top of page

Things happened by reasons, not random. We use the word "random" to simplify those complicated systems, in order to easier understanding our world.

Designs are made for reasons. There are many factors that will influence the outcome of a design. Can we use the word "random"  to simplify the complicated relationship of those factors, in order to easier designing a product? 


The original documentation of all the draft materials

Oct.2019 - in progress
Thesis project by Yutong Shen

click to see each of the research's detail








At the very very beginning, I have no specific idea about what I am going to do for my thesis project. I only have very little knowledge, no theory, no methodology, no idea about the real outside world. The only thing I know is I love industrial design and I want to make something "useful".

So,where to start?

I looked through all the project I made before: the jellyfish floating humidifier,the belly frier, the ballon garden,the needle cutlery, 100 steps pictures... and my daily drawing,all these were just random silly thoughts,came into my mind, naturally,unconsciously,  with no mind map, no design methodology,even not much "researches". However, the outcome of those random ideas worked pretty well, at least in my mind. 

I enjoy my design process. Though it is random, works well: based on my experience, the brain randomly picks up elements for me. I just need to take a shower, get a silly idea, and then make it. 

Therefore, in theory,  if I have experienced a lot,  read a lot, learned a lot,I will have enough elements in my head to finally generate some good ideas.

Sounds good.  


So I started from here -- The random idea generator.




01_The random idea generator

In order to make this random idea generator. Each of the three keywords here requires deeper researches.


The first thing is to understand: What is random?

According to WIKI, if the thing is lacks of pattern or predictability, we call it random.

Today, people use the word "random" in many conditions, however, most of the things which we call it random are actually not random. For example, the coin toss- if we have the ability to calculate all the factors that will influence the coin, we can predict the result before it happens.


In our daily live, real random things are rare ( Ramsey theory )A lot of time we say random, is because of the lackness of our knowledge and technology, which impede us to see the pattern and sequence behind that thing. Sometimes, even if we have the ability to give a predictive result, it is so difficult that we still say it random to simplify the complicated causal, so that we can have an easier way to understand things. For this reason, many scientific fields, such as physical, mathematics, and biology, are concerned with randomness.

Take the mutation in biology as an example. The mutation is random. Mutations can be beneficial, neutral, or harmful for the organism, but mutations do not "try" to supply what the organism "needs." With natural selection, evolution happens.

It is surprised to look at how similar the patterns are between the evolution of a product and the evolution of a creature. One is designed on purpose,  another is just random. But they end up in the same result. I might say that maybe it is because our decision is not always right, therefore, even if we try to supply what people's needs, we end up with products that are beneficial, neutral, or harmful for people. 


We know that the world tends to work in a way that requires the least energy( ■ The principle of minimum energy ) which makes It a high-efficiency model  Article: Nature is an Efficient Model for Productive Societies ) that last sustainably

Back to the earlier idea of predicting the coin-toss. Think about it in the product design field. If we can know all the factors related to a described scenario, we can actually make the perfect design for that scenario.


Yes, people already doing that. That's why we need a lot of researches during the design process. It seems like, the more research we do, the better result we might have. 

But ... it that the best way to design? Since the scenario for each design project various because of the different problems, target groups or other factors, the process of learning is costly. Doing perfect research for each project is not possible. At least, not very efficiently. 

Ideally, the more efficient way to design is to let those people who already in that scenario design things for themselves. The elder people design products for elders, disabled people design products for the disabled. Children design products for their peers.  Wait... that's weird. The flaw here is obviously - not every user group has the ability to design. That's why we sitting in this bright large studio, still worrying about how would the newborns survive in Africa (  Project: NeoNurture )

Thinking about my own design experience. I always find that doing traditional user research and understanding other people is suffering. My design relies on random ideas that pump up in my head. I only choose the idea which I truly love and have a passion for it. My theory here is, in most of the parts, I am a very normal person. So, even if I  only do the thing I personally love,  it has a high possibility ( almost 100%, definitely ) that it is also loved by someone else.

Though there is no research part in my design process, it does not mean that my design has no research. The random ideas that come into my mind are generated by my previous experience and knowledge. Therefore the elements in my head make up a research pool of myself. 


If we take all individuals on our planet as an entity. As this entity's research pool is combined by adding up everyone's experience and knowledge, we can say it includes every perfect research of every scenario. If we ask it to design a product, even without doing researches,  just start with generating random ideas,  it can finally come up with the product which suits that scenario perfectly.

That's exactly how evolution happens in biology - by mutation and natural selection.


For a quick test, I used a dictionary as an input ware to generate two randomly words. (■ Random idea generator ). It works, though needs some imagination. It does give me some inspirations, by just simply running this quick and rough test. For example, the "low vessel" that inspired me to design a vessel that aims to put under, or inside the table.

The problem here is if the outcome is too random, it is hard for us to really take use of it.

The History of Random

In ancient history, the concepts of chance and randomness were intertwined with that of fate. 

The Chinese were perhaps the earliest people to formalize odds and chance 3,000 years ago. The Greek philosophers discussed randomness at length, but only in non-quantitative forms. It was only in the sixteenth century that Italian mathematicians began to formalize the odds associated with various games of chance. The invention of modern calculus had a positive impact on the formal study of randomness. In the 19th century, the concept of entropy was introduced in physics.

Random in biology 

Evolution of a product / Evolution of a creature

屏幕快照 2019-10-06 下午9.34.37.png
→ Detail
未命名作品 3 (1).jpg

Possibility of evolution 

→ Detail

​Simplify the evolution pattern

pasted image 0.png

Speach: Where good ideas come from-Steven Johnson

Project: NeoNurture-Design That Matters


In developing countries, not only is there limited access to modern, high-tech incubators, but a lack of infrastructure and replacement parts render such devices worthless.

NeoNurture takes advantage of an abundant local resource in developing countries: car parts. The incubator leverages the existing supply chain of the auto industry and the technical understanding of local car mechanics. 

What if - Designs are random

Screen Shot 2019-12-11 at 7.21.39 PM.png
Screen Shot 2019-12-11 at 7.21.50 PM.png

Random idea generator - quick test

Rule:  Use to generate a random number( 1-400) as the page, and another random number(1-40) as the order to locate a random word.  Find 2 random words and combine them as one concept.

Testing Result:

1. palatable cart

2. rocker oral

3. urgent engrave blood

4. heat defog sextant

5. immature firebox

6. awe canvas

7. folder gosling

8. dig lousy modern

9. graphic jollity

10. low vessel

02_The dictionary of every product 

In the novel  The library of babel. The idea of just randomly combing 23 letters to generate a library of infinite books, which contains all the possibilities in the world inspired me a lot. 

If I change the 23 letters into all English words, then I could narrow down the library of babel to a smaller library that contains all randomly readable English books.

If I go further, just put in all the elements we usually use to describe a product,  we can have the library of definitions of every product. Both for the products that are already existing or the product that hasn't been designed or will never be designed in our world.

Then I start programming my  dictionary of every product.

One big challenge is how to form a random but readable product definition.  

I first use the Tracery Javascript to form some simple gramma. But the problem here is the gramma in some way might influence the randomness.

Also, which word needs to be put in this system is another big question.

The library of babel(novel)

The library of babel (website)

"The Library of Babel" is a short story by Argentine author and librarian Jorge Luis Borges (1899–1986), conceiving of a universe in the form of a vast library containing all possible 410-page books of a certain format and character set.

The dictionary of every product.

Screen Shot 2019-12-11 at 1.11.01 AM.png

The idea of the Dictionary of Every Product is that by using the tracery Javascript to form every product's definition. Ideally, if I can put all the factors into this system, it can generate all the products' descriptions.  

For a quick test, I used grabbed the name of ideas from Kickstarter for 1000 pages. I sorted them in excel and came up in 5465 the most used words in product design definition on Kickstarter. Then I input them into the Html. 

→ Detail


use  Tracery javascript to form the description sentence

Grab information to excel from a website: such as the Amazon, Kickstarter.

Github reference:

a related project by nicolehe/the-best-art/

Family trees of product


By classifying the words into different family trees on a transparency card( eg. furniture, office supplies, electronics). You can put different categories together, rotate the card to generate random ideas.

→ Detail

Another thing that draws my attention when I am working on the dictionary of every product is that if we do have this dictionary, which means every possibility of the product is already exists, it only needs to be lookup. 

But, how to do that?

How to search the thing you even don't know?

This reminds me of the same disappointing feeling when I using google and already searched for many keywords but

still can't find what I need. 

We already emerge in a world that has too much information. 

Search engine & Big data

Screen Shot 2019-12-11 at 3.12.04 AM.png
→ Detail
03_3d Galton board 

Besides, researching how randomness happens in biology, and how it connects to the design world. I  also started to think about randomness in the mathematics world. I tried to find the connection in design, mathematics, biology, maybe some universe rules in randomness.

→ More

Galton Board is a device invented by Sir Francis Galton[1]:63f to demonstrate the central limit theorem, in particular, that with sufficient sample size the binomial distribution approximates a normal distribution. Among its applications, it afforded insight into regression to the mean or "regression to mediocrity".

The 3D Galton Board

→ Detail
04_cylender world 

A series of experiments about random design and the meaning of random.​

未命名 [转换]-01.jpg
→ Detail

Dasher-The alternative way of typing 

→ Detail
→ Detail
→ Detail
→ Detail
bottom of page