Advanced Database and UI builder

[adame]

Building a site with a database backend is still way too much work in 2019 imho.

I am thinking about a fully managed online service that:
- Lets you build a database scheme in the browser
- Offers an (GraphQl) API for this database
- A fRontend UI is generated out of the box for the tables and properties for the common CRUD actions.
- The UI is 100% customizable. Think of it as a React app where the componenten and Styles are fully editable. This UI is connected to the API.
- The UI will work offline. When not connected the changes will be stored offline and submitted when the connection is back.

I thougt about this because i saw myself building more or less the same functionalities over and over again for different customers/products.

What kind of sites will this service service?
- Adminpanels
- Webshops
- Full featured SAAS products

I want to offer a solution that offers people to build such a system in a day.

Besides the technical challenges for building such a service: will this work as a viable product that people will use to build their sites/webapps in?

 

[hu7231]

@adame I agree with you fully that there should be better options for quickly spinning up a basic db + crud ux experience.

Though, I would suggest that you really look at who your target market should be.

I could be wrong, but f you are targeting developers, backend as a service BaaS projects don’t seem to fare well in the market? Re: GraphQL, you can look at the GraphQL BaaS services as case studies. And unless you’re Amazon AWS, it looks to be a tough road. (Amazon bought out one of these GraphQL BaaS services and relaunched their own browser-editable GraphQL BaaS service inside the AWS ecosystem.)

Most developers, it seems to me, aren’t as concerned about features or ease of use nearly as much as they are concerned about longevity.

“Will this platform even exist 4 years from now? I’m not going to spend the time to learn something — or suggest it for projects and look like a fool and have to re-architect it — if it won’t even be here in a few years.”

This is why open source is so attractive. Less about the cost and more about the stability.

Developers even look at a giant Google suspiciously when the tech is closed source like Firebase.

BUT, there are other markets that would love to have this I’d bet. Someone mentioned Zoho Creator. There are a lot of these internal process app building SaaS tools targeting non- or jr-programmers.

But even these seem to mostly target internal systems where employees are the user.

Then there are super crude, yet somewhat? popular “visual programming” services like Bubble.

You then have the option of cobbling together WordPress forms, membership and other plug-ins. There’s also the cobbling together of existing services via Zapier pipes. All messy, crude, restrictive stuff.

I could be wrong, but a well-crafted, no-code or low-code system targeting marketers, non-technical startup founders and others wanting to build prospect-facing and customer-facing apps seems like a big unmet need.

Something like Airtable but where end users (not just admins) can crud — not just create (ie, not just submit a form into a black hole.)

If your startup’s competitors are Zoho Creator, Bubble and Frankenstein WordPress sites — and not Amazon AWS — my hunch is that you may have a fighting chance.

 

[fullspinzoo]

@adame There are some decent services that do half of this job already - GraphCMS lets you make a schema through their admin section and exposes a GraphQL API, and Prisma has some pretty neat code generation capabilities similar to what you describe (but without a UI).

In my opinion, the real product would be something to address UI generation. It sounds like you're describing a component library generator, which could be extremely cool. As far as actually building out an app in it, I've personally had such bad experiences with other WYSIWYG editors that I'd be skeptical to try anything with too much magic behind it, but if you can make it work (at least consistently across multiple uses) I think it's a viable product.

The backend-as-a-service market is pretty full (not that any existing solution is perfect, there are just a lot of them). I think you might have trouble getting people to adopt both your frontend solution and a service that reinvents one of the others they already use, at least if you're asking for full buy in from the beginning.

If you want to test out the idea, I'd say start with the UI generation, and document the hell out of its integrations with other services. Maybe even automate some of them. If you have success with that, then maybe think about launching a backend service and see if it works off the name recognition you build on the front end. This sounds like a massive project, but there's a ton of potential. I'd love to try it out if you get a prototype going.

 

[nogueira527]

@adame Neat idea. Think about open sourcing the project so it is more widely adopted and offer professional services/hosting to generate revenue.

 

[adame]

@nogueira527 I like that idea

 

[colbaltheart]

@adame You can already get packages for the likes of Laravel that will do this for you, things like Laravel Nova.

Who would you audience be? To me it seems like it would only interest junior developers?

 

[adame]

@colbaltheart Why only juniors? I want to target everyone that needs a webapp/site. Ultimately also non-techies because everything can be done from within the browser.

 

[colbaltheart]

@adame Non-techies are unlikely to be putting together their own web applications. It's not because they haven't got the tools, it's because they haven't got the knowledge or experience to do so. For those that have a little bit of knowledge there are things like Zoho Creator.

As for techies, those above a junior level will already have workflows and snippets in their locker to make themselves more productive and will want more control than a tool such as yours would afford them.