If it wasn’t to confuse people, they could just show both bits and bytes. There’s no reason not to. I’ve seen this occasionally with ethical ISPs who care about their customers (extremely rare) - they would display it like this:
Typical evening speed: 25Mbps (approx 3 megabytes per second)
Almost everyone I know who isn’t in tech is extremely confused by this distinction. Absolutely they keep the bigger number because it directly benefits them. If they’re offering the 3 speed, but competitors are offering the 25 speed, of course customers want the bigger number! Normies don’t even know what a bit is.
Since 1975 or so, the word byte has come to mean a sequence of precisely eight binary digits
I don’t think it’s a big deal or a hill to die on or anything. Since everyone does it, at least it’s consistent. But the industry is kind of shooting themselves in the foot because they get ten thousand service desk calls per day complaining about slow internet speeds that are less than one eighth of the advertised number.
But what prefix? You could end up showing 4 units: Megabits, Mebibits, Megabytes and Mebibytes. Although there’s also the JEDEC 100B.01, under which binary prefix is used with KB, MB, GB, TB, where the only clear difference is between kB and KB.
In that case, you get the insane 1TB = ~1.1TB or 1TB = 0.9TB that you see on Windows.
My favorite example of unit confusion is the “1.44MB floppy”, which is neither true in MB, nor MiB, because it’s actually 1,440KiB, so it’s either ~1.47MB or ~1.41MiB.
There’s no reason not to include a helpful subtitle that gives you an approximation of file download speeds ( other than consumers mistaking it as a promise the company can’t actually keep), but that’s a completely different animal than advertising the bandwidth itself in bytes. Because simply converting the bits into bytes will not result in the number that the customer ultimately sees.
All you would have in that scenario is the same confused consumer calling the service desk asking why the number of Megabytes they paid for is not the number of Megabytes they get when they download a file.
This as absolutely a hill worth dying on for me. I don’t think it would be a good thing for companies to stop advertising their products based on their precise and accurate capabilities in order to sell to a consumer based on a guesstimate around one subset of use of that product.
Uh no by defending the completely honest and objectively accurate advertising. As opposed to the inaccurate and misleading advertising that would be advertising based on file download speed.
Modern networks pass data serially bit by bit over a wire. So on the lowest level networking works on a bit level, not a byte level. There is some logical stuff in upper protocol layers that will actually group things by bytes and such, but when you strip everything down to the smallest detail of what networking does, transmit data over a wire via electricity or lights, or otherwise wirelessly, its bits.
The Mbit number your your ISP sells to you is actually lower than what you actually get, to ensure that you get what you paid them for. If you track your bandwidth usage closely you’ll notice you burst higher than what that number it’s “up to” because there are variables outside of their control as well. So it’s not a hard ceiling, but a rate they promise to be available to you.
I repeated what you wrote here because I only now scrolled down, but thank you for being accurate about this.
It really is a hill worth dying on, in my humble opinion.
No matter how you flip it, when it comes down to the lowest level of bits travelling over the wire, using bits/s is really the only option to use for ISP providers.
If it wasn’t to confuse people, they could just show both bits and bytes. There’s no reason not to. I’ve seen this occasionally with ethical ISPs who care about their customers (extremely rare) - they would display it like this:
Almost everyone I know who isn’t in tech is extremely confused by this distinction. Absolutely they keep the bigger number because it directly benefits them. If they’re offering the 3 speed, but competitors are offering the 25 speed, of course customers want the bigger number! Normies don’t even know what a bit is.
I don’t particularly think the historical value of a byte is really relevant either. It’s been 8 bits for half a century now.
I don’t think it’s a big deal or a hill to die on or anything. Since everyone does it, at least it’s consistent. But the industry is kind of shooting themselves in the foot because they get ten thousand service desk calls per day complaining about slow internet speeds that are less than one eighth of the advertised number.
But what prefix? You could end up showing 4 units: Megabits, Mebibits, Megabytes and Mebibytes. Although there’s also the JEDEC 100B.01, under which binary prefix is used with KB, MB, GB, TB, where the only clear difference is between kB and KB.
In that case, you get the insane 1TB = ~1.1TB or 1TB = 0.9TB that you see on Windows.
My favorite example of unit confusion is the “1.44MB floppy”, which is neither true in MB, nor MiB, because it’s actually 1,440KiB, so it’s either ~1.47MB or ~1.41MiB.
There’s no reason not to include a helpful subtitle that gives you an approximation of file download speeds ( other than consumers mistaking it as a promise the company can’t actually keep), but that’s a completely different animal than advertising the bandwidth itself in bytes. Because simply converting the bits into bytes will not result in the number that the customer ultimately sees.
All you would have in that scenario is the same confused consumer calling the service desk asking why the number of Megabytes they paid for is not the number of Megabytes they get when they download a file.
This as absolutely a hill worth dying on for me. I don’t think it would be a good thing for companies to stop advertising their products based on their precise and accurate capabilities in order to sell to a consumer based on a guesstimate around one subset of use of that product.
Found the marketer.
Not a marketer, just a network engineer who cares about honest and accurate advertising.
By defending the least honest or accurate advertising in existence?
Okay…
Uh no by defending the completely honest and objectively accurate advertising. As opposed to the inaccurate and misleading advertising that would be advertising based on file download speed.
Bet you think a gigabyte is a thousand megabytes too.
How is the Mbit number an accurate value? In my understanding it was also an upward bound of an estimate
Modern networks pass data serially bit by bit over a wire. So on the lowest level networking works on a bit level, not a byte level. There is some logical stuff in upper protocol layers that will actually group things by bytes and such, but when you strip everything down to the smallest detail of what networking does, transmit data over a wire via electricity or lights, or otherwise wirelessly, its bits.
The Mbit number your your ISP sells to you is actually lower than what you actually get, to ensure that you get what you paid them for. If you track your bandwidth usage closely you’ll notice you burst higher than what that number it’s “up to” because there are variables outside of their control as well. So it’s not a hard ceiling, but a rate they promise to be available to you.
I repeated what you wrote here because I only now scrolled down, but thank you for being accurate about this. It really is a hill worth dying on, in my humble opinion.
No matter how you flip it, when it comes down to the lowest level of bits travelling over the wire, using bits/s is really the only option to use for ISP providers.
Follow up, is this rate limited by the link from your home to their local node primarily, or from their local node to the broader internet?
It’s from your home to their local node primarily.