[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Counting and so on
- To: John Cowan <cowan@SNARK.THYRSUS.COM>, Eric Raymond <eric@SNARK.THYRSUS.COM>, Eric Tiedemann <est@SNARK.THYRSUS.COM>
- Subject: Re: Counting and so on
- From: Chris Handley <cbmvax!uunet!GANDALF.OTAGO.AC.NZ!cuvmb.cc.columbia.edu!CHandley>
- Date: Tue, 28 Jan 1992 09:19:01 GMT+1200
- Reply-To: cbmvax!uunet!otago.ac.nz!cuvmb.cc.columbia.edu!chandley
- Sender: Lojban list <cbmvax!uunet!CUVMA.BITNET!cuvmb.cc.columbia.edu!LOJBAN>
Hi All
WRT the grammar of numbers and arithmetical/mathematical
expressions, one needs to decide whether one is going to be a litle
sloppy and follow the natural languages or whether one is going to
be precise and formal and follow the programming languages. No PL
that I know of allows one to elide an operator, nor to have more
than one (decimal?) point in a number.
However, I have alway found the Loglan/Lojban interpretation of a
string of digits as being a number rather than as a string of
digits. The '2' in '23' is different from the '3' and that
distinction is vital and should be retained. I admit that for
someone saying (particularly reading) a number, the current version
is easier, but I suspect that things are not that easy for the
listener, who has no idea until the end how big the number is, and
by then will have forgotten how it started or how many digits it
had, particularly if it had more than about 6 or 8. If one thinks
about it, we do not normally write long strings of digits, we break
them up. Telephone numvbers are divided into area code, exchange
code, and number; other numbers are divided by spaces or commas
depending on your background and I deplore the fact that most PLs do
not allow one to do the same - I certainly cannot easily comprehend
a string such as 18359807463 without some thought, and I am
convinced I could not do it on the fly in the middle of speech.
However the equivalent string 18,359,807,463 is a doddle.
Hence my vote is for some way of distinguishing the positions of
digits in a string so that large numbers can be assimilated easily.
This would also mean that I could distinguish between a string of
digits and a number or break a 'nameber' up in different ways -
there is a big difference between 'Pennsylvania 64 thousand' and
Pennsylvania 64 oh oh oh for those of you that can remember that far
back.
In addition I would like to support Ivan's position on complex
numbers, the construct {value kau value} is short and precise and
translates easily to any of the other ways of dealing with complex
numbers.
Chris Handley chandley@otago.ac.nz
Dept of Computer Science Ph (+64) 3-479-8499
University of Otago Fax (+64) 3-479-8577
Dunedin, NZ