Milana Glebova
Cartographer

IN THE BEGINNING THERE WAS…

…the map showing the age of buildings in Saint Petersburg – I remember reading about how it was made and feeling huge respect, with a hint of envy, for the author. It was written in a casual manner, but nevertheless I could imagine how much time and effort it took to create something that cool and exciting. There’s very little chance that the cartographer behind that map, Nikita, was aware of that. Nonetheless, later he invited me to join his team at the new stage of the project. I didn’t think twice.

Update: my assumptions regarding skills, time and effort appeared to be enormously underestimated.

It is not a completely novel idea to visualize the age of Moscow buildings: the mapping agency “Mercator” created a similar map several years ago, Strelka KB released an interactive map quite recently. But both projects focus on residential buildings only, while our intention was to give a broader and more thorough overview.

INGREDIENTS

There are several sources containing information on Moscow buildings, but, apparently, they all are coming from the same root – the real property register containing approx. 30 000 houses. Good, but not good enough.

Therefore, we decided to use the ‘Rosreestr’ (2016) – it contains almost all above-ground objects with an incredibly detailed footprint. More importantly, most of the features are attributed with the year of construction, so, after removing overpasses, manholes and other rubbish we get a decent base layer of about 200 000 buildings for our map. Let’s keep searching.
The Rosreestr layer
The next candidate – OpenStreetMap. It has neither detailed attributes we would need, not very precise geometry, but it’s certainly up-to-date: our basic cadaster layer is valid for 2016, so in some places instead of recently built fancy apartments it shows us monstrous factories which have already been demolished. This means that we can use OSM to fill in the gaps in the base layer, solving, however possible, spatial conflicts between past and present.

Rosreestr and OSM layers

By now we have collected all necessary information – buildings’ polygons with date of construction for most. But in addition to this we needed something interesting and meaningful, at least names and photos.

The Ministry of Culture of Russia is supposed to have proper names and photos for the objects under its protection. There are about 5000 of those in Moscow; however, amongst them the number of sculptures, gates and tombs is not proportionally larger than the number of buildings. Also, the information is clearly a bit… old-fashioned since a lot of objects are named and described in their relation to Vladimir Lenin. After filtering, only one half of the dataset is left viable. Not enough.

Base layer (Rosreestr and OSM) + Ministry of Culture points

Wikimapia is another possible source of additional information. Export of centroids gives us about 1 googol of objects, the wheat mixed with the chaff. The categories of objects, as well as their attributes, are rather messy; addresses are mixed with names, and ditches come together with towers. It takes a lot of ingenuity and patience to structure all this chaos. While ‘cleaning’, we can manage to grab some information on construction dates which is valuable for the new buildings not included in the Rosreestr 2016.
google the googol

Base layer + Wikimapia points

Overall, the Wikimapia experiment turned out to fit Thomas Edison’s quote about ‘finding N ways that won’t work’ – a lot of work resulted in an outcome of rather questionable value.
Now, we have a polygon layer containing 250 000 objects with a moderately chaotic attribute table.
GEOPROCESSING
We are using ArcGIS and QGIS software – both being Mr. Good/Dr. Evil depending on the situation. Empirically it turned out to be easier to adjust format and size of dataset to make it agreeable with one of the products.

We’ll look at our hybrid base layer: cadaster, corrected and updated with OSM. It will be our target layer to which we join info from different sources to fill, if possible, the following fields:
- name
- construction date
- address
- style (to be optimistic)
- architect (to be super-optimistic)
- photo
- links to Wikipedia and other external sources.

Cadaster data provides us with the construction data information for 129 000 buildings, the rest is to be found. Starting from the Ministry of Culture data, we have a point layer with attributes; joining them to the base layer to get addresses, names and links to the photos. Getting rid of unnecessary information regarding the life and times of the communism leader when it happens to be in view.

Some buildings have historical and cultural monuments connected to them, and trying to delete the extra data points presents a problem. We are training to solve it, however possible, without damaging valuable information.

Now we go on with structuring the Wikimapia data to derive a neat table filled with addresses, names, years, sometimes styles and photos where it’s possible. Tossing filtered points upon the base layer.

By now the table has several fields filled with information from different sources – some objects have their names and addresses filled in different formats three times. In the case of names the choice is not so wide, getting things in order following the simple principle ‘take what you’ve given’. We can prioritize the addresses though: OSM format is the most accurate; it’s a first choice followed by cadaster where the address field is a little clumsy yet it has a broader coverage. The remaining features get their addresses from Wikimapia. For instance, many of the houses in ‘New Moscow’ have their addresses derived from Wikimapia – this input happened to be way more valuable than photos, most of them of such a poor quality that we’ve decided not to use them at all.

Yet it’s still important to get nice photos. By the laws of the adventure genre, at the last minute the great Wikidata shows up and delivers not so much data in absolute numbers (12 000), but the photos and links of a very good quality. It also has a curious feature: the quality of data is better for less popular objects since well known historical spots and tourist attractions have too many intersections upon them, consequently causing irrelevant information to stick.

So, from these pieces we created a Frankenstein monster – cute and by no means scary.

Distributing all our buildings according to their age, which we discovered for 129 000 buildings:
Here we see the same 1917 peak explained by Strelka KB, the same was relevant for Saint Petersburg: all the houses with unknown construction dates were assigned to 1917 after the Revolution. Apart from that, one can notice decade peaks – 1910, 1900, etc., but this is an effect of rounding the dates of old buildings.

Since the beginning of the Soviet era the pattern has become logical: a decline after WWII, the Khrushchev project in the 1960s, then – stadial decline until the 1990s. Yet the tiny numbers of the recent years are not to be believed: in fact, renovation rocketed the numbers of newly built homes to the Soviet levels.

FIFTY SHADES OF DESPAIR

In a design phase there’s a trade-off between aesthetic and content-meaningful approaches. On the one hand, continuous distribution seems showier, on the other, it blurs content-wise breaks between different eras.

After a series of fittings it turns out that continuous pattern doesn’t fit the age patchwork of the center of Moscow, and doesn't give that impression of a spilled light which looks so hypnotising on the Saint Petersburg map. Maybe that’s for the best – no need to sacrifice content for the sake of form.
‘Moscow party’ and ‘Moscow after acid rain’ look funny, but it would be nice to keep an adequate association with the city.

Not so obvious is how to delineate historical periods of construction. Our predecessors dealt with the same issue and their solution seems absolutely fair: buildings in Moscow are linked either with Soviet leaders, or, in our days, with city mayors:

- Russia before Peter the Great
- Russian Empire
- Lenin
- Stalin
- Khrushchev
- Brezhnev, Andropov, Chernenko
- Gorbachev
- Luzhkov
- Sobyanin

Searching for the right colors. This time referring to different epochs by discrete distribution, but still trying to let them smoothly transition from the warm glow of the past to neon of new developments while still complementing one another.
The chaotically-associative approach wins: coloring the pre-revolutionary buildings in a deep red shade of old manufactories, Stalin’s skyscrapers and their peers – into a bright yellow, like the walls of Moscow State University’s main building in sunlight. Concrete 1970s are dull green. Cold electric blue fits to the modern buildings:
Advice from colleagues: it would be better to split the pre-revolutionary era – after all, the gap between the 14th and 20th century is too big to ignore it, and it’s also fair to distinguish between the old and the ancient.

Looks cool – one can clearly see the Khrushchev project blocks occupying the whole districts, how the monolith scar of New Arbat and Stalinist facades of Tverskaya Street stand out.

Stop, why is the Kremlin also Stalinist? The color scheme highlights the geoprocessing flaws: every join operation delivers a new set of data and a new source of mistakes simultaneously. Especially in the city centre, where the density of buildings and points of interest is as high as the possibility that attributes from a neighboring object will be linked to a polygon by mistake. But I’m still wondering how we’ve got the Gelmgolz Moscow Research Institute of Eye Diseases instead of The Four Seasons Hotel – there’s about 3 kilometers between them.

Uploading the result dataset to Geosemantika, the platform we use for a web map. Once we have set the style and the basemap, configured the object card's appearance, and added the timescale, our map is ready to go online.

Geosemantika interface
GOING POSTAL
Designing a poster, in comparison to dealing with big data, is a familiar and cozy process. Big data processing is not always an obvious procedure, often it is accompanied with a lot of swearing, but is still creative – in its own way.

Designing a poster, in comparison to dealing with big data, is a familiar and cozy process. Big data processing is not always an obvious procedure, often it is accompanied with a lot of swearing, but is still creative – in its own way.

Showing only the area within the Moscow Ring Road, slightly correcting the colors of the base map to adjust the dark background to printing; everything else falls into place. A bit of manipulation with the scale of the horizontal chart showing the distribution of the number of buildings built per year and we’ve got a paper map which is not only a nice tool for covering a hole in a wall, but also an interesting thing to review closely. It’s possible to order it here.

A smiling lady from Printful is holding a virtual poster
THANK YOU
for all who helped me to create this map and participated in the user study, especially to

● Nikita Slavin, the author of the project, for the invitation to play my part and learn a couple of new tricks – it was fun!
● Simon Pavlyuk for the contacts with the best of Moscow urbanists, and for still talking to me after all those phone calls at 1 a.m. asking to help me until yesterday;
● Ross Clark for hunting down the lost articles and generally enhancing the Englishness of this text.