I’m currently about halfway through the fcc curriculum, working on my dungeon crawler at the moment, and was looking through other people’s submission for inspiration. Anyways like every other project passed the front-end challenges there were very few posted. So Im wondering if it’s because a lot of people are dropping out before this point or do most of them simply not post links on this forum. Does anyone have any data on this?
You can see the relative completion rates of the certs here:
Wow thanks for the link. So less then a tenth go on to complete data vis and backend certification and only 89 people so far have completed all three. Given the level of activity on these forums I thought the number would be a lot higher. Well I think that will make me even more determined to reach all three and put myself among that small percentage.
You have to realize there are no data-viz challenges or tutorials, and the ones the back-end does have are not nearly sufficient. The people who are passing either already knew the skills and easily passed them, or they learned the skills somewhere else and hence are not as committed to doing the course on FCC.
Without seeing more detailed stats, I would assume there’s a dropout surge around the ‘Record Collection’/‘Profile Lookup’ JS challenges, since that tends to be a bit of a sticky section amongst newcomers.
I reckon some of those later intermediate algorithms probably cause another surge in dropouts.
As Isaac points out, the lack of content in Data Viz puts off a fair number, I reckon. And backend is sufficiently tricky that you just slow down a whole lot. I could easily believe that there are plenty of campers chugging along in the background making quiet progress with those. I just finished one of the bigger backend apps that took me a month, where the one just before only took a week.
Another consideration is that some just get jobs and stop! A guy at the local FCC meetup here in Melbourne got a dev job and was only half way through the intermediate algorithms!
I’m sure your right but I think its kinda sad. To me, when things get “sticky” is the best part of coding. Those are the ones that feel the best to complete. I’m sure when you finished that last backend challenge you were almost jumping up and down ;). As far as Isaac’s point about the lack of more tutorial elements in the data vis projects, I personaly think thats a good thing. It forces you to get out there and learn without it being spoon fed to you. I’m sure thats an invaluble skill when you get out there in the “real world”. I do realize that for some people tho if its not easy its not worth doing. Like I said kinda sad.
I think another thing to consider is that by the time someone has even a decent full-stack skill set he or she is very employable. People who get a job halfway into data visualization or back end are unlikely to have time to come back and finish their FCC certs as they will be too busy learning the specific skills needed for whatever job they got.
Will there be any more Backend challenges later do u know ? Oh boy this is not good news,i was planning on going all the way with this,guess im going to have to wait and see !
Well i finished my JS calculator app a month ago,and still haven’t made any progress with the new project i.e Pomodoro Clock.I want to lean about HTML5 canvas thoroughly and some cool CSS animations before i start with this project as i saw some amazing submissions for this project and i also want my project to be amazing .So what i think is when you look around for more resources apart from FCC to grasp a concept or to strengthen your already acquired concepts,it takes a little bit of time to get back on the project.Front end projects doesn’t have that much of a learning curve and most of the basic concepts you require to start with a project are available as part of the curriculum but for Data viz and back-end projects it isn’t the case.
I also work full time so i get a lot less time to work on fcc projects but as you said when things get “sticky” is the best part of coding,and are usually the ones you’re most satisfied with.
I would make a guess (backed by no real data) that the reasons are often one of the three below:
As noted, many see the backend challenges as more difficult because there are fewer lead-in challenges. It requires more of a search-ask approach.
Some people may only be interested in front-end and/or may get a job at this point doing front-end web development.
By the end of the front-end projects, campers feel more capable about their coding ability and then set off on their own to build things they want and/or get involved on other projects. I, for one, am slogging through the backend projects at a snail’s pace because I’ve also started contributing to the FCC codebase on github.
I do think it’s always worth reflection to consider ways to improve the site. I think the curriculum expansion will greatly help.
I’ll throw my anecdotal hat into the ring:
The front end certificate is fairly straight forward. All you need is HTML, CSS, and JS. You can use other technologies, but they aren’t required. You can do all those projects with Vanilla technologies.
Once you finish that certificate, the very next thing looking you in the eyes is a mountain of software you need to learn with no guidance.
- React or some other FE framework
- Sass or some other CSS pre-processor
- A build tool - Webpack, Browserify, Grunt, Gulp, etc.
- Managing dependencies and figuring out how to include external libraries using your build tool
- It may be the first time people have ever had to use their terminal/shell
All of these things you have to learn in tandem. Because you can’t write your first line of React code without bootstrapping your project using Webpack. You won’t get Webpack, React, or your dependencies working without understanding npm. You can’t use npm if you don’t understand how to use your terminal.
The difficulty level goes from relaxing jaunt into the hills for a picnic lunch to climbing Mount Everest without a safety rope. I think a lot of people get to that mountain of stuff and just say…“No”.
When I got to the DataViz projects, my progress ground down to a near stand still for 2-3 months (This alone would be enough to break a man’s spirit). I could no longer sit down in an afternoon and see any real progress happening because I would spend the majority of that time figuring out why all of these other things weren’t working right.
In short, when you get to DataViz, there’s a big chunk of time where you have to stop “learning to code” and spend that time learning your tooling and build process and that is extremely frustrating, obtuse, and doesn’t feel like progress anymore.
I hit the same wall. Just trying to figure out what I needed was pain. The approach I ended up taking was to continue doing the projects in codepen.io, this avoids all those problems you pointed out, while also working on my local machine to setup and run all those tools. This way I still feel like I’m making progress and learning the tool kits. Granted both progress at a slower rate because I’m dividing my time, but it helps keep me motivated.
I actually never used codepen even for the FE certificate projects. I find it to be clunky and hard to work with. Coding in my own editor I find to be a much more enjoyable experience. So, you’re right, some of my pain is self-inflicted in that I’ve decided I’m going to learn this the “right” way.
I feel like in the long run, I’ll be happy I went that route, but that doesn’t calm down the anxiety-voice in my head that I’m moving too slowly/not making progress.
The biggest reason for this disparity is that the Back End and Data Viz certifications were released well after the Front End certificate was released. The second biggest reason is the lack of supporting challenges for Back End and Data Viz - but these will be released soon.
As a person who kind-of-not-quite “dropped out” (as in slowed down and haven’t completed in challenges in 2-3 months) after the Front-End certification, I can confirm what people have predicted about people getting stuck with Data Viz and Back-End. I already know Sass and have been learning React on my own, but I still find myself completely lost when I go to start any of the challenges in that section.
I also tried skipping the Data Viz section and going straight to Back End and understood it…up until the “Voting App” project which requires connecting the back-end to the front end (i.e. taking user input and using it or displaying the results of calculations and operations to the user). No one seems to understand me when I say that I just don’t “get” how to do this when using Node, Mongo, Express, etc.
So I haven’t given up but I am stuck and looking around for resources outside of FCC for now, until I understand/the challenges are updated.
Thanks! It looks good. I’m not sure how Angular will go since I tried it before and didn’t get it, but maybe this course will explain it in a way that makes sense for me. If not I’m sure I can just use React in its place when I really get going.
Similar to others perhaps I’ve been in the Data Viz section for more than 6 months now. I finished the front-end then started React, that alone is enough to slow you down. I took a sidetrack into D3 and finished those quite quickly then came back to React. At the moment I’ve done 3 of the React challenges. While the last 2 I’m sure are useful, I’m not so much motivated by making games as opposed to web apps so I’ve been putting them off.
I felt whilst I knew HTML, CSS and basic JS pretty well I didn’t know it really well. So, I took advantage of the 3 month free Pluralsight and went back to basics to try to learn them better. At the moment I divide my time between re-doing some of the basic projects in React, learning Redux, trying to develop more locally (webpack, node etc.) using Pluralsight courses and then Codewars to reinforce the JS.
I’m still a part of FCC just off exploring other things I suppose.
I’d agree that the setup costs for React, Express, and full-stack challenges are huge. I have prior experience with the terminal and programming in general and it still took me weeks of fiddling, which was still insufficient. I finally found react-slingshot on github, a setup for React/Redux that is ready to go. (And even that won’t be sufficient for a full-stack app!)
Fortunately with create-react-app out now and very actively maintained, I think the barrier to entry to React/Redux should be lower in terms of setup. Beyond that, it’s still a challenge to pick up React and then Redux. At present the wider Web is the only way of learning this – fCC doesn’t have the necessary resources for teach it yet. React/Redux are such an expansive topic, you will need outside reading regardless.
I think the original idea was that we were to learn while we research and pair with others. On the Let’s Pair room in gitter, I find a lot of campers so far ahead of me (I’m on pomodoro and updating my previous work), and I can’t wait until I’m far enough along to pair with them. Of course, by then they will probably be onto their next certification level.
Also, you may want to check to see if there is a local group that has meet-ups in your area. There are also new FCC cohorts that are joining together to work on projects that help with their current projects.
Don’t give up on all 3 certs - they are definitely possible and well worth pursuing.
Oh, yeah - totally! It’s like diving into a snake pit!
I found the Back end easier to start since there was a little more teaching material and I only needed to learn Node first, then a little Express, then a little Mongo.
Like @taariya I have little idea how to link up the full stack apps well. I’ve done the Voting and Nightlife apps, but the front end for each of those is a horrible, hacky mess of EJS that I’m sure is not actually how you are supposed to do things.
I’m thinking of trying out Ember for the Stock app…hopefully their ‘convention over configuration’ philosophy will teach me a few best practices…