Digital local plans take two

We’ve written before about our initial work on local plan standards, and how it began a series of increasingly basic conversations that left us a bit baffled. Fortunately we have a long tradition in PAS of having lots of extremely clever people as friends and colleagues and picking their brains to make ourselves look more competent. We held our first session last week- there was lots of excitement and creativity. What follows are my initial thoughts – so you can be the judge of whether our tradition is working. 

It’s not the book it’s the library
It’s tempting to think of a local plan as a book – so a digital local plan is just an online version of that book. It’s true that ultimately the local plan process ends up producing something that looks like a book, but it is also at different times and for different audiences a whole set of proposals, data sets and conclusions drawn from the data. I live in the wonderful borough of Lambeth – you can see their local plan evidence base neatly marshalled in a library. The library is used to inform the debate and choices through consultation and ultimately examination at which point its importance fades. 
For me, there is limited benefit in doing a make-over of a local plan. Sure it would be interesting to see a “before” and “after”, and improving engagement and accessibility of the finished product is definitely important but if our goal is engaging, responsive and more open plans it’s got to involve the underlying data – that’s why it’s got to be the library and not just the book. 

A “rules based” system of assessing sustainable development involves everyone
To go back to basics for a moment the job of a local plan is to answer the question “is this application sustainable development?”. The goal of the current reforms to the planning system is to provide greater certainty to the answer of this question by shifting the balance away from discretion (planning judgement) and towards rules. 
Clearly the process of applying rules is going to work better for some concepts than others. Traditionally it’s where “building control” sits allowing “planning” to be a bit more human. 
But for a system of rules to work they need to be relatable to everyone. The rules have to be debated at local plan stage by local people, they are codified in the plan, they are acknowledged by architects and applied in a context in applications. It’s an end-to-end thing that involves everyone speaking the same language. This is a big ask, and our starting position is that we don’t have a national way of measuring floorspace or counting bedrooms – let alone rules about [cough] “beauty”. 

Eating the elephant
Where should we start? This is going to be the most important question – and is far more difficult question in this context than for any other project I’ve ever thought about. For example, I have begun many “transformation” projects on how planning applications are dealt with. Almost all planning applications take a few days to complete – you can just begin trying new things and the feedback loop tells you how its going in a week or two. The cycle time for an application is a few days. By contrast, a local plan cycle time is measured in years, and worse the cost of a plan is measured in millions. At some point we are going to roll up our sleeves and try new things but what we can’t do is to try a whole plan in one go and wait to see what happens. 

The startings of a manifesto?
So – a great conversation started to unpack some of the issues, hopefully to make me sound a bit cleverer. And, it’s clear that this is an interesting problem with long-term benefits around openness, cost reduction and understanding / managing development better. But what next? We could keep having sparky conversations and there are lots of interesting projects already underway out there – but perhaps we need a manifesto or some other way of organising the next year or two. 
Alongside the obvious stuff like “find out what the rest of the world is up to” and “ensure we understand the capabilities of what we already have” it feels like the next phase is going to need to deal with three topics:

1. Establish the design patterns for local plans. One of the most important things the Government’s Digital Service did in its very early days is to establish what were known as design patterns. These are the rules for how to deliver core elements that crop in many different contexts (eg how to ask for a credit card). We need to work out the common elements that inform a local plan, so we can see the opportunities for a standard approach and platform. I’m imagining patterns might include things like “ask people what they think about a rule expressed as a number” and “make site information findable” and even “allow people to suggest changes to a shape on a map”. 
2. Make an example “rule based system”.  I can remember a long time ago (12 years!) I tried to make an XML “expert system” to describe permitted development. It was hard. It neatly demonstrated that until the regulations are written in a rule-like way the judgement has to be deferred to a human operating the rules (“is this the principal elevation?”). We need to out a rule-based system end-to-end. How about adverts? Neat, specific and lots of experience and competence out there. New regs, new rules, and a language that connects them to a site. 
3. Talk numbers. All this stuff is exciting and there is already some great work underway, but at some point we are going to need to understand the costs and benefits. Getting datasets and policies to a defined standards (and quality!) is an enormous undertaking for LPAs and we should be clear about how much this transformation is going to cost. And, to re-use my “library” metaphor, is it all data equally valuable and are there any opportunities to reduce cost by maintaining some of it nationally?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s