Widgetizer 0.9.8 and What Comes Next

Version 0.9.8 is out and I’ve pushed it publicly, so I can’t hide behind “still in dev” anymore.

Up until now I could break things freely, change direction, rewrite parts of it without thinking too much. That phase was useful, but it’s over. People will start using it, some of them seriously, so updates need to be a bit more careful from here on.

If you haven’t seen the original idea, I wrote about it here. That hasn’t changed. It’s still a visual builder that outputs static sites. No databases, no layers pretending to be simple, no systems that make sense only if you already know how to build websites.

What changes now is everything around it.

Hosting

Right now you build a site and export it. That works if you know what you’re doing. Most people don’t.

So the next step is hosting.

The idea is very simple. You upload a ZIP and the site is live. No setup, no decisions to make.

There will be a free option where you get a random subdomain with some limits, just enough to get going. Then a small paid tier where you can use a custom subdomain and later your own domain. I’ll also add backups and rollbacks there because they make sense as part of the same thing.

I don’t want to turn this into another control panel with a hundred options. It should feel like “I uploaded my site and it works”.

Forms

Forms are always a mess. You either embed something, or you wire a third party service, or you deal with some half-broken plugin.

I’m adding a forms widget to Arch so you just drop it in, define your fields and that’s it.

Submissions will be stored, you’ll have a simple panel to view them and you’ll get emails when someone submits something. No API keys, no integrations, no extra setup.

It’s a small thing on paper but it removes one of the main reasons people get stuck.

Analytics

Then analytics.

Not the usual dashboards with twenty charts that nobody understands. Just the basics, visits, pages, sources, presented in a way that makes sense without a manual.

The interesting part comes later. Once there’s enough data, this can start suggesting changes. Move this section higher, simplify this page, adjust this content. Small things that improve the site without asking the user to figure everything out.

Most people don’t know what to change on their website. That’s where this helps.

Hosted editor and themes

At some point I’ll move the editor online as well, for people who don’t want to install anything. Same approach, just accessible through the browser.

Alongside that I’ll introduce a set of free and premium themes. Nothing surprising there, it’s just how the ecosystem grows.

Appointments and bookings

I’ll also add a small system for bookings and appointments.

This will work for regular users who want to accept bookings through their site, but also for AI agents interacting with the site. Same idea as everything else here, keep it simple and make it work without requiring setup gymnastics.

Basic availability, simple booking flow, and a way to manage requests. Nothing more than what’s actually needed.

Generating websites

And finally the obvious one. You provide your content if you have it, or just describe what you need, and the system generates a website as close to finished as possible.

Not a rough draft. Not a wireframe. Something you could realistically publish with minimal changes. From there you tweak things if you want to. But the goal is that you don’t have to. That’s the bet at least.

Updating via email

There's one more thing I've been exploring. The ability to update the website simply by sending an email with the changes. You email what you want updated, the system generates a preview link, and you just approve or decline the change. No login required, no interfaces to learn. Just the same way you'd communicate with a human assistant.

No need to spend more than two minutes and a single email just to change the opening hours of your business for the next bank holiday. It should be that simple.

The agentic era

We are entering a phase where AI agents will do a lot of the browsing and interacting. For a small business owner, their digital footprint needs to be clear, structured, and easy for these agents to parse.

Instead of adding more complex layers and dynamic widgets, we can help them best by taking a few steps back. Clean HTML, Markdown versions of their content, simple structures, and straightforward forms mean an agent can easily read the site, book an appointment, or extract information without getting lost in a mess of scripts and popups.

Why all this?

This is not a random list of features. It’s the same direction I’ve been moving towards for years.

Most tools are built for people who build websites. Then they’re marketed to small business owners. That’s where things go wrong.

I’m still focused on the same group I’ve been working with all this time. People who just want a site that works and don’t care about systems, components, or any of that.

What 0.9.8 actually means

The biggest change is not technical. It’s that I need to treat this as a real product now. Updates need to be stable. Changes need to make sense long term. I can’t just rewrite things every week because I feel like it. That’s fine. It’s a good problem to have. This is where things actually start.

Feeling alive after a long time again. Onwards.