Post No.: 0459
A fantastic way to discover things or create something new is to experiment. We can conduct an experiment or two to check to see if what we or other people claim is true, or make some prototypes and possibly invent something that no one has ever seen before.
Don’t just unquestionably follow myths, old wives’ tales, traditions or someone else. Just because something has been done the way it has been done for many years, or even generations, it doesn’t mean that there might not be a better way? We do absolutely learn a lot from others but occasionally the blind could’ve been leading the blind, something may be severely outdated for today, or assumptions or misinterpretations might’ve been made along the way. For instance, if you copied your mother (who copied her mother) by chopping off the ends of the legs of fowl before putting them in the oven, you might assume that this was done to improve the flavour or something like that; when the only reason your grandmother did that was because her oven dish wasn’t long enough for a full-length chicken(!)
There are a few, more commonly believed, cooking myths that have been relatively recently debunked via scientific experiments, such as needing to constantly stir risottos from the start (in a non-stick pan at least), needing a large amount of boiling water when cooking pasta, storing eggs in the fridge (if the chickens have been vaccinated against salmonella), and plastic chopping boards being more hygienic than wooden ones. So if we question things or experiment by trying something different, we might learn something that’s easier or better. (Granted, the last two could not have been safely experimented without a laboratory though!)
‘Mondegreens’ can result when people mishear and therefore incorrectly copy something, such as the lines of songs or poems. Copying is an effective and efficient way to learn but in cases of folklore or cultural cross-pollination, such copying errors can be passed onto the next generations, like in a real-life game of ‘Chinese whispers’/telephone. Traditions are important parts of cultural identities, but when some start to appear outdated then we shouldn’t be afraid to be experimental by trying something new – perhaps create a new tradition? After all, even those old traditions didn’t exist at one time. Wedding traditions like taking the husband’s name, asking for the father’s permission, carrying the bride over the threshold, tossing a bouquet or garter, expecting the bride’s parents to pay, and more, certainly need to be questioned nowadays. Woof.
When it comes to experimenting in a scientific context – the difficulty of measuring some things doesn’t stop one from holding certain beliefs, only from testing them – but I think that everything that allegedly has an effect can ultimately be measured and therefore tested. It’s just a matter of how we choose to do it (our operational definitions and methodologies). So if someone, say, believes that wearing a particular pair of socks brings them good luck (which is an effect that can be measured) then we can conduct an experiment to test if this is true by comparing them wearing those socks one day with them not wearing those socks another day. Experiments can be conducted to answer such questions of belief.
How we operationalise and measure constructs to satisfy validity and reliability is critical. In the social sciences in particular, ‘intelligence’ as a construct is impossible to objectively or definitively operationalise so how shall we measure that, for instance? One way is via a standard IQ test, but another could be a creativity task, a general knowledge quiz, or solving a real-world problem? Even if we break intelligence down to mean someone’s ‘problem solving ability’, this can be broken down further (e.g. social relationship problems, financial problems, abstract spatial problems or word problems), which in turn could possibly be broken down even further again (e.g. ciphers, crosswords, word searches, rebus puzzles or anagrams). You can check out Post No.: 0225 to read more about the limitations of IQ tests.
Relying on proxies, such as the frequency of search engine queries for ‘flu-related terms’ when we really want to know how many people in a country actually has flu, doesn’t always lead to reliable correlations; with many false positives (e.g. people who are well just searching for information on a flu after reading a particular media story) and false negatives (people who are unwell who don’t search online for information about flu at all).
When designing an experiment, you naturally want to set up the experiment so that everything between the treatment and the control is equal except for the independent variable that’s of interest. But be careful to check if this controlling of external variables is valid and fair. For example, if you want to find out if playing games with a keyboard and mouse is better than with a gamepad – here, using the same game to test them both seems fair, but this game might specifically suit one method over the other?
Experiment, prototype, test, critique, improve and iterate. And get external feedback early…
When it comes to experimenting in an inventive context – start with rough prototypes first and never expect your first attempts or versions to be the final attempts or versions. Accept failure frequently – fail faster in order to learn and improve faster. Don’t just start once you think you’ve found the ‘perfect’ idea. Don’t just wait until it’s ‘ready’ before you start testing. The later you fail, the more costly it’ll be to change something and the less likely you’ll want to change things too.
Some argue that instead of trying to plan our lives so far ahead, we should prototype through life by doing small experiments (e.g. trying a new hobby) to see where that leads us. Trying to predict things over a year ahead is often futile, and this includes trying to predict what we’ll still want ourselves in the future!
Don’t be precious with any idea. Consider making big changes to help you understand what’s going on as you experiment? Dump what’s not working so that you can start something else that might work. This might mean that you’ll need to measure and record things over time to know what’s working and what’s not (metrics). Don’t upfront over-engineer what isn’t necessary for simply testing your idea at first because this work may be scrapped if the idea doesn’t work i.e. optimise and polish afterwards. And make it work before you make it pretty.
Take things out or simplify things rather than keep adding things (feature creep). Don’t spend too much time on the things that no one will ever notice either – spend most of your time on the parts that’ll make the most difference. This requires knowing what your goals or priorities are. Fix the biggest problems first. And complete what you start! Only finish once there’s nothing more to be learned from something.
You cannot ever impartially assess your own creations with fresh eyes so seek independent external opinions. Preferably not from friends though because they can either be too kind or too protective of you taking risks. It also doesn’t mean that you should always do as market research says (e.g. those who were surveyed may say that they want ‘a blue background’ but what they were really trying to say was that the words aren’t clear). And be aware that even independent testers can be too nice by trying to not hurt your fluffy feelings, so just let them self-narrate on their experience with your creation without them thinking about you in the room.
And you cannot always (if ever) obtain perfection so stick to a deadline because you’ve got to get your creation out there. There’ll be a point when trying to implement another change will potentially introduce a new defect/bug or a step backwards. Unless dangerous or critical, you can always evaluate and learn what went well and what you’d do differently the next time.
Everyone, no matter how old they are, should consider themselves like a child (or like a puppy!) because there is still so much for everybody to learn. If you’ve lost the love of playing then you’ve lost the love of learning. And a key part of playing is experimentation. The smartest animals in the animal kingdom love to play, be curious and experiment. Curiosity is expressed via experimentation. Continue playing and learning to slow down age-related decline. Life expectancies have generally increased over time thus we must extend the amount of time we spend being curious about new things – even better is to remain a learner for life!
The instincts we’re born with are wonderful, effective and efficient most of the time but are sometimes fallible, and genetic evolution is typically slow. But we can adapt by personally learning about things to somewhat overcome those shortcomings. And probably the principal reason why humans are so marvellous at adapting is because humans love to experiment! Creativity and science are both related because they’re both about experimentation and evaluation.
Woof! Nothing ventured, nothing gained. For me, this blog incorporates a few experimental things itself.