Over the weekend I shared some thoughts on the proliferation and maturity of "no code"
https://medium.com/@rrhoover/the-rise-of-no-code-e733d7c0944d
Over the weekend I shared some thoughts on the proliferation and maturity of "no code" tools and why this is great news for makers like many of the people here on IH. 😊
@rrhoover awesome read man
When COBOL was invented in the 50's they said - no need for programmers any more, just write it out in english
In the 80's there was a program called "the last one" - supposedly the last program you'd have to get as it could "easily" make any other
And in the news today https://venturebeat.com/2019/01/28/kite-raises-17-million-for-its-ai-powered-developer-environment/
a system that will write the python for you
Tools to reduce or get rid of code have been around for years. But no sign of less people doing coding
While these mat be great investments for @rrhoover, this trend is nothing new. Even when I was a kid, things like RPG Maker existed. Flash was great, but it never was "no code". It was where I really started programming.
The thing is tools can automate making specific kinds of software, but people want to customize and as customization becomes more complex and involves more logic it slowly becomes coding. And from personal experience I've learned that programming in RPG Maker/Bubble/Zapier/Excel is a horrbile experience compared to a tool actually designed for programming.
Until strong AI arrives and out-competes humans at everything, there will be value in acquiring technical skills. Saying otherwise might be great marketing, but it's essentially selling a lie people desperately want to believe!
I completely agree with you on this.
No one is going to make a product, that doesn't involve customization via coding. Currently, it cannot be done, unless we are talking about generic, internal enterprise applications.
I think that there is a sweet spot for these tools. I believe we can make a tool like this FOR DEVELOPERS. The goals of it would be to eliminate a large % of the standard behaviours that every app does, while still allowing for customisation where required.
You're right about those tools being a horrible experience for programming. I would add thought, to be fair, they do these common features & boilerplate well - but suck at proper developer focused customisations.
This is what we are attempting at Budibase.
An open source tool, where you can get an app running a a few hours/days. Then, download the source, and start programming plugins against it, for example.....
PS @alchemist ... I would really like to know your honest opinion about this idea.
Yes, acquiring technical skills is valuable, but I certainly don’t see anyone taking the position that it’s not. Learning to code (what I think you’re really trying to say) is super valuable, but it’s also something not everyone can do. Some of these tools simply allow the people with a vision and creativity to create, without coding, that which people like me would normally would create with code. Yes, there will be some limitations and lack of atomic flexibility inherent to that style of creation, but the best of these tools severely minimize those drawbacks. Again, don’t see a lie being put forth or sold here.
The problem is that "the best of these tools" end up essentially being coding. The idea that building complex rule-based systems involves "no code" is mirage. Defining data schemas via Air Table and putting business logic into IfThisThenThat is coding. It's just an inefficient way of doing so for anything but a trivial project.
I am so, so glad I didn't listen to pieces like this one 9 years ago when I was a non-tech entrepreneur "with a vision and creativity to create", as you put it. Learning to code was one of the best life choices I've made, and it's opened up innumerable opportunities.
If I'd been sold on the dream of being able to build anything without any difficult learning, it would have been a loss.
If these tool were positioned more as a "introduction to coding", would it help? I.e. be up front about introducing the non-tech creator to coding concepts, that help them start their skills?
(Rather than sell the dream of a no-code app)
I agree, coding is the balls. The example in the list above which I am thinking of specifically is VoiceFlow. It's getting super robust and the user can create some pretty rich, awesome voice experiences, without coding. And (to my initial dismay), the complexity can be quite high. Yes, there are tons of technical nuance to be learned, but the overhead of dev environments, building, release management, and syntax is not one of them.
This is a good point. A separate thing I've said a lot here is that there are way too many people here who go into what I call "devops architecture astronaut" mode. Much of the "overhead of dev environments, building, release management" is self-imposed.
I launched my last project without any of that. Of course if you're doing something mission critical or used in a hospital, you need some process around releasing new features (whether you created them in a terminal or with VoiceFlow)!
I'm all for using the quick and dirty solution early on, and I'm also a fan of special purpose tools like VoiceFlow or even RPG Maker, for that matter. I'm just cautioning against buying into the "no code" dream. I'd also caution against or the "4 hour work week" dream, the "nomad travel blogger on the beach dream" or "VC-funded zillionaire" dreams too, for that matter.
If anyone is interested, we a building an open source low code platform, specifically for saas businesses.
You will be able to get something started fast, but still customise (front and backend) with javacript and frameworks of your choice.
Sign up to https://budibase.com if interested!
This looks really cool. Are there any technical docs?
Working on it :)
<argument with myself>
Me1: Would we really say in an hour? Can we really create anything actually useful in an hour with tools like bubble etc?
Me2: No, not really. It takes much longer than that. Especially when you're new to it.
Me1: Hmm. Ok what would be realistic?
Me2: Probably 2-5 days when we consider polishing things up.
Me1: Really that long!
Me2: Yeah, but still, does that really matter? We don't need to tell people that part. Let's just get them started. Because "technically" it is possible to make something in an hour. The more that get started the better. We want more makers!
Me1: Yeah... It's for the greater good.
Me2: Yeah... "For the greater good."
Me1: Ok, an hour it is.
</argument with myself>
https://www.youtube.com/watch?v=yUpbOliTHJY
Hi Justin.
As per my comment below, we are bulding https://budibase.com
We like your product!... Its a great resource for us.
To your point about it actually taking a few days to learn the frameworks... youre right. However, for me, the point of these is to be able to produce products in volume.. like all those little ideas uou think just might work.
In which case, achieving the "1-hour-saas" might be possible.
I also think its important to allow as much frontend and backend customisations as possible, allowing the product to be unique and keep evolving.
This comment was deleted 8 years ago