For years I gave students lifetime access to everything: the material, the updates, the support. I thought I was being as helpful as I possibly could.
Turns out it did the opposite.
My last couple of programs have been limited time, with limited support, and engagement and results have been vastly better because of it.
So I'm right with @RinodeBoer on his new Pagebuilder Exit challenge (which looks great!): livingwithprompts.com/progra…
30 days, everyone starting the same week, building one real WordPress site together without a pagebuilder. More like a class than a library of videos.
Doors close tomorrow.
"SaaS is dead” is just a bunch of bullshit.
The idea that everybody is going to build their own software and be happy with it is a tech bro bubble take. Or a sales pitch from tech CEOs selling you tokens.
I just spent a full day building my own video editing Mac App in Codex. And it is becoming incredibly good and useful.
Anything I throw at Codex, it builds it. Most changes are done within a minute. But that’s not where the hours go.
I ask for a feature, I wait, I try. I realise it doesn’t quite work how I imagined it. I figure out why (which takes skill), and then I figure out what to prompt next to fix it.
Over and over.
This is not easy. You need to understand how to describe what you want and need properly. And then you need to be able to judge whether the result is actually good. Having access to the best model in the world doesn’t automatically give you those skills. And the model can't know what I need for my specific workflow. It can't predict that. I need to describe that, after conceptualizing my own workflow in my mind.
I’ve been obsessed with software for 20 years, studied UX/UI design, I teach software, consult companies on their interfaces, and built multiple webapps. And it's still challenging for me.
So it is really cool that this is possible, and I absolutely wasn’t able to do this a few years ago, but it's still quite an exercise for my brain to get it right.
Now imagine doing all of this without that experience. How is a normal person going to pull that off? It can't. They don't have the same prompting and software judgment skills. That's what 20 years working in software gives you.
Will a big part of the group that works in tech build their own software? Sure, but most of the world is not part of that group. And software isn't just used by tech bros.
So tech bros that are promoting this idea of "Saas is dead" are just in their own bubbles and assume everybody is like them. But that's not how it works. And the next model doesn't change that.
Most people just want software that works so they can get on with their actual job. And we haven't even talked about the lack of free time that the modern human has.
So I don’t buy the “SaaS is dead” BS anymore. It just doesn't make sense.
These image optimise & compression plugins on the WordPress repo are always so annoying, trying to upsell you to their Pro plans, just to do some basic upload & bulk compression.
So I build one myself 😎
Compressino...
So far:
- Auto convert to WebP on upload
- Compress & restore from Media Library popup
- Instant visual preview of the compression with easy slider (with load in your own image from library)
- Bulk compress with 1 button
- Reduction preview and history
- Keep & restore original
- All free, no external server involved
What a time to be alive
High agency / tech builder types dramatically overestimate regular people’s desire to save time, money or be more productive.
They don’t want to make their own software let alone design and assemble their own furniture, even if it’s a cost savings of $15k.
They want to order the table from poliform so when their friends ask where they got it they can say poliform. They don’t want to think about designing something custom for their space because they have no idea what that should look like.
Spend time watching people order at a restaurant and you’ll realize most people want to make the least amount of decisions as possible
Claude Code can design furniture part by part and then give you a bill of materials to order each part and put it together yourself
And way cheaper than big interior brands sell you their furniture
Furnishing my house I discovered how absolutely insane prices are for the most basic cheap materials put together with a fancy brand tied to it
One shop had a sofa for $30,000 😂
I think with Opus 5.5 and other models easily operating 3d programs like Blender and FreeCAD (both free) it makes more sense to cut out the middle man and go straight to suppliers and design and build your own furniture
Which in a way makes customers more powerful (they can get exactly what they want and usually for cheaper without the markup)
Anyway @theannalux designed this table from scratch, she's not a furniture designer, not a coder, she just had an idea of a style and worked with Claude Code in the terminal (far superior to the app I will keep saying this)
If you design your own furniture and buy the raw materials directly, you can save 85% compared to premium brands like Poliform
That means the $30,000 sofa becomes $4,500! And the $8,000 dinner table becomes just $1,200!
You can see where the idea comes from that AI will make everything 5x to 20x cheaper, and it's already starting!
ChatGPT SOL 6.1 really cooked with this one.
After a few hours of prompting, I got my own AI Video Editor Mac app, and it looks like it can already compete with other video editing software I use.
It can now:
- Save and reuse your own scene layouts
- Trim by dragging and camera & screen stay in sync
- Smooth scene transitions with (adjustable) easing and duration
- Layers panel with resize, corners, shadows, etc settings
- Local project files that any LLM can work with, so fast agentic editing, but manual control for details
I am making this because I saw what was possible with Remotion in combination with Claude. But Remotion Studio (their browser editor) didn't have great manual editing experience. And I found with video editing you often need those small adjustments to get the audio cut right. That's not a task you want to do via prompting. You need a proper timeline editor.
So I think this is the best of both worlds. Fully agentic editing, but also a UI for the small details.
I tried to build my own AI Video Edit Mac app that talks to Claude or ChatGPT, and after a few hours I have this.
It already works and I've made my first MP4 export 🤯
How is building your own AI video editor a weekend project now?
Crazy times
@RinodeBoer this morning: "Hey Gio, I couldn't find Gemini Desktop in @novamira_ai ."
It's there now. We've released the official Gemini Desktop integration. Thanks for the suggestion, Rino!
novamira.ai/novamira-1-12-15…
Someone wrote an article about Pencil (I don't know him even though he is Dutch).
Joost van der Schee comes from the Hugo world (different framework for building websites), and he landed on the same idea: design in code, clients edit content.
Pretty cool to read: usecue.com/blog/pencil-by-ri…
In a few hours I’m going live on YouTube to interview the founder of Independent Analytics.
Why analytics, right now?
For years, design and development were our moat. They were hard enough on their own, so a lot of us could afford to not fully focus on results. Even though I have always focussed on good design, which means starting with understanding what the client needs, if I'm honest I did ignore the analytics a bit too much. I just installed Google Analytics and never really looked at it. I think I'm not the only one.
Now AI is making design and development easier and easier. So the value of that part is going down, and so I believe the value is shifting towards results, so that will become the new moat for web designers. I don't think we really have a choice anymore.
But how do you actually prove a website works? Analytics. But also a great setup. Because do you know how people move through your website? Are you tracking which buttons people click and which ones they never touch? Without knowing what’s going on, you can’t prove anything.
I already knew Independent Analytics, because it was the only analytics plugin that didn’t need a cookie banner (GDPR). But then a month ago I learned it tracks (and keeps) all the data inside your own WP website, which means your AI agent can read and analyse it, build plugins using that data. That’s when it all clicked for me and I became super interested in their product.
So I reached out, and now Ben from Independent Analytics, is going to show me what is possible, live. So we can all learn together.
youtube.com/live/3f5T4otodf4… (this link will automatically turn into a replay)
Btw: this livestream is not sponsored at all. I bought the plugin myself and reached out because I like it.
4 months ago I cloned Airtable for a client in 5 days because they refused to pay Airtable $500/mo
months later, I’m still maintaining it
every two weeks it’s another edge case, a UI tweak, or a "quick 10-minute request." They’ve easily spent 10x more in headaches than the $2,000 they would’ve paid Airtable
just proving that SaaS isn't dead. Cloning isn't enough, if you want it to be same quality as original it becomes a full time job
and at this point isn't it easier to just pay?
I have 98 plugins installed on @wpzoom. Everything runs fine.
Some plugins are literally a single PHP file.
A plugin count tells you absolutely nothing about performance, code quality, or security.
But I guess counting to 47 is easier than understanding how WordPress works.
"WordPress plugins are the past, Claude can do everything now"
Now that AI is here, so many people are against having a list of WordPress plugins. But it doesn't make a lot of sense.
A plugin is just a piece of code that adds functionality to your website, which you can activate or deactivate. When Claude generates that functionality, it's also code. There's not much of a difference.
The main difference is who maintains it. Plugins are often managed by companies that release security fixes and new features. Not always, of course. You can vibe code your own plugins too.
Those updates can cause problems, but so can your AI-generated code. Using Claude with another CMS, like Payload, doesn't make vulnerabilities disappear either. This screenshot shows vulnerabilities flagged in a Payload site's Node.js packages.
WordPress plugins have this reputation for being bad, but having code you can switch off, with a company maintaining it, is actually pretty useful. You can reasonably expect a good company to work on keeping its plugins secure and up to date. Especially when the plugins are paid tools.
Your vibe coded solution can quietly break in the middle of the night, leaving you to reschedule your morning to fix a client's website. With a maintained WordPress plugin, the fix might already be available.
You litterally click "Update", and you're done. How is that not better?
And yes, we're moving toward a time when we won't need plugins for every little thing, especially on the frontend. But using code from companies that keep it maintained still makes sense.
So plugin updates are a blessing, not a problem.