Message passing syntax for objects | OOPv2

Chris Angelico rosuav at gmail.com
Sat May 11 21:10:36 EDT 2013


On Sun, May 12, 2013 at 10:50 AM, Gregory Ewing
<greg.ewing at canterbury.ac.nz> wrote:
> Dennis Lee Bieber wrote:
>>>
>>> I also believe in a path of endless
>>> exponential growth. Challenge: Create more information than can be
>>> stored in one teaspoon of matter. Go ahead. Try!
>
>
> If that's your argument, then you don't really believe
> in *endless* exponential growth. You only believe in
> "exponential growth for long enough that I won't be
> around to suffer the consequences when it runs out".

Technically, according to the laws of thermodynamics, there cannot be
any actually endless growth, yes. (Anything beyond that is the realm
of religion, not science.) But in that case, the term "endless" is
like "infinity" - a concept only. Like the Infinite Monkey Protocol
Suite description in RFC 2795, there will be many numbers that come up
that are plenty huge but fall pitifully short of infinity (Graham's
Number, for instance, is pretty small in those terms).

So long as storage capacities keep on increasing, we can keep
increasing the world's information at the same rate. So long as the
number of computers connected to the internet increases, we can keep
increasing the internet's information at the same rate. Put both
together - and neither shows any sign of ceasing any time soon - we
can continue with the corresponding growth. How long before that runs
out? A *looooong* time. We're not talking here of the Year 2000, a
couple of decades after the software was written. We're not talking
about the 2038 issues, roughly half a century after the software was
written. We are talking timeframes that make the Y10K problem look
like a serious lack of foresight.

ChrisA



More information about the Python-list mailing list