Question regarding Code Quality & Bloat

Hi everyone! I was just wondering about the quality of the code on the backend and what you’d get if you were to export the code.

Of course I’m sure that a dedicated team of developers at a startup would create more optimised code if they were starting from scratch with JS etc, but when using a no-code tool like WeWeb, how much bloat/overdeveloped code would you find? And if I was to show that code to a team of developers, would they almost want to start developing from scratch or would they be happy working from this base.

I hope this makes sense :slight_smile:

1 Like

Maybe this topic about code export can be a starting point to answer this.

1 Like

In general, the issue is less that the generated code isn’t optimal and more that it is largely unreadable. The idea that you will build on it is largely a fantasy - you will find that the business knowledge you’ve gained and the value of the next-best-tool on the market mean you will rebuild it faster from this template rather than trying to directly modify it in an exported state.

I tell people to worry about code export if you want to separate the concerns of editing and hosting. Like wanting CDN features, advanced DNS, or edge functions to create a more advanced experience. State Change members have done this multiple times with remarkable results - I think it’s a great production path for weweb builds. Most no-code platforms are excellent editors and only middling hosts.

But if you want to abandon weweb as the editor, you will probably have the shortest and most profitable path back by rebuilding in whichever new platform you want to use leveraging the business insight you’ve gained from the prior path. Mental Model Principle #1 at State Change is “you will build this again.” The knowledge you’ve shipped along the way and the continual improvement in the tool landscape make this reality not only inevitable, but desirable - the new builds are faster and unlock more economic value.

5 Likes

That definitely makes a lot of sense! I think our plan will be to use WeWeb as a PoC to get our idea to market and then when/if the product begins to scale and dedicated development becomes worthwhile, we can translate over to code. All the while users are still able to use the PoC and we’re generating revenue.