As an example of creative house address planning possibilities, we plan the address grid of an small community using a coordinate system with only one axis. We curve our "grid" following the general trend of the streets. The result looks like a familiar contour map, but is an "address contour" map.
Small bent final parts of streets have one end rooted in our "address surface", from which their numbers then proceed independently.
Goals:
100% Free Software is used.
Keywords: address planning
This paper was included in the Taiwan Geographic Information Society 2005 conference.
This is an example of "custom made" addressing to fit a particular community. As several streets there are to be assigned first time numbers, we took the opportunity to make an integrated numbering plan for the whole community. The result is a highly specialized example of one possibility of addressing tailor made for a certain community.
![[Image: Shangcheng address contour map]](images/twist.jpg)
1:5000 orthophotomap of Shangcheng Community, Dongshi Township, Taizhong County, Taiwan. This paper describes a proposal for that community's addressing.
Pictured are "address contours", similar to familiar elevation contours. Getting an approximate value from his position amongst the contours on the map, a worker then looks along the road in the direction of increasing values (always south (down) here), then assigns numbers, odd on the left side of the road, even on the right.
We only specify the Y axis. The X axis will be little used for this community.
On the map we see smaller numbered pairs,
They are exceptions for bent ends of streets. Here we have three streets each with two bent ends. For example, on the first street, numbers 110 to 407 will fit nicely into our contour system. But what about the bent ends? At house number 110 we begin independently numbering until 38. That is, 110 is still part of our overall system, but from 109 to 38 the system is ignored and we create house numbers based solely on distance along the street. Same from 407 to 437.
With these splices in place, we can appear more seamless to the man in the street.
Of course in the end our system must be easy for the man in the street to use for it to be meaningful.
We establish a grid to assign values with, as is done in North America, but as the major roads are all north-south, we don't even bother making an east-west component of the grid. Our half grid is in fact a "house number contour map". We also twist the contours to better match the street pattern.
We expect any east-west lanes to use the usual Taiwan style lane number subordinate to their "parent" north-south road. The house number is then computed by distance along the lane, as below.
For final segments of north-south roads that bend to meet other north-south roads, we "cheat" and just extend the numbers starting at where the curve begins, ending where we hit the other road. For that part we use our often used [J] 400 numbers per kilometer standard. (Upon travelling one kilometer, the house on the right is numbered 400, and the house on the left is 401.) We also use 400 numbers per km. along our main system axis. In the big picture, we achieve our goal of lockstep numbering of parallel roads.
The reader might wonder how we achieve numbering continuity with the communities to the north and south. Well, in fact our axis road is actually to be called Dongguan Rd. Section 2. Numbered independently, as is Taiwan custom, vs. Section 1 to the north, and Section 3 to the south. In the sectionalized system, one needs not delve deeper into continuity with other sections. Sectioning of long roads is contemporary Taiwan "house numbering language", hence we also follow this practice.
The non-sectionized roads of our study area are to be numbered in lockstep with Dongguan Rd. Section 2.
In Taiwan up till now even hoping the numbers on the left and right sides of a single road could stay coordinated for very far was dreaming. One hopes upon reaching the next Section the numbers would do better, but no, before long they start slipping again. Section 3 same story. Section 4 same story. In this paper, we not only require the left and right sides of the same road to be lockstep, parallel roads are all to be lockstep.
In Taiwan up till now, "just as the sun rises in the east, house numbers start at 1." We throw this in the dumpster, if we are to accomplish anything. In a total community plan, not every road will have a number 1. You might have to look for it in the museum.
We use Free Software: GRASS GIS[G]. Program code: http://jidanni.org/geo/house_numbering/programs/twist/makefile.txt . Highlights:
At this point we have already drawn by hand:
![[Image: rib cage]](images/ribcage.png)
The horizontal trend guide curves are just arbitrary curves we placed onto the orthophotomap reflecting what we feel are the trends the contours to be produced should follow. We can put as many as we like on the map. The final contours produced will generally be shaped similar to these trend curves.
We draw our vertical axis curve following the street we chose as the axis. There numbering will indeed be 400 addresses per kilometer, exactly. We also chose an origin point.
That ends the choices we make by eye. The rest is more mechanical.
There are unlimited opportunities for unique creative solutions in address planning applied to a single community.
[G] GRASS GIS: http://grass.itc.it/
[J] Jacobson, D. Taiwan address planning: http://jidanni.org/geo/house_numbering/
This article was written Summer, 2005.
Copyright: GNU Free Documentation License: http://www.fsf.org/licensing/licenses/fdl.html
Last modified: 2007-06-12 03:28:31 +0800