One advantage of no code tools is that it has created many jobs. I'm a freelancer and about 20% of my gigs todays are in fixing or rebuilding apps that were built using no code tools. It's either it doesnt work as expected, or some function stops working because the client forgot to renew some subscription or that the no code provider no longer exists. The more no code tools they release the happier i am.
@pythonantole9892 I am a freelance Social Media Manager and I was thinking of learning a No Code development tool called "Webflow" to increase my range of services and get some projects in this domain.......Any advice ?
The problem with nocode is you absolutely need to know what you need to do before you start and you need to determine which nocode tool will actually fit your needs. Each tool has different abilities and shortfalls. No code has made me a king in my career. When I need something custom I just code it myself. No code can take 5 minutes to do what code takes a week.
Agree with dumdum above 👆 if you thrive with no code platform, you must be able to negotiate with the client to never use customized components that you can not build
Yeah, the same for me too. I work for agency team that create Enterprise application for small to medium company in my local area. the no code tool helps us a lot but the only reason it works for us is because our boss has 12 years' experience in software development.
If you really know what you are doing, why would you want to not code. The fact that you are considering no code simply tells us you have no idea what you are doing. I can bet it's faster to code than the drag and drop on a long run, so why waste your time. Drag and drop are for businesses that don't want to hire a developer and do things themselves. Why would you call yourself a developer if you don't want to code? Are we joking here!?
As a “founder” the goal is to validate there is demand for the idea or solution as quickly and as inexpensively as possible. This can and should be accomplished with no code tools. If 90% of all the functionality was complete, and there was a true demand for the product, there should be no problem onboarding the first 15% of the target market. This 15% will be enough to validate the idea and raise your first round of funding. Then you will have the money to pay for a scalable version of the application. Don’t fall into the trap of building a scalable version of an app for an idea that has not been validated. Good luck!
Much agreed, im a big fan fan flutterflow and i can tell, they are moving in correct direction and they really care about making their platform solve the "no code platform issue". i beleive in the coming years flutterflow will become even better then what it is at the moment.
by far the best answer here. FlutterFlow is a brilliant option for majority of the types of apps on the market (e-commerce, social media, chat, etc). I wouldnt recommend it if your needs are very specific (video streaming, AR/VR, etc).
FF is not even useable in an effiecient way in any case... the web version is cumbersome in the extreme and the new Local Run/Test feature is paid version only... #FAIL
I have NEVER, in 18 years developing products, found a no code platform that we didn't have to go and develop components, customisation, special parameters, you name it. PoC? OK. Go ahead with no code. Simple landing page or personal site? OK as well. My personal experience with any use case other than these ones is bad!
No code is like foam board, an exact knife and elmers glue for architects. I You build scale modeks, not production code with it. Nobody builds a building using the same materials and techniques they used to build the demo model. No code is for a rapidly developed mock up.
Slightly longer career, but a lots of IDE a very close to no-code experience, I think the closest I got was with MS Access, of course as mentioned customization is non existent. Moving forward, Dreamweaver generated rather ok code, and from a very superficial point of view, a good no-code experience (but still flawed and open to a lot of vulnerabilities). Of course these are rather old experience, web and application landscape have vastly changed, and expectations have also changed (for the better). My take would be that no-code/low code should help get you kickstated in learning more code, reducing the amount of boilerplate code. The problem pointed out in this video, is that these no-code platform did not account for some degree of play nor customization, or customization that is so obscure that it requires a very high level of skill. I have some interest in low-code platform that help you with the boilerplate code and offer a way to connect your interface and your customized logic.
This is also why AI will not replace programmers, it can generate some simple code but enterprise grade applications are pretty complex with a lot of business focused nuances. I know everyone has bought into the AI hype, just like they bought into the Blockchain and Big Data hype. But crypto hasnt replaced world wide fiat currencies and AI will not replace complex programming jobs as easily as our corporate overlords want to believe.
I think the real issue with no code is the spaghetti. I mean sometimes with libraries you still have to make custom components because it just doesn't do exactly what you need but libraries are not generating custom code behind them.
@@BillClinton228I'm glad I've finally found someone else that sees AI the same way as me. It all stems for the 80/20 rule, you spent 20% of the time getting something to 80% ready, then 80% of the time to get that last 20%. It's super easy to make a prototype, it's really difficult to make it work 99.9% of the time. Coding isn't hard, problem solving is, most of what software engineers do is the problem solving part, which AI/no code can't do for you.
I think everyone commenting here must be really young. I'm nearly 60. I've built websites (off and on) for 30 years. Back in the day, you had to code to develop a website. When the first rudimentary tools came out that would "allow anyone to build a website", many programmer/developer types said the same things folks are saying here. Slowly they were proved wrong and now, 30 years later, you can make a wordpress site do almost anything you can imagine without touching a line of code. Sure, when really serious issues develop (and they will), you'll still need to hire a developer, but locking onto wordpress 20 years ago, after coming from dreamweaver and front page, and geocities before that, was hardly a trap. If you stuck with wordpress, you're now an amazing website builder in high demand making a great living -- and still not coding. I expect the app world to follow the same trajectory but, of course, true developers will always be in demand.
Interesting perspective about web development evolution. It's worth to mention one more thing when comparing WordPress to modern no-code tools: The fundamental difference between WordPress and no-code platforms isn't about whether you need to code or not - it's about freedom and control over your business's digital presence. With WordPress, you're not locked into someone else's ecosystem. No-code platforms often have hard limits you can't overcome, while with WordPress if you need a custom feature, you can simply add it through plugins or custom development. Your data is truly yours - want to move hosting providers or export your data? No problem. That's not always possible with no-code platforms where your data lives in their ecosystem. Cost is another big factor. While no-code SaaS platforms typically increase their prices as you grow, WordPress platform costs stay more stable. We've seen cases where no-code platforms doubled their prices overnight or even shut down, leaving businesses stranded. With WordPress being open-source, this simply can't happen - your website keeps running because you own the platform and the only cost that could go up is the server price. In my opinion WordPress which can be used as a no-code platform has much more advantages than other no-code solutions. We've written an in-depth article exploring these differences coditive.com/blog/is-wordpress-the-right-tool-for-building-mvp-a-comprehensive-guide/ What are your thoughts on this perspective, especially regarding data portability and vendor lock-in?
@@pragmatedevwhat is anyone’s thoughts on integrating micro saas AI product on Wordpress then? I’m saying this as a web dev who wants to integrate logic, backend but isn’t trying anything too front end fancy but presentable enough to display the results
The important feature here is open source. Closed source providers like to harp on disadvantages of open source (such as security) as if their closed source addresses those problems. They don't. They just make problems harder to find and fix. If you're ashamed of of your code, or it's got proprietary parts that you can't open, I'd rather go elsewhere. Don't trap me in your ecosystem, especially when you can't predict which big company will be gone tomorrow.
No code platforms provide a limited set of functionality predefined. Is pretty much a template. The moment you step out of the "template", your run into customizations which can rapidly become chaos integrating with the said "template"
@@paulscottrobson Exactly right ... the first No code app development tool I've ever used was Visual Basic in Windows 95. I remember devs saying "oh, with this you may not need to code for much longer"
@@ldandco At best it moves it. I (you may also) remember designing dialogue boxes and windows manually or with a very simple tool on Windows 3.x This is now much easier from a prototyping POV
@@paulscottrobsonthere is another threat for coders right around the corner, AI😂 I’m a coder myself and I see some limitations but duh in 2 years, chatgpt or another llm and their increasing contextuality will make some juniors jobless lol
Agreed. Senior engineer here - would never encourage a no code solution. Props to you for exploring and trying to figure it out - it’s a valuable lesson for everyone
Senior here, today is first day I saw "no ode" for something more advanced like mobile apps... And why am I not surprised it's basically just templates... With bad code as output...
IBM Visual Age was a 98% completion product. The missing 2% would take a team of programmers many years to complete. Every time IBM was asked when the missing features would be added and critical issues resolved, the answer was always 18 to 24 months.
Was this Visual Age C++ or the original Visual Age Smalltalk. Always wanted to use the Smalltalk to write OS/2 and WinNT but the price tag of IBM was more insane then the bugs and killed the project.
0:58 Flutter is a framework, Dart is the language. Apart from that, yes, no code is a trap. It always has been, and probably will be for many years to come.
@@somedude6420 when you say anyone can build with AI, traditional software developers still have the upper hand since they can understand the code and correct and guide the AI. but yes, the democratization process of knowledge is insane.
She didn’t even say what she couldn’t do. I feel like this video was more emotional that actually providing a logical argument. Flutterflow to my startup has been amazing we are able to build something super basic and gives us the foundation to scale in the future. You have to know exactly how you want the app / product to function . When I went to my CTO I had a massive PowerPoint deck and flow charts and a basic UI that I had made. No code is not bad it just requires a different skill set
She is dummy. I work for this digital bank and your app is dev on FlutterFlow. A bank. Again, a bank. So sorry, but this video is a argument for people that don't want to study and think that no code is simple. And FlutterFlow is LOW code. Ok.
She did say it you just weren’t listening - she mentions she had a roadblock at creating widgets and couldn’t add them. The code was broken, and not doing its job at that point. When she downloaded the code base she was able to see where it went wrong. She states many others had this issue, and only those who have knowledge in coding would’ve known this. Those who are new to coding in general or app making won’t find this to be an issue, so of course it’ll feel like sunshine for you. Nothing emotional, she’s talking from a developers point of view and she’s 100% right. Any newbie would think this is amazing. It’s not lol.
No code that works for specific limited role should not be taken as no code for everything. It can work for simple apis but in most cases would require custom code to make it work.
@@johnyepthomi892 do you want some real usecase or fundable startup that only run with no-code ? Like : Closify, Synthflow, Beev, Prello, Cuure, Betterlegal, ...
I am a very experienced developer, and my advice to people is to avoid no-code solutions. I think they are a fallacy. If you are serious about creating software, you are much better off putting in the time and learning how to code and other related development processes.
What about someone coming from a strictly design based background with little programming knowledge (the very basics). Isnt there some merit to producing the front end and then fix it up towards the end? I don't expect to never learn code but I've always build visually. Also from your experience, could you explain to me some examples of customisations that are limited with no code builders? What kind of scope are we talking here. What kind of features would you say make it practically impossible to work with no code?
I totally agree with you! I'm working with .Net , C# ! I've just got recently got a new client because the client tried use WordPress and it failed with many plugins! I can see people are struggling with Wix too! Avoid drag-drop stuff to build app.
@@deechonadaive worked with webflow, once you move away from its built in features youll start having a bad time tm. An example would be numbered pagination/cms filtering would only take 1 day for a good developer to implement from scratch but with low code it would take a lot of jumping hoops i see the value tho, for very simple sites and designs nocode/low code is really good just dont use it on projects that get scope creeped hard
@@deechonada, yes there are use cases for low code, regardless of what "experienced" and "senior" developers tell you. Many low code apps have hooks and other optional ways to augment them, add customize behavior, allow changes to their internal DB to further customize. I've got 45 years as a true senior and experienced developer, so I know. To be fair though, the number of use cases is small. But to make the blanket statement "no code is all bad", is just showing one's lack of experience.
@TheSilverGlow Thanks for your advice! ive actually decided to learn how to program using dart and flutter. I like the idea of drag and drop, but I honestly think for the scope of my app (lottie animation heavy, and highly reliant on unique UI), i thought it be best to just bite the bullet and learn how to program. I don't think it will be easy, but as far as I have learned since I did my last message i am actually able to execute short blocs of code! I'm starting to understand the syntax and the purpose of each tool. On top of that, I was planning to leverage packages for the app when i wanted to use flutterflow, but that would require me going into the source code and manipulating it, which I think I would need to know how to code for anyways. And again, so Ive heard from developers, the code flutterflow produces makes animation heavy apps noticeably slow. I still think FF and other low code builders are worth their salt, but since im already design minded, I just feel like low code is too restrictive for what I want to build. As far as I can see it, and please tell me if I'm wrong, low code is good at providing services with quite boilerplate UI, and I think i just need to learn how to code. I've been putting it off for ages, and saw low code as a get out. It taught me that I might aswell try seeing as I'm going so far as to be willing to pay a service to essentially code for me. I just have to do it. Wish me luck!!
The problem with no-code is that they sell it for non-technical people and each of them have their own learning curve. and other problem is that. no-code actually doesn' t exists. it's all low-code. I work with Webflow, and everytime I'm doing some simple javascript. Low code has their merits and it's a powerful tool in the hands of a specialist. but you need to learn how to deal with each one limitations.
@@kevinsedwards I'm not using Webflow because of lack of skill to code, I can do a website with code anytime I need to. It's just that I did not choose Webflow, Webflow choose me and it pay my bills.
@@kevinsedwards It's good when you want to make animations that would probably take a lot of time if done manually, it's better to save time when it comes to something like html css lmao
@@waltermelo1033 it only pays the bills because webflow has good marketing. There's always some potato head client who falls for marketing then wants you to use it
To save you time, summary of the "trap": "I can't do this one specific function(s) so the whole no-code platform is trash cuz I didn't understand what the no-code platform was actually capable of when I planned my app." This video was the real trap. 😒
Yes and no. You will never be able to do everything you need in FF. It's a backwards model. You still have to understand technology but now you're layering the code with a drag and drop UI which causes additional headaches like remember what things do and how a widget is made etc... Better just learn to code and have full control. Even if the code is janky, it will work better and you will be able to modify as you need. FF is a layer of abstraction that is not a transferable skill.
I had a similar experience with building my website. I used a GUI to make my business website the first time, and it took maybe 1-2 weeks.. Now I am not much of a coder, but I understand theory enough to seriously take advantage of coding with ChatGPT. The second go-round I threw together a website going straight into the html, ccs, and javascript with heavy help of chatGPT in 2-3 days. It was a faster website because as described in the video, the GUI to code algorithm creates a mess of code that is nowhere near optimized, and creates multiples more than necessary. With AI advancing so fast there is barely a need for these no code solutions anymore.
I've worked with Power Automate that is another low code/no code platform and while it is quite powerful and allows you to do A LOT there were still close to zero use -cases where we didn't need to build our own custom APIs to automate the things our client needed and we automated more than 60 different business processes with a bunch of systems involved so it was a wide range of things that had to be solved. Version control is also an issue with low code platforms. Also- you should still get experienced software developers to do the low/no code projects of you want to make them more or less maintainable. Non-technical users will make spaghetti almost every time.
WordPress also suffer from this. As a WordPress developer, I have built many websites app using WordPress but always fall short into scalability issues and limitation. Another problem with this is that your are enforce to learn the page builder of their choice. It can either be gutenburg, elementor, divi, newspaper, or so many too mentions. You are forced to learn it but they are technically not transferable to any other programming language or migrating from another platform. No code solutions are not inherently bad; You just have to understand their limitation. Before committing to building an app with no-code tools, your client and team should identify these limitations and agree on the terms.
@@juanamillo You never heard of WordPress developer was a thing? You are probably still in college when that role sprung up. Let me lecture you. WordPress devs don't just deal with no-code, we fill the roles for Theme/Plugin customization or creating one from scratch. We also create or extend widgets to these page builder so that we meet client specification. We also incorporate WP CLI and REST API. These things involves coding and a lot of WordPress shenanigans to account for. Working and mastering a page builders is one the biggest thorn of my career. It is hard to negotiate a client and tell them to build everything in code because they will think it is an expensive route and they will have hard time with maintenance or making enhancements since most clients has no coding experience. We sometime laugh at something we don't understand, I feel you. Even myself laugh the hell out when I first heard "prompt engineering" was a thing.
@@juanamillo Lol who do you thinks builds the plugins? Writes the themes? Tell me you're inexperienced without telling me, also immature and condescending.
Don't let anyone tell you 'no-code'. Low code yes, I've been working with Microsoft Power Apps (canvas apps) for several years now and have had great success. No issues scaling, but it does take code! To do anything more complex than a basic CRUD app you will need (or need to learn) development experience.
As someone maybe getting into Microsoft Power Apps. How is the data input validation? Ex. I want the user to input a correct formatted email address. Needs coding?
@@henrikbergman4055 Plan to do coding for validation. If you're working with Dataverse on the back end some of the validation is easier - but just plan on coding for everything you want beyond the very basic and don't make too many assumptions. Code/test/code/test to be sure it all works like you expect.
I work at a very big non-tech company. We have at least 5 no code platforms for our internal tools that I know of. It is always the same. They buy a platform for huge amounts of money. Then they realize it has the same problems as all the others. Then they end the contract and buy a new one. And the best thing is that every time they do this, everybody who used the platform has to redo everything on the new platform because you can't just migrate your projects from one to another. None of these platforms are focused on developers - they are made for non technical people. However, developers are the ones who end up using them and it always blows up in their face.
As someone developing from scratch on web and mobile, I agree and disagree. I use flutter flow to turn my figma design into the code for my UI and it saves an inordinate amount of time compared to coding it up by hand. I also love using wordpress for building static sites as I can do in a day what would take me a week or longer otherwise. Does that mean you mean you should build the whole of a fully featured app in them? Obviously not. They might have obvious limitations, but the nature of their shortcomings is exactly that. Obvious. Some wisdom on when to apply them will help you much more than scrapping them completely.
Have you tried the WWX stack? Webflow, Wized, and Xano or Supabase or Firebase for that matter. It's not exactly no code, more like low-code but you can build customized web apps.
Would've been nice if she mentioned what kind of developer she is, and what programming languages she knows. Maybe i'll take a look at some of her videos for an answer...
As a seasoned Webflow developer, I saw this video in my feed and immediately was upset by the clickbaity title - I decided to watch the full video and here are my thoughts: 1. The video should be renamed “No Code MOBILE App Development is a Trap” - you can build web apps and even scalable ones. The title is a clickbait and I really hate it. Some people might call me “nitpicky”, but Dee never mentions “mobile” in the description and title - even in the video, for the most part. She makes no-code look bad in general. I believe it’s a really bad thing to do. 2. While I didn’t have personal experience with building mobile apps, I do agree with Dee on one point - these platforms aren't that much honest in their marketing. I think it’s the issue even with non-mobile no-code related platforms, sadly. That being said, it’s not like they are blatantly lying - you can, in fact, build production-ready apps with these mobile app no-code development platforms. However, they do not tell you about all the nuances you might encounter such as lack of customization (which is fine, even Dee admits it). So calling “no-code mobile app development” a trap is a stretch. 3. No-code in general isn’t bad as Dee tries to make it look like - moreover, as I’ve mentioned before, she talks SPECIFICALLY ABOUT MOBILE APPS. I use Webflow and it is amazing! Does it have limitations? Obviously, as any other platform, but you can build so much with it - by utilizing 3rd party and integrated apps too, you can build web-apps even. A popular combo stack is WWX - Webflow + Wized + Xano. You can build production and scalable apps. Sure, not everything can be built with this stack or no-code platforms in general, and it makes sense. It’s not a panacea. 4. Anyway, people must do prior research and learn about limitations of tools they use. It’s hilarious to see people coming expecting a panacea from every tool they use - then they go to Reddit and Twtiter and start shitting on a tool. I’m not saying that Dee lies about stuff in the video - she does not, and I believe here, I’ve heard about no-code mobile app dev platforms, they aren’t astoundingly great, but they do the job for the most part. Many apps can be built with them. Not all, not super custom, yes. But still many. And I believe it’s amazing. Sure, these companies kinda lie in their marketing about possibilities, but it’s not like these tools are garbage or a TRAP. 5. I’ve seen many people complain about Webflow on Reddit and (a few times) in Telegram group chats related to the tool, that it is bad, trash, garbage, and so on. Turns out, in 99% of cases, that these people didn’t learn about the tool, didn’t do prior research, just deep dived into it and expected everything on a platter without moving a muscle - hilarious! These tools, Webflow too, aren’t ideal, but the amount of possibilities they give to the users is really empowering. I have way more thoughts about this video and similar ones - maybe I’ll make a video addressing them one day, but for now I think this long-ass reply is enough. I just want to add that coders should stop gatekeeping development from other people - no one is stealing your jobs, and even if they do, then you need to step up your game and adapt. Gosh, the amount of elitism is insane. The comments just reek of it. Disgusting, honestly. P.S: Keep in mind, that I’ve talked only about APP DEVELOPMENT here - there are so many other things like websites/frontend development, automation/integrations (n8n is amazing!), etc.
The main use case for no/low code is the "citizen developer" that wants to put forms and lists over a spreadsheet and create more of an "app" for internal use.
Yes. The limitations are scalability which is unlikely to impact most internal apps. The next issue though is multi-tenant, multi-user, the My Account subsystem needed for a classic SaaS. Should not be that hard to do, but just a few of the many no/low code tools appear to be giving it a go.
As a Developer (Java since 2009, LowCode since 2021) i can disagree. We build huge Applications with OutSystems (LowCode) that are much much more than just a Spreadsheet or internal use.
Here's a different question though: Do you think you could have an MVP **without** those missing 10%? i.e. Could you validate the market demand with only 90% finished? Or were those missing 10% so vital that it wouldn't even make the whole thing into an MVP? Could you have done a few tweaks to get a functioning MVP with just 90%?
This is the absolute right question. I'm not a no-coder, but I hate strawmen arguments, so it's important to know what ended up on these 10%. If it's some snazzy CSS animation that really won't make an iota of functional difference, it is quite different from not being able to pull data from a system of record.
I was just asking myself the exact same question. Could you attract financial and dev partners with an MVP that was 90% of what the final launch product would be? The biggest issue is that there is no way to really know where the 90% lies. Is it just before those features that really set your product apart, or after they're in and working?
No offense but this is a gather point for insecure developers. Being both a person that is a developer and commissions software devs …thank god for no-code. To be able to Prototype working software in days is the best way to get the developer and client on the same page. I think I have lost way more time and money on software developers not delivering what is scoped. Please don’t lecture me on how to scope. Our days are numbed anyway, which is cool by me.
"To be able to Prototype working software in days is the best way to get the developer and client on the same page." She literally said no-code platforms are good for that
@@flaviaoliveira5949 just noticed your comment. I didn’t say that she was wrong i was referring to several people making comments. I think she is great and it takes a lot of guts to put yourself out there. Ill be more careful in my wording.
If you just want to prototype, use Figma. It can deploy to mobile and will allow anyone to build UI/UX. When coding something more than a Hello World, then it's better to code by hand than to use a WYSIWYG because FF will have limitations and in your project, the trouble is, when you find these limitations with FF for your project, you'll be too invested so you're either going to get fired, lose the market opportunity or start again from scratch using real coding. moral, don't waste your time with FF. It's like the saying goes, if you think hiring a professional is expensive, try hiring an amateur
It definetly is... at least no code web tecnology is going waaaayyy far... so far I honestly decided to stop fighting it and learn it, as Devs we can do so much more than people with no coding experience in these platforms, we just need to take our time and LEARN the tools.
No code is pretty much one evidence that “devin” or whatever these AI developers are called will not replace developers anytime soon. They either write horrible code that’s not customizable or it’s buggy code that still needs knowledgable people to debug it. I’d say, give it about 5 decades before it’s truly a thread to us engineers
Every programming language has some limitations too and depending on the dev skills it can still be difficult to scale or not secured. In no code, it just have a lot more limitation and needs a know how to make it more scalable and secured.
Flutterflow's no code is mostly in the UI side. Still have to jump into the dart file often to get custom functions to work properly. Good thing is that many of the AI's out there are semi-decent at writing the solutions you need.
The thing with no code solutions is that you need to be very willing to adapt your business/project to the platform, and not the other way around that is the normal approach when developing apps with code. You always need to settle for 90 - 95 % in the best case where you have picked the best platform for your project. For most projects there is no way to know exactly what you want ahead of time, and definitely not where you are going to end up after you talk to your actual users. The chance that you will able able to get anywhere near 90 % is therefore pretty low. The only way to make good apps is to develop them iteratively while you gather as much info and data as possible from user users and target audience. That is pretty much impossible with no code platforms and is the largest flaw with no code platforms in my experience. You will end up when a pretty bad app that will be really hard to sell, so it is probably hard to use even as a PoC or MVP. It's usnually way better to try to make a pretty polished a PoC or MVP with a very limited feature set and let real users use it.
It kind of reminds me of the drag-and-drop Web Builders that were pushed Years ago! They would get you started, but required you to have to go through the code to fix the bad spots! And there were a plethora of issues! While they have gotten somewhat better, you still need to know the code. Programming Code is more intense and there are so many paths one can go down that drag and drop would automatically be limiting and if the code is bad so will the app be bad!
Drag-n-drop web builders are used by enterprise companies all over the world on a daily basis. There are agencies, freelancers, and an ecosystem of plugins and add-ons for website builders like Webflow and Framer. All websites on these platforms are hosted on secure servers, deploy similar code, and follow best practices. In some cases, a website on Webflow is more stable and performant than a website you commission to a random developer. Not all platforms are the same, and the idea of "limitations" is quite abstract. There are limits in everything you choose, including hiring a developer. Cost limits are real limits, the dev skillset is a real limit, etc.
@@Pensandoci9 True! But I was talking about the Web Builders in their infancy and I know they have gotten a lot better, but they still are not perfect, and knowing how to code, is still a good idea.
No-Code just can't cover all use cases. It's like borrowing someone else's toolbox. Writing the code yourself removes these limits. Edit: Great video by the way, been wondering when someone would discuss this.
I wonder as well. 99% of all IT solutions are just basic CRUD stuff. No idea what people are building that requires all this "customisation". The real downfall of no code are licensing issues and vendor lock-ins, but that wasn't even mentioned.
I have basic idea of coding and I've been using bubble for well over 5 years. I think it depends on what u are building. I have had some hurdles to overcome when customizing but overall my experience has been good.I have built at least 15 apps both for clients and myself.
@@anantghere1 One thing I can say is that if you've never ever built an app before on any nocode platform or on wordpress, and you are not tech savvy, hire someone, because the learning curve might be high for u. Otherwise go for it It is one of the best, when it comes to customization. You can look up some apps that some people have done to confirm
@anantghere1 it depends on what you intend to perform in your app. For most CRUD apps, Bubble will suit you. However, if you grow over time, you might need to switch database, to plug to apis and/or to run custom scripts.
The problem is that people think they can stsr building right away with zero previous knowledge, but no code tools also requires you know how to use them Right
I am a computer teacher for students in a British international secondary school. Next school year I will teach the basics of how to make apps. Since it is going to be one of these no-coding alternatives, I am grateful that I saw your presentation.
Twas ever thus. Years ago they used to usual Visual BASIC 3.0 drag and drop ; that's pretty much all they did, but the "wow he/she wrote a windows app" fooled examiners.
As a dev, please teach them to use chat bots. And to use them only after unsuccesful trying doing it manually or asking for help/suggestions. And teach about process, not just the final result. Or just for simple, repeatable tasks. Or data summary or further explanation.
In my evaluation, most "no code" are garbage that leaves out the critical parts of actual software development like proper debugging and testing support, or even version tracking. Avoid at all cost. But, there are some "low code" platforms that are more similar to Rapid Application Development from the 90's that don't pretend that actual coding and debugging is not needed. In my evalulations, so far, AppSmith seems to be the best balance and is under an Open Source license.
Problem is that managers think that AI and No-Code are the future of programming, and if you try to convince them otherwise then they tell you that you're old and not *want* to move to modern techniques. They tell you maybe it's better to hire young "developers" that have more knowledge of the modern "revolution". Let's face it after 100 years we still have rubber tires. Old can be good for a long time ....
I like the Concorde in the back. Where did you get such a big one? I once travelled by British Airways (when I was a kid) and they gave a “build a concorde” paper toy and it turned out very nice. I have two little matchbox concord now - one is BA and the other is Air France. So sad that Concorde was discontinued.
Every time I’ve tried low/no code platforms, you always hit a wall. Trying to get past that dead end is a complete waste of time and it makes you wish you’d just have done it properly from scratch. The problem is that it’s impossible to know what the limitations are until you get there and are backed into a corner. Don’t do it.
Great review and loved the storytelling. What's an alternative for those who want to develop an APP without taking years to learn how to code tho? I was looking forward to the solution
Not to take from the validity of her overarching point, but to spend over a year with Flutter Flow and not learn the difference between Flutter the framework vs Dart the underlying language, and call Flutter "the backend", reflects poorly on her comprehension.
At no point did she argue against Flutter the framework or Dart the language, only against Flutter Flow the RAD tool (and possibly its default integration with the Firebase backend), so if we code in Dart/Flutter using VS Code we're good to go Dee? RAD tools have generated spaghetti code since their inception 30 years ago, with Motif/C toolkits on Unix. This is a story of no skills meets no code. I do agree on her point though, as one is largely sold a lie, which can still work if one stays inside the limited scope of the box.
I am a flutter dev and I remember once using flutterflow, the main benefit I found was producing complex UIs with drag and drop. I hate writing UI code without drag and drop so it did help me there. I downloaded and used the UI code and slotted in my own busines logic etc. So yeh, I use it sometimes to make complex uis. I cant recall if the code was bad or not.
One other BIG problem you did not mention, I think, is that you are completely dependent on this platform for the future. Apps regularly need updates, because requirements for the app store change (and many other reasons), now you have to HOPE they have this new requirement available and it still plays nice with the rest of your app. I am a 15+year unity3D dev, and I hardly use plugins at all for this specific reason. You are now reliant on it, I update the plugin and the app breaks (since they changed too much), or they just stopped updating it and it doesn't work anymore etc. No this is just one plugin, with these tools your whole app would rely on a third party.. that is a major red flag for me.
You are exactly right. I would also add that on many of this No Code platforms you are vendor locked and you can’t migrate your app to another platform, so you find your self trapped and have to pay for it.
You didn't even explained what your problem is. It could be that you were wrong with your implementation of flutter flow. No code is definitely a solution that saves a lot of money if you know how to do it right :)
I'm a typescript programmer and I could tell from the title where this was going lol. All of these "ZOMG, I make $20k/month with a no-code app I built in 2 days" videos on RUclips were red flags. No-code is fine for simple things. Like Wordpress, Squarespace, Wix, etc. But when you have an idea for something that isn't supported on their platform, you're stuck and eventually get around to having to learn how to code lol...or hire someone who knows the language already. That destroys this whole illusion that you can't "build anything" with no code.
You're right. The "I build an app in 2 days" pitch is absolutely wrong. But you're wrong in that it is fine for simple thing. I've build a financial marketplace for alternative investments that comes with financial performance analysis and modeling all with nocode. It took 6 months, but it can be done. We've been operating with paying customers for two years.
Adding my 2 cents here. I've been a web developer for 15 years but, due to greater forces (media hype), I needed to carry out some projects on low-code platforms, such as Microsoft Power Apps. I can also prove that no-code/low-code only works for simple projects. No-code/low-code is REALLY a trap. But one thing we can't deny here: their marketing is very strong.
Literally I don't understand why it is so hard for people who want to do "programmers stuff" to actually learn programming. It's really not that hard. No one asks people to play with memory allocation, pointers and low level C stuff. And there are great libraries for Python, Java and much more. Also, it was a time I was obsessed with abstractions. But I simply realized that the more abstraction you add to code, the less it can do. While it serves you good for a starting point, when you try to do more fancy stuff, you'll fight the abstractions 100x worse than if you wrote a little more complex code. That's one reason why I use vanilla CSS and I've never even tried CSS frameworks. Vanilla CSS is a bit more tedious in the beginning but it rewards you later if you're building a complex web application. Also, sometimes when I need to build GUIs (and the design matters) I use 2D libraries where I create my own widgets.And it's not really that hard.
@@guillermosanchez1224 Does it? And does it work efficiently? Will it scale under higher user load? Does it actually do 100% of what you intended it to do? The closest I've had AI come to generating usable code included a bug that caused data corruption. This bug was difficult to detect. Most of the code generated by AI looks like the garbage produced by a programmer with a shiny new master's degree in computer science but lacking real world experience in a professional setting. If the code works, then the approach taken is very inefficient (especially when it comes to operations against a relational database management system).
Just a small detail - "Flutter" is not a language, but a UI framework, based on the programming language "Dart". I tried it once and found it quite cool - got something nice running with very little code (compared to horribly cluttered coding in good old Java). But then when I re-opened that project 3 months later, a lot was broken and did not build anymore, once my IDE had updated Flutter. After looking at the release notes for Flutter, I realized that they seem to introduce breaking changes all the time... and this is a no-go for me. I guess I will keep looking for a better alternative.
As a computer scientist I can't help but think if you are thinking a no code solution can be a backbone of a startup of any kind you aren't technically skilled enough to be running a startup to begin with.
Her mistake was knowing how to code. If she didn't, she would have hired coders and did what needs to be done and got on with the business. Instead, fell into a trap. Sometimes not knowing is better than knowing something. Wouldn't you agree dear Scientist?
I'm no programmer, but all of the negatives I am reading are from programmers. Seems like a battle between the Dairy and the beef industries in that one is trying to protect their interests. No code will appeal to certain people who need something no as complex. And in the not so distant future complex coding will also be possible.
Great and honest review ... No code is like: Make a cake... without learning how to cook Go places... without learning how to walk Build a house.. without touching a brick Bring ideas to life... without learning how to do nor think a get-code-quick scheme sounds alarmingly similar to get-rich-quick scheme... they all get you 10% of the way and leave you there..
As an oldie I remember one of these that ran on a tape drive on the PET (1981) which claimed to be the last program you'll ever need. There've been many panacea since. I can't think of an older one than that.
It depends on the user. If you have code knowledge - transitioning to low-code is like having a superpower. Low code with traditional code knowledge is like this: Go places... knowing how to walk - but you have a bike
Hi all, What are your favourite tools to make a 'code' web app? I am a developer but worked for a no-code company for 8 years. Now, that I am out again I am rapidly noticing there are far to many tools to learn them all. Any suggestions? For example, What service/tool would you suggest for: - A great looking interface designer (I have wasted too many hours in HTML and CSS) - Where do you run your logic layers? (AWS? or something more specific) - What Database would you suggest? - Anything else? (I don't own any servers)
This goes for any third party library or component, tools that allow you to do things more quickly. They get you 90% of what you want and have you spend the rest of the life cycle time of the application working around its limitations and dealing with its side effects.
No code just means you are following someone else's way of processing, living within their process, and limited by their ability to provide drag and drop services to limited experience people. For example: Excel is a no-code solution. It lets you input data, apply formulars, set presentation options. If provides charts and pivot tables at the click of a button. But if you want a business accounting system connected to your bank and inventory system, you can't use the no-code options of Excel. My point is, 'no-code' has been around in various names for over 25 years. Software engineers are still needed to create real business and game software.
It depends on what you're trying to build. For enterprise level highly custom apps, no-code platforms are terrible. But Ive had plenty of small business clients who have found them ideal for their needs.
As an architect, when I'm interviewing clients, if I find no-code anywhere near a real product, it will cost them twice my normal rate if I'm feeling nice.
Thanks for this honest feedback. I'll still keep using FF though to test and validate my concepts. I think we shouldn't forget a very important point: these tools have existed for a very short time, and FF is improving extremely fast. If the progression goes along with AI's (which btw is becoming incredibly good at coding), I guess we could soon (under 3 years) have revolutionary ways to build apps. Maybe could you make a feedback video about these no-code tools each year to see if you progressively change your mind 😜
No-code like Bubble is the Canva for designers, it won't do even close to what Photoshop can do, but it gets you to point A to B more intuitively and efficiently, and if you know what you want and your point B is in Canva's capabilities, that's all you ask for. I'm happy things like Canva and Bubble exist, it gives more power to us to work without paying someone else to do it or teach you; the future is not no-code, it's agency, no-code is just like AI in many ways and they'll probably intertwine very soon.
Disagree. FF has been phenomenal for me for visual frontend ui development. I use APIs, cloud functions and firebase for the rest of my App's functionality. You have to know what you're doing and adapt to the tool you are using. If you want to use an Axe to cut paper, well sure you're going to have problems
AI writes VERY BAD CODE. Barring insertion of simple functions, it typically takes longer to fix the issues with AI generated code than to just write the code by hand. If you aren't proficient at writing code, then you either won't see the flaws in AI generated code or won't be able to fix them.
@@gaiustacitus4242on the other hand, ChatGPT seems to be more knowledgeable about React esoterica than 95% of devs out there, including myself, so that’s a plus
I work for a company where the CTO adopted no code solutions for a lot of stuff they want. He's a great guy but he's no coder. So sometimes users cannot work anymore, integrations stop working or random errors occur on production - there's no ticket, no source tracking and lots of sudden technical debt. Architecture gets worse and performance gets worse with every change and everybody is losing track of what's happening. This is no code done wrong.
Really apreciate the advice, im learning to code and have plan to learn no code tools also, I was specting no code tools to be insuficient, but It surprises me they have the isues you have mentioned, everione wants to sell the new magic money maker but I hear no body talk about the limitations, thank you for sharing your expirience, Is encouraging to hear hard working and honest people.
I am not a developer, but I made an app in appsheets and it works nice. I would not say it was no code though. It was a lot of hustle, lots of problem solving nad sometimes complicated formulas. Also I combine my sheets with java scripts. So they are not basic at all. I chose appsheets for 2 reasons. My employer already had appcore plus google is pretty stable in the market, so there is less worry if the platform will survive. And as for AI, you really need to be good a problem solving and know what you want to achieve and see the mistakes AI made to point them out and correct it. So you need some knowledge to work with AI. It can make your life easier but will not replace thinking.
I think the point is that you can iterate faster on no code so you learn more about the problem that you’re solving in a shorter amount of time. If you know exactly what you’re building and you know how to code…..sure, use code. On scalability, you’re going to rebuild your app a bunch as you scale regardless of what stack you use…again unless you know exactly what you’re building from the jump. All of these are tools at the end of the day, you need to choose the right one for the job. But I also agree with you that No Code marketing makes it seem like it solves all the world's problems.
Hi, do you think Webflow has the same problem? I want to create a learning platform for people to improve their Excel, Python, SQL, … skills My idea was to create the layout of the page using no code in Webflow and then hand it out to a developer so they can do the back end create a functioning learning platform.
No-code is definitely work in progress, but also definitely the future of development. Customization and integration limitations are being sorted out gradually, and AI won't but boost this revolution of coding from a visual or natural language interface. The article on Bubble is from 2021, a lot has changed since. I don't think it's a trap, although it might not yet be good enough for your project.
Fully agree, I've built a whole document management system using no code (focus on UI is no code but back end is Linux) and it's integrating nicely using no code API connectors.... so yes they shouldn't shoot it down because it didn't work for their particular use case. There is no time for learning how to code now... AI is taking over and people still want to learn how to code....
@@maphuthumanemasilela8934exactly the point. If AI is to take over the code, why would you even need a low code platform when AI can write the code? Low code platform just doesn’t work. You will end up writing more code just to label your product as “low code”
I think this is a circljrk of seasoned engineers who are discouraging the new tool on the block. A tool not meant for them. Its like a premier pro power user bashing capcut or the basic phome video editor for its lack of features.
I'm not a premiere pro guru, but I'm decent. I tried Capcut and haven't really gone back since. For what I need, it can output good enough quality In a fraction of the time and computing resources. The democratization of different fields in the past 20 years has made a lot of the old guard gatekeepers really annoyed. Because for a lot of average practitioners, they found themselves being out competed by relative newcomers with really good tools.
Hi I really appreciate your sincere review. I am python programmer and a low code power Apps developer. I am really i terested inlearning mobile app development. Would you recommend me only to learn flutter? Have you learned already flutter?
I’m glad you talked about that. What those platforms say about themselves is a wishful thinking or rather straight lying. I’m a developer, and I participated several times extending applications that someone like yourself created with “no code” tools and hit the wall. I can say, it’s always a huge pain for developers and it takes times more time than actually rewriting from scratch. As a developer and a startup partner I continue hoping for some tool that would help building faster, but nothing so far. They all barely maintainable, no proper tools for testing nor debugging, they’re opaque so, they just not working sometimes as expected and you don’t have straightforward way to fix it. Just trying something differently and hoping they would regenerate the app fixed. You will always end up hiring the best developers (expensive) to dig out of that trap 😢
If anyone is still considering no-code solutions after this video and is looking for an assurance to continue with them anyway, don't. You will just waste your time. When I started working on my side project (as a person coming from Sales and Marketing), I thought that no-code was great. Then I started using one of them and it was so complicated and so rigid in the way it worked, that I gave up on it in a few weeks and decided to learn how to code instead. It took me several months to get the basics and to start to code full-stack apps but now I can practically make anything (as long as I put the effort in it). Other benefits of doing it yourself: You own the product, you have control over deployment, which makes it times cheaper when you start having real users.
I have never had practical experience with these issues but ALWAYS preached this to people that would ask me about no code platforms. Btw, you said you're a dev, what kind? Dart is super easy to pickup and flutter can be a joy once you get used to the widget tree. I am faster with coding flutter than i would be in flutter flow (with the + of it being good scalable code)
"Application Development Without Programmers" by James Martin 1981. These days, the development user interface has adopted modern platform features, but the underlying paradigm for abstracting application features is the same as 1981. No Code (4GL back in the day) works for applications that conform to the assumptions of the platform and its widgets. If you have a design that's outside of those assumptions, you hit the wall. You end up dropping the abstraction and writing code. When do you find out that your requirements will hit the wall? After a lot of wasted time and effort. Another example of this is "User-developed" applications. Maintainability or even documentation is often an afterthought. They work for a while... until the smart cookie who built it leaves the company. Heaven help any business that is dependent on these. Anyone seen any of these written with Microsoft Access? Think AI will solve this problem?
Just because YOU see it as a roadblock doesn't mean it is. It just means that YOU are not a good troubleshooter. I've been using No Code for about a decade without too many problems. And I know a bit of coding, too.
Just admit that you're doing apps with 2 view - one for login/register and second one with a big button, that does nothing :V And never needs to be updated.
I imagine they're great for people who want to drag and drop their interfaces (like Xamarin which promised no-code d&d apps back in the day). But when it comes down to scalability, unit and E2E testing (what, you mean you're just gonna ask someone to test on their phone?), version control for collaboration between devs, etc. they fall short. Plus I actually enjoy coding, that's why I became a coder.
Why people are seeking no code solutions? If the app / service takes off, how are you going to support it later, scale it, extend it? No code solutions are for fun or educational purposes only.
What seems to happen is that the nocode marketing makes people believe they can help you achieve more than you actually can. This attracts naive freelancers to sell apps to naive small businesses.
Perfect for building internal tools in businesses up to 1,000 employees using them as web-Apps. They enable quicker faster decision making, a robust relational database and ease of automation. This gives you visibility, collaboration and efficiency. You just leverage the functionality they give. But I agree for some Apps it’s not going to work. The right tool…
This is the same issue that came about in website design about 10 years back with no code designers like Wix and WordPress. Stay away from these crazy companies. If you download your WordPress code, you will see that it uses 1,000s of lines of code just to draw a simple block layout, that a web developer could do with Javascript, CSS, and PHP or other basic scripting languages (or real programming like Python, Java and C++/C#) in less than a day! Kudos to you for taking on such ripoff technology, wanna be, programming companies!
FYI for people watching this - The Bubble section here is complete bullshit & based on a random Medium article. I'd recommend talking to a few pro Bubble devs & asking them to show what they've built.
In all honesty, it has been a growing tendency to use no-code apps or systems for all sort of development. A wicked evolution of the old CMSes. Those who are still programming are becoming part of the framework herd... and believe me when I say, if you step even a little bit outside of the framework tutorial path .. it is all spaghetti 🍝 code over spaghetti and more spaghetti on the inside. And there's also the new trend "yaml programming" ... which leads not just to simple spaghetti code.. but to MAGIC spaghetti code. I the end, everyone thinks you can become a SW developer just by watching RUclips videos or hiring self thought individuals... which is OK. But it should never match the proficiency of a career SW development. Nobody will trust a doctor who's only a doctor after watching a few RUclips videos, right ? So why do people think SW development should be any different?
No-code is not a trap, it's just new. With time it will become quite powerful but obviously not to the extent of a fully coded app. Choose your type of development based on the use case. Comparing both head to head is just dumb.
It's not that new, and almost by definition, it can never be that powerful. The whole point of a high level abstraction like no code is that you're trading control for convenience.
No-code is as old as code itself. Definitely is not new. It's just unrealistic. For anyone who have worked on any company that has more than a few clients knows how fast everything changes. Clients require new features or changing in the existing ones all the time. Libraries get updated, platforms changes their APIs. It's just too fluid for any no-code promise to catch up.
We tried a no code product back in the late seventies, it was awful! Tried them several times since and every time one very quickly runs out of functionality as soon as you get deep into the requirements. I’d REALLY like it to be a thing, but there comes a point in every app where you exceed the capabilities of the no code platform and then you are down a deep rabbit hole bridging the gap between what is needed and what is possible 😵💫
Uh oh! All the same, I wouldn't be giving up on my dream of learning to code my own apps anytime soon, because I would love to learn to solve problems.
Are you kidding? If you are using a nocode tool because you cannot code, how are you going to analyse and supervise the AI tool code? Coding with a AI tool is for people who have at least fundamental of coding. If you cannot code at all, you will not make something worthy for production using AI tools.
In short, NO CODE app development is still a work in progress. But they are going to rule this segment in the near future as they are investing a lot in streamlining these gaps.
One advantage of no code tools is that it has created many jobs. I'm a freelancer and about 20% of my gigs todays are in fixing or rebuilding apps that were built using no code tools. It's either it doesnt work as expected, or some function stops working because the client forgot to renew some subscription or that the no code provider no longer exists. The more no code tools they release the happier i am.
I suspect that is the case with many (most?) no-code platforms. To really do something meaningful you need to pay for "special" components.
@pythonantole9892 I am a freelance Social Media Manager and I was thinking of learning a No Code development tool called "Webflow" to increase my range of services and get some projects in this domain.......Any advice ?
@@kelvinleigh fun fact: most of the special components can be recreated by fully functional human fingers with fully functional keyboards
How do I find you.
😂
The problem with nocode is you absolutely need to know what you need to do before you start and you need to determine which nocode tool will actually fit your needs. Each tool has different abilities and shortfalls. No code has made me a king in my career. When I need something custom I just code it myself. No code can take 5 minutes to do what code takes a week.
Exaclty bro no code is here changing lives 😊😊
It also requires the app to never require custom modifications for it's entire lifetime
Agree with dumdum above 👆 if you thrive with no code platform, you must be able to negotiate with the client to never use customized components that you can not build
Yeah, the same for me too. I work for agency team that create Enterprise application for small to medium company in my local area. the no code tool helps us a lot but the only reason it works for us is because our boss has 12 years' experience in software development.
If you really know what you are doing, why would you want to not code. The fact that you are considering no code simply tells us you have no idea what you are doing. I can bet it's faster to code than the drag and drop on a long run, so why waste your time. Drag and drop are for businesses that don't want to hire a developer and do things themselves. Why would you call yourself a developer if you don't want to code? Are we joking here!?
As a “founder” the goal is to validate there is demand for the idea or solution as quickly and as inexpensively as possible. This can and should be accomplished with no code tools. If 90% of all the functionality was complete, and there was a true demand for the product, there should be no problem onboarding the first 15% of the target market. This 15% will be enough to validate the idea and raise your first round of funding. Then you will have the money to pay for a scalable version of the application. Don’t fall into the trap of building a scalable version of an app for an idea that has not been validated. Good luck!
Much agreed, im a big fan fan flutterflow and i can tell, they are moving in correct direction and they really care about making their platform solve the "no code platform issue". i beleive in the coming years flutterflow will become even better then what it is at the moment.
by far the best answer here. FlutterFlow is a brilliant option for majority of the types of apps on the market (e-commerce, social media, chat, etc). I wouldnt recommend it if your needs are very specific (video streaming, AR/VR, etc).
great advice
FF is not even useable in an effiecient way in any case... the web version is cumbersome in the extreme and the new Local Run/Test feature is paid version only... #FAIL
@@boombustinvest they created a service that expedites development. If you don't want to pay, just use flutter manually. 🙂
I have NEVER, in 18 years developing products, found a no code platform that we didn't have to go and develop components, customisation, special parameters, you name it. PoC? OK. Go ahead with no code. Simple landing page or personal site? OK as well. My personal experience with any use case other than these ones is bad!
No code is like foam board, an exact knife and elmers glue for architects. I
You build scale modeks, not production code with it. Nobody builds a building using the same materials and techniques they used to build the demo model. No code is for a rapidly developed mock up.
Slightly longer career, but a lots of IDE a very close to no-code experience, I think the closest I got was with MS Access, of course as mentioned customization is non existent.
Moving forward, Dreamweaver generated rather ok code, and from a very superficial point of view, a good no-code experience (but still flawed and open to a lot of vulnerabilities).
Of course these are rather old experience, web and application landscape have vastly changed, and expectations have also changed (for the better).
My take would be that no-code/low code should help get you kickstated in learning more code, reducing the amount of boilerplate code.
The problem pointed out in this video, is that these no-code platform did not account for some degree of play nor customization, or customization that is so obscure that it requires a very high level of skill.
I have some interest in low-code platform that help you with the boilerplate code and offer a way to connect your interface and your customized logic.
This is also why AI will not replace programmers, it can generate some simple code but enterprise grade applications are pretty complex with a lot of business focused nuances.
I know everyone has bought into the AI hype, just like they bought into the Blockchain and Big Data hype. But crypto hasnt replaced world wide fiat currencies and AI will not replace complex programming jobs as easily as our corporate overlords want to believe.
I think the real issue with no code is the spaghetti. I mean sometimes with libraries you still have to make custom components because it just doesn't do exactly what you need but libraries are not generating custom code behind them.
@@BillClinton228I'm glad I've finally found someone else that sees AI the same way as me. It all stems for the 80/20 rule, you spent 20% of the time getting something to 80% ready, then 80% of the time to get that last 20%. It's super easy to make a prototype, it's really difficult to make it work 99.9% of the time. Coding isn't hard, problem solving is, most of what software engineers do is the problem solving part, which AI/no code can't do for you.
I think everyone commenting here must be really young. I'm nearly 60. I've built websites (off and on) for 30 years.
Back in the day, you had to code to develop a website. When the first rudimentary tools came out that would "allow anyone to build a website", many programmer/developer types said the same things folks are saying here. Slowly they were proved wrong and now, 30 years later, you can make a wordpress site do almost anything you can imagine without touching a line of code.
Sure, when really serious issues develop (and they will), you'll still need to hire a developer, but locking onto wordpress 20 years ago, after coming from dreamweaver and front page, and geocities before that, was hardly a trap. If you stuck with wordpress, you're now an amazing website builder in high demand making a great living -- and still not coding.
I expect the app world to follow the same trajectory but, of course, true developers will always be in demand.
Interesting perspective about web development evolution.
It's worth to mention one more thing when comparing WordPress to modern no-code tools: The fundamental difference between WordPress and no-code platforms isn't about whether you need to code or not - it's about freedom and control over your business's digital presence.
With WordPress, you're not locked into someone else's ecosystem. No-code platforms often have hard limits you can't overcome, while with WordPress if you need a custom feature, you can simply add it through plugins or custom development. Your data is truly yours - want to move hosting providers or export your data? No problem. That's not always possible with no-code platforms where your data lives in their ecosystem.
Cost is another big factor. While no-code SaaS platforms typically increase their prices as you grow, WordPress platform costs stay more stable. We've seen cases where no-code platforms doubled their prices overnight or even shut down, leaving businesses stranded. With WordPress being open-source, this simply can't happen - your website keeps running because you own the platform and the only cost that could go up is the server price.
In my opinion WordPress which can be used as a no-code platform has much more advantages than other no-code solutions. We've written an in-depth article exploring these differences coditive.com/blog/is-wordpress-the-right-tool-for-building-mvp-a-comprehensive-guide/
What are your thoughts on this perspective, especially regarding data portability and vendor lock-in?
@@pragmatedevwhat is anyone’s thoughts on integrating micro saas AI product on Wordpress then?
I’m saying this as a web dev who wants to integrate logic, backend but isn’t trying anything too front end fancy but presentable enough to display the results
Well said.
The important feature here is open source. Closed source providers like to harp on disadvantages of open source (such as security) as if their closed source addresses those problems. They don't. They just make problems harder to find and fix. If you're ashamed of of your code, or it's got proprietary parts that you can't open, I'd rather go elsewhere. Don't trap me in your ecosystem, especially when you can't predict which big company will be gone tomorrow.
As a dev, Well said Sir.
In few months they'll remember your words, its just the matter of time. Theres no limits in coding.
No code platforms provide a limited set of functionality predefined.
Is pretty much a template.
The moment you step out of the "template", your run into customizations which can rapidly become chaos integrating with the said "template"
Every time. I remember the first of these in 1981 ! "the last program you'll ever need" .... it wasn't.
yesssss
@@paulscottrobson Exactly right ... the first No code app development tool I've ever used was Visual Basic in Windows 95. I remember devs saying "oh, with this you may not need to code for much longer"
@@ldandco At best it moves it. I (you may also) remember designing dialogue boxes and windows manually or with a very simple tool on Windows 3.x This is now much easier from a prototyping POV
@@paulscottrobsonthere is another threat for coders right around the corner, AI😂 I’m a coder myself and I see some limitations but duh in 2 years, chatgpt or another llm and their increasing contextuality will make some juniors jobless lol
Agreed. Senior engineer here - would never encourage a no code solution. Props to you for exploring and trying to figure it out - it’s a valuable lesson for everyone
Senior here, today is first day I saw "no ode" for something more advanced like mobile apps... And why am I not surprised it's basically just templates... With bad code as output...
@@akirathedog777 only none smart individuals who don't have the learning capacity who will glorify no-code lmao get smart and get good in life bozo
with no code and ai guess yall aint as special anymore huh?
@@London_Gee its total cope
IBM Visual Age was a 98% completion product. The missing 2% would take a team of programmers many years to complete. Every time IBM was asked when the missing features would be added and critical issues resolved, the answer was always 18 to 24 months.
They will abandon it like Apple car.
I knew a guy who worked on Visual Age. He was NOT having a good time....
Was that like 20+ years ago? I remember installing it and only using the C compiler. Lol.
@@albirtarsha5370 Yes, it was. Visual Age had great potential but too many missing features to make it truly useful.
Was this Visual Age C++ or the original Visual Age Smalltalk.
Always wanted to use the Smalltalk to write OS/2 and WinNT but the price tag of IBM was more insane then the bugs and killed the project.
0:58 Flutter is a framework, Dart is the language.
Apart from that, yes, no code is a trap.
It always has been, and probably will be for many years to come.
No code is basically the modern snake oil.
I think its meant like saying you code in react, when really its JS or jsx. Just semantics I think, but I see your point.
but, you can code with claude now.
@@littledovecitydust true, with AI, anyone can build apps. They can learn through the AI if they lack the knowledge as well
@@somedude6420 when you say anyone can build with AI, traditional software developers still have the upper hand since they can understand the code and correct and guide the AI. but yes, the democratization process of knowledge is insane.
She didn’t even say what she couldn’t do. I feel like this video was more emotional that actually providing a logical argument. Flutterflow to my startup has been amazing we are able to build something super basic and gives us the foundation to scale in the future. You have to know exactly how you want the app / product to function . When I went to my CTO I had a massive PowerPoint deck and flow charts and a basic UI that I had made. No code is not bad it just requires a different skill set
Exactly 💯
🎯
Facts I think people are mad that tech can do it
She is dummy. I work for this digital bank and your app is dev on FlutterFlow. A bank. Again, a bank. So sorry, but this video is a argument for people that don't want to study and think that no code is simple. And FlutterFlow is LOW code. Ok.
She did say it you just weren’t listening - she mentions she had a roadblock at creating widgets and couldn’t add them. The code was broken, and not doing its job at that point. When she downloaded the code base she was able to see where it went wrong. She states many others had this issue, and only those who have knowledge in coding would’ve known this. Those who are new to coding in general or app making won’t find this to be an issue, so of course it’ll feel like sunshine for you. Nothing emotional, she’s talking from a developers point of view and she’s 100% right. Any newbie would think this is amazing. It’s not lol.
I've always wanted to hear someone speaking about this.
No code that works for specific limited role should not be taken as no code for everything. It can work for simple apis but in most cases would require custom code to make it work.
@@johnyepthomi892 do you want some real usecase or fundable startup that only run with no-code ?
Like : Closify, Synthflow, Beev, Prello, Cuure, Betterlegal, ...
I am a very experienced developer, and my advice to people is to avoid no-code solutions. I think they are a fallacy. If you are serious about creating software, you are much better off putting in the time and learning how to code and other related development processes.
What about someone coming from a strictly design based background with little programming knowledge (the very basics). Isnt there some merit to producing the front end and then fix it up towards the end? I don't expect to never learn code but I've always build visually.
Also from your experience, could you explain to me some examples of customisations that are limited with no code builders? What kind of scope are we talking here. What kind of features would you say make it practically impossible to work with no code?
I totally agree with you! I'm working with .Net , C# ! I've just got recently got a new client because the client tried use WordPress and it failed with many plugins! I can see people are struggling with Wix too! Avoid drag-drop stuff to build app.
@@deechonadaive worked with webflow, once you move away from its built in features youll start having a bad time tm. An example would be numbered pagination/cms filtering would only take 1 day for a good developer to implement from scratch but with low code it would take a lot of jumping hoops
i see the value tho, for very simple sites and designs nocode/low code is really good just dont use it on projects that get scope creeped hard
@@deechonada, yes there are use cases for low code, regardless of what "experienced" and "senior" developers tell you. Many low code apps have hooks and other optional ways to augment them, add customize behavior, allow changes to their internal DB to further customize. I've got 45 years as a true senior and experienced developer, so I know. To be fair though, the number of use cases is small. But to make the blanket statement "no code is all bad", is just showing one's lack of experience.
@TheSilverGlow Thanks for your advice! ive actually decided to learn how to program using dart and flutter. I like the idea of drag and drop, but I honestly think for the scope of my app (lottie animation heavy, and highly reliant on unique UI), i thought it be best to just bite the bullet and learn how to program. I don't think it will be easy, but as far as I have learned since I did my last message i am actually able to execute short blocs of code! I'm starting to understand the syntax and the purpose of each tool.
On top of that, I was planning to leverage packages for the app when i wanted to use flutterflow, but that would require me going into the source code and manipulating it, which I think I would need to know how to code for anyways. And again, so Ive heard from developers, the code flutterflow produces makes animation heavy apps noticeably slow. I still think FF and other low code builders are worth their salt, but since im already design minded, I just feel like low code is too restrictive for what I want to build.
As far as I can see it, and please tell me if I'm wrong, low code is good at providing services with quite boilerplate UI, and I think i just need to learn how to code. I've been putting it off for ages, and saw low code as a get out. It taught me that I might aswell try seeing as I'm going so far as to be willing to pay a service to essentially code for me. I just have to do it. Wish me luck!!
The problem with no-code is that they sell it for non-technical people and each of them have their own learning curve. and other problem is that. no-code actually doesn' t exists. it's all low-code.
I work with Webflow, and everytime I'm doing some simple javascript. Low code has their merits and it's a powerful tool in the hands of a specialist. but you need to learn how to deal with each one limitations.
If you're using webflow you should feel shame
@@kevinsedwards I'm not using Webflow because of lack of skill to code, I can do a website with code anytime I need to.
It's just that I did not choose Webflow, Webflow choose me and it pay my bills.
@@kevinsedwards It's good when you want to make animations that would probably take a lot of time if done manually, it's better to save time when it comes to something like html css lmao
@@daphenomenalz4100 if you're skilled and have a good framework, you're much faster than webflow and better.
@@waltermelo1033 it only pays the bills because webflow has good marketing. There's always some potato head client who falls for marketing then wants you to use it
To save you time, summary of the "trap": "I can't do this one specific function(s) so the whole no-code platform is trash cuz I didn't understand what the no-code platform was actually capable of when I planned my app."
This video was the real trap. 😒
On every project it seems like there is that one function though.
@@michaelderobio5256 sounds like a problem w/researching the no code platform *prior* to starting the project
Yes and no. You will never be able to do everything you need in FF. It's a backwards model.
You still have to understand technology but now you're layering the code with a drag and drop UI which causes additional headaches like remember what things do and how a widget is made etc...
Better just learn to code and have full control. Even if the code is janky, it will work better and you will be able to modify as you need.
FF is a layer of abstraction that is not a transferable skill.
@@sigma_z sounds like another way to say "i dont know to properly plan my project"
I had a similar experience with building my website. I used a GUI to make my business website the first time, and it took maybe 1-2 weeks.. Now I am not much of a coder, but I understand theory enough to seriously take advantage of coding with ChatGPT. The second go-round I threw together a website going straight into the html, ccs, and javascript with heavy help of chatGPT in 2-3 days. It was a faster website because as described in the video, the GUI to code algorithm creates a mess of code that is nowhere near optimized, and creates multiples more than necessary. With AI advancing so fast there is barely a need for these no code solutions anymore.
I've worked with Power Automate that is another low code/no code platform and while it is quite powerful and allows you to do A LOT there were still close to zero use -cases where we didn't need to build our own custom APIs to automate the things our client needed and we automated more than 60 different business processes with a bunch of systems involved so it was a wide range of things that had to be solved. Version control is also an issue with low code platforms. Also- you should still get experienced software developers to do the low/no code projects of you want to make them more or less maintainable. Non-technical users will make spaghetti almost every time.
WordPress also suffer from this. As a WordPress developer, I have built many websites app using WordPress but always fall short into scalability issues and limitation. Another problem with this is that your are enforce to learn the page builder of their choice. It can either be gutenburg, elementor, divi, newspaper, or so many too mentions. You are forced to learn it but they are technically not transferable to any other programming language or migrating from another platform.
No code solutions are not inherently bad; You just have to understand their limitation. Before committing to building an app with no-code tools, your client and team should identify these limitations and agree on the terms.
WordPress "developer"? 😂😂😂😂 That was a good one.
@@juanamillo You never heard of WordPress developer was a thing? You are probably still in college when that role sprung up. Let me lecture you. WordPress devs don't just deal with no-code, we fill the roles for Theme/Plugin customization or creating one from scratch. We also create or extend widgets to these page builder so that we meet client specification. We also incorporate WP CLI and REST API. These things involves coding and a lot of WordPress shenanigans to account for. Working and mastering a page builders is one the biggest thorn of my career. It is hard to negotiate a client and tell them to build everything in code because they will think it is an expensive route and they will have hard time with maintenance or making enhancements since most clients has no coding experience.
We sometime laugh at something we don't understand, I feel you. Even myself laugh the hell out when I first heard "prompt engineering" was a thing.
@@juanamillo Lol who do you thinks builds the plugins? Writes the themes? Tell me you're inexperienced without telling me, also immature and condescending.
😂😂@@juanamillo
😂 easy@@Arjun44598
Thanks a bunch for this! You confirm my suspicions without me having to suffer thru the process.
Don't let anyone tell you 'no-code'. Low code yes, I've been working with Microsoft Power Apps (canvas apps) for several years now and have had great success. No issues scaling, but it does take code! To do anything more complex than a basic CRUD app you will need (or need to learn) development experience.
Same here!!
As someone maybe getting into Microsoft Power Apps. How is the data input validation? Ex. I want the user to input a correct formatted email address. Needs coding?
@@henrikbergman4055 Plan to do coding for validation. If you're working with Dataverse on the back end some of the validation is easier - but just plan on coding for everything you want beyond the very basic and don't make too many assumptions. Code/test/code/test to be sure it all works like you expect.
Man, I despise the scripting language so much. But that's just me.
Which low code platform do you use?
I work at a very big non-tech company. We have at least 5 no code platforms for our internal tools that I know of. It is always the same. They buy a platform for huge amounts of money. Then they realize it has the same problems as all the others. Then they end the contract and buy a new one. And the best thing is that every time they do this, everybody who used the platform has to redo everything on the new platform because you can't just migrate your projects from one to another. None of these platforms are focused on developers - they are made for non technical people. However, developers are the ones who end up using them and it always blows up in their face.
As someone developing from scratch on web and mobile, I agree and disagree. I use flutter flow to turn my figma design into the code for my UI and it saves an inordinate amount of time compared to coding it up by hand. I also love using wordpress for building static sites as I can do in a day what would take me a week or longer otherwise. Does that mean you mean you should build the whole of a fully featured app in them? Obviously not. They might have obvious limitations, but the nature of their shortcomings is exactly that. Obvious. Some wisdom on when to apply them will help you much more than scrapping them completely.
Have you tried the WWX stack? Webflow, Wized, and Xano or Supabase or Firebase for that matter. It's not exactly no code, more like low-code but you can build customized web apps.
Hi, informative video.....
Just a very (probably insignificant pick)...
The language is *Dart*
The app development framework is *Flutter*
Would've been nice if she mentioned what kind of developer she is, and what programming languages she knows. Maybe i'll take a look at some of her videos for an answer...
@@mitigamespro8757 bam, she got you engaged in her channel
As a seasoned Webflow developer, I saw this video in my feed and immediately was upset by the clickbaity title - I decided to watch the full video and here are my thoughts:
1. The video should be renamed “No Code MOBILE App Development is a Trap” - you can build web apps and even scalable ones. The title is a clickbait and I really hate it. Some people might call me “nitpicky”, but Dee never mentions “mobile” in the description and title - even in the video, for the most part. She makes no-code look bad in general. I believe it’s a really bad thing to do.
2. While I didn’t have personal experience with building mobile apps, I do agree with Dee on one point - these platforms aren't that much honest in their marketing. I think it’s the issue even with non-mobile no-code related platforms, sadly. That being said, it’s not like they are blatantly lying - you can, in fact, build production-ready apps with these mobile app no-code development platforms. However, they do not tell you about all the nuances you might encounter such as lack of customization (which is fine, even Dee admits it). So calling “no-code mobile app development” a trap is a stretch.
3. No-code in general isn’t bad as Dee tries to make it look like - moreover, as I’ve mentioned before, she talks SPECIFICALLY ABOUT MOBILE APPS. I use Webflow and it is amazing! Does it have limitations? Obviously, as any other platform, but you can build so much with it - by utilizing 3rd party and integrated apps too, you can build web-apps even. A popular combo stack is WWX - Webflow + Wized + Xano. You can build production and scalable apps. Sure, not everything can be built with this stack or no-code platforms in general, and it makes sense. It’s not a panacea.
4. Anyway, people must do prior research and learn about limitations of tools they use. It’s hilarious to see people coming expecting a panacea from every tool they use - then they go to Reddit and Twtiter and start shitting on a tool. I’m not saying that Dee lies about stuff in the video - she does not, and I believe here, I’ve heard about no-code mobile app dev platforms, they aren’t astoundingly great, but they do the job for the most part. Many apps can be built with them. Not all, not super custom, yes. But still many. And I believe it’s amazing. Sure, these companies kinda lie in their marketing about possibilities, but it’s not like these tools are garbage or a TRAP.
5. I’ve seen many people complain about Webflow on Reddit and (a few times) in Telegram group chats related to the tool, that it is bad, trash, garbage, and so on. Turns out, in 99% of cases, that these people didn’t learn about the tool, didn’t do prior research, just deep dived into it and expected everything on a platter without moving a muscle - hilarious! These tools, Webflow too, aren’t ideal, but the amount of possibilities they give to the users is really empowering.
I have way more thoughts about this video and similar ones - maybe I’ll make a video addressing them one day, but for now I think this long-ass reply is enough.
I just want to add that coders should stop gatekeeping development from other people - no one is stealing your jobs, and even if they do, then you need to step up your game and adapt. Gosh, the amount of elitism is insane. The comments just reek of it. Disgusting, honestly.
P.S: Keep in mind, that I’ve talked only about APP DEVELOPMENT here - there are so many other things like websites/frontend development, automation/integrations (n8n is amazing!), etc.
best comment so far. Totally agreed
Awesome contribution
The main use case for no/low code is the "citizen developer" that wants to put forms and lists over a spreadsheet and create more of an "app" for internal use.
Absolute game changer for someone who wants something more robust than a convoluted over-sized spreadsheet.
Yes. The limitations are scalability which is unlikely to impact most internal apps. The next issue though is multi-tenant, multi-user, the My Account subsystem needed for a classic SaaS. Should not be that hard to do, but just a few of the many no/low code tools appear to be giving it a go.
And that's how a lot of corporate applications are born and later become huge problems to mantain and evolve.
As a Developer (Java since 2009, LowCode since 2021) i can disagree. We build huge Applications with OutSystems (LowCode) that are much much more than just a Spreadsheet or internal use.
@@axeo123 Maybe you picked the wrong LowCode Platform. We handle multi-tennant, multi-user with ease.
Here's a different question though: Do you think you could have an MVP **without** those missing 10%?
i.e. Could you validate the market demand with only 90% finished?
Or were those missing 10% so vital that it wouldn't even make the whole thing into an MVP?
Could you have done a few tweaks to get a functioning MVP with just 90%?
This is the absolute right question. I'm not a no-coder, but I hate strawmen arguments, so it's important to know what ended up on these 10%. If it's some snazzy CSS animation that really won't make an iota of functional difference, it is quite different from not being able to pull data from a system of record.
I was just asking myself the exact same question. Could you attract financial and dev partners with an MVP that was 90% of what the final launch product would be? The biggest issue is that there is no way to really know where the 90% lies. Is it just before those features that really set your product apart, or after they're in and working?
I have 13K installs and have had no problems and estimate I won’t pull away from this nocode platform until 50-100K.
No offense but this is a gather point for insecure developers.
Being both a person that is a developer and commissions software devs …thank god for no-code. To be able to Prototype working software in days is the best way to get the developer and client on the same page.
I think I have lost way more time and money on software developers not delivering what is scoped. Please don’t lecture me on how to scope.
Our days are numbed anyway, which is cool by me.
"To be able to Prototype working software in days is the best way to get the developer and client on the same page." She literally said no-code platforms are good for that
@@flaviaoliveira5949 just noticed your comment. I didn’t say that she was wrong i was referring to several people making comments. I think she is great and it takes a lot of guts to put yourself out there.
Ill be more careful in my wording.
If you just want to prototype, use Figma. It can deploy to mobile and will allow anyone to build UI/UX.
When coding something more than a Hello World, then it's better to code by hand than to use a WYSIWYG because FF will have limitations and in your project, the trouble is, when you find these limitations with FF for your project, you'll be too invested so you're either going to get fired, lose the market opportunity or start again from scratch using real coding.
moral, don't waste your time with FF.
It's like the saying goes, if you think hiring a professional is expensive, try hiring an amateur
It definetly is... at least no code web tecnology is going waaaayyy far... so far I honestly decided to stop fighting it and learn it, as Devs we can do so much more than people with no coding experience in these platforms, we just need to take our time and LEARN the tools.
No code is pretty much one evidence that “devin” or whatever these AI developers are called will not replace developers anytime soon. They either write horrible code that’s not customizable or it’s buggy code that still needs knowledgable people to debug it. I’d say, give it about 5 decades before it’s truly a thread to us engineers
And by that time it will be a threat to the system as we know it anyway so we'll either be set or cooked anyway.
5 decades??? More like 5 years.
@@chrisc9725 I see why you would think that but remember news about not having data anymore to train these models?, that's why I'm expecting decades.
This will happen within 2 years.
@@MarkBetterDevYoooo you right about the running out of data thing. Dayum
Every programming language has some limitations too and depending on the dev skills it can still be difficult to scale or not secured.
In no code, it just have a lot more limitation and needs a know how to make it more scalable and secured.
Thanks!
Flutterflow's no code is mostly in the UI side. Still have to jump into the dart file often to get custom functions to work properly. Good thing is that many of the AI's out there are semi-decent at writing the solutions you need.
The thing with no code solutions is that you need to be very willing to adapt your business/project to the platform, and not the other way around that is the normal approach when developing apps with code. You always need to settle for 90 - 95 % in the best case where you have picked the best platform for your project. For most projects there is no way to know exactly what you want ahead of time, and definitely not where you are going to end up after you talk to your actual users. The chance that you will able able to get anywhere near 90 % is therefore pretty low.
The only way to make good apps is to develop them iteratively while you gather as much info and data as possible from user users and target audience. That is pretty much impossible with no code platforms and is the largest flaw with no code platforms in my experience. You will end up when a pretty bad app that will be really hard to sell, so it is probably hard to use even as a PoC or MVP. It's usnually way better to try to make a pretty polished a PoC or MVP with a very limited feature set and let real users use it.
So it is like SAP or MS Business Central .
It kind of reminds me of the drag-and-drop Web Builders that were pushed Years ago! They would get you started, but required you to have to go through the code to fix the bad spots! And there were a plethora of issues! While they have gotten somewhat better, you still need to know the code. Programming Code is more intense and there are so many paths one can go down that drag and drop would automatically be limiting and if the code is bad so will the app be bad!
Yes exactly! Those web builders were horrible. Definitely learned my lesson ☠️
Drag-n-drop web builders are used by enterprise companies all over the world on a daily basis. There are agencies, freelancers, and an ecosystem of plugins and add-ons for website builders like Webflow and Framer. All websites on these platforms are hosted on secure servers, deploy similar code, and follow best practices. In some cases, a website on Webflow is more stable and performant than a website you commission to a random developer. Not all platforms are the same, and the idea of "limitations" is quite abstract. There are limits in everything you choose, including hiring a developer. Cost limits are real limits, the dev skillset is a real limit, etc.
@@Pensandoci9 True! But I was talking about the Web Builders in their infancy and I know they have gotten a lot better, but they still are not perfect, and knowing how to code, is still a good idea.
@@Pensandoci9I agree with you 💯
No-Code just can't cover all use cases. It's like borrowing someone else's toolbox. Writing the code yourself removes these limits. Edit: Great video by the way, been wondering when someone would discuss this.
Can you show us the features you couldn't create with nocode?
Anything that needs some customisation or more than basic business logic above stupidly read and write data to a database.
Like what...@@silentwater79
@@silentwater79 lies everything you can do with flutter you can do it with flutterflow
I wonder as well. 99% of all IT solutions are just basic CRUD stuff. No idea what people are building that requires all this "customisation". The real downfall of no code are licensing issues and vendor lock-ins, but that wasn't even mentioned.
@@Dav-jj2jb Exactly, flutter-flow rocks for fast and decent development
Being 100% dependant on one framework or 3rd party environment is a dumb move if you're trying to establish a business or in this case create an app.
I have basic idea of coding and I've been using bubble for well over 5 years. I think it depends on what u are building. I have had some hurdles to overcome when customizing but overall my experience has been good.I have built at least 15 apps both for clients and myself.
dear sir ..can i go for bubble for my some basic apps?plz tell me to save my time....waiting for your reply
@@anantghere1 One thing I can say is that if you've never ever built an app before on any nocode platform or on wordpress, and you are not tech savvy, hire someone, because the learning curve might be high for u. Otherwise go for it It is one of the best, when it comes to customization. You can look up some apps that some people have done to confirm
@@anantghere1 i can answer on behalf as i am a bubble user(3 yrs) as well, basic apps shoudnt be a problem i wud say,
@anantghere1 it depends on what you intend to perform in your app. For most CRUD apps, Bubble will suit you. However, if you grow over time, you might need to switch database, to plug to apis and/or to run custom scripts.
The problem is that people think they can stsr building right away with zero previous knowledge, but no code tools also requires you know how to use them Right
I am a computer teacher for students in a British international secondary school. Next school year I will teach the basics of how to make apps. Since it is going to be one of these no-coding alternatives, I am grateful that I saw your presentation.
Twas ever thus. Years ago they used to usual Visual BASIC 3.0 drag and drop ; that's pretty much all they did, but the "wow he/she wrote a windows app" fooled examiners.
Maybe look into MIT''s App Inventor. It's geared towards education.
As a dev, please teach them to use chat bots. And to use them only after unsuccesful trying doing it manually or asking for help/suggestions. And teach about process, not just the final result. Or just for simple, repeatable tasks. Or data summary or further explanation.
In my evaluation, most "no code" are garbage that leaves out the critical parts of actual software development like proper debugging and testing support, or even version tracking. Avoid at all cost.
But, there are some "low code" platforms that are more similar to Rapid Application Development from the 90's that don't pretend that actual coding and debugging is not needed. In my evalulations, so far, AppSmith seems to be the best balance and is under an Open Source license.
Problem is that managers think that AI and No-Code are the future of programming, and if you try to convince them otherwise then they tell you that you're old and not *want* to move to modern techniques. They tell you maybe it's better to hire young "developers" that have more knowledge of the modern "revolution".
Let's face it after 100 years we still have rubber tires. Old can be good for a long time ....
I like the Concorde in the back. Where did you get such a big one? I once travelled by British Airways (when I was a kid) and they gave a “build a concorde” paper toy and it turned out very nice. I have two little matchbox concord now - one is BA and the other is Air France. So sad that Concorde was discontinued.
Every time I’ve tried low/no code platforms, you always hit a wall. Trying to get past that dead end is a complete waste of time and it makes you wish you’d just have done it properly from scratch. The problem is that it’s impossible to know what the limitations are until you get there and are backed into a corner.
Don’t do it.
Could be useful for prototyping, maybe ?
Great review and loved the storytelling. What's an alternative for those who want to develop an APP without taking years to learn how to code tho? I was looking forward to the solution
Not to take from the validity of her overarching point, but to spend over a year with Flutter Flow and not learn the difference between Flutter the framework vs Dart the underlying language, and call Flutter "the backend", reflects poorly on her comprehension.
At no point did she argue against Flutter the framework or Dart the language, only against Flutter Flow the RAD tool (and possibly its default integration with the Firebase backend), so if we code in Dart/Flutter using VS Code we're good to go Dee? RAD tools have generated spaghetti code since their inception 30 years ago, with Motif/C toolkits on Unix. This is a story of no skills meets no code. I do agree on her point though, as one is largely sold a lie, which can still work if one stays inside the limited scope of the box.
I am a flutter dev and I remember once using flutterflow, the main benefit I found was producing complex UIs with drag and drop. I hate writing UI code without drag and drop so it did help me there. I downloaded and used the UI code and slotted in my own busines logic etc. So yeh, I use it sometimes to make complex uis. I cant recall if the code was bad or not.
One other BIG problem you did not mention, I think, is that you are completely dependent on this platform for the future. Apps regularly need updates, because requirements for the app store change (and many other reasons), now you have to HOPE they have this new requirement available and it still plays nice with the rest of your app.
I am a 15+year unity3D dev, and I hardly use plugins at all for this specific reason. You are now reliant on it, I update the plugin and the app breaks (since they changed too much), or they just stopped updating it and it doesn't work anymore etc. No this is just one plugin, with these tools your whole app would rely on a third party.. that is a major red flag for me.
Not true, flutterflow can push to GitHub and all code is available and you can continue off platform if you like.
You are exactly right. I would also add that on many of this No Code platforms you are vendor locked and you can’t migrate your app to another platform, so you find your self trapped and have to pay for it.
You didn't even explained what your problem is. It could be that you were wrong with your implementation of flutter flow.
No code is definitely a solution that saves a lot of money if you know how to do it right :)
I'm a typescript programmer and I could tell from the title where this was going lol. All of these "ZOMG, I make $20k/month with a no-code app I built in 2 days" videos on RUclips were red flags. No-code is fine for simple things. Like Wordpress, Squarespace, Wix, etc. But when you have an idea for something that isn't supported on their platform, you're stuck and eventually get around to having to learn how to code lol...or hire someone who knows the language already. That destroys this whole illusion that you can't "build anything" with no code.
You're right. The "I build an app in 2 days" pitch is absolutely wrong.
But you're wrong in that it is fine for simple thing.
I've build a financial marketplace for alternative investments that comes with financial performance analysis and modeling all with nocode. It took 6 months, but it can be done. We've been operating with paying customers for two years.
Flutter isn't a language, its a framework. Dart is the programming language Flutter is based on.
If one says it was written in React, is it also wrong to say or must it be JavaScript? Ever heard of the flutter IDE?
@@lavenduct2001 cool story.
I would love to hear your thoughts on Cursor and Bolt for No coding. As the game changed with these new Apps?
Adding my 2 cents here.
I've been a web developer for 15 years but, due to greater forces (media hype), I needed to carry out some projects on low-code platforms, such as Microsoft Power Apps. I can also prove that no-code/low-code only works for simple projects.
No-code/low-code is REALLY a trap.
But one thing we can't deny here: their marketing is very strong.
Literally I don't understand why it is so hard for people who want to do "programmers stuff" to actually learn programming.
It's really not that hard.
No one asks people to play with memory allocation, pointers and low level C stuff.
And there are great libraries for Python, Java and much more.
Also, it was a time I was obsessed with abstractions. But I simply realized that the more abstraction you add to code, the less it can do. While it serves you good for a starting point, when you try to do more fancy stuff, you'll fight the abstractions 100x worse than if you wrote a little more complex code.
That's one reason why I use vanilla CSS and I've never even tried CSS frameworks.
Vanilla CSS is a bit more tedious in the beginning but it rewards you later if you're building a complex web application.
Also, sometimes when I need to build GUIs (and the design matters) I use 2D libraries where I create my own widgets.And it's not really that hard.
Not everyone has the intelligence and/or dedication required to learn how to write software.
@@gaiustacitus4242 Then don't expect to be able to make software.
@@alexale5488 I'm explaining why not everyone learns to program. I've been writing software since 1976 when I started by writing in machine language.
bro is it still worth learning to code today with Ai around? I just made an extension with gpt 4o and it actually worked O.O
@@guillermosanchez1224 Does it? And does it work efficiently? Will it scale under higher user load? Does it actually do 100% of what you intended it to do?
The closest I've had AI come to generating usable code included a bug that caused data corruption. This bug was difficult to detect.
Most of the code generated by AI looks like the garbage produced by a programmer with a shiny new master's degree in computer science but lacking real world experience in a professional setting. If the code works, then the approach taken is very inefficient (especially when it comes to operations against a relational database management system).
Just a small detail - "Flutter" is not a language, but a UI framework, based on the programming language "Dart". I tried it once and found it quite cool - got something nice running with very little code (compared to horribly cluttered coding in good old Java). But then when I re-opened that project 3 months later, a lot was broken and did not build anymore, once my IDE had updated Flutter. After looking at the release notes for Flutter, I realized that they seem to introduce breaking changes all the time... and this is a no-go for me. I guess I will keep looking for a better alternative.
As a computer scientist I can't help but think if you are thinking a no code solution can be a backbone of a startup of any kind you aren't technically skilled enough to be running a startup to begin with.
That’s because you fallaciously believe that startups are about knowing technology and not about knowing how to run a business.
The future of NoCode is Code + AI
Her mistake was knowing how to code. If she didn't, she would have hired coders and did what needs to be done and got on with the business.
Instead, fell into a trap.
Sometimes not knowing is better than knowing something.
Wouldn't you agree dear Scientist?
Not sure what is meant by never being able to download the code and deploy the app. Why wouldn't you be able to do that?
Saying no code app will fail is like saying cloud networking will fail. Everything has its place and niche
Just people mad they spent so much time doing something , now it’s easier but a learning curve to what they already know
I'm no programmer, but all of the negatives I am reading are from programmers.
Seems like a battle between the Dairy and the beef industries in that one is trying to protect their interests.
No code will appeal to certain people who need something no as complex.
And in the not so distant future complex coding will also be possible.
Great and honest review ... No code is like:
Make a cake... without learning how to cook
Go places... without learning how to walk
Build a house.. without touching a brick
Bring ideas to life... without learning how to do nor think
a get-code-quick scheme sounds alarmingly similar to get-rich-quick scheme... they all get you 10% of the way and leave you there..
As an oldie I remember one of these that ran on a tape drive on the PET (1981) which claimed to be the last program you'll ever need. There've been many panacea since. I can't think of an older one than that.
It depends on the user. If you have code knowledge - transitioning to low-code is like having a superpower.
Low code with traditional code knowledge is like this:
Go places... knowing how to walk - but you have a bike
Hi all, What are your favourite tools to make a 'code' web app?
I am a developer but worked for a no-code company for 8 years. Now, that I am out again I am rapidly noticing there are far to many tools to learn them all. Any suggestions?
For example, What service/tool would you suggest for:
- A great looking interface designer (I have wasted too many hours in HTML and CSS)
- Where do you run your logic layers? (AWS? or something more specific)
- What Database would you suggest?
- Anything else?
(I don't own any servers)
This goes for any third party library or component, tools that allow you to do things more quickly. They get you 90% of what you want and have you spend the rest of the life cycle time of the application working around its limitations and dealing with its side effects.
No code just means you are following someone else's way of processing, living within their process, and limited by their ability to provide drag and drop services to limited experience people.
For example: Excel is a no-code solution. It lets you input data, apply formulars, set presentation options. If provides charts and pivot tables at the click of a button. But if you want a business accounting system connected to your bank and inventory system, you can't use the no-code options of Excel.
My point is, 'no-code' has been around in various names for over 25 years. Software engineers are still needed to create real business and game software.
It depends on what you're trying to build. For enterprise level highly custom apps, no-code platforms are terrible. But Ive had plenty of small business clients who have found them ideal for their needs.
"Physical code"?
Does this mean you print it out, so you can pick it up and hold it in your hand?
Love your summary and your story telling. Thanks for your experience with flutter flow, this probably saves me some headache.
Are you looking to build a web application anytime soon
As an architect, when I'm interviewing clients, if I find no-code anywhere near a real product, it will cost them twice my normal rate if I'm feeling nice.
Thanks for this honest feedback. I'll still keep using FF though to test and validate my concepts. I think we shouldn't forget a very important point: these tools have existed for a very short time, and FF is improving extremely fast. If the progression goes along with AI's (which btw is becoming incredibly good at coding), I guess we could soon (under 3 years) have revolutionary ways to build apps. Maybe could you make a feedback video about these no-code tools each year to see if you progressively change your mind 😜
I love your response bro. She needs to make her research on FF solid enough.
I totally agree, I am doing the same
No-code like Bubble is the Canva for designers, it won't do even close to what Photoshop can do, but it gets you to point A to B more intuitively and efficiently, and if you know what you want and your point B is in Canva's capabilities, that's all you ask for. I'm happy things like Canva and Bubble exist, it gives more power to us to work without paying someone else to do it or teach you; the future is not no-code, it's agency, no-code is just like AI in many ways and they'll probably intertwine very soon.
Disagree. FF has been phenomenal for me for visual frontend ui development. I use APIs, cloud functions and firebase for the rest of my App's functionality. You have to know what you're doing and adapt to the tool you are using. If you want to use an Axe to cut paper, well sure you're going to have problems
But the issue was marketing I guess. Those who are new to development would get stuck.
I once tried to write my own Word Processor...after spending countless hours, I chose to go the no code solution: I bought a copy of Microsoft Word.
I think a low code platform with an AI copilot could be a better solution in the future than a no code platform
AI writes VERY BAD CODE. Barring insertion of simple functions, it typically takes longer to fix the issues with AI generated code than to just write the code by hand.
If you aren't proficient at writing code, then you either won't see the flaws in AI generated code or won't be able to fix them.
@@gaiustacitus4242on the other hand, ChatGPT seems to be more knowledgeable about React esoterica than 95% of devs out there, including myself, so that’s a plus
Just a small nitpick. Flutter is not a language. Flutter uses the Dart language.
There are plenty of good no code systems. Geometry Nodes in Blender, Blueprints in Unreal, Bifrost in Maya....
there aren't
@@scharlesworth93 ok…what is wrong with the examples I mentioned?
I work for a company where the CTO adopted no code solutions for a lot of stuff they want. He's a great guy but he's no coder. So sometimes users cannot work anymore, integrations stop working or random errors occur on production - there's no ticket, no source tracking and lots of sudden technical debt. Architecture gets worse and performance gets worse with every change and everybody is losing track of what's happening. This is no code done wrong.
Really apreciate the advice, im learning to code and have plan to learn no code tools also, I was specting no code tools to be insuficient, but It surprises me they have the isues you have mentioned, everione wants to sell the new magic money maker but I hear no body talk about the limitations, thank you for sharing your expirience, Is encouraging to hear hard working and honest people.
I am not a developer, but I made an app in appsheets and it works nice. I would not say it was no code though. It was a lot of hustle, lots of problem solving nad sometimes complicated formulas. Also I combine my sheets with java scripts. So they are not basic at all. I chose appsheets for 2 reasons. My employer already had appcore plus google is pretty stable in the market, so there is less worry if the platform will survive.
And as for AI, you really need to be good a problem solving and know what you want to achieve and see the mistakes AI made to point them out and correct it. So you need some knowledge to work with AI. It can make your life easier but will not replace thinking.
I think the point is that you can iterate faster on no code so you learn more about the problem that you’re solving in a shorter amount of time.
If you know exactly what you’re building and you know how to code…..sure, use code.
On scalability, you’re going to rebuild your app a bunch as you scale regardless of what stack you use…again unless you know exactly what you’re building from the jump.
All of these are tools at the end of the day, you need to choose the right one for the job.
But I also agree with you that No Code marketing makes it seem like it solves all the world's problems.
Hi, do you think Webflow has the same problem? I want to create a learning platform for people to improve their Excel, Python, SQL, … skills
My idea was to create the layout of the page using no code in Webflow and then hand it out to a developer so they can do the back end create a functioning learning platform.
No-code is definitely work in progress, but also definitely the future of development. Customization and integration limitations are being sorted out gradually, and AI won't but boost this revolution of coding from a visual or natural language interface. The article on Bubble is from 2021, a lot has changed since. I don't think it's a trap, although it might not yet be good enough for your project.
They have been telling this to us under various names for at least 30 years…
Fully agree, I've built a whole document management system using no code (focus on UI is no code but back end is Linux) and it's integrating nicely using no code API connectors.... so yes they shouldn't shoot it down because it didn't work for their particular use case. There is no time for learning how to code now... AI is taking over and people still want to learn how to code....
It will never be the future of development. Companies have been trying to sell this kind of thing forever with it never really working out.
@@maphuthumanemasilela8934exactly the point. If AI is to take over the code, why would you even need a low code platform when AI can write the code? Low code platform just doesn’t work. You will end up writing more code just to label your product as “low code”
@@CaptRespect But back then AI was not well developed right? Maybe things will be different in future. No code could be the future no?
The trap is also that they can change their whole pricing system like Bubble did a year ago
You talk a lot of not being able to complete it and being restricted, but not a lot of the actual issue you was facing. What couldn't you do in FF?
exactly.
I think this is a circljrk of seasoned engineers who are discouraging the new tool on the block. A tool not meant for them. Its like a premier pro power user bashing capcut or the basic phome video editor for its lack of features.
@@beavenjnr4187 Not even close, capcut actually makes videos and pros use it too, genius, if you want to use nocode go for it.
She explained exactly the problem, she got most of the way there and couldn't complete the app because of code errors knucklehead
I'm not a premiere pro guru, but I'm decent. I tried Capcut and haven't really gone back since. For what I need, it can output good enough quality In a fraction of the time and computing resources.
The democratization of different fields in the past 20 years has made a lot of the old guard gatekeepers really annoyed. Because for a lot of average practitioners, they found themselves being out competed by relative newcomers with really good tools.
Hi I really appreciate your sincere review. I am python programmer and a low code power Apps developer. I am really i terested inlearning mobile app development. Would you recommend me only to learn flutter? Have you learned already flutter?
I’m glad you talked about that. What those platforms say about themselves is a wishful thinking or rather straight lying. I’m a developer, and I participated several times extending applications that someone like yourself created with “no code” tools and hit the wall. I can say, it’s always a huge pain for developers and it takes times more time than actually rewriting from scratch.
As a developer and a startup partner I continue hoping for some tool that would help building faster, but nothing so far. They all barely maintainable, no proper tools for testing nor debugging, they’re opaque so, they just not working sometimes as expected and you don’t have straightforward way to fix it. Just trying something differently and hoping they would regenerate the app fixed.
You will always end up hiring the best developers (expensive) to dig out of that trap 😢
If anyone is still considering no-code solutions after this video and is looking for an assurance to continue with them anyway, don't. You will just waste your time.
When I started working on my side project (as a person coming from Sales and Marketing), I thought that no-code was great. Then I started using one of them and it was so complicated and so rigid in the way it worked, that I gave up on it in a few weeks and decided to learn how to code instead. It took me several months to get the basics and to start to code full-stack apps but now I can practically make anything (as long as I put the effort in it). Other benefits of doing it yourself: You own the product, you have control over deployment, which makes it times cheaper when you start having real users.
Great analysis once more Dee. At the moment, no code is not mature enough for high grade, highly customizable applications.
Definitely. I’m sure it will get better in future but I don’t think it will be anytime soon. Thanks for watching!
@@codingwithdee That's why I have stayed away from no-code, when the time is right if ever, will jump in
Actually the declarative programming is the no code I use.Did I mentioned that microservices are perfect for no code?
I have never had practical experience with these issues but ALWAYS preached this to people that would ask me about no code platforms.
Btw, you said you're a dev, what kind? Dart is super easy to pickup and flutter can be a joy once you get used to the widget tree.
I am faster with coding flutter than i would be in flutter flow (with the + of it being good scalable code)
She's in data analytics so probably python and sql or R, etc.
"Application Development Without Programmers" by James Martin 1981. These days, the development user interface has adopted modern platform features, but the underlying paradigm for abstracting application features is the same as 1981.
No Code (4GL back in the day) works for applications that conform to the assumptions of the platform and its widgets. If you have a design that's outside of those assumptions, you hit the wall. You end up dropping the abstraction and writing code. When do you find out that your requirements will hit the wall? After a lot of wasted time and effort.
Another example of this is "User-developed" applications. Maintainability or even documentation is often an afterthought. They work for a while... until the smart cookie who built it leaves the company. Heaven help any business that is dependent on these. Anyone seen any of these written with Microsoft Access?
Think AI will solve this problem?
Just because YOU see it as a roadblock doesn't mean it is. It just means that YOU are not a good troubleshooter. I've been using No Code for about a decade without too many problems. And I know a bit of coding, too.
exactly
Just admit that you're doing apps with 2 view - one for login/register and second one with a big button, that does nothing :V And never needs to be updated.
I imagine they're great for people who want to drag and drop their interfaces (like Xamarin which promised no-code d&d apps back in the day). But when it comes down to scalability, unit and E2E testing (what, you mean you're just gonna ask someone to test on their phone?), version control for collaboration between devs, etc. they fall short. Plus I actually enjoy coding, that's why I became a coder.
Great video Dee, very well explained. No code app dev is full of holes
Why people are seeking no code solutions? If the app / service takes off, how are you going to support it later, scale it, extend it? No code solutions are for fun or educational purposes only.
What seems to happen is that the nocode marketing makes people believe they can help you achieve more than you actually can. This attracts naive freelancers to sell apps to naive small businesses.
Perfect for building internal tools in businesses up to 1,000 employees using them as web-Apps. They enable quicker faster decision making, a robust relational database and ease of automation. This gives you visibility, collaboration and efficiency. You just leverage the functionality they give.
But I agree for some Apps it’s not going to work. The right tool…
This is the same issue that came about in website design about 10 years back with no code designers like Wix and WordPress. Stay away from these crazy companies. If you download your WordPress code, you will see that it uses 1,000s of lines of code just to draw a simple block layout, that a web developer could do with Javascript, CSS, and PHP or other basic scripting languages (or real programming like Python, Java and C++/C#) in less than a day! Kudos to you for taking on such ripoff technology, wanna be, programming companies!
FYI for people watching this - The Bubble section here is complete bullshit & based on a random Medium article. I'd recommend talking to a few pro Bubble devs & asking them to show what they've built.
In all honesty, it has been a growing tendency to use no-code apps or systems for all sort of development. A wicked evolution of the old CMSes. Those who are still programming are becoming part of the framework herd... and believe me when I say, if you step even a little bit outside of the framework tutorial path
.. it is all spaghetti 🍝 code over spaghetti and more spaghetti on the inside. And there's also the new trend "yaml programming" ... which leads not just to simple spaghetti code.. but to MAGIC spaghetti code.
I the end, everyone thinks you can become a SW developer just by watching RUclips videos or hiring self thought individuals... which is OK. But it should never match the proficiency of a career SW development.
Nobody will trust a doctor who's only a doctor after watching a few RUclips videos, right ? So why do people think SW development should be any different?
No-code is not a trap, it's just new. With time it will become quite powerful but obviously not to the extent of a fully coded app. Choose your type of development based on the use case. Comparing both head to head is just dumb.
It's not that new, and almost by definition, it can never be that powerful. The whole point of a high level abstraction like no code is that you're trading control for convenience.
No-code is as old as code itself. Definitely is not new. It's just unrealistic. For anyone who have worked on any company that has more than a few clients knows how fast everything changes. Clients require new features or changing in the existing ones all the time. Libraries get updated, platforms changes their APIs. It's just too fluid for any no-code promise to catch up.
We tried a no code product back in the late seventies, it was awful! Tried them several times since and every time one very quickly runs out of functionality as soon as you get deep into the requirements. I’d REALLY like it to be a thing, but there comes a point in every app where you exceed the capabilities of the no code platform and then you are down a deep rabbit hole bridging the gap between what is needed and what is possible 😵💫
Uh oh! All the same, I wouldn't be giving up on my dream of learning to code my own apps anytime soon, because I would love to learn to solve problems.
If you think no code is new then it just means that you are young.
Has anyone tried a hybrid approach where an AI coding assistant is used to analyse the generated code and implement the last 10%?
Are you kidding? If you are using a nocode tool because you cannot code, how are you going to analyse and supervise the AI tool code?
Coding with a AI tool is for people who have at least fundamental of coding. If you cannot code at all, you will not make something worthy for production using AI tools.
In short, NO CODE app development is still a work in progress. But they are going to rule this segment in the near future as they are investing a lot in streamlining these gaps.
Nah don't think so, No code will struggle in large scale apps
Bubble was launched in 2016. And it mostly did not change since then.
Not good enough.