Scriptcase By Jamie

[Midas Souza]

Thisis not a developer bashing no-code just because I can write code and dislike no-code for obvious reasons. Rather, I am writing this post after wasting months with no code. So much so had I written my own platform from the bare bones it would be live by now.

Here is my experience:

I have been using a few no-code course platforms to run my courses over the years and I've wasted months (> 6) trying to figure out the right platform. I hopped on from one platform to another taking my courses and there was this one or few things always missing.

It's impossible to get all the features you want in a single place and then they say "Hey use Zapier and connect with XYZ for a certain feature" which in my opinion will break my bank once I go full throttle with my business. I do not want to pay 40$ a month for every tiny additional feature I need. And again, those integrations come with their own limitations.

Besides, these no-code platforms won't allow me to make basic UI changes like adding custom CSS to specific pages. Which is crazy in my opinion. The ones that sort of do lack other features that the initial platform offered. The security to content is minimal, users are sharing credentials, the reports are flawed, images are not optimized and I can't do anything about it.

No-code may appear quick to production but not in the long run. It's the opposite in my opinion. As a developer, you'll always do better with writing your own code. Atleast you can tweak your application and make subtle changes without having that fear in your mind, "Oh God, I wonder if the platform is gonna let me make these changes."

But the no-code marketing doesn't say so. According to them, we can build apps on their platform that can be as complex and scalable as systems like Instagram, Facebook, and the like. This is not only ignorance at its peak but also laughable. I wish they really knew how distributed systems work internally. Can we build graph, wide-scale, document-oriented data models with no code? No code platforms feel everything can be stored in an excel sheet and connected via an API.

That same person could have just used Shopify by the sounds of things and been already built to scale 'out of the box' from a technology perspective...I don't think no-code makes sense for anything that a dedicate platoform CMS like Shopify or Wordpress have already solved in ways that scale.

I think a lot like crypto, no-code is selling a dream, and the main people making money from it are the platforms/people that sell the dream - not actually any real benefit. At least with no-code I can see loads of utility for those tools for streamline business ops stuff and connecting commercial SaaS applications together, or for small sized business that don't need a complicated software set up, but they do need to stop trashing their reputation and fostering cynicism by pretending that people can build the next big tech company with no code tools. Maybe it could happen one day, but it's not soon.

Non-tech founder and I quickly found the limits of no-code when it comes to data security. The hoops to jump through and eventual lack of scalability were far more daunting than learning (and eventually) deploying a full stack MVP.

Am surprised people haven't come to this realization yet: "You can't beat code". If you're making something light...no code may do. But for something complex...code is the way to go. And I prefer vanilla code

No code isn't there yet, that is true. But it creates a bridge into tech that was not there before for non-technical people. The tribal part is quite annoying though putting coders and non-coders against eachother...

I believe the ideal solution would work on two layers. There would be a code based platform that you can code yourself, with a nocode editor on top.

Use the nocode editor 90% of the time, when if it fits your problem at hand. When it's not sufficient, dive down into the code layer and implement it there.

I could build a course creation platform from scratch in Bubble within 10 hours or less. Add subscription payments, drip marketing campaigns, MUX api for enhanced video processing/hosting, and that'd be another 10 hours.

Might sound like a bluff, but I've already built two SaaS companies and a hiring marketplace in Bubble that are all profitable, and all of which I built during nights and weekends over the past 18 months.

Plus, here are some other metrics that I've accomplished for the enterprise nocode applications that I've built for my day job (yes I work professionally as a nocode developer and have for the past year):

All built in Bubble, with dozens of API integrations. Add Xano into your stack, and those loading speeds go down to milliseconds. Don't believe me? I just published a case study on this that could be found here: =20&t=DYqEb3rS0KmcG72vw6Xbjw

Point of this post is not to come at you. But instead, to recommend that you take a breath and find other ways to explore your technical frustrations instead of bashing an entire industry that you clearly don't know much about.

I see that my post has singed you a bit :) And I choose not to enter into a discussion with you since you totally do not get the point I am trying to convey across. Take care of your health and good luck with your business running on no-code.

We started as a headless api for marketplace and commerce solutions. However the market is too niche specifically focusing on developers that are building marketplaces. We did provide pre-built mobile apps from the start but web templates are started last 1.5years. Because the market trumps everything

True. I believe the primary issue is those no-code misleading marketing campaigns that promise to build systems at par with coded distributed systems like YouTube, Spotify and the like. This is crazy.

Yep, this has been my thoughts exactly on no-code platforms from day 1. IMO, it can get you up and running quickly but soon there are tons of customization needed which just can't be occupied by a one shop stop no-code. Good for showing some progress in early stage for investors though.

1) No-code platforms are for non-coders only. False.

I start programming when I was 8, I have a PhD in Computer Science and I really enjoy coding.

However, I use DaDaBIK whenever I can. Why? Because I can build an application in a fraction of the time needed without it and because I can automate the most tedious, repetitive, coding tasks and focus on the development parts that matter (and that are more fun).

2) No-code platforms are for prototypes or for simple applications only. False.

I have personally created and I have seen in use complex applications developed with DaDaBIK at work, in production, for large companies, serving their employees everyday for years.

And yes, security is taken very seriously and yes, you can easily add your own CSS (I'm replying to the OP here).

3) You end up spending more time. False (if you use the right tool and you know how to use it).

Let me answer with a real use case. I have just posted this video:

where I create, in about one hour and a half, a helpdesk system without coding. I have used a few line of code only at the end, but it's so simple and high-level that we could call it pseudo-coding.

Well, probably if you remove all the speeches I make during the video, the actual development time is 30-45 minutes. Can you implement it in vanilla code in the same amount of time? I don't think so.

We also have to keep in mind that, at least at the moment, these platforms are not equally valid for all types of applications.

I have a lot of experience in what most people call "internal tools" (because DaDaBIK is mainly used for internal tools) and I can guarantee they are very good for these applications. Typically, data-driven applications, having (even complex) forms as input interfaces and grids, reports and the like as output are very suitable for no-coding. Other types of applications may be not as suitable.

In DaDaBIK (and I think in some other applications as well), this is not the case. There are a certain (and growing) number of things that you can implement without coding, if this is not enough you can always add your own PHP/Javascript code (low-code) through hooks, custom functions and other means. Maybe you only need 30 lines of code instead of 30k, but those 30 lines are essential because they cover a specific business process that you (and maybe only you) have.

Well said. I have used a different PHP generator (scriptcase) for many years and build very successful applications. It can save a zillion hours of work to stand up the "standard" type of app features. Then I use my programming skills and add the PHP/JS business logic where it really counts.

My advice for "no coders" out there is to find some tool for RAD (Rapid application dev) that does both. Use it top generate the basics using your data tables, etc; BUT having a way tpo add code will become important when you keep going. I call it adding the "special sauce" to an app.

As a developer, one of the issues is that invariably I want no-code tools to do things in the way that I would code them, because I know what's possible with code, and I know what's best for my requirements. Invariably, the nocode tools don't allow that.

I'm obviously biased, but this is why we built Autocode. Rather than obfuscate the code, we let you dig in and customize all of it, while still giving some of those "no-code" niceties like event and API integration, instant hosting and endpoint availability etc.

If your primary concern is fine-grained technical control then of course no-code stuff is going to frustrate you, but like some others mentioned, it's a inroad into automation for non-technical founders or smaller teams who don't have the time/resources to be managing devops infrastructure.

Cost is always going to be a tradeoff: either you're paying an employee a salary to solve these problems in a very specific way, or you're paying a SaaS company that's invested in their own engineers to solve this problem more broadly at scale.

3a8082e126