🚀 go-pugleaf

In article <KANZE.95Oct9113030@slsvgqt.lts.sel.alcatel.de>,
kanze@lts.sel.alcatel.de (James Kanze US/ESC 60/3/141 #40763) writes:
>In article <1139933150.15676234@ortel.org> Paul_Long@ortel.org (Paul
>Long) writes:
>|> This is a common misunderstanding. A byte, in Standard C, is not an 8-bit
>|> unit of
>|> storage:
>In this, the standard is only following normal technical usage.  A
>byte has never been defined as 8 bits, but rather the smallest
>addressable size, when that size is less than an word.  In fact, the
>very first use of byte was not for an 8 bit entity.
>
>(This information comes out of a previous discussion of the meaning of
>byte, which I didn't save, which is why I don't know how many bits
>were in the original byte, nor what machine it was.  The jargon file
>will, however, collaborate my pretention that a byte is not always 8
>bits.)

Historically, a byte may have first been something other than 8 bits,
and technical dictionaries may not define it in terms of a partilcuar
number of bits, but I'd be willing to bet that most people, and
specifically programmers, think that a byte is synonymous with
8 bits. That was my point. If someone says that their new disk
drive has 1 Gigabytes of storage, do you stop and ask how
many bits are in their bytes?

>The correct word for a (must-be exactly) 8 bit unit is an octet.  This
>is the word normally used in data comm's, precisely for this reason.
>(If the protocol specifies octets, you had better send 8 bits, even if
>you hardware bytes are 9 bits.)

I've got a background in telcom, too. I started describing "octet" in
my post, but deleted it because I thought that it might be too
tangential to the subject. IMO, "byte" implies a unit of storage,
which is typically 8 bits, while "octet" implies a _set_ of 8 bits
which is not usually talked about in terms of storage, but in
terms of transmitted data. I'm sure you can find something
wrong with this impression, though. :-)

Paul_Long@ortel.org
plong@perf.com