Видео

I would say stay within your ecosystem - and stick with Redux Toolkit query (unless you rely on SSR, i hear React-query might work better with tools like Next.js)

​​​​​@@techgirlinstyleThank you for your reply. It's actually a React Native application, so I'm not too worried about SSR for now. We already use redux-toolkit and a bare api layer with axios and I couldn't decide if to use React query (since it's so popular) on top of that or just use RTK query. But like you said it might be better to stick with RTK query. I want to avoid a setup where I end up fighting with the libraries esp when things start getting complex. But would you also say that RTK query is just as robust as React Query or is React Query better in some ways ?

Sounds like Redux maintainers as well as Rect Query maintainers suggest to stick with RTK if you are already using Redux for state management (according to Reddit:)) But here is a short comparison: * Caching: React Query: Automatic and optimized. Redux Toolkit Query: Built-in and Redux-powered. * Data Fetching: React Query: Declarative hooks (useQuery). Redux Toolkit Query: Declarative hooks (useQuery). * Optimistic Updates: React Query: Built-in. Redux Toolkit Query: Built-in. * Polling: React Query: Supported. Redux Toolkit Query: Supported. * Prefetching: React Query: Supported. Redux Toolkit Query: Supported. * DevTools: React Query: Advanced, dedicated dev tools. Redux Toolkit Query: Integrated with Redux DevTools. * SSR Support: React Query: Strong. Works well with Next.js. Redux Toolkit Query: Decent, but can be more complex. * Customizable Caching: React Query: Fine-grained control. Redux Toolkit Query: Tied to Redux caching mechanisms. * Middleware Support: React Query: Not applicable (standalone). Redux Toolkit Query: Leverages Redux middleware.

@@techgirlinstyle Wow. Thanks a lot for this. I have a better understanding now. I appreciate your time🙏

happy it was useful! I am myself excited to use Zustand in my next project

Yay! Short but hopefully useful overview

@@techgirlinstyle by the way, what would be the best way to get in touch?

Thanks! I wonder this myself. However at the same time, i remind myself that JavaScript is a high abstraction language in itself. I don’t really understand what happens in the browser when i call a good handful of JS native methods. But it does not slow me down, it gives me more brain power to lean into higher abstraction technologies (like React) instead of memorizing the 1s and 0s being transmitted to the machine. It might just be the new reality. Languages like Python, JS, Java are written in English language. We are seeing the next evolution where the syntax might not matter anymore and you can code in conversational style

No lol, front end

@@tech-adventurer thanks for watching! New video coming out today :)

Never used it before. I'll put it on my list

sure thing! here you go: ruclips.net/video/RqiNsQd0hjQ/видео.html

Will zoom in! Thanks!!!

Hello! It depends on what you are looking for. I would go on Dribbble, Behance, Deviant Art. I am sure there are also a lot of communities on Discord

@@Ss-zg3yj you’re entitled to your opinion 🙃

@@Branstrom i use it every day for work. People ask this question hence the title

@@ousmand742 lol gotta try this some time

Of course I am doing these in purpose :) these are actual technical screening questions you get asked (and I was actually asked) in 2024 in one of the top 20 paying companies (www.levels.fyi/leaderboard/Software-Engineer/Senior-Engineer/country/United-States/)

I hope it was helpful! Happy to review more interview questions if you’re interested

I am extending the prototype later in the video - and adding a new method for the Array prototype than later becomes accessible on the newly created array. This demonstrates the inheritance principle

The video is interesting and the tipic is not very common, good idea.

V0 is on my list to review next - however it is a slightly different use case. With V0 you can generate a Shadcn/ui based components - and yes it will let you preview the code but it will be a Radix UI primitive/Shadcn-ui/Tailwind baby. It is a great way to grow your component library that already has this stack - but not a way to built the entire web page from scratch especially if yuo are choosing a different stack for one reason or the other (e.g. i don't like Tailwind)

Tailwind is a collection of utility classes, it is not a universal panacea. When i know exactly what i want and (especially) when working on a code someone else wrote (even when someone else is AI), I will go in manually writing my stying. if i am just trying to make a few containers to have a flexbox layout - sure i can use Tailwind for that

@@techgirlinstyle We all know what Tailwind is, and I’m not here to compare it with plain CSS. I’m here to suggest what could have made the task simpler. For instance, why use React instead of vanilla JS? Why use AI instead of writing all the code yourself? It’s about taking advantage of tools and how they add simplicity to development. Creating a flex div in Tailwind takes a second, but with plain CSS, you’d have to navigate to another file, think of and create a unique, meaningful class name, add the properties, go back to the JSX/HTML, find the div, and apply the class. That process could take almost 10 seconds. This video could have been shorter if you had used Tailwind, and it would have made it even easier for AI to generate the code

@@techgirlinstyle Moreover, I wouldn’t say you’re inexperienced, but maybe you’re a bit of an old-fashioned dev. I know plenty of frontend developers who dislike Tailwind 😅, but they probably don’t realize how much time they waste on styling. By the way, I haven’t used CSS/SASS since 2021. I only use it when I need more control over specific properties

@@okkashaally2115 like with any tool, it is a matter of preference. Any btw not everyone knows what Tailwind is, I would not assume that. I personally won’t even start with Tailwind because inevitably i will end up with an unreadable string of classes. I can instead have a single class and easily see if there is a conflict in my CSS. Whether it is a separate file or a CSS in JS - is again arguable and depends on the project. You can have it in the same file if you want - just switch to CSS in JS. In the long term I absolutely hate an idea of using Tailwind. For you it might be easier to use a class - I on the other had can recall most of the CSS rules by heart (and my IDE will help me too with autocomplete) while Tailwind requires me to memorize class names instead. Knowledge of CSS is transferrable project to project, company to company. Knowledge of Tailwind classes is only relevant if the team had agreed to use it. That’s why I won’t even show myself using Tailwind in a greenfield project - because of my PERSONAL preference to write CSS directly in any shape or form.

@@okkashaally2115 haha thank you for not calling me "unexperienced" - I was not aware we were talking about personal levels of proficiency here, but discussing a tool preference. Trust me I built apps in pure CSS, SASS, SCSS, CSS-in-JS (styled components, emotion), worked with CSS modules, monolithic css/scss files and Tailwind. I began working with CSS in 2011 - when CSS3 began rolling out. I might be old fashioned in a way that i prefer to have more control over my styling and have a way to have a clear picture of what styles are being applied (and where) without having to go down the rabbit hole. If you love Tailwind - all power to you. I am not a fan of it and won't be using it in my projects.

Hi friend 🙃

@@jonhobbssmith yeah i struggle with it as well… Tailwind is just such a sad reflection of today’s reality that people can’t write CSS anymore, and instead plop on 20 class names in their markup

@@techgirlinstyle , in my experience every person I've spoken to who loves Tailwind is somebody who thinks CSS is hard. The kind of person who does z-index:10000

@@jonhobbssmith 😭I love CSS, it is my first love even before i switch from UX design to Front End back in 2012

@@jonhobbssmith I use Tailwind not because CSS is difficult, but because it saves me a lot of time.

@@vercieli simple as that.

Yes I’ll add it to my list. What about it is specifically interesting to you?

Hi thanks for your comment! happy to help you, you can reach me also on Linked in and we can chat there privately about your experience: www.linkedin.com/in/yanaha/ I am still figuring out content for my channels, advice for recent graduates is on my list. On a high level, we are all figuring out what the industry will become now with every newly introduced model. o1 preview is pretty amazing and can truly write end-to-end applications. However the fact of the matter is that third of the web is still running on jQuery. What that means is there there are a lot of legacy systems and systems that need to be maintained, upgraded and improved, rather than built from scratch. There is a still value in human front-end specialists that focus on system design and understand best practices, as well as can develop and upgrade/improvement plan for legacy systems. AI can help with that, but you have to have an opinion to make sure AI is guiding you in the right direction. My main advice would be - do what you are passionate about. I am passionate about design and user experience. 15 years ago that meant becoming a web developer. Today it might mean embracing AI while also having a deep understanding of what it means to write high quality maintainable and scalable front end. The future is in scalable system design in my mind. If you happen to have subscription, give this article a read: www.technologyreview.com/2024/08/28/1096515/ai-interfaces-ux-growth/

Hi thanks for your comment! happy to help you, you can reach me also on Linked in and we can chat there privately about your experience: www.linkedin.com/in/yanaha/ I am still figuring out content for my channels, advice for recent graduates is on my list. On a high level, we are all figuring out what the industry will become now with every newly introduced model. o1 preview is pretty amazing and can truly write end-to-end applications. However the fact of the matter is that third of the web is still running on jQuery. What that means is there there are a lot of legacy systems and systems that need to be maintained, upgraded and improved, rather than built from scratch. There is a still value in human front-end specialists that focus on system design and understand best practices, as well as can develop and upgrade/improvement plan for legacy systems. AI can help with that, but you have to have an opinion to make sure AI is guiding you in the right direction. My main advice would be - do what you are passionate about. I am passionate about design and user experience. 15 years ago that meant becoming a web developer. Today it might mean embracing AI while also having a deep understanding of what it means to write high quality maintainable and scalable front end. The future is in scalable system design in my mind. If you happen to have subscription, give this article a read: www.technologyreview.com/2024/08/28/1096515/ai-interfaces-ux-growth/

Hi Nedal! thanks for your comment. Let's unpack this! First, of course I am using GPT to help me structure and wordsmith my scripts, however the facts are coming from my personal experience. 1. There are pros and cons in both cases. If you would like, say have an ability to occasionally update your components (aka. pull the latest version from shadcn/ui and overwrite the one you have locally: `npx shadcn-ui@latest add -a -y -o`) you should wrap them in a HOC. Hoverer some components are known to have issues even after the initial download. e.g. the select component actually have a functional bug that i had to fix in the source code before being able to use it. Overwriting my fix would mean revering to the broken version. This is a lot to keep in mind 3. Since Shadcn/ui is not an npm dependency, it is not added to your project.json upon installation. But the peer dependencies do! This means Radix-ui packages get added directly to your project.json file. Now it is entirely up to you how to manage them (fix the version, always get the latest, stick to a major or a minor version, etc) Here is my codesandbox that shows a clean project with only shadcn/ui initialized - you will find Radix UI dependencies populated in the package.json: codesandbox.io/p/live/f95df568-4d7f-46d7-8679-a205523e2fbe 2. You are actually right. I had experience using Radix UI design system github.com/radix-ui/design-system that was using Stitches for the styling of all components and theming. Looks like it was deprecated earlier this year and I missed the mark on this fact, I will publish a follow up to clear this out, thank you for this catch!

@@techgirlinstyle 1. The example with the select component is quite insightful. From my experience, when dealing with buggy components, rather than editing the source code directly, I usually clone the component, apply the necessary fixes, and once the issue is resolved, I simply point my higher-level components back to the original version. This keeps things clean and manageable. (it's the same philosophy of shad, you build your own composable components) 2. You're absolutely right! Also, it's important to note that the shadcn CLI handles the setup and updates of its dependencies, many of which are tied to specific versions. For instance, "clsx: 2.1.1" is defined as a fixed release. So if clsx 3.0 is released and I manually force it, I must blame myself and not shad. To me, shadcn feels like a wrapper around an excellent set of low-level primitives and solutions (react-hook-form, radix, recharts, etc.). Approaching it with the mindset of learning what's underneath provides valuable insights and skills-something I can't say for other component libraries. Thanks again for the clarity, and I look forward to your upcoming content. Wishing you all the best!

Try setting up an entire application from scratch - it might be more efficient that way

Haha thanks for the support! These are educational for myself - so hopefully more folks find them useful.