There's an ongoing conflict on the meaning of words such as "megabyte", "gigabyte", and so on.

Historically, the computer industry had agreed that rather than being based on powers of 10 like the metric system, these values should be based on the nearest power of two, which can be obtained by multiplying things by 1024 instead of 1000. So, for example, a "kilobyte" was 1024 bytes instead of 1000, and a "megabyte" was 1024 kilobytes instead of 1000 kilobytes.

Then the metric fags came and said everybody had to change it. They said that a kilobyte had to be 1,000 bytes, not 1,024, and the 1024-byte number should be called a "kibibyte" (abbreviated as KiB to differentiate it from KB for kilobyte). The same convention was applied all the way up the scale, and the metric fags thought they had "fixed" the system.

The problem is that only some of the tech industry went along with the new convention, so now when somebody says "megabyte" you can't be sure if he's thinking of original megabytes, known as "mebibytes" by metric fags, or the new metric megabytes.

Also, all documentation and standards from before the metric fags came along has to be read with the knowledge that those words didn't have the metric meaning back then.

The website might be reporting in megabytes (for the file size limit) while your OS is reporting in mebibytes (for the file’s size). That would cause you to hit the limit “without” hitting the limit.