Halifax data projectHow old is the play equipment in Halifax parks? A project on the municipality’s own file
Halifax Regional Municipality publishes an Outdoor Recreation Equipment file as open data: swings, slides, hoops and play structures, one row each. Our teen students download it and ask a plain question that turns out to need careful code. Every figure below is our count, not the municipality’s.
Our count of the Halifax Regional Municipality Outdoor Recreation Equipment file, downloaded 22 October 2026 (2,218 rows). In-service rows only, with an install year rated confidence 1 or 2 by the file| Equipment (decoded from ASSETCODE) | Rows used | Median install year | Oldest year |
|---|
| Climber | 60 | 2007.5 | 2000 |
| Spring Toy | 41 | 2012 | 1998 |
| Play Structure | 274 | 2013 | 1999 |
| Slide | 19 | 2014 | 1997 |
| Swing | 149 | 2015 | 1995 |
| Basketball Hoop | 101 | 2018 | 1994 |
| Other (our group: triple hoops, gym equipment, teeter totters, ramps and items coded OTHER) | 68 | 2013 | 1998 |
Here is the script that prints the table. It needs only Python 3 and the downloaded CSV in the same folder; the equipment names come from the coded values the municipality publishes with the layer:import csv
import statistics
NAMES = {"BBH": "Basketball Hoop", "PYST": "Play Structure", "SWNG": "Swing",
"CLIM": "Climber", "SPTY": "Spring Toy", "SLID": "Slide"}
years = {}
blank = 0
with open("outdoor-recreation-equipment.csv", encoding="utf-8-sig") as f:
for row in csv.DictReader(f):
if row["ASSETSTAT"] != "INS":
continue
if row["INSTYR"] == "":
blank += 1
continue
if row["INSTYRCONF"] not in ("1", "2"):
continue
name = NAMES.get(row["ASSETCODE"], "Other")
years.setdefault(name, []).append(int(row["INSTYR"]))
print("in service, no install year:", blank)
for name, ys in sorted(years.items(), key=lambda p: statistics.median(p[1])):
print(name, len(ys), "median", statistics.median(ys), "oldest", min(ys))
1. Decode the codes
The ASSETCODE column holds short codes such as PYST and SPTY. The layer’s own metadata maps them to Play Structure and Spring Toy, so the first job is a small dictionary that turns codes into words a reader understands.
2. Keep what is in service
Of 2,218 rows, 2,195 carry the status In Service. The rest are out of service, pending, proposed or blank, and a question about equipment in parks today should leave them out.
3. Ask how sure the year is
Each install year has a partner column, INSTYRCONF, rating it from 1 (a measured value or known fact) to 5 (inaccurate). We keep ratings 1 and 2 only, which leaves 712 of the 839 in-service rows that have a year at all.
The number the script prints first
Before any median, the script reports that 1,356 in-service rows have no install year. That is most of the file. A student who publishes “Halifax climbers date from about 2007” without saying the median rests on 60 rows out of 205 climbers has told a smaller truth than it sounds. Saying how much of the data an answer stands on is the habit this project exists to build.
Good next steps for a strong student: compare the ACCESSIBLE column (228 rows marked yes, 378 no, and 1,612 left blank across the whole file) by equipment type, then plot the equipment on a map with the X and Y columns. We use the file under the municipality’s open licence, and its required wording is: Contains information licenced under the Open Government Licence - Halifax (licence, dataset page). Nothing here is endorsed by Halifax Regional Municipality.
