WEBVTT

00:00:00.000 --> 00:00:30.015
One thing in relation to this that I want to talk about is the difference between
*puzzles* and *problems*.
I think if you've been practicing it for a while doing real design,
you'll understand this difference without thinking about it.
If you're perhaps newer,
perhaps you've done a university course and you're newer at that,
it might be less obvious.
Let's start with puzzles.

00:00:30.015 --> 00:01:01.709
The thing about puzzles is they have a *single right solution*.
The same is true of exam questions. Exam questions are puzzles.
They might say, "This is a problem," but they're puzzles.
There's one right solution, and in fact I get annoyed if I do the newspaper puzzle
and I realize there was a couple of right solutions. I expect there to be *one*.
But certainly, there is a right solution.
And what you're presented with in a puzzle is you have
only and all of the relevant information.
Indeed, one of the ways if you're doing an exam question

00:01:01.709 --> 00:01:33.507
to actually work out "Am I on the right track?" — you can say:
"Have I used all the information?"
You get the same with a whodunit book. You know, you open the book, the detective book.
When there's something in there, you think:
"The author's put that there for a reason. That's probably part—
It might be there to mislead you. But it's either *deliberately misleading*
or it's actually crucial.
The other thing about problems and puzzles is that
the solution is *fixed* – as in, you can't change the problem.

00:01:33.507 --> 00:02:00.526
So – I mean, you *can*: I can pick up my newspaper
and I can just color in the squares in the sudoku rather than solving them.
But that's not the idea of it.
And if you do that in an exam, you'd probably get no marks.
A few philosophy — you know, there are these stories about the philosophy questions, like:
The question says, "Is this a good question?"
and the student says, "If this is a good answer!"
I don't know if anybody really got top marks in philosophy for doing that sort of question!

00:02:00.526 --> 00:02:31.101
But – on the whole – if you've got a puzzle, you have to do what the puzzle says.
*Real-world problems* are not like that.
Real-world problems may have many, many solutions,
or they may be actually *insoluble* in the way you first approach them.
You may not have all the information you need.
And you've probably got lots of information available that's totally superfluous, that doesn't help you.
So, often part of problem-solving is *finding that information*.

00:02:31.101 --> 00:03:00.146
But actually, the problem might be insoluble; so,
partly, problem-solving is about *negotiating*
and working out what *is* doable and perhaps negotiating
and a reformulation of the problem
that is one that's both doable and solves the problem.
So, it's about *redefining* it
and it's about *understanding* it again.
And the two kinds of things – puzzles and problems – have different ways of addressing them.

00:03:00.146 --> 00:03:32.848
So, things you've learned about puzzle-solving don't always help you with problem-solving.
So, I mentioned a classic heuristic – an exam – is "Have I used all the information?"
Of course, that doesn't help in actual life.
Part of the reason I was talking about that is that
this redefinition of problems, of *understanding* a problem
is often the first stage to find a solution.
Indeed, I often find – and you've probably found this yourself –
if you really, really understand a problem well,

00:03:32.848 --> 00:04:00.387
often solutions are obvious.
I mean, sometimes they are really hard!
But often, once you've really got to the nub of "What's really going on with that? *Why* is that difficult?",
"Aha!" — and you get to solve it.
So, often the solution is trivial if you understand the problem.
But how do you get to understand the problem?
You might have a very abstract way of approaching it,
but often the way we understand things is through concrete instances.

00:04:00.387 --> 00:04:32.878
You actually want solutions – you want things that are solid and down-to-the-ground
in order to look at them, make sense of them and then understand the problem better.
But how do you *get* those concrete solutions
if you don't understand the problem?
So, actually understanding the problem is as important as solving the problem.
There's this quote – you might have come across some variant of this –
and I'm just going to give one variant that says:
"If I had an hour to solve a problem

00:04:32.878 --> 00:05:01.626
I'd spend 55 minutes thinking about the problem
and 5 minutes thinking about solutions."
Now, this is often attributed to Albert Einstein.
And it's actually the sort of thing you could imagine him saying.
But, actually, the evidence is that he never said it at all.
It's just one of those sayings that
seems reasonable, and it gets attributed to other people as well.
However, even though he didn't say it, it is true
and I'm sure it's something that he would have espoused as well.

00:05:01.626 --> 00:05:32.768
Spending time really understanding the problem
often is the first stage to solving it.
Once you really understand the problem, sometimes – and very often –
solutions find their way out.
Some years ago, I did a walk around the perimeter of Wales
– about a 1000-mile walk. And
there were some things that I was interested in from a research point of view:
How to deal with issues when you have low connectivity in terms of mobile signal and things.

00:05:32.768 --> 00:06:00.069
But part of it was about *learning questions*.
And the way I often phrased it to people before I did it was:
You know, if I've just got a problem, I can probably find a solution.
But actually knowing *what are the important questions* were part of what I was doing.
So, for instance, I came away with
questions about the nature of community when industry died in areas.
And you could observe — in some areas, the community seemed to collapse;
in others, it seemed to be resilient – and trying to *then*

00:06:00.069 --> 00:06:33.799
answer that question or seek more understanding of what it was
with the factors that made that difference.
Wicked problems have properties.
It's a whole list of properties; I've got some of them here.
One of them is about *being unique*
– that it doesn't mean you can't learn from previous ones,
but you can't just take the solution you had for a previous problem and apply it to new ones.
For wicked problems, each one you have to look at *individually*.
There's *no definitive formulation* – that is actually
even stating what the problem is is problematic.

00:06:33.799 --> 00:07:01.028
*Non-enumerable* – you can't just go through
a set of things and say, "Which one of these is better?"
The space of potential ways you might tackle it is unbounded.
So, you can't just sort of try the first idea,
try the second idea. You've got to go beyond that.
They talk about *one-shot operation* – and this is about the fact that
often you have to start executing the solution
before you know what it all is.
So, imagine you're in a desert and you've only got a certain amount of water

00:07:01.028 --> 00:07:33.245
and you don't know which way to go,
and you spot a very high dune.
What you might do is walk to the dune in order to spy the land.
But in walking to the dune – if it was a long way away –
you of course have used up some of your water; you've got hot and you've used up some of your food;
you've *committed* yourself. And real problems are often like that.
*No stopping rule* – You know: Have you finished?
And that's related to the fact that often in the real world
you don't just say, like with a candle – and you get to this and you say,

00:07:33.245 --> 00:08:04.426
"Yes, I've done it. I've got a solution that works – done!"
In the real world, there isn't a fixed solution – or the dots – I say:
"Yes, I've got a solution, with four lines that go to a dot."
You can't just say, "I've done it!", because – actually – you tend to have
things that are better or worse solutions, rather than the perfect solution,
the right solution, versus another solution.
And so, you're probably not trying to solve the economy.
But these kinds of characteristics you actually will find in a lot of real design problems.

00:08:04.426 --> 00:08:13.584
So, they are difficult – and that's why you need to think creatively.
But you're not unique in having to solve some of these issues.