As previously discussed, I am obsessed with the original blue house number plates in Regina. It's not the most popular hobby out there, but what would I do if I didn't have weird interests - watch TV? Play video games?? Relax???
Here's a map of the 104 known original house number locations that I've been able to personally collect, based on a lot of travel through the just-off-downtown Cathedral and Heritage neighbourhoods. From Streetview and other sources, I can confirm that original number plates also exist in North Central, and in fact it's very possible that North Central contains more than Cathedral and Heritage put together. (I also finally found one in the core... while researching this post.)
Let's dig a little deeper. We have:
- Photographs
- GPS coordinates
- Addresses
We don't have:
- Any relevant original sources
Things could be worse, but this is not ideal for a history project. Luckily, there's something else in our corner: the City of Regina Property Search.
Property Assessment Data
This spits out some public tax information about any property that we care to look up...
...including, crucially, the Effective Year Built.
The effective year of construction of a structure. This is used when major renovations have taken place and parts of the building now contain new material. (City of Regina Property Tax Glossary)
By the way, I can't stand when municipalities provide database information that needs to be manually looked up one entry at a time, so I wrote a simple Python web scraper to solve this problem forever.
import requests
import pandas as pd
from lxml import html
url = "https://atpublictaxweb.azurewebsites.net/SearchAccountByID?accountID="
headers = {"User-agent":"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/47.0.2526.80 Safari/537.36"}
numbers = range(10000000, 10400000)
try:
properties = pd.read_csv('regina_properties.csv').dropna(how='all')
properties = properties.set_index('Account Number')
properties.index = properties.index.astype('int')
except:
print("Regina property database doesn't exist, creating it now...")
properties = pd.DataFrame()
for account in numbers:
if account not in properties.index:
print("Processing account number " + str(account))
dest = url + str(account)
page = requests.get(dest, headers=headers)
tree = html.fromstring(page.content)
summary = tree.cssselect('#property_summary table tr td')
assessment = tree.cssselect('#property_assessment table tr td')
taxes = tree.cssselect('#property_taxes table tr td')
l = tree.cssselect('#property_characteristics_left ul li')
r = tree.cssselect('#property_characteristics_right ul li')
summary = [str(x.text_content()) for x in summary]
summary = dict([summary[i * 2:(i + 1) * 2] for i in range((len(summary) + 2 - 1) // 2 )])
for x in list(summary.keys()):
summary[x.replace(":","")] = summary.pop(x)
assessment = [str(x.text_content()) for x in assessment]
assessment = [assessment[i * 3:(i + 1) * 3] for i in range((len(assessment) + 3 - 1) // 3 )]
assessment = dict([[x[0], x[2]] for x in assessment])
summary.update(assessment)
taxes = [str(x.text_content()) for x in taxes]
taxes = [taxes[i * 3:(i + 1) * 3] for i in range((len(taxes) + 3 - 1) // 3 )]
taxes = dict([[x[0], x[2]] for x in taxes])
summary.update(taxes)
characteristics = l + r
characteristics = [x[0].text_content().split(': ') for x in characteristics]
characteristics = dict(characteristics)
summary.update(characteristics)
output = [summary]
if summary:
outputFrame = pd.DataFrame.from_records(output).set_index('Account Number')
else:
outputFrame = pd.DataFrame(index=[account])
properties = pd.concat([properties, outputFrame])
properties.to_csv('regina_properties.csv', index_label='Account Number')
Effective Year Built
With a dataset of 104 observations and access to all the building ages in the city, there is PLENTY that we can learn about the history of these number plates. Let's take a look at build year frequency in Regina's three main inner-city neighbourhoods:
Regina was founded in 1882, recorded a population of 2,249 in 1901, and was incorporated as a city in 1903. None of that matters to this dataset, which only records 11 existing properties as having been constructed before 1905, only 2 of which fall into our study area. Whatever happened in this era, it is very difficult to reconstruct from existing architecture.
But, what the heck, you're here anyway, so here's the full list of 11 properties. Take this with a grain of salt, the city tax roll was never intended to be a heritage inventory.
Building | Address | Effective year built | |
---|---|---|---|
St. Paul's Anglican Cathedral | 1861 | McIntyre Street | 1800 (hint: this is definitely a data error) |
RCMP Depot | 6101 | Dewdney Avenue | 1883 |
Al Ritchie area residence | 2500 | Edgar Street | 1900 |
Northeast area residence | 464 | Toronto Street | 1900 |
Centre Square area residence | 2250 | Scarth Street | 1902 |
Centre Square area residence | 2062 | Cornwall Street | 1903 |
Dawat India Restaurant | 2167 | Hamilton Street | 1903 |
Centre Square area residence | 2049 | Lorne Street | 1903 |
Cathedral area residence | 2354 | Garnet Street | 1904 |
Reign Hair Studio | 2334 | Lorne Street | 1904 |
Cathedral area residence | 3628 | Victoria Avenue | 1904 |
Imperfect as it may be, let's map that tax roll information out. Maps are the fastest way to tell if a dataset is junk or not!
Remember, this data only concerns the current building on every lot, but it still tells us plenty.
In general, residential development in Regina expanded outwards from the core... and then the Great Depression brought the province grinding to a halt, creating a lost decade where nothing was built anywhere in the city. Regina wouldn't boom again until 1945, the advent of the modern suburb.
A useful rule of thumb for this area: if anything is built after 1930, it is modern and doesn't fit the original age of the neighbourhood. In particular, you can see intense redevelopment around the major commercial arteries that surround both areas.
Sample vs. Population
So how do those number plate observations match up with the overall population of old buildings?
Bingo! Even though 104 is not that large of a sample, the distribution clearly reflects the two-peaked pre-war boom and interwar boom that we see in the city tax records.
The newest building I've found with an original number plate was built in 1955. So the city probably stopped handing them out sometime in the 50s, right?
NOT SO FAST. These old plates are usually attached by two rusty nails, and there is nothing to guarantee that they haven't been moved or tampered with. As long as people can't resist messing around with their property, no single observation in this dataset should be regarded as fully credible. (Perhaps they even bought a convincing replica of a historical fixture because of an inexplicable personal obsession?)
Remember, a month of hunting only revealed 104 original number plates, in two neighbourhoods that contain 4,195 addresses between them. That's a miserable 2.5% survival rate:
Neighbourhood | Total addresses | Observations | Survival rate |
---|---|---|---|
Cathedral | 2,557 | 41 | 1.6% |
Heritage | 1,638 | 63 | 3.8% |
Given that complete drop-off in all construction during the Depression, I suspect that the plates stopped being manufactured around 1930. Why keep maintaining a standard address system for a bankrupt city that's not growing? The rest of the story might look something like this.
Remember, Cathedral and Heritage were almost completely built out before the Depression: only North Central retained enough vacant development land to record large numbers of new properties in the early 50s. During the post-war boom, a few new builds reused the original plates. Maybe this was done under the direction of a new generation of owners, with the grandparents still remembering being issued their shiny new plate on their first house, twenty to forty years earlier? Or was it the responsibility of the general contractor, who was trained under some bylaw to make sure that the standard blue plate remained visible from the street? Eventually, with an explosion of post-war consumer goods on the market, the existence of the government number plates would have been forgotten, and buying your own insignia from the hardware store would have become the norm as it is today.
Here's an example of a plate that was clearly saved because a later renovator recognized its heritage value:
It's a nice story, but without any other sources, it's just a story. Still, there's one additional piece of information hidden in these blue enamel plates.
Paleographic Analysis
Paleography (Wikipedia) - "one of the auxiliary sciences of history". Analyzing the shapes of letters and numbers, which tend to shift from generation to generation, is used to date Biblical manuscripts more accurately than radiocarbon.
In a modern industrial context, we already have our century pinpointed pretty accurately, but there's one secret that these plates have left to give up: they were painted by two distinctive hands.
Can you see it? The Regular hand is well-practiced, but is definitely not a set of machine stencils. The form of the 2's varies a lot, and the kerning is all over the place.
The Narrow hand is a little tidier, usually a little thinner, takes up less of the plate, and always leaves a distinctive point on the top of every 1.
I've also found four oddballs that defy any classification. Two seem to match each other, while the other two are completely unlike any other plate in the city. Despite the dissimilarity, their dimensions and weathering patterns suggest that they are of the same original vintage as the other plates, and I strongly doubt that they would have come from a different factory.
The Regular hand is dominant, but not overwhelmingly so. Here's the frequency table:
Type | Count |
---|---|
Regular | 97 |
Narrow | 19 |
Oddballs | 4 |
What happens when we plug this typeface classification into our tax year data?
Bingo! There's a clear relationship!!
Here are my best guesses about how these signs were made. First of all, I think they were manufactured inside the city. Saskatoon is still home to an enamel sign manufacturer today, in an era when the technology is much less ubiquitous. I'll further assume that the consistent Regular hand was the work of a single employee, who would have had thousands of very rote address signs to produce over their long career. How much of the manufacturer's business would have been taken up by this single reliable public-sector contract, I cannot guess.
I am very confident in suggesting the following timeline for Regina house number plates:
- The original Regina sign painter worked from the earliest days of the city to 1928.
- The narrow hand took over from 1928 to 1930.
- Standard number plates were abolished as the Depression took hold.
- Later builds with old number plates are proof of the human tendency to become interested in the strangest things.
- The oddballs are so physically similar to the other plates, but so few in number. They were the result of other employees occasionally filling in.
Modern Hope
Clearly, some other people in Regina are interested in their original number plates.
One lovely house in Cathedral boasts a replica from Sosenco, the Polish supplier that I used for my own replica! What a shame that this guy just went with the off-the-shelf typeface, rather than spending days wandering the neighbourhood in order to reconstruct the original government-issue typeface. Having a real job must be really hard.
Elsewhere in Cathedral, the plates are tucked safely behind glass, never to touch another raindrop...
...or displayed right alongside other symbols of local pride.
It's not surprising that Cathedral people would fussily look after their old properties, but even in the rougher Heritage area, plates in situ showed evidence of repair...
...restoration...
...and even re-invention.
When a person eases off a period of intense obsession, it's tempting to rationalize it all by speculating that the object of the fixation is actually really important. I doubt it! I guarantee most people who hang out in the middle of Regina have gone by these things hundreds of times without noticing, and installing eight thousand new blue enamel number plates would be very unlikely to bring back the short-lived prosperity of Saskatchewan in the 1920s.
To Regina's credit, everybody who caught me taking pictures of their front door was polite and, by the end of the conversation, most of them were even curious. If you happened to catch a suspicious man on your doorbell camera in October: whoops! (I don't even want to imagine what's going around on Facebook about this.)
I like standardized things; they are a reminder that we live on a planet where co-operation is mandatory. If you like getting things delivered to your house, you probably want to participate in a local authority that can implement a consistent addressing system. If the city hires one central sign painter, it creates a good, reliable job, and less waste than sending everybody to drive to a store at the edge of town to improvise one themselves. Clearly, these are some crazy, radical thoughts.
If I've learned one thing on this project, it's that I am never going to be able to walk through an old neighbourhood for the rest of my life without trying to reconstruct a little mental breakdown of what exactly Western Canadian hardware stores were selling in 1925.