DEV Community

loading...
Cover image for The rise of no-code & low-code: are we out of a job yet? 😨

The rise of no-code & low-code: are we out of a job yet? 😨

Amelia Vieira Rosado 🐣✨
INTJ & knowledge hoarder 🧠 junior techie 🐣 blogging my πŸ’– out @ dev.to & technoglot.com πŸ’» lifelong learner πŸ€“ experimenter πŸ§ͺ anime-lover 😍
・2 min read

Photo by Luca Bravo on Unsplash

Hello fellow devs! πŸ‘‹πŸ» I hope you are doing well!

I'm sure you have come across no-code and low-code solutions before. It is by no means a new concept.

For those unfamiliar, low-code is described as a "tool" that enables developers of all skill levels to rapidly design and develop applications, while keeping code to a minimum. Such app development is typically accomplished by dragging and dropping visual blocks of code into a workflow.

No-code solutions also feature drag-and-drop and visual development, but unlike low-code, they cater to business people or others in IT who may not be familiar with programming languages but still want to develop an application for a specific use case. In other words, no-code solutions allow organizations to equip their teams with the necessary tools to create applications without any formal development training.

As you can see there's a subtle difference between the two. Now, let us ponder on this topic a little more, shall we? 🧠

Here are some questions to help you craft your comment:

  • What do you think of no-code/low-code solutions?
  • Do you think they will take our jobs? Why or why not?
  • What do you think are advantages and disadvantages of no-code/low-code?
  • Does no-code/low-code have a future? Or is it just a passing fad?

Drop your comments below, I love to read them all! 😁

See you in the comment section! πŸ‘‡πŸ»

P.S.: Try to stay on topic and don't nitpick too much on the wording or title. Thanks in advance πŸ˜‰




Still here? Catch me on Twitter or find me elsewhere! If you like my blogs and are feeling generous, kindly consider to πŸ‘‡πŸ»




My Redbubble shop is LIVE! Feel free to take a look 😊




technoglot footer banner

Discussion (40)

Collapse
zdne profile image
Z

Low-code and no-code is about the abstraction. Think of low-maintenance or no-maintenance cars. In early days of cars you needed to understand how car works to get somewhere. Now you just need minimal understanding (of the technology under the hood).

So where this takes us? On one hand it obviously helps less tech savvy developers to build apps. On the other hand, by hiding the complexity and moving up in abstraction you can build much bigger, more complex systems easily.

So to answer your original question, no, it will enable more people to develop software and to build previously unimaginable.

Collapse
jeikabu profile image
jeikabu

The game industry has had no-code tooling for at least a decade. Depending on the engine you use, the vast majority of materials (I.e shaders), gameplay/"scripting", and content in general is increasingly made by non-engineers. Programmers haven't gone away, the projects and problems have gotten bigger and more complex so their focus has just shifted.

WYSIWG/high-level workflows just make some things more accessible and empowers other departments to do what they want rather than having to open a jira ticket. None of this is new, it's been progressing like this for decades. We've got analysts that can do work in Excel that used to take a room full of people.

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author

Never said this was a new concept. Thanks for your contribution!

Collapse
jmfayard profile image
Jean-Michel Fayard πŸ‡«πŸ‡·πŸ‡©πŸ‡ͺπŸ‡¬πŸ‡§πŸ‡ͺπŸ‡ΈπŸ‡¨πŸ‡΄

I've been there before
In my former company, a mobile app agency, we bet a lot on Parse.com now parseplatform.org/
"Don't worry about infrastructure" they said
It belonged to Facebook, not your average startup

They decided to kill it anyway, and it ended up being much more work than if we had taken care of it from the beginning.

In order to make it look "easy to get started", Parse ended up being a complex beast.

Simple != Easy

I'm not afraid that we will looose our jobs. As longs as bugs are written, and project managers want more features, we will need developers.

Collapse
dracount profile image
Dave Sev

Interesting post!

No code and low code solutions are great! But they only do 1 job well.

As soon as you need anything that doesn't fit the mold, a developer will be needed.

They also open doors for clients who otherwise never would have been able to have such a service.

Eg take Shopify or WordPress - simple online stores and sites that can be done by anyone. But if you look at the jobs and openings for shoppify and wordpress devs there are hundreds.

Whole industries have sprung up from the existing solutions.

The truth is if you are switched on early, you can set yourself up as an expert in a newly created low code or no code tech, with little competition

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author

I think the first no-code/low-code solution I used was in my senior year of high school. I can't remember the name of the solution though. However, I can say with certainty that it had its limitations.

And to be very honest with you, the more I dive into the technicalities and learn about programming, the more difficult I find it to work with low-code/no-code. The limitations surely drive me nuts! And it makes sense, because these solutions are for a broader audience (not devs per se). An audience that most probably is not acquainted with programming and the like.

As you noted, there are several opportunities out there for developers, even with regards to low-code/no-code. So, in the grand scheme of things, it is not very likely that we'll be put out of a job any time soon πŸ˜‚

Thanks for sharing your thoughts! 😁

Collapse
kallmanation profile image
Nathan Kallman

At risk of self-promotion, I've written about just this! Spoilers: I don't think we are at risk of being out of a job (at worst our jobs will change from writing text-based code to some sort of graphic ui).

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author

Hahahah, I was just thinking about this blog of yours when I put this discussion together! πŸ˜‚ Just thought I'd make a discussion of this topic instead of a blog (I mean, you already did a great job!).

The self-promo is welcome! πŸ˜‰

Collapse
kallmanation profile image
Nathan Kallman

Thank you! I'm honored. And I've enjoyed reading all the comments and discussion so far

Collapse
theowlsden profile image
Shaquil Maria

Another great discussion!

As I see it, these platforms will not precisely put us out of a job. We will be needed to create other type of solutions, enterprise applications and more low/no-code platforms🀣.

I think that low-code platforms can enhance the development process by simplifying some aspects of the development that would otherwise be a pain to do. For example if you are a freelancer working on an app, but you are not that good at building UI, you can use such low-code platforms to build the initial screens, and then go ahead and tinker with the rest of the code and add logic with code. This can increase the development efficiency of the project.

On the other hand, you have the people that has no idea of programming. If these people start using no-code platforms to create applications, when there is a bug in the application, how do they solve these? That is what I fear. Most of the times the apps created with such platforms have boilerplate and overhead. Instead of creating the component with two lines of code when you code it, these platforms might create it with ten because it is a general component with parameters that maybe you didn't use.
This will make the codebase complicated, and more difficult for a noobie to grasp. And I wonder, if more and more applications are built with no-code platforms, will we have a new era of applications filled with bugs that are never addressed because the creators have no idea of how to code?

The solution is obviously to have developers look at the code and fix the bugs. But why go through the whole hassle if you can just let the developers create the application in the first place?

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author

You take an interesting approach here. πŸ€” I think most of the components in low-code/no-code solutions are well-tested in general. But despite this, the resulting apps' quality can be a concern (maybe not as huge a concern as we imagine).

Now, you know what I find funny? These solutions (no-code/low-code) typically offer paid features. There's virtually NOTHING you can do on the free versions (I get it, they are a business, they got profits to make yada yada). So I wonder; is paying a dev cheaper than paying for PRO features in a no-code/low-code solution? Food for thought!

I'm pro making our lives easier as devs, so I do welcome the low-code/no-code solutions, but when does it become too much? Where do we draw the line? Can we even turn everything into no-code/low-code?

Thanks for joining the discussion! I appreciate your contribution to the topic! πŸ™ŒπŸ»

Collapse
theowlsden profile image
Shaquil Maria

Now, you know what I find funny? These solutions (no-code/low-code) typically offer paid features. There's virtually NOTHING you can do on the free versions (I get it, they are a business, they got profits to make yada yada). So I wonder; is paying a dev cheaper than paying for PRO features in a no-code/low-code solution? Food for thought!

I wonder that too. Why go through all that? Is what you get after paying for such platforms more valuable than what you get after paying a developer?

I'm pro making our lives easier as devs, so I do welcome the low-code/no-code solutions, but when does it become too much? Where do we draw the line? Can we even turn everything into no-code/low-code?

That's definitely something worth discussing.

Collapse
hasnaindev profile image
Muhammad Hasnain • Edited

These technologies are usually assistive for developers rather than replacing. Code is essentially nothing more than a language in which we ultimately interpret and express the client's requirements.

Clients are usually confused about what they want and they don't know how to get it. Even if they spend considerable amounts of time to build something using a no-code platform, there will come a time when their website will becomes extremely slow because they kept on adding things and a number of other things. They'll want a custom functionality which this platform doesn't offer.

Do they change the platforms or use code to add the custom functionality? Changing platforms may entail changing DNS, correctly migrating website, does the said platform offer export options because Wix surprisingly don't. Does the other platform has the plugins or apps that the current platform has and you are using?

What about adding custom features using code? Of course, either the client will have to spend the next year or two studying how to code, or they'll simply hire you so you can create an app and host it for them on an EC2 instance.

THE PROBLEM: The problem is when you as a developer can't do better than a no-code tool. The development world is becoming quite complex. We don't just create websites now. We create SPAs, PWAs, SSRs. We have to worry about lazy loading, code-splitting and other optimizations. We have to keep up with the hype-train and use web-workers with Redux, hooks vs class states.

To be honest, the above paragraph was a little bit on rant too, especially the next half part.

Collapse
egilhuber profile image
erica (she/her)

I'll lead with a disclaimer - I've never actually used any no/low code solutions.

I also don't think they're here to take our jobs. They had to be built by someone, right? I think these solutions are going to become increasingly popular at a smaller scale, but there will always be devs slapping a keyboard for enterprise-tier software. If anything, no/low code might step on the toes of the freelance community that served smaller projects and web apps to their clients, who are now slowly being enabled to build their own solutions. Legacy and large code bases will always need their devs, but creating creating an app for your restaurant is becoming attainable for non-tech folk.

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author

If anything, no/low code might step on the toes of the freelance community that served smaller projects and web apps to their clients, who are now slowly being enabled to build their own solutions.

That's a great take on the topic! I agree with the points being made here. Whether these devs like it or not, they will have to adapt to these changes.

Trust me, I've seen some devs that are too comfortable where they are standing, so I fear for their future. By this I mean that some devs are not very willing to step out of their comfort zones and switch to another dev role in the field. Sooner or later, some will have to pay the price for being so "static".

Thanks for joining the discussion as usual! πŸ™ŒπŸ»

Collapse
hanpari profile image
Pavel Morava

Anytime when people import a thousand line library to write a few lines of simple script they practice low code.

Anytime, they use Excel instead of Python or R, they practice no code.

Business as usual I guess πŸ˜‹

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author • Edited

Perhaps you can address one of the questions mentioned in the discussion (blog above)?

Collapse
hanpari profile image
Pavel Morava

I've done that already.

I don't see Excel taking jobs from Pythonistas or less coding jobs because large frameworks exist.

What more is there to say?

Collapse
darkwiiplayer profile image
DarkWiiPlayer

I agree with Pavel Β­β€” his comment does address a few of those questions, albeit without directly naming them.

Collapse
code_doge profile image
jc566

I dont think "no" or "low" code will replace conventional methods for some time.

There are factors such as training and familiarity thats required. A larger firm might have the luxury of having employees "venture" out to train and learn these new tools, but a smaller firm might not have the extra resources or time to have training to adopt.

As with anything, the project in mind dictates the tools/services/language aka infrastructure used.

There is plenty of space for conventional and no/low code to exist in harmony as some teams would benefit from using one over the other more due to any number of reasons such as resource/financial/time constraints and etc.

Also, in my opinion, a good developer is the mind behind the tech stack. Someone who has a knack to be a developer will be successful in their field regardless of tech stack.

Collapse
miketalbot profile image
Mike Talbot

Well I'm building a low code/no code framework. It's certainly not going to put any developers out of a job (because they can add to it) - it's going to free them so that it isn't there job to "change a question on this questionnaire" or "add another chart to this dashboard", or indeed build yet another document workflow. What they can do is start really creating value (add AI for document recognition, or integrate or create some other amazing function).

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author

That's an interesting approach πŸ€” I also believe that low-code/no-code offers devs new opportunities. So it's not likely to put us out of a job.

I would love to see your tool in action when it's ready for use!

Thanks for dropping by! πŸ™‡πŸ»β€β™€οΈ

Collapse
leob profile image
leob • Edited

Interesting topic!

I think no - at least not in the short term (say, the next 5 to 10 years). I think for now, no/low-code will just replace a small part of the work, the most simple and boring part. If there will be a revolution then it will be a slow and gradual one at first, for now no/low-code is still a bit niche, except for website builders where people can create a basic "website" with a few mouse clicks, that's already pretty mainstream.

But in 10 years from now, yes who knows, it's not unlikely that our work will not have "disappeared" but will look vastly different.

I can't imagine that in 10 years from now we'll be working in exactly the same way as we do now - meticulously putting together our React apps in the way we do now, like medieval craftsmen. Look at the amount of "plumbing" that we still do (for instance auth and basic signup/signon), again and again, for every new app or website - I can't imagine that we'll still be doing that.

The whole process is probably going to be orders of magnitude faster and more efficient. Simple tasks will cost almost no effort (or will be automated away completely), and we'll only be doing the "hard" work (and the work that cannot be easily automated, like analysis and the more abstract "thinking").

SaaS will be bigger than ever, a large part of website development might just be clicking the mouse and drag & drop or "AI driven", and so on.

It's very likely that we'll need to adapt. But all of us being completely (or even largely out of a job), no I don't think so.

Collapse
darkwiiplayer profile image
DarkWiiPlayer • Edited

Before reading the post: No. Coding is to software development as stacking bricks is to building a house.


Okay; now on to the questions:

What do you think of no-code/low-code solutions?

The difference is purely psychological. Dragging boxes around feels more friendly and avoids the idea that "code is hard". Ultimately, people could just as well learn to code a scripting language on a very simple level, but have the benefit of potentially expanding from there to go deeper into programming.

Do you think they will take our jobs? Why or why not?

Definitely not. As I mentioned above, being a "programmer" isn't really just writing code; which is why some people dislike that word and prefer being called software architects or ~ engineers instead.

Unless your job consists of nothing more than stupidly translating program flow into actual code, these solutions should only shift certain tasks to non-programmers.

A personal anecdote: Where I work, we have a system that can, for the most part, be used by non-technical people, but some data transformation still requires the use of scripts, which effectively means a "programmer" has to do certain things that only require extremely simple code changes (say you want to address people differently if they're above or below 18 years of age; that's just an if statement with string concatenation in both branches). This is the kind of thing that could be achieved with a no-code system (in our case, it's already very close to being low-code, I'd say).

What do you think are advantages and disadvantages of no-code/low-code?

The obvious disadvantage is the limit in complexity; as systems get more complicated, code just becomes an easier representation to read; but that's mostly irrelevant, as the use-case of these systems is usually restricted to simpler logic anyway.

A more compelling argument for code is all the implications of it being a purely textual format. You can post a snippet of code in a chat and people can directly read it. Same goes for VCS: sure, no-code systems can totally be used in a VCS if they use a format like XML under the hood, but you still can't just open a file in an editor and read it. Maybe github would implement an online-viewer for a few big formats, but then we'd run the risk of slowing down the development of new technologies by making it much harder for them to establish themselves as viable.

Does no-code/low-code have a future? Or is it just a passing fad?

Did it ever have a future? Because the basic idea has been around for ages now, and it always seems nice on paper but every time it's really tried those systems end up failing.

Personally, I would prefer a sort of hybrid solution: a system that can read in code and produce (well-formatted) code as output, but provides the user a simple interface to set up logic without having to care much about code. The main difficulty here being, of course, that you'd need more than just a visual coding interface (that part would be easy), but a system that properly abstracts away all those programm-y concepts like loops, functions, methods, etc. (otherwise you're just coding with boxes)

Collapse
tonycletus profile image
Tony Cletus

Coincidentally, I published this article here yesterday. I tried to answer the question with my experience as a software developer and a no-coder.
dev.to/tonycletus/the-outcomes-of-...

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author

Thanks for sharing!

Collapse
hrmendoza96 profile image
Harold Mendoza

Try building the next Facebook, LinkedIn, Twitter, Instagram, Netflix, Dropbox, Reddit, Zoom, Skype, Google Drive, etc. with a Low-Code/No-Code tool. It's impossible.
Those tools (Low-Code/No-Code) are designed to create a very simple app (just like Wordpress, Wix, Squarespace do in the Web Industry). You can mimic or incorporate many things from the apps I mentioned above, but when it comes to large-scale enterprise level application development and management, the low-code/no-code tools will become insufficient.
Low-code has its place in the tech industry. It is necessary as demand for devs is so high worldwide. But when it comes to building more complex products, they will definitely fall short.

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author

Well said! Thanks for sharing your thoughts!

Collapse
jmkenn0 profile image
jmkenn0

Its very situational, but iPaaS (mulesoft/boomi) even for a seasoned programmer can be the easiest ways to manage integrations, especially for popular platforms with robust REST APIs. I mean yeah I could sit down and right a bunch of code and fuss over complex JSON objects but why not use this thing that has it all built in?

Especially for integrations, but one of the big advantages I've seen in low-code is debugging - its all right there, in graphics form, with a big "ERROR" sign on it with all the details. Again that's situational, but much easier than digging through logs.

I manage a PaaS and the first question is, can you create the back-end logic in the low-code part of the platform, and if not, why not? We are also slowly refactoring legacy JS to the no-code back-end, because the debugging is phenomenal and frankly the design forces a certain level of standardization and documentation.

Collapse
skhani profile image
Shahin Khani

Someone still needs to write the low code. No code platforms.
This is all about abstraction. But with abstraction you also give up portability and robustness. So low code and no code are great in some areas. But not I all

Collapse
vphreak profile image
Paul Sweeney Jr.

To me, it's like removing cashier at McDonald and replacing them with Self-Order touch screen terminal. But, we still need someone to maintain the system but that's another set of skill. Of course, what's next? more jobs being automated and putting people out of work.

I guess we will all adapt to new things. Universal basic income (UBI)? That's another solution we can think of. Robot taxes pay for UBI fund? I think UBI is great because it enable humanity to do more. We don't have to pay to live. If we want more to life, we work for it. I can't imagine how society is now... we have to pay to get water, use of the land, food and etc. This is basic necessities to be able to live. Without those, we will die! Water and food are from Earth naturally and it should be free. We are all human on Earth, it's our home and we should learn how to share resource and make sure everyone is able to live.

Wild dream? yes... I have dreamed of Star Trek philosophy where we work for the humanity not for money.

Collapse
jorzarios profile image
Jorge Zamudio

No code solutions are great for small projects and clients with small budgets. The no code or low code platforms are going to get better with final arquitecture and is ok to use them if the project "small".
Webflow is an excelent tool for websites and is ok to use it. the same thing will happen with apps. It doesnt mean that programming and code from scratch will dissapear

Collapse
juanfrank77 profile image
Juan F Gonzalez

Not only no-code & low-code. There are now AI's that write the code faster and 'better' than a typical developer could do. Just casually putting that out there....

Collapse
nefomemes profile image
Nefomemes

Nah, these AIs can do some simple tasks, like portfolio sites or blogs. But for complex apps nah.

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author • Edited

Panic mode ON πŸ˜‚πŸ˜‚πŸ˜‚

Collapse
kayis profile image
K

Can't see anything special here.

Low-code solutions are basically when we create libraries.

Just look at all the "webserver in 5 lines of JavaScript" tutorials. They're low-code, because the code is in the library.

The next step is obviously no-code, when the library gets an UI and some abstractions that non-developers understand.

I see the move from code to low-code to no-code as the goal of software development.

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author • Edited

Can't see anything special here.

It's not about this being special or anything.

What I meant here is mostly drag-and-drop low-code/no-code solutions and such. I did not mean libraries and frameworks (even if they qualify as no-code/low-code). Hope that clarifies where I meant to take this discussion.

Collapse
timocmd2 profile image
Timo Sarkar • Edited

Just think, stuff like Microsoftβ€˜s PowerFX is only going to automate the stuff, we would usually invest a lot of time in programming.

Collapse
technoglot profile image
Amelia Vieira Rosado 🐣✨ Author

Hmm, I'm not sure I understand what you mean. Can you elaborate, please?

Collapse
rahulsharmaeducator profile image
Rahul Sharma - SAP Guru

Nothing such will happen. I am 100% sure, tons of jobs will wait for the skilled person.
Regards,
sap pp online training