250 Tips from incompetent people

9 problems when working with neural networks like ChatGPT, Example of creating a unique icon using Midjourney, Blender and Figma, Powerful free storage tool, Why making AI safer could backfire.

Dec 22, 2023 - 08:58
Jan 26, 2024 - 21:35
 0  325
250 Tips from incompetent people

250 tips from incompetent people Hello, dear readers!

In this issue, among other things:

  • Quotes from Oliver Berkman's book Four Thousand Weeks: Time Management for Mortals
  • What is disrupt and how to make time for it
  • 9 problems when working with neural networks like ChatGPT
  • How to build a design process when designing complex products
  • Main principles of UX-design of children's digital products
  • How to conduct user tests of AI products
  • Example of creating a unique icon using Midjourney, Blender and Figma
  • Creating complex effects with Processing and After Effects
  • Powerful free storage tool
  • Why making AI safer could backfire
  • Quotes from the book
  • Articles and news
  • A short line
  • New technologies
  • Identity
  • Typography, calligraphy and lettering
  • Posters
  • Printing
  • Infographics and data visualization
  • Packaging
  • Websites
  • 3D animation
  • 3D graphics
  • Neuroart
  • Illustrations
  • History of Design
  • Portfolio of the Week
  • Lessons and lectures from VK
  • Education
  • Figma plugins
  • Tools
  • Fonts
  • Best of VKontakte
  • Dessert
  • Humor

Quotes from the book

Natasha: The end of the work week and the vacation season is the perfect time for Oliver Berkman's book Four Thousand Weeks: Time Management for Mortals. If you don't have time to read the whole book, here are a few quotes that will hopefully help you make your relationship with work and vacation healthier.

Ruth Schwartz Cowan, in More Work for Mother, says that when labor-saving appliances like washing machines and vacuum cleaners became available to housewives, it did nothing to help them save time, because accepted standards of cleanliness rose to the point where they undermined the engineers' efforts. Now that it was possible to get your husband's shirts in immaculate condition after every wear, you began to feel like you had to do it to show him how much you loved him. "Work fills the time given to it," wrote British satirist and historian Cyril Northcote Parkinson in 1955, thus formulating Parkinson's famous first law. But this is not just a joke, and this law applies not only to work, but to everything that needs to be done. In fact, it is the idea of what needs to be done that expands to fill all the time we have. ...even if you fill your life to the brim with enjoyable activities, it often doesn't bring the satisfaction we expect. It is simply an attempt to eat up as much of the experience as possible, to swallow as much of what the outside world has to offer, to feel that you have lived life to the fullest. But the experiences offered by the outside world are essentially limitless. So no matter how many experiences you take in, it will not bring you closer to the feeling that you have enjoyed all the possibilities of life. On the contrary, you will fall back into the trap of efficiency. The more you accumulate wonderful experiences, the stronger the feeling that you could - no, should! - to have more and more, which only increases the feeling of existential overload.
Gradually, I began to realize that in such situations, what is needed is, on the contrary, a kind of anti-skill: not a counterproductive strategy to become more efficient, but rather a willingness to resist such urges - to get used to living with the anxiety of being overwhelmed, of not having enough time to do everything and everywhere, without automatically reacting by trying to fit more things in. This approach means not getting things in order, but giving up and focusing on what's important, despite the uncomfortable feeling of having messages, errands, and other to-dos piling up on your list that you may never get to. You will have to work hard to get more things done, but only when you can't do otherwise. But this will not be your default mode of functioning, because you will no longer operate under the illusion that you will ever make time for everything.
Often, convenience doesn't just make an activity seem less valuable: it makes us abandon valuable things altogether in favor of convenient things. If there's an opportunity to stay home, order food, and watch TV shows, you take it - even though you may realize you'd rather spend your time meeting friends in the city or cooking a new recipe. "I love making coffee," writes law professor Tim Wu in an article about the pitfalls of convenience culture, "but instant coffee from Starbucks is such a convenience thing that I hardly do what I used to enjoy doing anymore. That said, the aspects of life that don't want to be smoothed over are starting to get repulsive. "When you can avoid standing in line and buy concert tickets on your phone, the line at the polling station is annoying," Wu notes. As convenience begins to fuel our daily lives, all our activities gradually fall into two types: those that are convenient but empty and don't match our real preferences, and those whose inconvenience is now terribly annoying.
Why do we feel so uncomfortable when we focus on the things that are important to us, the things we thought we wanted to devote our lives to? Instead, why do we do nothing but get distracted, that is, do things we clearly don't want to devote our lives to?<..
The solution to this riddle is that by distracting ourselves, we are trying to avoid a painful confrontation with our problem of limited time, and especially limited control over time. Because of this, certainty about what the outcome will be is impossible (except for the very unpleasant certainty that one day death will end it all). When you try to concentrate on what you think is important, you are forced to recognize your limitations, to undergo experiences that seem particularly unpleasant precisely because you value the task at hand so much. <...> Perhaps the creative project you cherish will be beyond your abilities; perhaps the difficult conversation with your spouse you've been preparing for will turn out to be an argument. And even if everything goes great, you can't know in advance, so you'd still have to give up feeling like you're in charge of your time. <...> That's why boredom can be so distinctly, aggressively unpleasant.
The most effective way to combat distraction is to simply stop expecting things to ever be different, to accept that this is the unpleasant feeling inherent in being a person who dedicates themselves to complex and important tasks that force them to realize that our control over our own lives is limited.
The problem is that trying to live in the moment is really just another version of treating time as a tool, a preoccupation with the future, even though it seems to be the opposite. You're still preoccupied with making the best use of your time, just not for the future outcome, but for the enriching experience in the moment - and this preoccupation overshadows the experience itself. Let's say you intend to live in the moment while doing the dishes-perhaps inspired by the words of Titus Nhat Hanh, Zen master and best-selling author, that you should immerse yourself in even the most boring activities. But you're failing because you're too preoccupied with the misplaced thought, "Am I absorbed enough in the present?" The phrase "being here and now" conjures up images of stoned, bearded guys in flared pants, completely disconnected from what's going on. But in fact, trying to be in the here and now leads not to detachment but to exertion: the very effort to experience the moment in its entirety guarantees that you will fail.
De Graaf pointed out one of the most tricky problems that arise when we treat time as a tool to be used in the best possible way. And that is that we begin to feel the need to use our leisure time as productively as possible. We can no longer enjoy rest in itself, which seems to be the meaning of it: we feel as if it is not enough for some reason. There is a vague feeling that if you do not treat time as an investment in the future, you are not living right.<..
From this point of view, anything that has no value for the future is just idleness. Rest is allowed, but only for the sake of recuperation for work or perhaps for some form of self-development. Enjoying a moment of rest in itself, without thinking about potential future benefits, has become difficult because rest without instrumental value seems wasteful.
Most people mistakenly believe that all it takes to stop working is to not work. The creators of Shabbat understood that the task is much more difficult. You can't make the switch easily and seamlessly, just as you fall into slumber at the end of a long day. To quote the Cat in the Hat, a character in a Dr. Seuss book, "having fun is fun if you know how." That's why the Puritan Sabbath and the Jewish Sabbath have so many requirements for people to observe precisely. One must prepare thoroughly for a day of rest, at the very least clean the house, stock up on food, and wash. The rules did not exist to torment believers. They were necessary to make people realize that the endless circle of work can be interrupted only by a great effort of will. And to make it easier to do so requires habit and social pressure.
Another important thing we can do on our own to get a full rest is to simply stop expecting to feel good right away. At least at first. "Nothing is so foreign to modernity as idleness," writes philosopher John Gray. And he adds: "What game can we talk about in an age when nothing matters unless it leads to something else? "In such an age, a real break - to rest, to break away from long-distance running training or meditation sessions for spiritual enlightenment - will almost certainly bring profound discomfort rather than pleasure at first. But discomfort doesn't mean you shouldn't do these things. It is a sign that you absolutely must do it.
We often think that a person who is seriously engaged in a hobby - painting miniature figurines, for example, or amassing a collection of rare cacti - is guilty of not participating in real life as vigorously as he or she might. In an age when time is valued solely as a tool, treating a hobby as something shameful is not surprising. The person who has a hobby looks like a saboteur: he or she insists that some things are worth doing for the sake of the process itself, even though they don't pay off in terms of productivity. The taunts we shower on the stamp collector or train-watcher may actually be a kind of defense mechanism so that we don't have to admit that they are genuinely happy. Whereas people who lead telic lives are constantly striving for future goods and are unhappy as a result. This also helps explain why it is much less awkward (and even fashionable) to have a side business that resembles a hobby that one engages in ostensibly for profit. Therefore, to be truly satisfying, a good hobby is and should look a little embarrassing. It is a sign that you do it for the sake of the activity itself, and not for the result approved by society.
In 2013, Terry Hartig, a Swedish researcher from Uppsala, together with several colleagues, elegantly proved a link between synchronization and life satisfaction. He had an ingenious idea: compare Swedes' vacation schedules with the number of antidepressants dispensed by pharmacists. The first of his two main conclusions was self-evident: when Swedes take vacations, they feel happier (judging by the fact that, on average, they need fewer antidepressants). But the second turned out to be a revelation. As Hartig demonstrated, antidepressant use depended on the proportion of the population that was currently on vacation. Simply put, the more Swedes were on vacation at the same time, the happier the nation as a whole became.
It's much easier to keep in touch with family and friends when they're not working, too. And if you know no one is in the office during your vacation, you're not bothered by thoughts of tasks piling up, emails filling your inbox, or conniving coworkers trying to steal your position. <...> Hartig showed that even retirees, although not working themselves, felt happier the more working people were on vacation. This finding echoes the results of other studies. They showed that people who are in a long idle, unemployed state feel the same rush of happiness on the weekend as working people on vacation after a busy work week, even though the unemployed don't have any work week. The reason is simple: weekends bring joy also because they provide an opportunity to spend time with other people who are also unemployed. In addition, the unemployed on weekends temporarily feel less ashamed of not working, even though they should.
It is suggested, the researcher noted, that people need not so much personal control over their schedules, but what might be called social regulation of time: more external incentives to use their time in a certain way. This means more willingness to submit to the rhythms of the community; more traditions, such as the Sabbath or the French grandes vacances, when almost all work stops for a few weeks in the summer.

yearly, judging by the quotes - not slag, surprising for an "international bestseller" and is in my book service

Articles and news

How to find time to disrupt in the routine of tasks

Ivan Kesel, team leader and CPO at Domklik, spoke about the concept of disrupt, which implies the creation of a fundamentally new product-competitor on the market or within the company. Using real Domklik cases as an example, he talked about launching such disrupt solutions, and also gave 10 proven practical tips on how to find time, where to look for ideas, how to test hypotheses and other subtleties.

Is there any way to get the font size in journolus back to the way it was before? From my phone it's small and uncomfortable ?

A guide to creating children's digital products

Children's app expert Luda Senshova gathered the main principles of UX design for children, and also prepared a mini-guide on creating children's digital products and shared useful links on the topic.

Top thoughts:

  • Children are curious and interested in different things at the same time. They are not ready to explore the world sequentially, but are willing to dive into individual phenomena. Therefore, children's applications are often based on the study of individual phenomena
  • If you really want to teach your children something, it has to be fun. You can't wrap learning in a pretty wrapper. It has to be a game
  • Games in children's apps should be based on spontaneity, without clear rules, time limits, opponents and victories. It should be the same as real games, only in digital form
  • Understanding child psychology is the most important thing when creating children's products. It is important to study children's games and communicate with children to understand their interests and logic
  • Children are not yet able to stay focused for long periods of time. Therefore, they need quick results, which means interactivity
  • When designing, it is important to consider the age range. Children two, five, or ten years old are completely different. They have different motor skills, cognitive abilities and emotional control
  • Each developer has their own unique style, just like directors. For example, Swedish apps have a lot of freedom and opportunities for experimentation, while Chinese apps like to introduce a competitive element
  • Motivation in apps can be divided into "white" and "black" motivation. While white motivation works through inspiration and self-actualization, black motivation works through manipulation and obsession. Examples of black motivation include Achievements that you have to fight for in order to match other players. It is best not to use this kind of motivation

How to build a design process when designing complex products

Ksenia Gavrilova analyzed the design process in great detail using the experience of the Selectel product team as an example. She talked about how to use the design process in practice and how it can differ from the theory, how each stage is organized and what criteria are used to determine readiness, what research and analysis methods are most often used, how to adjust individual stages to the team's resources, and much more.

Ksenia also gave a list of sources of information to pump up skills at each stage and a design process flowchart to use as a cheat sheet.

9 Human Challenges with Using AI Co-Pilots

Steven Anderson talked about 9 human problems when working with neural networks like ChatGPT.

9 problems:

  • Generators can do at best 80% of the work by creating a draft. Incompetent people may mistake this draft for a result
  • Incompetent people may not notice gross errors and inaccuracies in the text from the neural network
  • Neural network gives out general information, ignoring important nuances, which significantly lowers the value of such text
  • When receiving ready-made information from a neural network, instead of searching through various sources, a person tends to continue thinking in the direction already laid down in the text. This deprives us of the ability to reason independently and creates "mental blinders"
  • Self-writing is an important stage of the thinking process. By turning to a neural network, people outsource this stage to the machine
  • Neural networks are not well suited for research. Communicating with a machine is no substitute for communicating with real people, which is where empathy comes in
  • A neural network always responds very "confidently" even if it reports false information. This can play a cruel trick on those who trust it
  • Typically, people communicate with the neural network one-on-one. This eliminates teamwork and brainstorming sessions, which are an important part of the design process
  • When we create code with the help of a neural network, we cannot check the relevance of the solutions it contains and how people responded to them. Thus, a neural network can create code that is not worth using

"When creating code with a neural network, we can't check the relevance of the solutions it contains and how people responded to them. Thus, a neural network can create code that is not worth using" As if yes, the neural network makes up functions and methods that don't exist in the library.... but I don't expect the neural network to give me 100% results. I decompose the task, feed it to the neural network in parts, connect the pieces of code myself and get the result. Yes, the neural network does not use advanced ways of writing code (and not all people do), but in general - ok.

A short line

Icons8 talked about creating a 3D icon of their own plugin for Figma, which tells about its features and stands out among other icons. They used Midjourney, Blender and Figma to create it.

The evolution of the Barbie logo.

The U.S. Federal Trade Commission is proposing to set rules for paid subscription services that would require users to be able to describe themselves as easily as they subscribe.

Kirill: this event is a signal that people have a clear demand for simple unsubscribe methods and a strong aversion to the mechanics of artificially complicating unsubscribe that have proliferated in major products in recent years.

From the point of view of ethics and customer care the most correct solution would be - when the service instead of auto-discharge asks "will you renew - yes / no"?

  • ***, I got charged a month's worth of goole colab without notice.

How to Turn Good Design Direction into a Good System. Christopher Butler spoke about the importance of systematizing the design process, as well as how to divide the process into stages and how to properly allocate work resources between them.

Testing the Future: A Guide to Testing AI Products with Users. Dmitry Starkov talked about how to conduct user tests of AI products.

New technologies

Anthropic announced Claude 2 neural network, which may become the main competitor of ChatGPT. It is distinguished by its ability to reason, input up to 75,000 words (ChatGPT has 3000), knowledge relevance up to 2023 (ChatGPT has up to 2021), as well as superiority in intelligence tests, capabilities similar to Code Interpreter and much more.

At the time of publication, the chatbot is available for public testing in the US and UK.

tested an hour ago, to the question "What is the square of the modulus of the psi function?", after specifying from what field this psi function (quantum mechanics) gave the correct answer - 1 "This quantity characterizes the probability density of detecting a particle in a given point of space at a given time. Integration of |ψ|^2 over the whole space gives the total probability equal to 1." and ChatGpt started vaporizing about integrals, which is also correct, but did not give the final answer

AI Safety and the Age of Dislightenment

Jeremy Foward conducted a study that showed that regulating AI for security purposes could backfire. He interviewed 60 experts in law, economics, AI technology and other fields, then compiled the main thoughts into one article.

One conclusion is that if neural networks are restricted, a huge advantage is gained by those who would have access to unrestricted models (employees of development companies, hackers and other criminals). Such neural networks could be used to harm people or to compete unfairly. This could lead to those with greater resources being able to train or steal models and then dominate the rest.

Ilon Musk has launched his own AI company xAI, which, according to him, should help understand the true nature of the universe. The project team includes specialists from DeepMind, OpenAI, Google Research, Microsoft Research, Tesla and other companies.

Company Website.

Identity

Kili. A nice rebranding of a company that focuses on AI training. The style is based on a bright corporate color, animations and an abstract mascot that acts as a guide. With the help of such a friendly identity, the authors tried to get away from the usual boring image of a technology company.

Guardian Art Book Fair. An original experimental book fair identity with a colorful palette and modern layout. As an original graphic move the authors used images of eyes, which speak about the importance of visual perception and symbolize different points of view of people.

We would like to note the scale of the project and the number of media on which the identity was used. An excellent example of the practical use of the style.

Stylish informal Onslow restaurant identity with quirky typography and a vintage palette.

Typography, calligraphy and lettering

A selection of excellent lettering from Cyril Dosnon from France.

The Alphabetical Room. An unusual typographic experiment by Liad Shadmi from Germany, in which he experimented with the construction of letters within mathematical three-dimensional grids.

Posters

Borrowed Spaces. A variety of experimental works created at the screen printing workshop.

Printing

Two Hundred Fifty Things an Architect Should Know. Pentagram have published a beautiful gift book that contains 250 application and philosophical tips for every architect. The book is based on essays by renowned architect and critic Michael Sorkin.

Infographics and data visualization

The grass courts are now in session. Reuters article about the peculiarities of grass on tennis courts with unusual illustrations and graphics, which are stylized as embroidery on athletes' polos.

Packaging

Nice Sahar Sun dairy product packaging with friendly lettering and cartoon illustrations.

Aesthetically pleasing pseudo-vintage Cervezas Salvajes Uncules beer labels with monochromatic color combinations and brutal typography.

TeaMajor's stylish utilitarian Chinese tea packs with a vintage-inspired layout and atypical hiking tins.

Websites

Brutal site of Good Times studio, designed as an explorer with folders and files. The site even allows you to choose the display: as icons, list or gallery.

An example of design for the sake of beauty, not convenience. Studio, and they themselves have a site with errors.

Intercom Fin. A literate Intercom landing page about their new AI bot Fin, which automates support communication with customers. The page features nice mesh gradients, large typography and a clear narrative structure.

Spectacular site by creative studio Kuya with animations, very large block layout and project video cards.

3D animation

The Third & The Seventh. An impressive short Full CG movie dedicated to the majesty and beauty of architectural art. The film features hyper-realistic graphics, as well as masterful work with composition, light, color and camera. The author of the project is Spanish artist and designer Jorge Seva, also known under the pseudonym Alex Roman.

The video was released way back in 2010, but it still looks relevant today. Years later, Jorge released a 4K remaster of the movie and also sold it as an NFT.

Separately, I would like to note that the project was created with 3DS Max and Vray, long before the advent of modern fast GPU renderers.

It's good, but I think I'm overdoing it with the nervous GRIF.

Duet. An elegant series of animations for the Sussex crane manufacturer.

3D graphics

Colorful abstract compositions by Alexey Zabrodin from Krasnodar.

Neuroart

Gianasgana. A research project, the authors of which tried to cross objects of architecture, design and fashion with natural forms. The resulting concepts are very unusual, including buildings, costumes, interior items and even fonts.

Illustrations

Futuristic illustrations by Viktor Sukhochev from Kursk with a complex and distinctive style.

A stylish series of character illustrations in the style of Japanese comics for the author's illustration exhibition Everyone is RPG.

History of Design

A selection of original posters and covers by Italian graphic designer Franco Grignani, who is one of the highlights of 1960s op-art graphics.

Portfolio of the Week

Antibody. An impressive portfolio from the creative team of filmmakers Patrick Clair and Raoul Marx, in which they create gorgeous screensavers for TV series.

During its run, the team received 19 Emmy Award nominations.

The sound design is also top notch

Lessons and lectures from VK

If your VK is blocked, use a VPN. For example, Tachyon or Mullvad.

Lessons and Lectures:

Figma:

After Effects:

Adobe Illustrator:

Blender:

Other:

A basic video tutorial on working with different grids in Figma and InDesign.

Konstantin Novikov has published several new lessons on creating complex effects using Processing and After Effects programming language:

Tools

Figma plugins

New plugins for Figma

TinyFaces - allows you to quickly insert neural network-generated images of people into your project

Group to Frame - converts groups to frames

Capacities. An advanced tool for storing information and taking notes. It differs from its analogs by an unusual system of objects, which can be pictures, contacts, links, locations, ideas, meetings and much more, including your own custom objects. Objects can be added to text or parameters of other objects, thus creating additional links in the database. A powerful tagging and linking system can also be used.

The library can be viewed both as cards and as convenient graphs that show relationships with other data. Additionally, you can view summaries of the records created on a particular day.

The service supports integration with email, Telegram, WhatsApp and other services, which allows you to send data to the storage without leaving other applications. Synchronization between devices is also available.

At the time of publication, the service is available for free with a full set of options. The Pro version is not available yet, but in the future it will include AI assistant, access to the iOS app and other features.

Kandinsky 2.2. Updated AI image generator from Sber. The new version improved the quality of generated images, added ControlNet functionality, increased the resolution to 1024 px, added aspect ratio control, and the ability to generate stickers without background using a special "sticker" promt.

Details of the update in an article on Hubra.

The generator is also available as a chatbot on Telegram. Weights are available on HuggingFace, GitHub, Diffusers and MLSpace.

Replit. An AI service for creating code in any programming language using text queries. Communication with the neural network takes place via the Ghostwriter chatbot, which can also integrate into your development environment. Supports teamwork, and allows you to instantly deploy applications and connect your own domains. Available as a web app as well as mobile apps for iOS and Android.

There is a free limited version. Price - from 7$ per month.

Kirill: one more reason to start your pet project without waiting to learn programming languages.

for example, chathpt makes up (hallucinates) functions and methods of libraries and produces clean code, but.... you can't get far without libraries... that's why you still need to take a course in python from scratch in a month without any problems to understand WHAT to edit

Stable Doodle. A new feature in StableAI's Clipdrop app that turns simple hand sketches into pictures.

Review.

Fonts

Cemento Nero. A narrow accent font with light decoration and low contrast. Available in 8 lettering styles with saturation gradation from Thin to Heavy. The case includes uppercase and lowercase characters of the Latin and Russian alphabets, as well as numbers and punctuation marks.

The price of individual lettering is 35$ and the whole family is 150$

Connector. A fine decorative typeface with unusual elongated bar serifs, monospaced characters and experimental forms that resemble handwritten fonts. The cast of characters includes uppercase and lowercase characters, stylistic alternates, as well as modern numerals, old-style numerals and other symbols.

Can be used for free for personal projects. Price - from $30.

Best of VK

A selection of interesting posts from our collection of design blogs from the last week. The top ones are highlighted in color.

Dessert

Eastern philosophy says there is no "self." Science agrees. Interesting article about why there is no complete "self" in the human mind, why Eastern philosophy has been saying this for 2500 years, and why modern science has come to agree with this view.

Why are you unhappy? Because 99.9% of everything you think and do is for your ego, and it doesn't really exist.

The 10 finalists for the annual optical illusion competition, Best Illusion of the Year Contest 2023, are presented.

Archaeologists have discovered 168 new geoglyphs in Peru that are about 2,000 years old.

Humor

Lastly, a meme about stages of perception.

What's Your Reaction?

like

dislike

love

funny

angry

sad

wow