Hello, dear readers! 👋
In this issue, among other things:
- 10 tips to help launch a successful product
- What skills are necessary for a digital designer
- Reviewing of design trends
- Why good design takes time
- "No AI" movement
- Common misconceptions of the world's map
- Online archive of vintage interior magazines
- Stable Diffusion shares some inspirations in a gallery
- Converting Figma designs into adaptive HTML/CSS or Flutter layouts
- Quotes from "Ruined by design" book by Mike Monteiro
📚 Book quotes
Found some interesting quotes from Mike Monteiro's book "Ruined by design". Read them and decide whether to read it in its entirety:
If you believe you need to do what your boss wants because they’re paying you, you also need to believe the doctor should provide the oxycodone if the addict is willing to pay for it. The exchange of cash for services doesn’t supersede ethics. Following unethical orders won’t keep you out of jail.
Ethics cannot be a side hustle.
It should freak you out that gangsters can agree on a code of behavior but designers can’t. Crime is more organized than design.
Empathy is a pretty word for exclusion. I’ve seen all-male all-white teams taking “empathy workshops” to see how women think. If you want to know how women would use something you’re designing, get a woman on your design team. They’re not extinct. We don’t need to study them. We can hire them!
The world isn’t broken. It’s working exactly as it was designed to work. And we’re the ones who designed it. Which means we f*cked up.
I intend to show you that design is a political act. What we choose to design and more importantly, what we choose not to design and, even more importantly, who we exclude from the design process—these are all political acts.
I write a bunch of form labels on the whiteboard, in a nonsensical order, along with a bunch of randomly sized input boxes. I include first name, last name, address, gender, city, state, email address, etc. Then I tell the interviewee that we’re designing a form to sign up for an email newsletter and to arrange them in the right order. Only people who ask me why I need the users’ gender, or physical address, or really, anything but their email address get a second interview. I won’t hire a designer who doesn’t ask why, and I won’t hire a designer whose desire to arrange boxes is more important than their desire to protect users’ data.
For the record, I have nothing against art. I love art. I’m an artist myself. I’m also a designer. I understand the difference between the two. Design is the solution to a problem, but that problem is never your self-esteem.
People don’t see the things they’re rewarded for as problems to fix.
(As designers) We need to advocated for the people who aren't in the room, and stand up to those who are. That's the job. We must be engaged in the process of what gets designed way before it enters the phase where traditionally and erroneously thought of as design.
The vast majority of design programs across the world still live within art schools. Not to shit on art schools—they’re a fine place to learn how to make art; but art has as much in common with design as a lobster has with a carrot cake.
🗞 News and articles
Tyler Angert described a set of basic skills that are necessary for a modern digital designer and entrepreneur, and also advised specific courses on topics. It describes areas such as graphic and information design, programming and thinking.
Important fundamental skills:
- Ability to work with primitives and abstractions. This will help to break complex systems into fundamental components and understand their structure, as well as facilitate the creation of new ones. For example, Lego primitives are plastic blocks, Photoshop layers, and music notes
- A general understanding of such basic concepts as information and visual hierarchy, color psychology, accessibility standards, typography, as well as data visualization
- , the ability to speak and describe something in an understandable language. This will help in negotiations, documentation and even design. It is also important to be able to pick up understandable analogies and examples to explain complex things in simple language
Martin Spinnangr described 10 tips that will help launch a successful product, and also shared links to useful books and articles on the topic.
Tips from the article:
- Often the product starts with the hypothesis of marketers, sales department and other specialists, but they may be wrong. Before you start doing something, talk to your potential users
- Communication with future users is an important stage, but people do not always know and understand what they really need. Use alternative means to test your hypothesis
- There is uncertainty when launching any product, so don't waste time making rigid plans. Use a flexible approach and make changes by getting new information from customers
- Be prepared for mistakes and remember that a bad experience is also valuable. Take it into account and adapt, instead of continuing to sell a non-working idea
- After the launch, you may realize that your theory is not working as you planned. Create new iterations and improve the product until it starts to make a profit and solve user tasks
- Think not only about how to create the perfect product, but also about how to sell it. No one will buy anything without a good business model
- Prioritize your hypotheses and check the riskiest ones first
- The more accurate the test, the more difficult and expensive it is to carry out. Start with simple primary tests and move on to more thorough ones only when you realize that your hypothesis is confirmed
- At the early stages, it is impossible to build long-term profitability forecasts, but it is important to conduct an initial assessment of the market. This is necessary so that you do not spend resources on developing a good, but not profitable product.
- Demonstrate your progress with the help of the test results obtained, moving from simple tests to more complex ones. This will allow the team to understand that you are moving in the right direction
Design Takes Time. The creators of the iA Writer and iA Presenter applications talked about the three-year process of creating icons for their products and explained why good design takes time.
A great overview of design trends in packaging and branding from the publishing house Dieline. The authors write about the positioning of companies and their focus on ecology, packaging forms and materials, as well as fonts, NFT and much more.
Last weeks, a discussion about ethics and generative arts flared up with renewed vigor. It resulted in a massive protest by ArtStation users.
Some time ago, several AI enthusiasts trained neural networks on the works of specific artists, which allowed them to completely copy the style when generating AI art. In addition, the site has increasingly begun to publish works created with the use of neurons. This caused a wave of indignation among artists, the main site is full of pictures with the call "NoAI", and tweets with such a tag collect thousands of likes and huge reach. Artists demand both not to let neuroart on the site, and to stop using their work to train neural networks. Many claim that AI destroys artists.
As a reaction, ArtStation released an update that allows artists to tag their art with the NoAI tag. Such a tag automatically adds an HTML meta tag that prohibits artificial intelligence systems from using art in training. Whether ArtStation will ban the publication of neuroart is still unclear.
Note that this discussion is fundamental and will be long. Because it raises questions to which there are no ready answers. Who is the author? Is it possible to have rights to the style? Where is the boundary between the tool and the author? Is a person considered the author of an image if the light, color, effects, proportions and perspective were calculated by a machine? Is the author valuable without his own tool? Where is the line between plagiarism and inspiration? Where is the copyright boundary? Can the author forbid looking at his creation? Is there a fundamental difference between the training of neural networks on silicon and the development of visibility, that is, in fact, the training of protein neural networks?
This discussion will be long, and no one knows where it will turn and what laws it will lead to. But it is clear that the era of exointelligence is coming. Generative neural networks will continue to improve and will be integrated into all creative tools. To text, video, graphics, code and music editors. And those who learn how to masterfully apply them will gain a huge competitive advantage over supporters of "manual" labor without the use of exointelligence.
Finally, we recall two interesting materials on this topic from past issues:
- Article on the topic of copyright in AI training from past issues
- Article about generative networks and the ethics of copying style
AI Homework. A large article, the author of which reflects on how students will be able to do their homework with the help of AI, as well as what opportunities modern neural networks hide. For example, he was able to "launch" a virtual machine on Linux directly in a dialog with ChatGPT.
The author also explains the difference between an "honest" calculation on a computer and a probabilistic calculation of a neural network, which relies on hints, context, as well as information published on the Internet.
What is GPT-4? Mandar Karhade is thinking about when the new GPT-4 model may be introduced and how it will differ from the previous generation.
The identity of Swedish Tugg fast food restaurants is modern and very atypical for this niche. The style is built on defiant custom typography, 3D animations and overloaded cyberpunk layout.