Is No-Code (or Low-Code) the Future of Development
Vložit
- čas přidán 14. 06. 2024
- In this video, you'll learn whether or not you should worry about no-code or low-code solutions taking your job as a developer or freelancer. I'll also share my thoughts on some of the solutions that are out there and if it's something you should experiment with too.
Intro: 00:00
Viewer’s question: 00:38
What is No-Code and Low-Code: 00:55
Where do we go from here?: 05:33
No-Code solutions: 09:00
The Cons of No-Code:15:30
#nocode #lowcode #freelancing
Great points! The evolution of WordPress always comes to my mind when I hear no-code-low-code. At some point 10-15 years ago I thought that will kill my simple web design business but actually it did help businesses as you mentioned proof the idea faster at lower cost and allowed a lower entry point for businesses to have a website (then). Moving forward now we have "WordPress Developers".
Exactly
Donn - good video. Being a professional software developer today is not about writing within a programming language, and writing code is not the most crucial part of the job. Software development is mostly about the stack, the platforms, data sources, databases, network layers, APIs, security mechanisms and procedures that you use to build software solutions. Meaning anything you can do faster is a win.
Nailed it Anthony. I've learned to disconnect any emotional attachment to my code or the language or platform I'm using. Example: Right now I'm doing a bunch of TypeScript with React and then doing some Node and Python on the backend. For what I need to do with this client, its the right tool for the right job.
@@donnfelkeryt we use Linx (linx.software) - have you tried it?
Loved your phrase on Access.
They use low code where I work, but I personally I don't like it because it doesn't allow fine tuning that are required in doing certain tasks. Simple things such as getting the value from a drop down list is sooo needlessly convoluted, I find it way easier to do this through code.
I think the future should be a hybrid of both code and no code, a benefit of low code is that there is less risk of vulnerabilities because of poor code, and the backend is automatically connected and handled for you, you are also able to deploy applications instantly and run them on any device which is massively useful.
If low code is the future, we should embrace it, but there should be a way to allow developers to utilise their skills in code to fine tune things, there should always be room for fine tuning for those who already have experience with development, whereas those who aren't experienced are still able to benefit.
My worry is that managers are getting sold on the idea of using non developers for churning out critical production apps. I’m probably going to get roped into fixing the mess, much like what already happens with the ‘unofficial’ Access databases being written by end users.
They could try, until their problem probably isn't solved. Hence getting back searching for the traditional developer.
woah, fantastic discussion! Thanks for sharing
Glad you enjoyed it!
Thank you Donn for a great explanation
Very informative video! How about Outsystem vs Mendix? What are your thoughts, if any? Thanks
I don't like the typing noise in the background. But everything else was wonderfully executed. Good job on the vid. It was very helpful to me :).
We still need coders to use no code solutions. There is no way for someone without any basic understanding of HTML structure to understand what is a Container in Webflow or Wix
Moreover those wix/square space have a trash experience
Low code-only valid use case is a fast prototype. Cannot be used for prod. Hard to debug, has no community compared to programming, is nonscalable, and is hard to customize and extend. Having a small community is the biggest hurdle as AI does not have a large enough dataset to train on, so exit copilot, chatgpt, and its predecessors.
Hi Donn, I was wanting to know what you think about Unqork? I would really love to see a video about it.
Hi Donn, you'd mentioned that you ran a Training site on no code for 6 years, may i know if it is a Website or a Webapp? Where did you host it on? On web-hosting services like BlueHost, Hostgator or one of those Cloud services like AWS or GCP? Thx for answering, as i am quite confused as to which way to pursue new business, as in, a website or webapp....Thx:)
Good high level view!
One big con with low-code specifically, is the development environment. Usually no testing, no code navigation, no documentation quick-view, basic autocomplete, and few keyboard shortcusts. And then there‘s latency!
That's true. These are things that you'll run into.
I still find work with VBA/Access, especialy VBA, companies love their excel macros
While my developers are open minded and understand that our puropse is to solve business problems, there is some hesitation and comments "I'm not a SW Engineer to drag and drop boxes around". I cannot see my highly skilled and qualified SW Engineers to work with it, without feeling degraded. If business thinks they want to do it themselves, I'm happy to leave this field to them. Entirely! There is enough important work that requires skills and capabilites that business cannot provide.
I understand what you're saying. I feel its important to keep your eye on the trends though. Remember at one point companies paid tons of money for basic HTML sites with content management system. Now? You install WordPress and you're done. They can manage it themselves (sans some theme modifications/etc). All I'm trying to say is this - the industry is constantly changing. As new tools come out to help the average consumer do more, more advanced things will emerge on the scene of computer science. Things we haven't even fathomed yet. We're still very early in our industry (comparatively vs other industries).
@@donnfelkeryt Appreciate your response. I agree, there are constant changes (that's why work in IT, right? 🙂) and we constantly have to stay sharp without running behind every new gimmick like headless chickens. I always think it it is an opportunity to assess a new tool and understand it's capabilites and potential applications but also it's limitations it currently has. As you wrote; It's constantly changing. Biggest challenge for me is to understand and anticipate the impact of new services on our overall system architecture and how to keep the legacy applications working together with new platforms we implement. My engineers will be very busy to provide the integration layers!😁😁
I'm web developer, front and Back!! Is this how it's going to be? 😳 Scary!!
Thanks for the video. I might be biased, being the founder of one of the very first low-code no-code platform for Web applications. About the future of these platforms, I want to add a simple observation: starting from the 80s, we have seen a constant trend in software development: adding more and more levels of abstraction; the low-code no-code trend is adding yet another level of abstraction and I don't think we'll go back.
About integration and the limits of these platforms, you made good points; I would stress the fact that integration and possibilities to add features/customize with additional code change a lot from platform to platform. In dadabik for example you can add hooks in pure PHP, so any PHP programmer can do it without having to learn additional languages.
Where the final application must be hosted is another point to consider: does the platform provide an on-premises solution so you can host (and move) your app and db anywhere or you are forced to use their cloud? This can make a lot of difference e.g. if you have direct access to the database the platform is using you have much more options in terms of integration and tuning.
Hi @dadabik I've just looked up your website and found it quite interesting. Just curious: why choose PHP?
@@dungbui1341 back in 2001, when the first release was published, php was the obvious choice. Today there are many alternatives but I personally think that php is still one the best solution for the development of this kind of web applications; of course we could discuss hours about pros and cons of different languages / frameworks / technologies and at some point both personal preferences and trends about technologies have a big impact on these choices.
As a matter of fact, php is very popular and it is easy to find a developer that can add some code to customize a dadabik application and very easy to find an hosting provider that can run it (requirements are very low, footprint is very low).
I also think that sometimes people focus too much on the technology used by these platforms; let me give you an example: let's say you want to use dadabik to build a headless CMS and that you have a set of other applications (created with different technologies e.g. node, python, swift) that need to read the content created with the your CMS app. Those applications can simply call the dadabik API, get the content they need in json and use it. They don't even need to know the headless CMS you created was built in php because they just call some URLs.
The final users of this hypothetical headless CMS (the ones using the app to add / edit content) also typically don't care about the technology used, they only care if the app is fast, secure and provides the set of features they need.
Thank you for the very detailed and prompt reply. I have another question: you mentioned that it'd be nice for the users to have control of the database the platform is using for integration and tuning. Does Dadabik support this? Do you see any other platform that does this well?
@@dungbui1341 you are welcome. dadabik supports it. It stores raw data in your database and by default it doesn't change the tables schema you have designed. This means for example that you can just query in SQL the database from other applications to get the content you might need (or you can use the HTTP API, but that's another story). I don't want to say much about other specific platforms that I don't know so deeply as I know dadabik to avoid the risk of saying something wrong. Generally speaking, if the platform offers a SasS cloud approach, typically it doesn't provide a direct access to the database. For platforms offering an "on premise" alternative, this can be more frequent but you have to check if the platform needs to modify the schema of your tables and if the data is stored in raw format.
Good video....
How can i get a quote?
14:33 - "5 figures a MONTH". 14:55 - "5 figures a YEAR". Which is it?
Totally agree. It has it's disadvantages but alot of advantages
I've used no code tools for some time now with my own consulting/freelance builder.
I used to code websites but with all the tools available I decided to move more towards a "marketing person" than a "developer person".
2:06 is that js icon a humour or a mistake?
😜
What happens when no code plug ins or whatever is being used just gets discontinued. Does that not just completely break whatever has been built?
Yes. If that happens, its like the power getting turned off on your refrigerator and the water dispenser in the door no longer working because the "power module" got unplugged. Same thing happens with other software though. Email service provider goes down? App wont send emails. Database is down, cant read or write data. The thing is, with no-code you can and often do have multiple points of failure and that can be frustrating when things start breaking.
@@donnfelkeryt thank you for your reply! ☺️
I was hired as a full stack developer to build solutions using low code tools....Even though I like the Business solutions being built but I feel my skills of being a Full stack developer is getting useless .....my love for programming is still there and I don't know how to get back into my interest.
Push your company to get rid of the low code crap, convince them that switching back to custom code is actually faster, more secure and efficient.
What's your stack ?
@@marcobga2398 Java , React , Aws was my stack before joining my new company into Low code tools.
You can make website with Wordpress but it doesn’t mean html css js are not used and payed
Not following you. Payed?
I almost quit my CS degree because they started the first semester with access.
😂 that’s funny
Time to move up the abstraction chain.
Exactly
5:43
Linux is not unix, what the heck man
2:05 Isn't this the Java logo?
Good eye 😉 (you’re right)
Alright, but, who, is going to make a no-code platform? CODERS!
Agreed 😉
It's just another tool which as a good developer one needs to know🙂
I've tried SO hard to learn coding but my brain is just not made for those levels of abstraction.
Yet I have the passion for app development, and no-code platfors have changed my life.
If I could code, I really would be elbow deep in it but oh man I just can't crack it.
Hopefully it just helps the coding environment evolve faster or something, I dunno.
I'm all for no-code though.
I couldn't code either at first. Keep at it, you'll get it. I wrote about my experiences here: www.donnfelker.com/learning-program-sucks/
The problem is no code tools are pretty niche uses, limitations, drawbacks and almost always over promise. Low code tools are often development productivity tools and need a highly technical person to skirt around the tool's limitations. And I agree it will just shift how developers are used. Some may even be used to help people write the code for the business users who are using low code tools. The best of these tools allow flexibility for the exception cases. Disasters can happen, but it is usually a sign the tool is poor. I saw one tool create a dynamic db except it put everything in two tables and performance was understandably horrid. Thankfully, the tool allowed you to define a schema, but some highly technical guy had to do that.
Well said and agreed
To build a no-code tool, you need to write code. Simple!
Inception ... ;)
This is the trend that keep inflating applications to the point where you need at least 1GB RAM to run them.
This is also so limited that it cannot go beyond the "mindless" applications that do no logic - in other words, static or extremely limited applications.
BTW, no-code is not correct, code is generated by the IDE, they should rename it to no-typing or no-code-writing or no-smart
There are many points here you make, some of which are valid. However, what business owners want is the ability to connect things faster and no code solutions allow that. Are they the best? Often no, if they work and allow other problems to be worked on. It’s about tradeoffs and eventually things might have to be written in a real language or framework. It’s also important to note that at one point object oriented languages were made fun of because they were not close to the metal as raw assembly. History repeats itself, just in new ways sometimes ;)
So you are not really a software developer.
Lol. Sure.
Hello. This is future. The answer is no.
Hello Future
short answer: No
Ok 😂
Nope. No-Code is useless in developing complex apps for large enterprises. Just like many RAD tools.
Large complex systems… Absolutely… It won’t do well there. Or it does the well is very early prototyping and simple, MVPs, and so forth.
I seen multiple businesses run on top of Zapier, degenerate millions of dollars a year in revenue.
Are they simply just wire things together. Not saying that will work for every company… But there are companies out there like it.
No code low code will surely affect developers... Not now but atleast in future... The rise of Al and ML .. is not a joke...
Low code is a crap, no code is a lie.
Haha Ok
Well if it works then it works. Lol
Aww did bots steal your job?
You're wrong. if you are a coder, you are obsolete.
This has been said sooooo many times over the years. So many. Human intelligence and effort will shift to higher problem domains while technology handles roles previously handled by some programmers. At one point we wrote assembly, 1's and 0's. We used punch cards. Then we got low level programming languages. Then we got high level languages, then we got website builders, and code generation. This is just the next thing. Humans are amazing at evolving, and we'll evolve with this too.