Video details

Why I Disagree On The State Of JavaScript Survey...

JavaScript
02.26.2022
English

🚨 Are you stuck on code, or your career? Get mentored by me! - 1 Week Free Trial -https://mentorcruise.com/mentor/erikhanchett/
The state of JavaScript survey is out for 2021, and I have some thoughts on it! In this video I talk about Vue.js, React, Angular and Ember and which one is on top! #javascript #vuejs
👉Check out my last video on Advanced Vue https://youtu.be/SoPo7bnhbqA
👉 Sign up and get free Vue cheat sheets and updates! https://www.vuecourse.tech/
Need some mentoring, help with a project, get a career in tech, level up your skills? Check it out 👉https://mentorcruise.com/mentor/erikhanchett/
Need to Learn Vue or Nuxt? Check out my courses below!
http://bit.ly/2LalQka - Learn Nuxt.js Course! http://bit.ly/3aiYe8s - Quick Starter On Vue 3 https://bit.ly/2OETt0M - Full 6 Module course on vue!
🗂️ 🗂️ 🗂️
0:00 Introduction 0:27 Vue Ranked A Tier 02:04 Satisfaction Rankings 04:15 NPM trends 07:49 Demographics 10:41 Backend frameworks 11:35 People
Links https://2021.stateofjs.com/en-US/

Transcript

Hey, hey, developers, the State of JavaScript survey is out, and there is some surprising results. So in this video, we're gonna talk about some of those results. We're gonna talk about how View is ranked, how Angular is ranked, and is it still worth using Ember today? So we're going to look at to all these rankings, and I think you're going to learn something. This is a survey that's created every single year, and I think it really shows where the JavaScript ecosystem is going. So I'm going to give you my opinion on it. So let's deep dive into it. All right. So here is the State of JavaScript survey. So first I want to show you this library is changing over time. It's kind of an ugly diagram. But what I think is more interesting is this library tier list. And what they did is they ranked all the different technologies and frameworks into this tier, S being the top A, B, and C. And this is all based on Satisfaction. So what I thought was interesting, if we look at the front end framework specifically, they have swell at 89%, React at 84%, and View at 80% just on the cusp between the A and B tier. So when I first looked at this survey, I was trying to understand how did it come up with this? Because React is 84%, it's only 4% higher than View. I thought React was the most popular JavaScript framework. How does that deal with satisfaction and usage? So when you deep dive into that a little bit more, they have this satisfaction versus usage. So obviously you can have a framework or library in the JavaScript ecosystem that has really high satisfaction, but if not many people are using it, it kind of skews one way or the other. So in this case, we have React right here, which has high usage and has high satisfaction. Angular is right here, which has low satisfaction and low usage, which is kind of like the worst of all of them, I suppose. And Ember being the lowest of this graph here, and View is right here. So it has higher satisfaction but lower usage. And so this is what I saw here. So they have this graph that shows versus all the data points and all the information they got from the survey. So there are four categories, Satisfaction, Interest, Usage, and Awareness. So you can see Solid is number one, and then Spelt is number two, and React is number three, and View is number four. And it actually does it through time, too. So you can see like in 2018 what it was and how the satisfaction is dropping. So you can see here, View was 91% in 2018 and then it dropped 87%, 85% in 2020, and now it's down to 80%. So at first I thought, wow, that's really not good. But if you look at React, it's pretty much the same thing. 89% in 2019, 88% 84%. It's dropping a little less fast but start dropping just as much as viewers. So I thought that was interesting to see. I know Fire Ship did a video on this and he was saying that he thinks the drop right here is because of the View, two to View, three migration. But then what would cause the drop in React satisfaction? It's only 4% here and 5% here, so I don't know if that fully explains the drop in satisfaction. Let's take a look. I also pulled another graph to kind of give us more ideas and see if this all lines up. So one other interesting tidbit is this usage now React pretty much hit the top usage at 80% and hasn't gone down at all. Well, View is actually the third month of the third place on usage this year with 51% and it's just slowly been going up. It's only like one or 2% at a time and it really hasn't hit over 50%. So that means 50% of people who have taken the survey have used View basically in the past. So I was trying to think that I know that definitely React is the most used JavaScript framework, but this seems to indicate that it's going to hit its Plateau. It's at its highest level of Saturation and while View is continuing to get more people using it, I wondered if this was true and things like Ember seems like the usage is dropping. So I went ahead and brought up the NPM trends to see if the data in the survey anyway relates to the actual data of NPM downloads. Now this tracks like popular library downloads per week, and it's also a good indicator of how healthy different libraries are. I feel I kind of ran the numbers to see the percentage increased. And in 2020 React actually gained 40%. It had a 40% increase. So it went from in this case like 7 million downloads. And this is between February in this case February 23, 2020 to February of 2021 to 10 million. So we gained about 40%. But in 2021, if we look at that from February 2022 it gained even a higher amount. It actually had a percentage increase of 55%. So it went from roughly 10,220,000 to in one week it was getting 13 million and even even peaked at one at 15 million. So I had this huge increase from 40% to 55%, meaning that the library was getting more and more popular. While View on the other hand almost did the opposite. It actually had a higher increase in number of downloads. When you look at this 2020 count, it started much lower. Of course it would start at in this case 1.4 million and then it went up to 2.2 million. So it had a 52% increase in 2020, but in 2021 it actually decreased to 47%. It seems to me that even though the JavaScript survey is saying that the Saturation is the highest it's ever been for and React, and it hasn't really improved and View is slowly increasing. This NPM trend seems to be showing the opposite, where there's actually more and more saturation, more and more people are using React while View is still increasing, but it's not increasing at the same speed as React. Just to see how that worked on the Amber side, because this seems to indicate that Embers usage is falling from 11% to 9%. And that was definitely the case here. When I ran the numbers for Ember, it actually had a 58% increase in 2020, but in 2021 it had a negative 13% increase, just super low. It actually went from 100 like 173,180, 7000, and it dropped to $162,000. So it's definitely not gaining. It's actually decreasing. I think that's just kind of says that both View and React are gaining in popularity, but it definitely feels like React is gaining faster. I don't think it's super concerning for the View community, but it just means that we just need to keep working on getting more people into the ecosystem. It's also interesting to note that if you look at these numbers here, even if View has 3.3 million, React is literally five times as big like there's five times as many downloads. The only caveat I was going to say to this graph though is that there could be some bias because this only has NPM downloads. I know people have talked about China and other places they don't use NPM, so this might be View has a bigger popularity in China, so maybe that wouldn't make up for some of it. I think, all things included, they use tons of React in China too. I don't know how big of a difference that would make to this chart. Let's jump into demographics now and talk about kind of some interesting points I've seen in here. Every year I've looked at this and I've kind of seen who's been taking the survey. So this year it's mostly younger people. This is the first year I think I've seen age on here. There's also something interesting here. So it looks like most people taking the survey are between 24 to 34 and that the years of experience are mostly five to ten years of experience. They actually added higher education degree this year too, which is interesting that most people taking the survey had a degree, which seems to me if you're on Tech Twitter, everybody's a self taught developer or career switcher. So I find it interesting that this was the highest category that everyone had a related field degree. So maybe you could interpret this as like maybe if you had an electrical engineering degree, that's a related field. Maybe not necessarily if you had an English degree or this could be maybe this lumps in people who have computer science degrees and all of every other sort of related field degree. So that's why it's so high. But I think you'd have more people that either have no, they don't don't have any higher education or no or yes, but it's completely unrelated. Like someone got an English degree and now they switched careers and became a developer. It's interesting here. Maybe my bubble in tech Twitter is a lot different than people who take the survey or what the reality is, I don't know. One thing that I also found interesting is gender. When I first looked at this, I saw that, wow, there's only 71% males that have taken the survey. And I looked back in previous years. If I went back here and looked at the percentage, it was 91%. So I was like, that's amazing. We need to have more diversity. We need more diversity in people taking the survey. But then I realized that 4% of women chose as their gender here and then it was 5%. So we actually had less women taking the survey this year. But what was really interesting is that they have this huge category called no answer. Every single one of these demographics have almost a quarter of the people who took it didn't even answer it. So I feel like all these values are really skewed and it's really hard to make any differentiation if we are really getting more people into this field, nor to diverse people into this field because you could take the survey without putting in an answer. So now we have this huge no answer field. So I'm assuming most likely if we put this no answer back in the numbers will probably be just the same as the 2020, or maybe it's just a little bit better. But I have a feeling like you can't really trust these answers. And since most people didn't answer or at least a quarter people didn't answer, these aren't very actionable or not very interesting. Another interesting point is looking at the back end. So this definitely is skewed. Obviously, since it's a JavaScript survey, these are mostly obviously all JavaScript frameworks. And once again you always get these cool new hip frameworks at the top of satisfaction. But when you really go to usage, you can see the old trustworthy JavaScript frameworks that everybody loves, like back end frameworks like Express at the top. Next. Jet has done a great job. A lot of people love Next. Gatsby is still really high, but it looks like 81%. It's almost double of what next is. So it looks like if you're creating a back end in JavaScript, most people use Express. I really like Fastify. It's holding in there. It's draining some usage, but it's still pretty low. It's higher in the satisfaction. It's like 91%. So if you want to hip cool new framework that not a ton of people are using, I think faster is a great way to do it. It's actually been on the survey for two years. This is an interesting point. I saw some Twitter threads going around about this particular question on the JavaScript survey. It's people you read, follow or just want to highlight from their work. Dan Abramoff Obviously he's hugely popular. Kenzie Dodds I think we all know who he is but it was interesting there's only two women on this list Cassidy Williams and Sarah Drazner. There was a lot of threads about trying to promote other people than just men on these types of lists so that's an interesting conversation. I hope that we can get a more diverse amount of people that people follow and want to highlight their work in next year's list. Let me know in the comments if you know any great people to follow from that should be on this list so yeah, leave a comment below I think that'd be great to see more diversity on this list. There is some YouTubers fire ship which is really cool. He actually did a video on this. I would recommend checking that out. There's also Brad Traverse on this list. Scott Tulinski even though he used to do YouTube, he doesn't do it too much anymore. Just like west boss they're more well known for their podcast. Kevin Powell is amazing in CSS so I would check that out and I think the last thing I want to just hit on is awards so most adopted Feature knowledge Colescann I think a lot of people are using this. I've been using this in my projects. Optional chaining is 90% private fields 9.5% Adopted technology es build high satisfaction Vet I'm a huge fan of vet I've been doing a ton of videos on it. Maybe I should do a video on V where I just go through and explain how it works. Let me know if you guys are interested in that. I can do a video like that. Highest interest is what you can't see but it's v as well so I hope you guys learned something. Let me know if you agree or disagree with any of the state of JavaScript survey. Leave a Comment below I'd love to hear things.