) which is very similar to the meaning of "I couldn't care less" but still has makes an important distinction that I think should be preserved.

Wrong homophones (or otherwise similarly sounding words) when the correct one is not obvious

Mistakes derived from those are usually not an issue since it's very easy to tell which is the correct one... until it's not. For example, "brake" vs "break" when talking about cars, "ordinance" vs "ordnance" when the topic intersects bureaucracy and the military, and "raise" vs "raze" might lead to very unfortunate misunderstandings in construction. More generally, "hear" vs "here" can quickly make the meaning of a sentence incomprehensible especially if the mixup happens in a sentence where both are used, and "than" vs "then" can radically change the meaning of the sentence. Similar sounding words can have pretty significant differences without mixing them up being necessarily obvious, such as amuse/bemuse, persecute/prosecute or prescribe/proscribe. Ironically, the common mixups that people tend to find the most annoying to see (e.g their/they're/there, to/two/too, loose/lose, affect/effect, should or could of instead of should or could have, definitely/defiantly) aren't the ones that are likely to actually introduce ambiguity (I would suspect bad faith from anyone claiming a mixup between "angel" and "angle" is actually ambiguous, with one notable exception), or, if they do, not in a way that would radically warp the sentence's meaning (inflict/afflict is a common one and the two words are similar enough that it would be difficult to notice if the "wrong" one was used... but that goes both ways: they're so similar such a mixup would most likely be of little consequence to the overall meaning)

Leaving unclear links between clauses

While the above is mostly about word (mis-)use, another big category of mistakes that gives me a headache is made up of sentences where the ambiguity comes from the structure of the sentence itself. I would include Garden-path sentences and certain cases of dangling/misplaced modifiers in this category (though not all of them as context is often enough to clear up any possible confusion). For the former, news article titles that are too clever for their own good by trying to fit as much information in as few words as possible are notable offenders. I've actually given up trying to understand a news headline for this reason at least once. For the latter, there are already many examples out there of leveraging it for comedy, so I'll use the following as a more straightforward example: in the sentence "I need to invite my best friend, the CEO and the mayor", it is unclear whether I'm referring to a single person that is my best friend, the CEO and the mayor at the same time, two people one of which is both my best friend and the CEO and the second person is the mayor, or three different people. Ambiguities like these are something I consider important to be mindful of because they can quickly result in the meaning you intended to convey being completely warped.