3 ms·
I'm increasingly leaning towards algorithmic zoning based on pre-existing use. For example, allow adding x% of floors above the average of the surrounding buil
by vidarh 4y ago
I'm increasingly leaning towards algorithmic zoning based on pre-existing use.
For example, allow adding x% of floors above the average of the surrounding buildings, rounded down, might be a rule to consider. Similarly apply classifications and allow one step "further" in the classification system than the average of your neighbours. Allow more if all neighbours consent. Allow more if e.g. adjacent to transport hubs.
So e.g. if your neighbours are all single floor bungalows, you might be automatically allowed to put up a two floor building, and you might be automatically allowed to change use from purely residential to, say, run a shop on the ground floor without late opening (just to take an example of a minor low impact step up from purely residential), but you'd need to apply for an exception if you want to build higher or do something higher impact. If there are automatic permitted developments with reasonable limits, the threshold for exceptions can be fairly high.
Most of the time a model like that would allow a neighbourhood to change gradually and dynamically as people move in who want it to change, while not so fast that it ought to drive out people who like it roughly the way it is.
Neighbours on both sides adding a floor more than you presently have? Suddenly you might have an automatic right to add 2 if you like, but your neighbours might not be allowed to match you unless neighbours on other sides also gets with the program. So done carefully, changes are contained in pockets, and as a bonus you get neighbourhoods that are not all cookie cutter but can still retain a distinct character. Density would naturally "radiate" out from existing high density areas and transport as people take advantage of the more lenient limits of high averages of surrounding buildings, and you "automatically" get gradually declining heights rather than huge abrupt steps (e.g. near me there's been recent planning conflicts over a number of new 20-30 floor buildings almost directly adjacent to two-floor terraced single family homes. But the building pattern is like that in the first place only because planning for decades resisted gradual increases).
Getting the patterns right to e.g. protect pockets for parks and essential services would take a lot of work, but you'd also have the benefit of being able to program visualisations to show immediate impacts on rights for neighbours etc. of allowing exceptions or when tweaking rules. And it'd allow streamlining approvals if more of the planning process is fully predictable.