From: Hans Aberg (haberg@math.su.se)
Date: Sun Jun 04 2006 - 07:35:08 CDT
On 2 Jun 2006, at 23:33, Theodore H. Smith wrote:
> Moores law doesn't mean we should be more wasteful.
It says that one can waste space, if needed. For example, at the
beginning of the 1990'ies only computer scientist's beefed up
personal computers could hold a full fledged functional language
implementation; nowadays, a functional language implementation does
not take up any significant amount of space on a standard, newly sold
personal computer. There is a similar principle with the character
width issue.
> I think by the time we are citing Moore's law, this isn't really a
> Unicode discussion, but a computing in general discussion...
Right.
> It could just mean that we should waste less, and have less
> ecological impact. or it could mean that we just make more money.
>
> If you get a computer 4x as fast, instead of using UTF-32 of UTF-8,
> you could maybe make 4x the money by having 4x the throughput.
The way to do modern computer programming is to first implement it as
structurally well as is possible, as the helps up maintenance, and
then doing some profiling to find out the parts the needs optimization.
> My original point was that UTF-8 can be used for more than it is
> given credit for. ... I've built some tools for UTF-8, and I'd like
> to build more.
Otherwise, just use the encoding that is most efficient for the needs
at hand. If it is UTF-8 for you, by all means, use it.
Hans Aberg
This archive was generated by hypermail 2.1.5 : Sun Jun 04 2006 - 07:37:54 CDT