In my last post I made some observations on the project plan below but I had to ask myself what I would have done with the same building and a similar brief. From what I understand from the Architectural Record article, the brief was to get as many units onto the site as inexpensively as possible. The architects used the word cowboys to describe their property developer clients and the project is what it is, so whatever alternative I may come up with will be nothing more than an exploration of parameters of my choosing.

Nevertheless, I’ll try to get windows on both sides to as many apartments as I can, and I’ll try to get them as equal in floor area as I can. I’ll try not to have apartments whose only view is a light-well. I don’t mind apartments being entered via the kitchen but I’ll try to avoid generic solutions. How do people eat these days? Is the kitchen counter now an acceptable substitute for a table and chairs? In seven out of seventeen apartments it will have to be. I’ll give each apartment space for a table and chairs as well as a sofa.

For my North American readers, should I continue calling these dwellings apartments or should I call them something tenure-neutral like flats or units? Because from what I understand, in North America, calling something an apartment means it’s rented while calling it a condominium means it’s owned. I’ve always thought type and tenure were different so whenever I use the word apartment I’m really only talking about a building with multiple occupation and not about tenure. I’m only concerned with the environment that’s built and not how it’s owned, although I know they’re not independent. In Japan, a condominium is called a “manshion” while an apartment (“apaato”) refers to a timber building where dwellings have a wc but no bath. These are typically occupied by students.

This is my first attempt at downscaling my Sky Rectangle proposal to single apartments on one level. For the time being, I’m persisting with the staggered levels for two reasons. The first is that if the dwellings were structurally integrated and prefabricated, then they could simply be stacked a lá Habitat ‘67. It’d actually be better in that the roof of the apartments would create the walkway one level up and it wouldn’t be necessary to contrive a dedicated structure for access. (To be honest, this is a wish, and I don’t see it surviving for long.)

The second reason is that a rotationally symmetrical plan means the bathroom and kitchen soil pipes align, meaning only one per light-well. I’m relaxed about it passing in front of the kitchen window.

In these next three views, I’ve set the angle of vision at 75° so there’s more information but, in reality, the main room won’t seem as spacious. The window-street and window-window relationships are all good.

The rotationally symmetrical plan forced a layout logic in which the front and back of the apartments were the same but it also created a square living room in the middle,

Smallish square rooms aren’t optimal for the two activities of eating at a table and sitting on sofas doing some group activity. [If families no longer eat together around tables or sit on sofas and engage in some kind of group activity, then we have to rethink what we want from our dwellings. Otherwise, we’re really only dealing in representations of ways of living.]

The square room is the legacy of this project’s beginning with the 8.5 m x 8.5 m grid of a shopping mall car park and so the width of the living room and that of the walkways are both equal at 5.5 metres – a width that’s excessive for walkways and not the best for a 5.5 m long living/dining room either. Dimensions more like the Golden Mean or a length twice its width would be an improvement. But how wide? I don’t know, so I’ll make it 6m x 4m which ought to be more useable despite being one square metre smaller. Both bedroom and kitchen seemed cramped so I added an extra metre at each end. This is the improved version.

• Internal area increased by about 2 square metres but the voids are now 6 x 2.7 metres instead of 5 x 2, an increase of 62%.
• Most of the area increase went towards making the bedroom and kitchen more useable, but there is now additional storage in the entrance.
• The kitchen can be enclosed to accommodate the Asian preference.
• Alternatively, kitchen and entrance can be curtained off, if desired.
• Both the bathroom and kitchen exhaust to outside.

This is what happened next.

It didn’t work. Access was using too much of the limited area even without the staggered floors. The smaller end units I’d designed to “square-off” the building had about the same area as the target units so this was my next test.

The apartment sizes were okay (and there were eighteen!) but this would mean demolishing or gutting the the building and rebuilding. I couldn’t not take into account positions of light wells, the party wall and the elevators and fire stairs in the former building on the right. (The building on the left was only three storeys and never had an elevator.) This is what happened next.

• The basic idea was to shrink the smaller module I’d arrived at and to use the voids for access as they are underused for both illumination and ventilation. (The window openings into the left void appear to have been blocked off. Relocating access to inside the light wells means there is more sellable area left outside the light wells – a net space gain.
• Having habitable rooms open onto light wells isn’t ideal but there’s nothing bad wrong with using them to ensure a minimum amount of illumination and ventilation to kitchens, bathrooms, access corridors and rooms (including bedrooms) that would otherwise be artificially illuminated and ventilated.
• Hopefully the main light wells have driveways beneath them somewhere on the ground level to ensure airflow. Hopefully it won’t be where I’ve relocated the fire stair.
• I shifted the exterior fire escape to the end of the left light well. Its existing position compromises the width of 18 apartments. As is now, all apartments have a width of 4.2 meters more or less. Lengths vary, but only for the length for the living area and bedroom. Twelve out of 18 apartments have the same layout with only differences in length.
• The living/bedroom area is unpartitioned but an inner bedroom could still be ventilated by what is shown as a light well outside the kitchen and bathroom. These are also a legacy of the project’s beginnings but, since the ground floor is commercial space, they’re probably not going to happen. I suspect that floors are timber because, with the current arrangement there must be risers all across all typical floors. At what level these collect I don’t know. Probably the best I can do is to have bathroom, kitchen, bedroom and entrance opening onto an outside space fronting the two existing voids and take soil pipes down there. [This is what I did settle for, as shown below.] The bedroom window uses the closet space as a buffer between private and semi-private space.
• I notice, a bit belatedly, that the party walls in the as-built building, align with those free-standing columns. Whether this alignment was enforced or simply expedient I have no way of telling and nobody is telling. Fingers crossed.
• The fire escape in the building on the left looks original because the apartments in the frontage remaining. each have two windows. They won’t in my redistribution. This becomes a question of facade retention vs. profit. In the building on the right both top and bottom, I have six apartments where the structural bays suggest four (as was built). If my proposal were to be built, some of these windows would have to be divided or blocked which would be a shame since the importance of illumination is a priority everywhere else.

I was surprised the total area for access in my proposal is exactly the same at 154±0.5 m2. However, mine is naturally illuminated and ventilated, easy to navigate and also lets you know you’re not alone in the building. Access area diverted to light-wells represents a net increase in sellable area but I used some of it to enlarge and illuminate the elevator lobby.

• I know nothing of the pre-existing structure and construction. I haven’t followed the grid implied by the columns, whatever they’re made of, but I was lucky they didn’t intrude into the apartments. I maintained the minimum 1.2m shared corridor width (that produces that lone column below the elevators).
• Because of this, I wasn’t so lucky with the window spacing.
• As for fire escape, apartments 17 and 18 might be too far from the fire escape stairs. If so, adding another door to the stairwell is better than adding a third stairwell in the light well. Shifting access to the light-wells unified access and better utilized the light-wells for what they’re there to do.

• • • 

• It didn’t work out how I imagined it would. It’s not possible to know all the conditions and constraints the architects had to work with. I made assumptions about the construction but I also saw things in the as-built plan that contradicted them. Budget is another constraint but only the client and the architects know what the budget allowed and what it did not. There’s also the unknown of market expectations. I think the entry “courtyard” is a worthwhile idea but it only works if there’s a light-well or some other external access. Apartments 11, 12, 13 and 14 didn’t have this so I designed them as active band apartments with the outside is the necessary void. The architects have designed generic apartments so I suspect there is no appetite in Winnipeg for active band apartments.
• Only apartments 15 and 16 have separate bedrooms that make use of that difficult space between the core and the light-well. This is made possible by using the light well to access apartments 17 and 18.
  • In all, I’m pleased but not smug. Any of several known unknowns and an unknown number of unknown unknowns could instantly invalidate this proposal but it was a useful exercise all the same.

• • • 

Sky Rectangle