Could you replace it with the calculator buttons? Does the
relationship of symbol to actual numeric value harm the
dissertation?
Yes, but calculator buttons would lead to a much less
clear differentiation, in my opinion.
The Strowger-switch example can be
taken to much greater detail:
In the original direct-dial telephone, the
digit-symbols and letter-symbols of the "telephone number" are individually
translated into a unary code, represented by a pulse train. This encoding
– visual symbol to electrical symbol – is what drives the strowger
switch exchange. This is a trivial symbol mapping, requiring no lookup. The
mapping is built into the dial mechanism, completely.
It's possible to bypass
the dial and generate the pulse train directly, as I learned as a kid. That
doesn't change things. In current practice, in any exchange that's responsive
to pulse train encoding, each digit's encoding is mapped, directly, to DTMF (or
maybe an MF) or to the telco variant of BCD (where '0' is coded as 0xA, decimal
10) for further use.
The point is that the older technologies used the
encoding of the visual symbol as a direct mapping to the network hardware, which
is exactly analogous to memory addressing and index fiddling. The symbol is
trivially modified by the technology into something directly useful. This is
the kind of usage that the CS term of art 'direct' encompasses, and which is
absent in the 'symbolic' term of art.
In later network technologies, staring
with IIRC, panel switching, the mapping from symbol to physical is less trivial:
I don't know panel-switch tech well enough to make a firm statement. The
current technologies, however, definitely treat the 'telephone number' string
('directory number') as a symbol used by a lookup mapping to the
physical.
IIRC, the ESS-1 used a table made from ferrite ROM for the lookup:
any change in the mapping required a change in the table's wiring. But it was
still a direct mapping, of the trivial kind. That's not the current technology.
The advent of 'number portability' made it necessary to do a database lookup of
the directory number in order to determine anything about the destination RJ-11
or cell-phone ESN. That's a nontrivial lookup, as the front end to more
processing — thus, the 'telephone number' is now properly treated, in CS
terms of art, as a symbol. --- --Bill. NAL: question the answers,
especially mine. [ Reply to This | Parent | # ]
|