Spent the weekend optimizing Blazorise's fluent utilities like Margin, Flex, and Gap. The results? 2x faster execution. 66% less memory allocation. Not bad for a weekend project. Next Blazorise version should feel a little snappier.
Mladen Macanovic
@mladenmacanovic.bsky.social
✨ Creator and Founder of Blazorise, crafting the web with code magic! Dad of two adorable mini-humans.
Tip for Codex CLI users. Don't enable `/vim`. You'll go from vibe coding to fighting a 50-year-old text editor.
Mixed feelings about Bootstrap 6. Love that it's still evolving. New APIs and components look great. But the default design? It looks like AI slop. I'd expect much better from some of the best designers in the industry. Am I being too harsh?
That feeling when you finally remove another JavaScript dependency. This time, we replaced Pickr.js with our own native C# and Blazor color picker. One less dependency. More control. Less JavaScript. Beautiful.
What is Daybreak and why would I need it? It keeps asking me to set it up, and at this point I'm afraid to ask. Am I missing something everyone else already knows?
.NET team has done a great job with this release. Updating Blazorise to .NET 11 was one of the easiest upgrades we've had. Mostly version bumps, a few small changes, and everything just works. Blazorise will be ready for November. github.com/Megabit/Blaz...
Chore: Add .NET 11 RC1 support by stsrki · Pull Request #6828 · Megabit/Blazorise
Adds .NET 11 alongside existing library targets and upgrades demos, docs, tests, and MAUI to .NET 11. Updates package versions, CI, publish profiles, and applicable language settings to latest.
github.com
Sometimes the comment section is more entertaining than the actual post. Here’s one from me commenting on a post about The Last of Us. I regret nothing.
I'd love to buy a new PC just to make my integration tests run faster. Then I look at the prices of new components. Suddenly, my tests don't seem that slow.
Blazorise 2.3.3 is out. A small custom color request turned into a refactor of our entire color system. Now you can use any CSS color directly with `Color` across components. Plus, many fixes and a completely redesigned website. blazorise.com/news/release...
Blazorise 2.3.3 - Custom Colors, Scheduler Improvements, and a New Website
Blazorise 2.3.3 adds arbitrary colors across components, improves Scheduler, DataGrid, NumericPicker, and Charts, and introduces a redesigned Blazorise website.
blazorise.com
One thing I wish C# had is the `With` statement from Visual Basic. Maybe it would make the code cleaner. Maybe it's just nostalgia. Either way, I still miss it.
This has been on my backlog for years: strongly typed ZIndex utilities. Probably not exciting to most people. But for my strongly typed obsession, it means a lot. `<Div ZIndex="ZIndex.Fixed" />` Small feature. Big satisfaction. github.com/Megabit/Blaz...
ZIndex: add utility with named layers and numeric values by stsrki · Pull Request #6823 · Megabit/Blazorise
Closes #4823 This adds a strongly typed ZIndex utility for controlling element stacking through the standard utility API, such as <Div ZIndex="ZIndex.Fixed" />. It includes named la...
github.com
GPT-6 Sol is crap compared to GPT-5.6 Sol. Absolute garbage.
That feeling when you find money in a pair of pants you haven't worn in months. Technically it's your money. But somehow it feels like free money.
Every side project starts the same way: "This should only take a weekend." That's how they get you.
The biggest difference between junior me and me today: Junior me wanted to prove I could build it. Today I first ask whether we need to build it at all.
Writing the code: 10 minutes. Deciding what to name the thing: 45 minutes. Renaming it three months later anyway: inevitable.
You can spend days trying to make an API perfect. Then a customer uses it in a way you never even considered. That's usually when you find out whether your API is actually good.
The longer I build software, the less impressed I am by how many features a product has. I'm much more impressed when I can't think of anything that needs to be removed.
A feature isn't finished when it works. It's finished when someone who didn't build it can understand how to use it without asking you. That's much harder.
AI makes me write code faster. Experience makes me delete half of what AI wants to write. Both are useful.
Open source changes once people depend on your project. That "small breaking change" suddenly isn't small when thousands of applications are using your API. You start thinking very differently about API design.
Developers love rewrites because the new code doesn't have any legacy problems. Yet. Give it five years.
If you were expecting the Education platform to be the new thing for a while now, think again! As is our founders way, Mladen is always exploring new ideas, building and thinking out new features and systems. Now he has given our Blazorise platform a completely new look! Check it: blazorise.com
One of the strangest things about building software: Customers will ignore the feature you spent 3 months building... and absolutely love the small thing you added on Friday afternoon.
I think every developer should maintain the same product for at least 5 years. Not build it. Maintain it. You start seeing code differently when you know every shortcut you take today might still be waiting for you in 2031.
I think developers underestimate how valuable boredom is. Some of my best ideas came when I had nothing urgent to fix. No deadline. No bug. No customer request. Just enough time to think: "What if we built this?" Then somehow I'm three weeks into a new feature.
I’ve spent the last few days giving the Blazorise website some much-needed beautification. Cleaner, fresher, and hopefully a lot nicer to browse. Hope you like it. blazorise.com
Blazorise - Blazor Component Library
Battle-tested since 2018 and actively maintained. Build with 100+ Blazor components, your choice of CSS framework, flexible theming, and commercial support.
blazorise.com