🚀 go-pugleaf

shen@ecf.toronto.edu (SHEN  XINHANG) writes:

>It can be proved that the natural area coding system has the minimum
>characters to represent a location with the same resolution using the most
>common characters:  digits and Latin alphabet without the chance to form bad
>meaning words or phrases.

Re: "bad meaning words or phrases"??  You mean like 4-letter words?  :^)

I think whatever proof you have in mind that your system uses a minimum
number of characters to locate a place will be subjective, no matter how
well-intentioned you are.

For example, since 75% of the earth is covered with water, aren't you
wasting a lot of code-space?  Islands do need to be located, but since they
usually don't move around like ships, they can be grouped, and located
within some name-space.

Your "natural" code system is not natural.  The system needs to be more
fractal-like and less geometric.  Such a system needs to be capable of
varying in/decreased densities, not just as added optional digits.

Why can't the "code" be meaningful and useful in itself?  Why not just have
a better address that needs no artifical code?  Is the English system
discussed based on the "New Town" concept introduced in the rebuilding of
London after WWII?

As an example in an earlier response, I gave a hierarchy based on PLACES --
natural and built environment: Person, Dwelling, Watch Group, Neighborhood,
Village, Community, Town, Area, Region, Nation, Continent, Planet.

The "long form" of the address I gave could be abbreviated to:

	JM/8959/BFR/CV/BG/SW/PDX/PNW/USA

where each code is a unique abbr of the entity, such as PNW meaning Pacific
NorthWest.  The meaning and the long form need only be known locally.  In
such an address, those levels above the area of delivery could be omitted
since they would be irrelevant.  Therefore to deliver something within my
own part of the city, I could leave off every level above 'town'.  Also, if
a particular part of a town had only one community (for example) then the
community name would likely be the same and therefore redundant and not
used, and so on.

I claim that "codes" based on actual places are more useful than articfical
ones that change with the grid lines of the earth.  You need to admit that
your codes do change abruptly and do not acknowledge rivers, freeways,
gulleys, etc, that NATURALLY separate neighborhoods, communities, regions,
etc... and that human development does follow natural patterns and patterns
established by earlier generations of development... i.e. "places".

Your (Shen's) system must be augmented with computations that show the best
delvery routes!  The address system I have cited guides the delivery of each
item all the time, without the aid of any computation.  Of course, all the
place-names would need to be complete, but I claim 1) that this is already
the case in many parts of the world, particulary outside of the USA, and 2)
this is an irreversible trend that we are observing today, and needs only
encouragement to accellerate it.  Very low cost, able to adapt over time.

The problem of dispatching emergency services in rural areas is related to
existing roads.  No matter what coding system people had, the computation of
a best route would have to be a function of the roads, and shortest
distances routes.  Any digital map would have to be very detailed to give
the correct answer for point-to-point routes using your system in all
cases.  Each location, having its unique cell-like code would need to be
recomputed -- you couldn't develop any sense of locality.

Rural roads are very web-like, but ruralites do have rural communities and
neighborhoods even though on a different scale, and these are often ignored
by all but the residents, unfortunately.

I very much sympathize with your motives and justifications.  Your system
is not "user-friendly", however, and I predict that it will never be adapted.

John
--
"What's ahead?"