This is interesting for sure. Kudos for bringing this capability to the web!
One issue the demos reveal is, it doesn't _feel_ like the web. That is, I can't hit Ctrl+F to find text on a page. I can't select text with my cursor. I can't copy the address of a hyperlink. On my phone, I can't hard press on an image and share it to others. Screen readers can't handle it. I can't press a shortcut key to make everything larger.
These all may seem pedantic, but they contribute to the feeling "this is not the real web."
This is the same problem with Java applets in the late '90s, Flash and Silverlight in the early 2000s. They are islands of richness within a web page, but those islands are, well, opaque to browsers, search engines, and virtually all web tooling.
That's not pedantic at all! Indeed, without these capabilities, it is by some definition not the real web.
This hits into that concept of what exactly the "web" is. Is it just a media transport system? Or is it something more than that. Of course, we could cite Tim Berners-Lee here or Roy Fielding in this discussion.
But at minimum, I think a lot of us are tired of the app-lification of the web and somewhat wish we could have a bit of the old.
It's also an interesting question, why, in traditional rich desktop applications, I can't say I have ever missed the ability to select and copy text from the UI chrome - whereas on the web I'd definitly miss it and in badly designed mobile apps, I often do.
I think some part of UI design degraded with the web, where there used to be a clearer distinction between "user data" and "app chrome" areas than there is today.
I'd also like if we could get back to selections of more complex data types at some point and not just treat everything as text. UI toolkits have all kinds of lists and treeviews to model selectable entities, whereas in the browser, there just a single huge wall of text for everything.
> why, in traditional rich desktop applications, I can't say I have ever missed the ability to select and copy text from the UI
I do miss this on an almost daily basis and I have stopped paying for services that force me to use an app without offering a website.
The last instance of this was just a couple days ago when I could not copy a tracking number from an e-commerce app (to then paste it into the shipping company website) but at least this e-commerce company has a web UI so I could rely on that.
Oh and the other one that I miss almost daily is cmd-F / ctrl-F
"E commerce apps" are very much not the sort of traditional desktop application they were referring to. Note that they add "in badly designed mobile apps, I often do."
They're referring more to things like "you can't copy the text labeling the brush width field in Photoshop" (but you CAN copy the text out of that editable field). It's a part of app design people are extremely lazy with today, as you note.
In any sensibly designed desktop package tracking app that number would've been selectable or copy-able text, like how an email subject is in a desktop email app. (Thunderbird, say.)
(Interestingly, ctrl-f to find is one that many apps/OSes have now borrowed back, with the ability to "find" items in menus through a Help menu -> Search action.)
While I do occasionally miss it there as well, I think the main difference is that I very rarely use desktop applications for information gathering.
I never "read" a desktop application, whereas that is mostly what I use a browser for. And if I can't properly interact with text on a website, then I would likely reach for something else.
Back in ye olden days desktop applications for information gathering like Encarta let you select and copy text because they were thoughtfully designed and knew that "information you were gathering" should be different than "application chrome" - that's the distinction being made here.
Information-oriented desktop apps still do this - any good email client, for instance, should make it trivial to copy a subject line or "to"/"from" address even if it's in the UI chrome.
> in traditional rich desktop applications, I can't say I have ever missed the ability to select and copy text from the UI chrome
I forgot what desktop application it was, but there was a time that I repeatedly needed to copy texts from a dialog, which didn't support text selection. It frustrated me so much, that I put together a script to do OCR on the dialog.
Supporting complex data types for copy & paste is good; but it is almost trivial to also support plain text copying as a fallback when it already supports copying of other mimetypes. The problem is that some UI has no support of copying in any format at all.
If it was a standard Windows dialog box by any chance, you could just have pressed Ctrl+C with the dialog in focus to copy the message. It's one of these subtle things that go almost completely overlooked.
Thanks, doesn't help me, but you're right, a good tip to know. Though I'd still prefer a similar option to start selection directly in the UI instead of finishing the job in a text editor, this would also help highlight text in a screenshot without having to do image post-processing! I'd even accept some arcane finger-breaking ctrl-alt-win-x-y-z (which I could rebind) for the privilege
All the more annoying when such years-old fundamentals are broken in all the new "supposedly better" frameworks
IMHO there's no gatekeeper of what the "real" web is or should be. It grew organically - regular people building things they liked or needed. It's certainly more of a life necessity than it used to be, but that happened organically too.
I know there are strongly held opinions about this, but I for one see no reason why the "application web" can't peacefully coexist, and interlink with, the document web. In my opinion it therefore makes sense to allow for different models for the application web, ones that do not revolve around a document.
On the other hand, if we're just bashing on javascript being the lingua franca of the web, that's a train I'll happily board!
Not using the standard web stuff usually means it's also an accessibility nightmare, tried using a screen reader on the demo and it doesn't work at all unfortunately
I wonder if at any point browsers will offer a low level accessibility API for you to manually describe components. I’ve worked in the web for years and I’m a big believer but it’s also indisputable that Canvas offers more performant UI rendering than HTML when done correctly. I don’t think it should ever be used for web “documents” but web apps already bastardize HTML and CSS to achieve their aims anyway. Accessibility remains the missing component.
OTH, we are still failing to provide a bare minimum for accessibility. Heck, we even needed a law (in the EU, that than needed to be translated to national law), so that companies providing crucial end user services would care about accessibility.
It's HTML imagemaps from the 90s, when we could not style buttons and navbars where GIFs with links in the right places. Browsers still have the code to render them.
I'm not aware of any screen reader that works by continuously feeding screenshots of user interfaces into a remote expensive image LLM, which is an absolutely insane and impractical idea for many reasons, but I used standard TalkBack on Android
MAUI was never intended for the web. This is not what Microsoft wants you to use it for.
WASM is just one of the platforms that Avalonia supports and so, if you run MAUI on Avalonia, you can run it on WASM.
If you do that though, it is going to be like rendering any other desktop GUI toolkit in WASM. It is not a web app. I mean, it is cool you can do it and MAUI in WASM is better than no web capability at all I guess. But you would never set out to create a web app in MAUI.
MAUI on Avalonia on WASM is really a modern replacement for Silverlight. And it will likely be about as popular.
The really cool thing is being able to target the Linux desktop finally. A lot of people will love that.
And, while MAUI was meant to use native controls on each platform, many people may prefer the Avalonia approach of having your app render the same everywhere.
I think it's the same problem that flutter web has, and probably any other canvas/wasm based backend? Those features still need to be implemented, while still missing out on accessibility?
Yeah. I think you need to render to actual DOM nodes when targeting the web if you want a first class experience.
We're betting on this over at https://github.com/DioxusLabs/dioxus where we're building a cross-platform UI solution that enables you to do this by having a web-centric API (we are developing our own custom HTML/CSS renderer for native platforms).
>One issue the demos reveal is, it doesn't _feel_ like the web. That is, I can't hit Ctrl+F to find text on a page. I can't select text with my cursor. I can't copy the address of a hyperlink.
That's because MAUI is intended for mobile and desktop apps.
If you want to use .NET for front-end web SPA, you can use Blazor which will behave exactly like you asked.
I get the value in this and realize it's not for your polished -$500 ARPU consumer social apps, but man this is weird.
(Also if anyone who worked on it is here, it's crashing for me on OSX 26, Chrome 142.0.7444.135, if I run an animation and hit back as the animation finishes)
I'm currently in the process of creating a piece of development software and needed a cross-platform GUI environment to do it with.
I first tried Maui because its seemingly so highly rated and recommended. It turned out to be a complete nightmare. Even just installing it didn't just work, I had to spend 2-3 hours just to get that going and then when trying some basic things for my UI it turned out to just not be supported.
The next on the list was AvaloniaUI and wow, it was the complete opposite experience. Installation was a breeze and with only a marginal bit of googling I managed to do all kinds of things such as making a top-most only borderless transparent window etc.
The interface in Visual Studio could be a bit better overall, but compared to Maui its light-years ahead.
In the .NET ecosystem, I have noticed people to shame .NET MAUI because Microsoft themselves don't use this framework - Microsoft Team is built on Electron and not MAUI.
Why build a product on MAUI when Microsoft aren't too sure about it.
The answer to that is well known: Windows division builds WinUI/Win32 as their native C++/COM API, Office division went to React on their path to the web and the dev division fills gaps (WPF) and provides tools for external and internal devs (Maui for cross platform uis).
It is history not the lack of will. At one point the windows division was in shambles (remember vista) and WPF pops up. At another point, the windows and dev division have no answers to the office group (because you know who uses non win tech) so they went react. And then external devs screamed: where is the .net cross platform story so Microsoft acquired xamarin and later form Maui out of it.
It is history not lack of trust. But the outcome is the same: lackluster support for all UI toolkits.
Well summarized, and just as shocking today as it was every minute while it developed.
Someone needs to remind those cats that they own the platform. Being able to sanely develop apps for and on that platform should be possible, and UI kinda-sorta matters for that. At a certain point with the MFC they had it dialled in, while pioneering asynchronous browser tech, with many best in class tools. Decades later with a cross-platform cloud-centric stack they have a shrug emoji as big and wide as the eyes can see, and no sense this basic question of development will ever get improved.
Ballmer chanting ‘developers, developers, developers …’ springs to mind.
These days, I think Microsoft's web-based desktop apps mostly use WebView2 directly instead of Electron, so they don't have to bundle a browser. I think for Teams it happened at the same time that they moved from Angular to React.
The point about them not using MAUI still stands though. From what I understand, the .NET world has either adopted different abstractions like Avalonia, or stuck with tried and tested solutions like WinForms with proprietary controls. After all, they've seen this before with WPF which was never fully adopted by MS either, or with the debacle around Metro/WinRT. You're never quite sure what Microsoft wants you to use or will support in the long term. They also make Blazor, which is a different (and likely more accessible) way to build web apps with .NET.
Since we're on the subject of companies not dogfooding their shiny tech, is Google really using Flutter for their own apps? I feel like the evolution of the Android ecosystem towards Kotlin and Jetpack Compose implies otherwise.
Ive been building these apps (cross platform web based ui, C# backend) for years, and yes its finally good to see MS catch up and validate the architecture ive pushed since Xamarin. I wrote once wrote an electron version of this archand thought wtf are people doing? Things can be so much easier when you use a platform that knows how to multi-thread. At stages i had to build adapters/upstream patches for Chromeiunium directly onto Mac and Linux, and its was a major pain having to debug C calls.
Ive been using the same framework now for 10+yrs on apps in the stores, i wrote a small layer infront of the webviews and can swap out webkit, chrome, edge on demand. You really dont need much, just a constand way to boostrap logic and UI. 90% of code is shared across all platforms, there are def differences in WebView engines that you sometimes come across but those parts just get swapped out with browser specific JS. Ive found bugs and worked with browser teams at all vendors doing this and to see how simple this is with Dotnet these days compared to when i started is refreshing. Its easily the most stable cross-platform framework around, if you are stuck using something like flutter i pitty you, its just eletron with another skin. I can swap out and integrate directly with OS libs when i need to do stuff that the dotnet team hasnt gotten around to yet without re-writing. This has mean i really havnt used MAUI at all, but if i need to or could take advantage of it i can mix it on an Ui element by element basis. I prefer webUIs though, i have the chose to handle anything with either JS, WASM, or a combination. I can use traditional JS frameworks or traditional Native UI frameworks.
If i had started this process later, avalonia seemed to have the closest thing i required. It was just a bit a more complex /based on WinUI (which i dont really enjoy) but it supported all platforms and gave lower level api access. MS were smart, that canabalanised all these community effort and brought them into the fold. Every dotnet webview impl was a successfull community driven project before. They didnt write anything themselves from scratch.
> Microsoft Team is built on Electron and not MAUI
Microsoft Teams was released in March 2017. .NET MAUI was released in May 2022. In 2021, Microsoft replaced Angular with React and moved away from Electron to WebView2 (using the OS' built-in renderer rather than a bundled Chromium). So even the rewrite was a year before MAUI (and they probably started the rewrite before 2021). Plus, part of the point of using React there was that they could basically replace Angular bit by bit.
Microsoft Teams is just older than MAUI. It's like asking why Hadoop is written in Java and not Go or Rust or why Kafka is written in Scala and not Kotlin. Kafka was open sourced in January 2011 and Kotlin came out in July 2011. Kotlin wasn't an option given that they were developing it years before the language was released.
That's not to say that Microsoft's attitude toward MAUI doesn't leave concerns. There was some news a while back about a bunch of layoffs around MAUI. It's always concerning when there doesn't appear to be any dog-fooding going on - is this just some junk they're throwing at us that they don't want to use? I think some hesitation also comes from the Blazor side where it's looking like Microsoft doesn't really see Blazor as a React competitor so much as a way for internal company apps to be made quickly - in contrast to the Google IO presentations on WASM support for Dart/Flutter where they were emphasizing better-than-JS performance.
That said, Microsoft hasn't really released a lot of new (green field) stuff over the past 2-3 years. What product should they have made in MAUI, but didn't? You can't say Teams because that was a giant product way before MAUI even existed. Most of what Microsoft is doing is work on existing products - things they released before 2023/2024 and were in development before MAUI existed. Flutter had a 5 year head start on MAUI.
But there certainly is a feeling that Microsoft doesn't feel committed to it or at least not enough to put its weight behind it.
Never build a frontend on a .NET technology. Period. They always end up unsupported in the end. Just use standard web technologies and thank yourself later. I've been a .NET dev for a decade now and that's what I've learnt.
Blazor is pretty great. It is mature at this point and MS is using it internally more and more. Trying to go back to something like React makes me shudder. It's not perfect, but it's better than many alternatives.
I agree. For Blazor there is hope. It is standard based (web assembly, HTML, css) and it feels very intuitive particularly when compared to other spa frameworks like react. Also you can reuse all your html, css and design systems you have. Which is huge because like that it hooks up with the whole web development stack.
Tauri is pretty awesome. Rust backend, WebView front end. Nothing uses native desktop elements of course.
To be fair, there is no practical way to write native desktop applications using stylistically consistent UI elements AND have it be portable AND in a language that you enjoy using.
As far as I can tell, Windows 11 doesn't even have a toolkit with platform UI elements.
GTK on Gnome is pretty okay and GTK-rs is not dissimilar to React. Who know what MacOS uses but something something Swift XCode.
But I agree, just use web technologies. Write once, ship everywhere (and hum loudly when people complain about poor performance - joking, it's the vendors' fault we have to use web technologies).
In my experience with Tauri, it's pretty good on Windows, but not so much on other platforms, especially Linux. The decision to target different browser engines on each operating system means you still have to deal with a bunch of different OS-specific bugs.
For Windows you're dealing with Edge (so Chromium), on macOS you have Safari, and on Linux you have WebKitGTK. WebKitGTK has honestly abysmal performance, and you're missing a lot of modern standards.
The Tauri devs are looking at bundling a Chromium browser with it to deal with that, but that's still some time off, and leads to the same issue Electron has, where you have large bloated applications.
I don't know much about it but it seems like a weird combination. If you want high performance and low memory usage, you don't want HTML, if you want fast code writing, you don't want Rust.
I don't believe you generally end up writing a lot of Rust with something like Tauri. It is mostly web dev. While it is true that browser based UIs are slower than native, it isn't clear that .NET based UIs would be any faster while being very niche.
- HTML is the main way of designing interfaces (whether we like it or not)
- Rust is the main language promoted by intelligence agencies with multi-billion dollar budgets because it is laced with their backdoors (whether we like it or not)
I could see a case where the core logic needs to be performant, but the UI does not. The front end could be some menus, displaying (not a giant amount of) data, and a progress bar, while the back end does heavy computing.
And furthermore, if you want fast code writing, you write in the language you already know. For some people, that is Rust.
As wrong as it feels to have to use Electron for a desktop app, it really is the safest approach for most applications.
Qt also seems to be a good option, though there are licensing considerations for commercial applications.
I’m excited for various upcoming Rust options as well, but right now Electron is the battle tested option.
I am curious though about Avalonia. I’ve heard good things, but it’s definitely a smaller player compared to Electron. I’d most likely choose it over Microsoft’s first party frameworks.
Does their market share back up your take of them as horrible apps?
Are there QT or GTK competitors crushing them?
I always hear how terrible electron apps are, but the companies picking electron seem to get traction QT or other apps don't and seem to have a good cross platform story as well.
They said horrible user experience apps, not horrible apps. You can still deliver an app with a horrible user experience and build a profitable business. Ever done an expense report?
Companies aren't picking Electron due to inherent shortcomings in other platforms, they're picking it because it's easier (and cheaper) to find JavaScript devs who can get up to speed with it quickly.
Users will happily deal with a suboptimal experience as long as there are other things attracting them to the product. That's why Microsoft can do whatever it wants with Windows without worrying their users will run off somewhere else. So if you care more about people than businesses, maybe it shouldn't be an excuse to pick "better dev experience" over the user's.
Oh yes, the great old "works for me". On a yesterday's supercomputer, I presume? I live in a "developing" (have doubts it's really developing) country, most people are running laptops with no more than 8 GiB of RAM (sometimes it's 4 or less), and all this Electron nonsense is running like molasses, especially if you're trying to use a computer like a proper computer and do multitasking.
And most of the world is like that, very few of us (speaking globally) have $2k to drop on a new supercomputer every few years to run our chat applications.
You are right that WinForms and MVC have been around forever. However, Microsoft has continuously told devs that they are the past. So, you would be forgiven for expecting them to go away.
WinUI is the current official desktop paradigm and it is basically UWP from an API point of view. So the idea that UWP went away is not 100% accurate either.
Microsoft does not really abandon their UI tech like people say they do. But look how many different frameworks they have.
All of the above is Windows desktop only. There are a completely different set of UI technologies for the web.
UNO Platform (Open Source) allows you to use the WinUI API to target almost anything.
.NET MAUI is the official "cross platform" UI tech from Microsoft. It is what you use to target iOS and Android. As a bonus, you can target macOS and Windows too. On Windows, it uses WinUI. You will notice that the Linux desktop is missing from that list.
Here comes Avalonia to build MAUI on top of the Avalonia framework. This adds Linux and WASM to the list of platforms that MAUI will run on. Adding Linux is awesome. A lot of people have wanted that and it really completes the MAUI cross-platform story.
Adding WASM is neat but MAUI was never meant to target the web. If you use it for that, it is literally just the modern version of Silverlight. But Microsoft did not design it for that at all. It is just a back-end that Avalonia supports.
I was going to suggest the same, just use WinForms. It's basically feature complete, and it's going to be the last UI framework Microsoft is going to yank out from beneath you.
No kidding - kind of wild that winforms is still kind of a gold standard experience today! I actually liked VB Forms - lots of easy rapid application development was possible.
Generally speaking, I wouldn’t take what Microsoft uses as guidance nowadays, given a lot of the software they produce. (This is not an endorsement of MAUI.)
That has been a problem since forever. Microsoft themselves rarely used the tools they gave to developers. SourceSafe, MFC, WPF and the .NET frameworks that followed were only for 3rd party devs. And when they used these tools, the software usually got worse. One example was Visual Studio. 2008 was really nice with great customization and good performance. Then they wrote 2010 with MFC and it was slow and lost tons of features.
I think it’s better on the server side with ASP.NET.
As far as I have heard MAUI is pretty buggy and has lost momentum. It will probably go on the long list of basically abandoned .NET UI frameworks
You are mixing your UI frameworks and versions. VS 2010 is written in WPF. WPF is / was Windows Vista's and 7's UX. Old Control Panel in Win 10/11 still is WPF. All the wizards like ClearType wizard is WPF. MFC is much older (1992).
Unfortunately Microsoft likes to jump into bandwagons and many engineers at the company seem to like to reinvent stuff rather than adopt. WPF, WinUI2 and WinUI3 all share the same Xaml based structure. So they could have adopted WPF.
It is not that Microsoft doesn't develop advanced UIs with their frameworks. WPF is still well-used by Windows and other Microsoft utilities like Windows Terminal. They are just stupidly abandoning their built up bases for silly industry fads.
They jumped into tablet / touchscreen / hybrid-mobile-desktop bandwagon in late-2010s and tried to force WinUI as an UWP-only feature. It resulted in low adoption. They didn't adopt WPF to have same theming.
When WinUI2 failed, they tried to make modern C++ a reality and tried to remove UWP restrictions which is a good decision. However they diverted quite a bit resources into AI slop generation now and WinUI3 just languishes.
Same for MAUI. They tried to get into multi-platform, multi-device framework as a way to generate leads into Microsoft ecosystem.
They try to use various frameworks and UI stuff to get people hooked into the ecosystem and find ways to upgrade them into Azure and M365 customers. It is meaningless and tiring. All of those could be only WPF.
It is like Google and its many Bazel-like build systems (but not full Bazel) for each of Chrome, Fuschia and Android.
microsofts own stuff never seems to be what gets momentum. there's a strong aftermarket for better ways like back in the borland era bcb and delphi, the more things change the more they stay the same!
MAUI was horrible when I tried using it about a year ago, tons of bugs, pretty iffy comms/support from the MAUI team as to timelines when things might get fixed, etc.
Eventually dumped it and moved to Kotlin Compose/Multiplatform, which is just so much better at achieving a similar goal (though, obviously, without being part of the .NET ecosystem).
I go to "Launch MAUI in your browser" [0] in section "Try It Right Now". I click on "Word puzzle - A word puzzle game", third of the demo apps. I click on the "Randomize" button. Tiles start shifting around. I click on the left arrow at the top to go back to the main menu. The whole thing becomes locked with the word puzzle on the background and the main menu on top of it. Nothing is clickable any more.
Yeah the shuffling takes ages. Once it's finished you can go back. Also the movement of the tiles feels very strange with the mouse. The Click-Drag sometimes works and sometimes doesn't. No visual feedback is given on click. The tiles don't move but teleport, which is strange for a click-drag interaction. A simple click would have been better, or a move animation. Someone else noticed it, too https://news.ycombinator.com/item?id=45893986#45897805
The sliding puzzle is really finicky, have to try multiple times to slide stuff, the top left (back) arrow stopped working altogether until I refreshed the page, which of course takes you back to the original page... The time picker looks awful with the numbers not centrally aligned to the control (they look too high), and it's finnicky to use / same with the date picker.
> Using .NET MAUI, you can develop apps that can run on Android, iOS, macOS, and Windows from a single shared code-base.
This new development adds Linux and Browser to that list.
I recently tried out .NET MAUI to see how easy it was to build a hello world app. It was quite messy getting it setup on Mac but eventually I got a simple hello world app working. Nice to use XAML again after all these years. I always liked it.
I clicked a total of three times in the demo page before the tab froze and locked up.
MAUI is nice, and I'm glad it's getting decent Linux support now, but I would never use it in the browser.
Avalonia's big money ticket is taking old WPF projects, making minor changes to the XML to load their tools, and instantly making the projects work on macOS/Android/iOS/etc. (for an enterprise fee, of course). Great for forcing internal .NET applications built to run on Windows Vista into the modern world, but I don't think their web platform is a good idea for new projects because of the massive overhead you end up with.
> We are collaborating with the Flutter team at Google to bring Impeller, their GPU first renderer, to .NET. That work is already in progress and as it lands, the MAUI backend will inherit those gains.
Interesting, I wonder how good Impeller is and if it's actually better than the new Graphite backend of Skia.
Predictable performance: Impeller compiles all shaders and reflection offline at build time. It builds all pipeline state objects upfront. The engine controls caching and caches explicitly.
or as described here [2]
Flutter’s Impeller renderer outperformed Skia. Impeller eliminates runtime shader compilation stalls, delivering lower frame times and more stable performance. For animation-heavy, graphics-rich apps, enabling Impeller significantly reduces jank and provides a smoother user experience.
This is absolutely horrible and I hope I never have to use it on the web - right click doesn't work, my plugins don't work, I can't even open DevTools and when I manage (before the app has loaded), it's just a <canvas> element in the page. I know a lot of people put work into this and something is probably better than nothing, but actually no, I think such solutions are disgusting.
If you're gonna claim cross platform, don't just stoop down to the lowest common denominator and write a bunch of pixels to the screen, use the damn system UI solutions, or in the case of web apps - HTML, CSS and JS/WASM. I already take an issue with Electron (Lazarus lcl did things right, in contrast) but realize that it yields a lot of utility, yet these canvas based approaches for web apps feel like a step too far.
Yet obviously nobody can admit that they can't do the heavy lifting needed to support multiple different target platforms for the same widgets in a way that respects the corresponding platforms, while also still wanting to ship something. In this case, I think nothing is better than something, if that something is so bad.
My kingdom for a UI toolkit that can be used to make real CAD programs, and not yet-more things that just look like webviews and could just be a webpage.
You can make CAD programs with any toolkit as long as you have a GPU surface to render your actual CAD output to. You can do anything with this toolkit from weather apps to video editors, map analysis, and PDF editors.
You can check out the Avalonia demo reel to see what you can already do with the .NET GUI stack that MAUI now uses on Linux and on the web: https://avaloniaui.net/showcase
Yes, that's my litmus test for GUI toolkits - is it suitable to develop Photoshop-like software? Or is it only good for yet another weather or todo apps?
This 100%. I hate the trend of UX/UI that got unleashed upon us in the last decade of the web. Everything is scaled up for touch interactions and has to have fancy animation and very "comfortable" spacing around elements.
I wish we can go back to UIs that focus on information density and usability. I love looking at Japanese websites because of this.
It's great to have .NET ecosystem expanding.
I am a .NET dev, and I still use web optimized for mobile as my mobile platform(where applicable). Tried MAUI and all the other stuff, just doesn't cut it for me. You can make an app using it, but I would rather not to. The best 'mobile platform' for me was blazor hybrid, but then again - if it's already blazor why not go full web...I guess it depends on the 'seriousness' of your mobile application. If I had to develop a complex mobile app, I might choose another language framework, cause MAUI uses XAML and MVVM stuff that is quite a big overhead IMO.
I wish it was a real "systems" language as well, i.e. not single-threaded/javascript-mindset one. That's why I hope someday we will have a nice cross-platform framework for Swift.
Go is frankly the polar opposite of C#. Go compiles to a native binary with no runtime dependencies and it relies on simple garbage collection and static linking. C# runs on the .NET runtime which is heavily managed, gives you JIT, reflection, dynamic code generation and so on. Go views concurrency as fire and forget, C# views it as compose and await. Go is extremely explicit while C# is extremely implicit.
Now, I understand that you may talk about it from a non-technical perspective, but even so, there are major differences. C# is a general purpose language for the cloud/web, and so is Go, but Go is also widely used in other areas like in embeded software. TinyGo is soooooooo much better than working with C/C++ or Rust as an example. Places like that where you wouldn't usually find a transpiled language (other than maybe Python with MicroPython).
C# has AOT compilation which creates a single, native binary. This has gotten so much better with .NET 10 and since the introduction of source generators to deal with reflection issues.
Also, check out nanoFramework for a .NET runtime that can run on MCUs like the ESP32 [1]
.NET can be compiled to native binaries. .NET can be used for embedded development.
C# is more expressive and .NET comes with batteries included. Go is more explicit and more verbose.
You can pick up Go faster and is easier to reason about Go code when you first encounter a new project but C# feels like it enables you to develop faster and be more productive.
For web both are excellent and performant, even if they have different philosophy.
What I like about C# is that it becomes more functional and I can even mix F# in the projects if I want even more functional programming.
Would be interested to know about the a11y story on web and other platforms. Does it integrate with screen readers, accessible navigation tools, a11y apis, etc.
This is somewhere quite a few tools fall down. For example, for web: Rendering to canvas doesn't create the dom nodes required to hook into the tools, and creating the nodes appropriately can have a significant performance impact.
What about native a11y tools and tech.. there are some ios specific ones that you'd expect on an ios app for example.
To clarify, it's using React Native XAML, where XAML is Microsoft tech for building UIs. So they are dogfooding but they have too many UI frameworks to choose from.
You wouldn’t use MAUI to build something for only one platform. You would just use whatever it’s an abstraction over for that platform which in the case of Windows is WinUI.
You can make your own better system which supports GUI and provides a reasonable runtime that works on all current major platforms and provides development tools, years of ecosystem development, etc. etc.
If it's better than what MAUI provides and you can support it for years, I'm sure that could take over and many people would use it instead. But... will you and why?
Anything would be better than another UI toolkit made by Microsoft. They will abandon it in the matter of a few years like they did about 10 times already.
I started moving to using GO and Fyne for cross-platform GUI tooling. GO 1.21 supports Windows 7 with Fyne. Ya still have to support Windows 7 for work. No more wasting time with managing a Windows installer.
QT Framework is still one of the best for cross-platform desktop applications when speed is key.
Like I said tooling. Means that the User is well defined where Accessibility is not an issue. The tooling I creating is for people that have to engage with physical equipment; from replacing parts to making sure a motor is working properly.
.NET works amazingly on the web. This is just not the UI framework you would use.
There is ASP.NET of course and Razor Pages. We all use apps built with these every day without even realizing it. There are other great frameworks as well.
I do not even see Blazor as a real web technology but of course it is positioned that way.
MAUI is a "cross-platform" and frankly mobile first UI framework. It was never meant for the web.
I wouldn't use it for consumer apps because it requires a Websocket connection to maintain state and probably doesn't scale very cheaply... but for business applications or personal tools it's actually kind of insane how much functionality you get out of the box (at least by the standards of statically typed languages).
I don't know if it has since improved, but .NET MAUI was really, really rough when I created a mobile app for my employer last year. I'm talking basic things - changing basic colors on the toolbar (1), putting non-text content inside a button (2), basic trigger behavior (3), to list a few. Not to mention that .NET UI has been years behind on hot-reload and developer tooling. Additionally, It was a fight to keep our app performant. The XAML compiler is a step in the right direction, but we had relatively simple views (in the dozens of components) absolutely tanking our FPS. I know there is probably some of my skill issue in there, but when I find basic things taking hours to optimize that I wouldn't even think about in React, I start to wonder about the framework. I spent a lot of time creating PRs on .NET MAUI but their team appears quite small and overloaded. I wish them the best - they're some talented folks, but I don't envy their job.
I can't help but think of Joel Spolsky's Things You Should Never Do (5) - the transition from Xamarin to .NET MAUI feels like a very similar mistake to Netscape. All of the battle tested Xamarin code, documentation, community examples, packages, etc. is now dead and has to be converted over to .NET MAUI.
On top of that, XAML just doesn't do it for me - having to deal with code-behind, MVVM view models, custom converters, and the actual XAML files themselves is insane for what is usually just a a single file in JS. The fact that you need to write a "InvertedBoolConverter" (4) just to flip a boolean is the most Microsoft thing ever. MAUI feels like it's designed just to keep a large development team busy. I'm not joking, we have a 42 line file that's only purpose is to flip booleans for XAML views.
We're a C# shop so it was nice to share our common C# with our desktop application, but I don't think it was worth it in the end. Sure JS has its problems, but I'll take those problems any day over MAUI.
I hope Avalonia can fix .NET MAUI - it'd be a massive kudos to them if they can smooth it over, but I can't say I'd willingly rely on this project long term.
Well, I guess Avalonia can solve 4 at least as you can negate a binding[1]. Good news for me as I recently started an Avalonia project, and thought you did still need an InvertedBoolConverter.
This is awesome… but:
On the Web, is Avalonia using Skia to render inside a Skia Canvas?
**insert inception meme here**
Joking aside: this points to MSFT moving away from the whole Mono/Maui investments and into Aspire or whatever they call it. Without MSFT backing this I am not sure if there is much more future left for MAUI (or dotnet on mobile in general).
If it runs on Avalonia anyway why choose MAUI? If you have the option just use Avalonia.
MAUI is unfinished and its docs is awful.
Do you want to build an app using MAUI? Unless you build an app that barely deviates from the template, expect to desperately search through decade old Xamarin documentation and figure out the details through painful trial and error.
It is a way to get people choosing the "official" path (ie. choosing MAUI) to experience Avalonia. They are hoping you come for the MAUI and stay for the Avalonia and become an Avalonia developer.
As for, why choose the Avalonia version of MAUI, there are three reasons:
- Linux support (the big one I think)
- Drawn framework (same renderer on all platforms)
- WASM support (probably useful sometimes but not the real draw)
They are making a big deal of WASM here because it is easy to demo. We can all go into it and run it. But do we want to use it for our apps?
As for, why not Avalonia directly? To loop back to the beginning, it is because you do not yet know Avalonia and trust it. The Avalonia team is hoping this helps with that.
Thank you for explaining this. I didn't even realize this was an Avalonia project until you pointed it out. I like Avalonia (and love .NET in general), but I think the messaging on this needs a lot of work. Avalonia is creating unnecessary cognitive dissonance by emphasizing MAUI, which is a competing project after all.
This is what mystifies me about this announcement. Avalonia already works fine on Linux, allowing anyone to build a cross-platform .NET GUI application.
MAUI is supposed to be a wrapper around native widgets. The fact that they had to use Avalonia under the covers to get it to work on Linux seems to defeat the point. (Avalonia is a complete UI toolkit, like Qt or Flutter, that owns the entire stack from XAML to pixels.)
A toolkit announcing “Linux support” is pretty ambiguous. Does it mean Xorg support? Wayland support? Framebuffer support? The announcement provides no clue about this.
They are not announcing Linux support for Avalonia. It has long supported Linux.
What they are announcing is being able to run MAUI on Avalonia such that you can run your MAUI apps on any Avalonia supported target. This includes both Linux and WASM, neither of which is supported by MAUI directly.
Since it's using Avalonia, I'd say it's just X support at the moment. They've announced that they intend to support Wayland, but that was a couple months ago[1], so I doubt that's ready.
Honestly, I feel that everything UI related has gone backwards to the stone age.
I wonder how hard would it be to go back to visual designers like we had with Delphi or VB6. There were flexible layout container components which helped a lot when adapting forms to varying screen resolutions.
Avalonia is also working towards using the new Flutter rendering backend Impeller which is being used to replace Skia, which is used by Chrome for rendering, turtles all the way down:
This is a huge win for people using MAUI. The only thing I worry is some performance loss from using Avalonia backend instead of using directly the target platform.
It seems like alot of people in this thread dont undestand the dotnet stack. MAUI, Native UIs, and WASM are all interchangable. Your cross platform app could literly be built out of everyone of of these components at once. MAUI is not required for use-cases generally when you have a webUI, its really targetting at the simple UI, write-once, run anywhere crowd. Once you outgrow it, you can move to native integrations using dotnet linux/mac/windows. Or you can just integrate with webviews, and have C# backend's or WASM backends. The combination is limitless.
Xamarin turned into dotnet, Xamarin.forms turned into turned into Maui. Name another large tech company that has embraced community projects like this and pushed provided enterprise support for community driven stuff over the same period.
I was writing cross-platform apps before Flutter, angular, or whatever other language you choose for 11yrs now. Find another framework thats done this before the scare tactics of "MS will abandon this" rhetroic. dotnet is the premier cross-platform enterprise ready framework full stop. No other framwork has the backwards compatiblty while maintaining paritity with the latest OS APIs. No other framework can serve millions of pages per/sec while supporting pixel perfect UIs and code-reuse. They have invested so much money in building automation that means as soon as the OS releases an API, your getting access to it. In real world terms, this is what counts if you want to build cross-platform stuff that your clients cant tell isn't native.
I run the same code from 2014, today, in apps in all the stores. Over the years all i have had to change is various namespaces to take advantage of the latest enhancements. Code that ran on dotnet4, silverlight, xamarin, still runs today on dotnet10. I share 90% of code across all platforms yet clients cant tell they are not native apps. Thats what i call return on investment.
I agree that, despite all the complaining, code re-use and longevity on .NET is second to none.
Where there is intense confusion in these comments is in thinking that MAUI is a web application framework. MAUI is "cross-platform" as you say but really mobile first with desktop brought along for the ride. It is in no-way intended to be a web application tech.
Avalonia has created an Avalonia back-end for MAUI which means running MAUI apps wherever Avalonia is supported, and that includes WASM. So now you can run MAUI apps in a browser. You can but that does not make MAUI web tech.
I can run Windows 95 in a browser. That does not make Win32 a web API.
Pathetic that MS doesn’t manage to do this themselves.
Even more pathetic that MS doesn’t actually use it for any of their products, so forgive me if I have very little faith in the future of MAUI with all its bugs.
Not in the industry I work in - AAA gamedev. Art folks typically would have two, or even three monitors - so good solutions for docking across them (and still working) are required.
A chrome browser by itself can't work that - it's great for many things, but not for Creative Tools.
MAUI is first and foremost a mobile UI framework. It was built for iOS and Android. The desktop is added on as a bonus so that you can target macOS and Windows as well.
If you are building a Windows desktop application though, Microsoft does not want you to use MAUI. You use MAUI because iOS and Android are your top platforms and you want to target macOS and Windows without writing dedicated applications.
Linux has always been missing. This Avalonia port fills that gap.
You would not target the web with MAUI either. I guess "you can" now because WASM is one of the platforms that Avalonia supports. Again, I guess you might if you already have a MAUI app and do not want to create one for the web. But you would never set out to create a MAUI app for the web.
1) In today's American political climate I think it's appropriate to express appreciation for immigrants to like Miguel de Icaza who dragged Microsoft kicking and screaming into cross platform .NET and is the godfather of .NET Maui
2) As someone who came up developing desktop apps for Windows and Mac, I never liked developing web applications. There was so much lacking, but now developing web apps is becoming like developing desktop apps. Now you get/put your data from HTTP calls instead of file system and database calls or with Blazor and SignalR you don't even have to think about those. This may seem obvious to younger programmers today, but it would have seemed like magic back in 2004 when Dymanic HTML and Ajax (both Microsoft) were being invented.
3) I'm grateful Microsoft has changed their old ways to be a forward thinking company. They still have problems that any GIANT, Inc. organization has, but let's not forget how far they've come.
This is interesting for sure. Kudos for bringing this capability to the web!
One issue the demos reveal is, it doesn't _feel_ like the web. That is, I can't hit Ctrl+F to find text on a page. I can't select text with my cursor. I can't copy the address of a hyperlink. On my phone, I can't hard press on an image and share it to others. Screen readers can't handle it. I can't press a shortcut key to make everything larger.
These all may seem pedantic, but they contribute to the feeling "this is not the real web."
This is the same problem with Java applets in the late '90s, Flash and Silverlight in the early 2000s. They are islands of richness within a web page, but those islands are, well, opaque to browsers, search engines, and virtually all web tooling.
That's not pedantic at all! Indeed, without these capabilities, it is by some definition not the real web.
This hits into that concept of what exactly the "web" is. Is it just a media transport system? Or is it something more than that. Of course, we could cite Tim Berners-Lee here or Roy Fielding in this discussion.
But at minimum, I think a lot of us are tired of the app-lification of the web and somewhat wish we could have a bit of the old.
It's also an interesting question, why, in traditional rich desktop applications, I can't say I have ever missed the ability to select and copy text from the UI chrome - whereas on the web I'd definitly miss it and in badly designed mobile apps, I often do.
I think some part of UI design degraded with the web, where there used to be a clearer distinction between "user data" and "app chrome" areas than there is today.
I'd also like if we could get back to selections of more complex data types at some point and not just treat everything as text. UI toolkits have all kinds of lists and treeviews to model selectable entities, whereas in the browser, there just a single huge wall of text for everything.
> why, in traditional rich desktop applications, I can't say I have ever missed the ability to select and copy text from the UI
I do miss this on an almost daily basis and I have stopped paying for services that force me to use an app without offering a website.
The last instance of this was just a couple days ago when I could not copy a tracking number from an e-commerce app (to then paste it into the shipping company website) but at least this e-commerce company has a web UI so I could rely on that.
Oh and the other one that I miss almost daily is cmd-F / ctrl-F
"E commerce apps" are very much not the sort of traditional desktop application they were referring to. Note that they add "in badly designed mobile apps, I often do."
They're referring more to things like "you can't copy the text labeling the brush width field in Photoshop" (but you CAN copy the text out of that editable field). It's a part of app design people are extremely lazy with today, as you note.
In any sensibly designed desktop package tracking app that number would've been selectable or copy-able text, like how an email subject is in a desktop email app. (Thunderbird, say.)
(Interestingly, ctrl-f to find is one that many apps/OSes have now borrowed back, with the ability to "find" items in menus through a Help menu -> Search action.)
While I do occasionally miss it there as well, I think the main difference is that I very rarely use desktop applications for information gathering.
I never "read" a desktop application, whereas that is mostly what I use a browser for. And if I can't properly interact with text on a website, then I would likely reach for something else.
Back in ye olden days desktop applications for information gathering like Encarta let you select and copy text because they were thoughtfully designed and knew that "information you were gathering" should be different than "application chrome" - that's the distinction being made here.
Information-oriented desktop apps still do this - any good email client, for instance, should make it trivial to copy a subject line or "to"/"from" address even if it's in the UI chrome.
> in traditional rich desktop applications, I can't say I have ever missed the ability to select and copy text from the UI chrome
I forgot what desktop application it was, but there was a time that I repeatedly needed to copy texts from a dialog, which didn't support text selection. It frustrated me so much, that I put together a script to do OCR on the dialog.
Supporting complex data types for copy & paste is good; but it is almost trivial to also support plain text copying as a fallback when it already supports copying of other mimetypes. The problem is that some UI has no support of copying in any format at all.
If it was a standard Windows dialog box by any chance, you could just have pressed Ctrl+C with the dialog in focus to copy the message. It's one of these subtle things that go almost completely overlooked.
> traditional rich desktop applications, I can't say I have ever missed the ability to select and copy text from the UI chrome
You've never had to type error code/message instead of copying&pasting? Or use search to jump to a specific settings section?
On Windows, with common messages boxes, you can just do Ctrl+C for copy and you get the message box text in the clipboard.
Don't know if that helps you particularly, but it is great when it works and little-known.
Thanks, doesn't help me, but you're right, a good tip to know. Though I'd still prefer a similar option to start selection directly in the UI instead of finishing the job in a text editor, this would also help highlight text in a screenshot without having to do image post-processing! I'd even accept some arcane finger-breaking ctrl-alt-win-x-y-z (which I could rebind) for the privilege
All the more annoying when such years-old fundamentals are broken in all the new "supposedly better" frameworks
I’ve never done it twice, I can tell you that much!
Not being able to copy text from UI interfaces is just normalization of deviancy. It should be the norm and it’s subpar when not possible imho
IMHO there's no gatekeeper of what the "real" web is or should be. It grew organically - regular people building things they liked or needed. It's certainly more of a life necessity than it used to be, but that happened organically too.
I know there are strongly held opinions about this, but I for one see no reason why the "application web" can't peacefully coexist, and interlink with, the document web. In my opinion it therefore makes sense to allow for different models for the application web, ones that do not revolve around a document.
On the other hand, if we're just bashing on javascript being the lingua franca of the web, that's a train I'll happily board!
If the “application web” can’t share the text to another app,
then forget that.
Not using the standard web stuff usually means it's also an accessibility nightmare, tried using a screen reader on the demo and it doesn't work at all unfortunately
I wonder if at any point browsers will offer a low level accessibility API for you to manually describe components. I’ve worked in the web for years and I’m a big believer but it’s also indisputable that Canvas offers more performant UI rendering than HTML when done correctly. I don’t think it should ever be used for web “documents” but web apps already bastardize HTML and CSS to achieve their aims anyway. Accessibility remains the missing component.
As far as standards is concerned, that API is ARIA [0].
W3C already offers guides for accessibility and canvas. But no one who opts for canvas turns around and remembers to do their landmarks.
Then I’m showing my ignorance… how do you add ARIA landmarks to Canvas elements?
You would create transparent DOM elements in the right places with the right ARIA attributes and content, I suspect.
I guess that’s what I’d like to see a better API for, then. Mapping on click events for invisible elements feels like a hack.
OTH, we are still failing to provide a bare minimum for accessibility. Heck, we even needed a law (in the EU, that than needed to be translated to national law), so that companies providing crucial end user services would care about accessibility.
It's HTML imagemaps from the 90s, when we could not style buttons and navbars where GIFs with links in the right places. Browsers still have the code to render them.
https://developer.mozilla.org/en-US/docs/Web/HTML/Reference/...
https://caniuse.com/mdn-html_elements_map
That's possible. But it is difficult to get right, and can have poor performance if you you have many such elements.
> I wonder if at any point browsers will offer a low level accessibility API for you to manually describe components
Accessibility Object Model:
https://wicg.github.io/aom/spec/
It's very slowly coming together, but it won't be rone for many years yet. Especially since what you want is Phase 3.
What screen reader? Over the last few years AI's ability to understand images has improved a lot.
I'm not aware of any screen reader that works by continuously feeding screenshots of user interfaces into a remote expensive image LLM, which is an absolutely insane and impractical idea for many reasons, but I used standard TalkBack on Android
That’s not exactly fast for people who need these tools though.
I can't code, I'll use an LLM to write one!
I can't use your app, I'll use an LLM to read it!
MAUI was never intended for the web. This is not what Microsoft wants you to use it for.
WASM is just one of the platforms that Avalonia supports and so, if you run MAUI on Avalonia, you can run it on WASM.
If you do that though, it is going to be like rendering any other desktop GUI toolkit in WASM. It is not a web app. I mean, it is cool you can do it and MAUI in WASM is better than no web capability at all I guess. But you would never set out to create a web app in MAUI.
MAUI on Avalonia on WASM is really a modern replacement for Silverlight. And it will likely be about as popular.
The really cool thing is being able to target the Linux desktop finally. A lot of people will love that.
And, while MAUI was meant to use native controls on each platform, many people may prefer the Avalonia approach of having your app render the same everywhere.
> I can't hit Ctrl+F to find text on a page. I can't select text with my cursor. I can't copy the address of a hyperlink.
I was intrigued before I read this. This stuff is a non-starter for me.
I think it's the same problem that flutter web has, and probably any other canvas/wasm based backend? Those features still need to be implemented, while still missing out on accessibility?
Agree. The examples feel a bit like I'm using a specific window in remote desktop session.
If this can’t support web standards it’s a nonstarter for me.
Yeah. I think you need to render to actual DOM nodes when targeting the web if you want a first class experience.
We're betting on this over at https://github.com/DioxusLabs/dioxus where we're building a cross-platform UI solution that enables you to do this by having a web-centric API (we are developing our own custom HTML/CSS renderer for native platforms).
Let's move the goalposts downfield. If you can't go into developer mode and mess with the DOM, and JS, it's not real web.
Unironically – yes.
>One issue the demos reveal is, it doesn't _feel_ like the web. That is, I can't hit Ctrl+F to find text on a page. I can't select text with my cursor. I can't copy the address of a hyperlink.
That's because MAUI is intended for mobile and desktop apps.
If you want to use .NET for front-end web SPA, you can use Blazor which will behave exactly like you asked.
Am I losing it or am I looking ClearType on OSX?!
I get the value in this and realize it's not for your polished -$500 ARPU consumer social apps, but man this is weird.
(Also if anyone who worked on it is here, it's crashing for me on OSX 26, Chrome 142.0.7444.135, if I run an animation and hit back as the animation finishes)
I'm currently in the process of creating a piece of development software and needed a cross-platform GUI environment to do it with.
I first tried Maui because its seemingly so highly rated and recommended. It turned out to be a complete nightmare. Even just installing it didn't just work, I had to spend 2-3 hours just to get that going and then when trying some basic things for my UI it turned out to just not be supported.
The next on the list was AvaloniaUI and wow, it was the complete opposite experience. Installation was a breeze and with only a marginal bit of googling I managed to do all kinds of things such as making a top-most only borderless transparent window etc.
The interface in Visual Studio could be a bit better overall, but compared to Maui its light-years ahead.
In the .NET ecosystem, I have noticed people to shame .NET MAUI because Microsoft themselves don't use this framework - Microsoft Team is built on Electron and not MAUI.
Why build a product on MAUI when Microsoft aren't too sure about it.
The answer to that is well known: Windows division builds WinUI/Win32 as their native C++/COM API, Office division went to React on their path to the web and the dev division fills gaps (WPF) and provides tools for external and internal devs (Maui for cross platform uis).
It is history not the lack of will. At one point the windows division was in shambles (remember vista) and WPF pops up. At another point, the windows and dev division have no answers to the office group (because you know who uses non win tech) so they went react. And then external devs screamed: where is the .net cross platform story so Microsoft acquired xamarin and later form Maui out of it.
It is history not lack of trust. But the outcome is the same: lackluster support for all UI toolkits.
This feels a lot like the Linux desktop ecosystem where many apps have a different look and feel (GTK, Qt and a bunch of others).
Well summarized, and just as shocking today as it was every minute while it developed.
Someone needs to remind those cats that they own the platform. Being able to sanely develop apps for and on that platform should be possible, and UI kinda-sorta matters for that. At a certain point with the MFC they had it dialled in, while pioneering asynchronous browser tech, with many best in class tools. Decades later with a cross-platform cloud-centric stack they have a shrug emoji as big and wide as the eyes can see, and no sense this basic question of development will ever get improved.
Ballmer chanting ‘developers, developers, developers …’ springs to mind.
I understood (can not confirm it though) that the new start menu in Windows 11 was built using React Native, so yet another ui framework in the mix.
These days, I think Microsoft's web-based desktop apps mostly use WebView2 directly instead of Electron, so they don't have to bundle a browser. I think for Teams it happened at the same time that they moved from Angular to React.
The point about them not using MAUI still stands though. From what I understand, the .NET world has either adopted different abstractions like Avalonia, or stuck with tried and tested solutions like WinForms with proprietary controls. After all, they've seen this before with WPF which was never fully adopted by MS either, or with the debacle around Metro/WinRT. You're never quite sure what Microsoft wants you to use or will support in the long term. They also make Blazor, which is a different (and likely more accessible) way to build web apps with .NET.
Since we're on the subject of companies not dogfooding their shiny tech, is Google really using Flutter for their own apps? I feel like the evolution of the Android ecosystem towards Kotlin and Jetpack Compose implies otherwise.
Ive been building these apps (cross platform web based ui, C# backend) for years, and yes its finally good to see MS catch up and validate the architecture ive pushed since Xamarin. I wrote once wrote an electron version of this archand thought wtf are people doing? Things can be so much easier when you use a platform that knows how to multi-thread. At stages i had to build adapters/upstream patches for Chromeiunium directly onto Mac and Linux, and its was a major pain having to debug C calls.
Ive been using the same framework now for 10+yrs on apps in the stores, i wrote a small layer infront of the webviews and can swap out webkit, chrome, edge on demand. You really dont need much, just a constand way to boostrap logic and UI. 90% of code is shared across all platforms, there are def differences in WebView engines that you sometimes come across but those parts just get swapped out with browser specific JS. Ive found bugs and worked with browser teams at all vendors doing this and to see how simple this is with Dotnet these days compared to when i started is refreshing. Its easily the most stable cross-platform framework around, if you are stuck using something like flutter i pitty you, its just eletron with another skin. I can swap out and integrate directly with OS libs when i need to do stuff that the dotnet team hasnt gotten around to yet without re-writing. This has mean i really havnt used MAUI at all, but if i need to or could take advantage of it i can mix it on an Ui element by element basis. I prefer webUIs though, i have the chose to handle anything with either JS, WASM, or a combination. I can use traditional JS frameworks or traditional Native UI frameworks.
If i had started this process later, avalonia seemed to have the closest thing i required. It was just a bit a more complex /based on WinUI (which i dont really enjoy) but it supported all platforms and gave lower level api access. MS were smart, that canabalanised all these community effort and brought them into the fold. Every dotnet webview impl was a successfull community driven project before. They didnt write anything themselves from scratch.
> Microsoft Team is built on Electron and not MAUI
Microsoft Teams was released in March 2017. .NET MAUI was released in May 2022. In 2021, Microsoft replaced Angular with React and moved away from Electron to WebView2 (using the OS' built-in renderer rather than a bundled Chromium). So even the rewrite was a year before MAUI (and they probably started the rewrite before 2021). Plus, part of the point of using React there was that they could basically replace Angular bit by bit.
Microsoft Teams is just older than MAUI. It's like asking why Hadoop is written in Java and not Go or Rust or why Kafka is written in Scala and not Kotlin. Kafka was open sourced in January 2011 and Kotlin came out in July 2011. Kotlin wasn't an option given that they were developing it years before the language was released.
That's not to say that Microsoft's attitude toward MAUI doesn't leave concerns. There was some news a while back about a bunch of layoffs around MAUI. It's always concerning when there doesn't appear to be any dog-fooding going on - is this just some junk they're throwing at us that they don't want to use? I think some hesitation also comes from the Blazor side where it's looking like Microsoft doesn't really see Blazor as a React competitor so much as a way for internal company apps to be made quickly - in contrast to the Google IO presentations on WASM support for Dart/Flutter where they were emphasizing better-than-JS performance.
That said, Microsoft hasn't really released a lot of new (green field) stuff over the past 2-3 years. What product should they have made in MAUI, but didn't? You can't say Teams because that was a giant product way before MAUI even existed. Most of what Microsoft is doing is work on existing products - things they released before 2023/2024 and were in development before MAUI existed. Flutter had a 5 year head start on MAUI.
But there certainly is a feeling that Microsoft doesn't feel committed to it or at least not enough to put its weight behind it.
Teams is not older than MAUI because MAUI was mostly a rebrand of Xamarin Forms.
Microsoft seems more committed to react native than MAUI.
Never build a frontend on a .NET technology. Period. They always end up unsupported in the end. Just use standard web technologies and thank yourself later. I've been a .NET dev for a decade now and that's what I've learnt.
Blazor is pretty great. It is mature at this point and MS is using it internally more and more. Trying to go back to something like React makes me shudder. It's not perfect, but it's better than many alternatives.
I agree. For Blazor there is hope. It is standard based (web assembly, HTML, css) and it feels very intuitive particularly when compared to other spa frameworks like react. Also you can reuse all your html, css and design systems you have. Which is huge because like that it hooks up with the whole web development stack.
Tauri is pretty awesome. Rust backend, WebView front end. Nothing uses native desktop elements of course.
To be fair, there is no practical way to write native desktop applications using stylistically consistent UI elements AND have it be portable AND in a language that you enjoy using.
As far as I can tell, Windows 11 doesn't even have a toolkit with platform UI elements.
GTK on Gnome is pretty okay and GTK-rs is not dissimilar to React. Who know what MacOS uses but something something Swift XCode.
But I agree, just use web technologies. Write once, ship everywhere (and hum loudly when people complain about poor performance - joking, it's the vendors' fault we have to use web technologies).
In my experience with Tauri, it's pretty good on Windows, but not so much on other platforms, especially Linux. The decision to target different browser engines on each operating system means you still have to deal with a bunch of different OS-specific bugs.
For Windows you're dealing with Edge (so Chromium), on macOS you have Safari, and on Linux you have WebKitGTK. WebKitGTK has honestly abysmal performance, and you're missing a lot of modern standards.
The Tauri devs are looking at bundling a Chromium browser with it to deal with that, but that's still some time off, and leads to the same issue Electron has, where you have large bloated applications.
https://github.com/tauri-apps/wry/issues/1064
> As far as I can tell, Windows 11 doesn't even have a toolkit with platform UI elements.
They do, it's called WinUI 3. It's barely used for all of the aforementioned.
I tried this. Their examples don't even compile lol
> Rust backend, WebView front end.
I don't know much about it but it seems like a weird combination. If you want high performance and low memory usage, you don't want HTML, if you want fast code writing, you don't want Rust.
Use the slower but easier to write languages for front end is the norm for complex apps. Many apps that passed the trial of time are like that.
Blender: frontend Python, backend C++.
Houdini: frontend Python(PyQt), backend C(presumably)
Sim City: frontend JavaScript, backend C++
The reason is very simple: frontend is more error tolerant, but less resistant to the product designer's whims (or the users' desire to customize.)
I don't believe you generally end up writing a lot of Rust with something like Tauri. It is mostly web dev. While it is true that browser based UIs are slower than native, it isn't clear that .NET based UIs would be any faster while being very niche.
There's Wails for Go backend and webview frontend; https://github.com/wailsapp/wails
A very reasonable combination:
- HTML is the main way of designing interfaces (whether we like it or not)
- Rust is the main language promoted by intelligence agencies with multi-billion dollar budgets because it is laced with their backdoors (whether we like it or not)
I could see a case where the core logic needs to be performant, but the UI does not. The front end could be some menus, displaying (not a giant amount of) data, and a progress bar, while the back end does heavy computing.
And furthermore, if you want fast code writing, you write in the language you already know. For some people, that is Rust.
I sure never will write a single line of JS/TS again now that Blazor exists and is stable.
As wrong as it feels to have to use Electron for a desktop app, it really is the safest approach for most applications.
Qt also seems to be a good option, though there are licensing considerations for commercial applications.
I’m excited for various upcoming Rust options as well, but right now Electron is the battle tested option.
I am curious though about Avalonia. I’ve heard good things, but it’s definitely a smaller player compared to Electron. I’d most likely choose it over Microsoft’s first party frameworks.
> it really is the safest approach for most applications.
It's also the option which gives your users by far the worst experience. Not worth it at all, imo.
Not really. The downsides are mostly overblown.
Plenty of category leading applications like Discord, VSCode, Slack, Figma, etc. use it quite successfully.
All of those are examples of overbloated, slow, horrible user experience apps.
Does their market share back up your take of them as horrible apps?
Are there QT or GTK competitors crushing them?
I always hear how terrible electron apps are, but the companies picking electron seem to get traction QT or other apps don't and seem to have a good cross platform story as well.
They said horrible user experience apps, not horrible apps. You can still deliver an app with a horrible user experience and build a profitable business. Ever done an expense report?
Companies aren't picking Electron due to inherent shortcomings in other platforms, they're picking it because it's easier (and cheaper) to find JavaScript devs who can get up to speed with it quickly.
Users will happily deal with a suboptimal experience as long as there are other things attracting them to the product. That's why Microsoft can do whatever it wants with Windows without worrying their users will run off somewhere else. So if you care more about people than businesses, maybe it shouldn't be an excuse to pick "better dev experience" over the user's.
They work great for me.
Oh yes, the great old "works for me". On a yesterday's supercomputer, I presume? I live in a "developing" (have doubts it's really developing) country, most people are running laptops with no more than 8 GiB of RAM (sometimes it's 4 or less), and all this Electron nonsense is running like molasses, especially if you're trying to use a computer like a proper computer and do multitasking.
And most of the world is like that, very few of us (speaking globally) have $2k to drop on a new supercomputer every few years to run our chat applications.
Hey, I found CEO of Discord
So you've never used WinForms, WPF and MVC
MVC is not .NET of course.
You are right that WinForms and MVC have been around forever. However, Microsoft has continuously told devs that they are the past. So, you would be forgiven for expecting them to go away.
WinUI is the current official desktop paradigm and it is basically UWP from an API point of view. So the idea that UWP went away is not 100% accurate either.
Microsoft does not really abandon their UI tech like people say they do. But look how many different frameworks they have.
All of the above is Windows desktop only. There are a completely different set of UI technologies for the web.
UNO Platform (Open Source) allows you to use the WinUI API to target almost anything.
.NET MAUI is the official "cross platform" UI tech from Microsoft. It is what you use to target iOS and Android. As a bonus, you can target macOS and Windows too. On Windows, it uses WinUI. You will notice that the Linux desktop is missing from that list.
Here comes Avalonia to build MAUI on top of the Avalonia framework. This adds Linux and WASM to the list of platforms that MAUI will run on. Adding Linux is awesome. A lot of people have wanted that and it really completes the MAUI cross-platform story.
Adding WASM is neat but MAUI was never meant to target the web. If you use it for that, it is literally just the modern version of Silverlight. But Microsoft did not design it for that at all. It is just a back-end that Avalonia supports.
> MVC is not .NET of course.
https://en.wikipedia.org/wiki/ASP.NET_MVC
I feel like they're both talking about MFC instead..
Except WinForms, spectacular for Windows-only utility GUIs.
I was going to suggest the same, just use WinForms. It's basically feature complete, and it's going to be the last UI framework Microsoft is going to yank out from beneath you.
> spectacular
Not exactly the word I'd use, since it really hasn't changed since VB4, but it's definitely reliable and stable.
No kidding - kind of wild that winforms is still kind of a gold standard experience today! I actually liked VB Forms - lots of easy rapid application development was possible.
WPF as well
IIRC, because of Wine, Mono has enough of WinForms to make a few things work. But who the hell wants to distribute an application with a Wine runtime?
Generally speaking, I wouldn’t take what Microsoft uses as guidance nowadays, given a lot of the software they produce. (This is not an endorsement of MAUI.)
That has been a problem since forever. Microsoft themselves rarely used the tools they gave to developers. SourceSafe, MFC, WPF and the .NET frameworks that followed were only for 3rd party devs. And when they used these tools, the software usually got worse. One example was Visual Studio. 2008 was really nice with great customization and good performance. Then they wrote 2010 with MFC and it was slow and lost tons of features.
I think it’s better on the server side with ASP.NET.
As far as I have heard MAUI is pretty buggy and has lost momentum. It will probably go on the long list of basically abandoned .NET UI frameworks
You are mixing your UI frameworks and versions. VS 2010 is written in WPF. WPF is / was Windows Vista's and 7's UX. Old Control Panel in Win 10/11 still is WPF. All the wizards like ClearType wizard is WPF. MFC is much older (1992).
Unfortunately Microsoft likes to jump into bandwagons and many engineers at the company seem to like to reinvent stuff rather than adopt. WPF, WinUI2 and WinUI3 all share the same Xaml based structure. So they could have adopted WPF.
It is not that Microsoft doesn't develop advanced UIs with their frameworks. WPF is still well-used by Windows and other Microsoft utilities like Windows Terminal. They are just stupidly abandoning their built up bases for silly industry fads.
They jumped into tablet / touchscreen / hybrid-mobile-desktop bandwagon in late-2010s and tried to force WinUI as an UWP-only feature. It resulted in low adoption. They didn't adopt WPF to have same theming.
When WinUI2 failed, they tried to make modern C++ a reality and tried to remove UWP restrictions which is a good decision. However they diverted quite a bit resources into AI slop generation now and WinUI3 just languishes.
Same for MAUI. They tried to get into multi-platform, multi-device framework as a way to generate leads into Microsoft ecosystem.
They try to use various frameworks and UI stuff to get people hooked into the ecosystem and find ways to upgrade them into Azure and M365 customers. It is meaningless and tiring. All of those could be only WPF.
It is like Google and its many Bazel-like build systems (but not full Bazel) for each of Chrome, Fuschia and Android.
Everything you say is using WPF is not actually using WPF, other than Visual Studio.
Oh yes. I mistyped. 2010 was written in WPF
Why invest strongly in a desktop-first framework when society aren’t too sure about it?
MAUI was to get them through to where everyone wants webapps - served from an Azure backend, of course.
microsofts own stuff never seems to be what gets momentum. there's a strong aftermarket for better ways like back in the borland era bcb and delphi, the more things change the more they stay the same!
MAUI was horrible when I tried using it about a year ago, tons of bugs, pretty iffy comms/support from the MAUI team as to timelines when things might get fixed, etc.
Eventually dumped it and moved to Kotlin Compose/Multiplatform, which is just so much better at achieving a similar goal (though, obviously, without being part of the .NET ecosystem).
In case anyone cares:
I go to "Launch MAUI in your browser" [0] in section "Try It Right Now". I click on "Word puzzle - A word puzzle game", third of the demo apps. I click on the "Randomize" button. Tiles start shifting around. I click on the left arrow at the top to go back to the main menu. The whole thing becomes locked with the word puzzle on the background and the main menu on top of it. Nothing is clickable any more.
This happens on various different browsers.
[0] https://brave-sky-0c7a41a03-preview.westeurope.3.azurestatic...
Yeah the shuffling takes ages. Once it's finished you can go back. Also the movement of the tiles feels very strange with the mouse. The Click-Drag sometimes works and sometimes doesn't. No visual feedback is given on click. The tiles don't move but teleport, which is strange for a click-drag interaction. A simple click would have been better, or a move animation. Someone else noticed it, too https://news.ycombinator.com/item?id=45893986#45897805
The demos barely work for me at all on Chrome.
The sliding puzzle is really finicky, have to try multiple times to slide stuff, the top left (back) arrow stopped working altogether until I refreshed the page, which of course takes you back to the original page... The time picker looks awful with the numbers not centrally aligned to the control (they look too high), and it's finnicky to use / same with the date picker.
Awful.
It also locks the entire app when you go back during shuffling
https://news.ycombinator.com/item?id=45893986#45897572
In case anyone is confused
> Using .NET MAUI, you can develop apps that can run on Android, iOS, macOS, and Windows from a single shared code-base.
This new development adds Linux and Browser to that list.
I recently tried out .NET MAUI to see how easy it was to build a hello world app. It was quite messy getting it setup on Mac but eventually I got a simple hello world app working. Nice to use XAML again after all these years. I always liked it.
I clicked a total of three times in the demo page before the tab froze and locked up.
MAUI is nice, and I'm glad it's getting decent Linux support now, but I would never use it in the browser.
Avalonia's big money ticket is taking old WPF projects, making minor changes to the XML to load their tools, and instantly making the projects work on macOS/Android/iOS/etc. (for an enterprise fee, of course). Great for forcing internal .NET applications built to run on Windows Vista into the modern world, but I don't think their web platform is a good idea for new projects because of the massive overhead you end up with.
> We are collaborating with the Flutter team at Google to bring Impeller, their GPU first renderer, to .NET. That work is already in progress and as it lands, the MAUI backend will inherit those gains.
Interesting, I wonder how good Impeller is and if it's actually better than the new Graphite backend of Skia.
More info here [1]
the big difference is this
or as described here [2] [1] https://docs.flutter.dev/perf/impeller[2] https://medium.com/@raiden.lpf666/skia-vs-impeller-a-perform...
Yeah, buts that's compared to the older "ganesh" version of Skia. The new "graphite" version purportedly brings all those same advantages.
This is actually really good news, as impeller was built to replace skia. Its one of the best technical bits in the flutter stack in my opinion.
Impeller is designed with mobile apps in mind while Graphite is designed with desktop apps in mind.
> Launch MAUI in your browser →
This is absolutely horrible and I hope I never have to use it on the web - right click doesn't work, my plugins don't work, I can't even open DevTools and when I manage (before the app has loaded), it's just a <canvas> element in the page. I know a lot of people put work into this and something is probably better than nothing, but actually no, I think such solutions are disgusting.
If you're gonna claim cross platform, don't just stoop down to the lowest common denominator and write a bunch of pixels to the screen, use the damn system UI solutions, or in the case of web apps - HTML, CSS and JS/WASM. I already take an issue with Electron (Lazarus lcl did things right, in contrast) but realize that it yields a lot of utility, yet these canvas based approaches for web apps feel like a step too far.
Yet obviously nobody can admit that they can't do the heavy lifting needed to support multiple different target platforms for the same widgets in a way that respects the corresponding platforms, while also still wanting to ship something. In this case, I think nothing is better than something, if that something is so bad.
My kingdom for a UI toolkit that can be used to make real CAD programs, and not yet-more things that just look like webviews and could just be a webpage.
You can make CAD programs with any toolkit as long as you have a GPU surface to render your actual CAD output to. You can do anything with this toolkit from weather apps to video editors, map analysis, and PDF editors.
You can check out the Avalonia demo reel to see what you can already do with the .NET GUI stack that MAUI now uses on Linux and on the web: https://avaloniaui.net/showcase
GPU surface is one thing, but I believe OP means the actual UI controls: buttons, tabs, combo boxes, data grids, multi-window support, etc...
I don't think that's a problem? The .NET library this is about, that now supports MAUI code unchanged, is already used in programs like https://github.com/sourcegit-scm/sourcegit/blob/master/scree... and https://github.com/GPUOpen-Tools/GPU-Reshape/blob/main/Docum...
Yes, that's my litmus test for GUI toolkits - is it suitable to develop Photoshop-like software? Or is it only good for yet another weather or todo apps?
This 100%. I hate the trend of UX/UI that got unleashed upon us in the last decade of the web. Everything is scaled up for touch interactions and has to have fancy animation and very "comfortable" spacing around elements.
I wish we can go back to UIs that focus on information density and usability. I love looking at Japanese websites because of this.
Why do you hate it?
It's great to have .NET ecosystem expanding. I am a .NET dev, and I still use web optimized for mobile as my mobile platform(where applicable). Tried MAUI and all the other stuff, just doesn't cut it for me. You can make an app using it, but I would rather not to. The best 'mobile platform' for me was blazor hybrid, but then again - if it's already blazor why not go full web...I guess it depends on the 'seriousness' of your mobile application. If I had to develop a complex mobile app, I might choose another language framework, cause MAUI uses XAML and MVVM stuff that is quite a big overhead IMO.
MAUI has felt like a barebones project for years. Forgive me if I don't believe this is the beginning of more robust support.
Microsoft acts when it feels competitive pressure. I think Google's Flutter has been validated so Microsoft feels the need to respond.
Dart is a wonderful language though. I'm not switching back to using .NET anytime soon.
I wish it was a real "systems" language as well, i.e. not single-threaded/javascript-mindset one. That's why I hope someday we will have a nice cross-platform framework for Swift.
What would be to point of switching from Kotlin + Composel (Multiplatform) to Dart + Flutter ?
Currently my favourite way to build applications (desktop or mobile) is Kotlin and Compose.
I never used Dart, how would you compare it to Kotlin ?
I love Dart, which I consider to be Google's C#. Either language is fine with me.
Go is google's c#.
Go is frankly the polar opposite of C#. Go compiles to a native binary with no runtime dependencies and it relies on simple garbage collection and static linking. C# runs on the .NET runtime which is heavily managed, gives you JIT, reflection, dynamic code generation and so on. Go views concurrency as fire and forget, C# views it as compose and await. Go is extremely explicit while C# is extremely implicit.
Now, I understand that you may talk about it from a non-technical perspective, but even so, there are major differences. C# is a general purpose language for the cloud/web, and so is Go, but Go is also widely used in other areas like in embeded software. TinyGo is soooooooo much better than working with C/C++ or Rust as an example. Places like that where you wouldn't usually find a transpiled language (other than maybe Python with MicroPython).
C# has AOT compilation which creates a single, native binary. This has gotten so much better with .NET 10 and since the introduction of source generators to deal with reflection issues.
Also, check out nanoFramework for a .NET runtime that can run on MCUs like the ESP32 [1]
[1] https://github.com/nanoframework/Home
.NET can be compiled to native binaries. .NET can be used for embedded development.
C# is more expressive and .NET comes with batteries included. Go is more explicit and more verbose.
You can pick up Go faster and is easier to reason about Go code when you first encounter a new project but C# feels like it enables you to develop faster and be more productive.
For web both are excellent and performant, even if they have different philosophy.
What I like about C# is that it becomes more functional and I can even mix F# in the projects if I want even more functional programming.
Dart is much closer to C# than Go is
tried the demo in Firefox on Windows and... it wasn't great? clicks didn't register in multiple subapps
Would be interested to know about the a11y story on web and other platforms. Does it integrate with screen readers, accessible navigation tools, a11y apis, etc.
This is somewhere quite a few tools fall down. For example, for web: Rendering to canvas doesn't create the dom nodes required to hook into the tools, and creating the nodes appropriately can have a significant performance impact.
What about native a11y tools and tech.. there are some ios specific ones that you'd expect on an ios app for example.
So like we're really just going to do this every decade or so with a new runtime until people remember they hate it and start over?
My exact reaction
Why Avalonia and not UNO?
If React is what’s powering the start menu now, I’m more curious about why Maui doesn’t power the start menu.
.NET Maui running on windows seems like a more logical first step to prove the organization buys their own dogfood.
> If React is what’s powering the start menu now
To clarify, it's using React Native XAML, where XAML is Microsoft tech for building UIs. So they are dogfooding but they have too many UI frameworks to choose from.
You wouldn’t use MAUI to build something for only one platform. You would just use whatever it’s an abstraction over for that platform which in the case of Windows is WinUI.
Is there anything we can do to stop it? Or will it come anyway?
Why tho? We really need Silverlight reborn.
Also C# and .net overall are so damn good.
Anything to abolish the js and constant hacks upon hacks
I want WASM+DOM so that we can use any language on the web
WASM and DOM in Rust: https://www.leptos.dev/
Uno can render to canvas or the DOM in C#: https://platform.uno/
Blazor renders to the DOM: https://dotnet.microsoft.com/en-us/apps/aspnet/web-apps/blaz...
You can make your own better system which supports GUI and provides a reasonable runtime that works on all current major platforms and provides development tools, years of ecosystem development, etc. etc.
If it's better than what MAUI provides and you can support it for years, I'm sure that could take over and many people would use it instead. But... will you and why?
Anything would be better than another UI toolkit made by Microsoft. They will abandon it in the matter of a few years like they did about 10 times already.
They may not be improved significantly now, but none of the toolkits got abandoned. You can still use anything from Win32 to MAUI today for a new app.
I started moving to using GO and Fyne for cross-platform GUI tooling. GO 1.21 supports Windows 7 with Fyne. Ya still have to support Windows 7 for work. No more wasting time with managing a Windows installer.
QT Framework is still one of the best for cross-platform desktop applications when speed is key.
Searched the docs: No results for "Accessibility"
(Which is ok for some situations, but not for wide deployment like .net provides)
Like I said tooling. Means that the User is well defined where Accessibility is not an issue. The tooling I creating is for people that have to engage with physical equipment; from replacing parts to making sure a motor is working properly.
It looks like it comes just as a wasm payload, so, you can only vote with your feet.
Took more than a minute to load on my macbook. Ouch!
I really love C# and the .net ecosystem, but they just haven't made it work for web.
.NET works amazingly on the web. This is just not the UI framework you would use.
There is ASP.NET of course and Razor Pages. We all use apps built with these every day without even realizing it. There are other great frameworks as well.
I do not even see Blazor as a real web technology but of course it is positioned that way.
MAUI is a "cross-platform" and frankly mobile first UI framework. It was never meant for the web.
Blazor is incredibly productive.
I wouldn't use it for consumer apps because it requires a Websocket connection to maintain state and probably doesn't scale very cheaply... but for business applications or personal tools it's actually kind of insane how much functionality you get out of the box (at least by the standards of statically typed languages).
To replicate this example in Typescript, I'd probably still be installing packages in the time it took to write the 20 lines of code it contains: https://learn.microsoft.com/en-us/aspnet/core/blazor/compone...
I don't know if it has since improved, but .NET MAUI was really, really rough when I created a mobile app for my employer last year. I'm talking basic things - changing basic colors on the toolbar (1), putting non-text content inside a button (2), basic trigger behavior (3), to list a few. Not to mention that .NET UI has been years behind on hot-reload and developer tooling. Additionally, It was a fight to keep our app performant. The XAML compiler is a step in the right direction, but we had relatively simple views (in the dozens of components) absolutely tanking our FPS. I know there is probably some of my skill issue in there, but when I find basic things taking hours to optimize that I wouldn't even think about in React, I start to wonder about the framework. I spent a lot of time creating PRs on .NET MAUI but their team appears quite small and overloaded. I wish them the best - they're some talented folks, but I don't envy their job.
I can't help but think of Joel Spolsky's Things You Should Never Do (5) - the transition from Xamarin to .NET MAUI feels like a very similar mistake to Netscape. All of the battle tested Xamarin code, documentation, community examples, packages, etc. is now dead and has to be converted over to .NET MAUI.
On top of that, XAML just doesn't do it for me - having to deal with code-behind, MVVM view models, custom converters, and the actual XAML files themselves is insane for what is usually just a a single file in JS. The fact that you need to write a "InvertedBoolConverter" (4) just to flip a boolean is the most Microsoft thing ever. MAUI feels like it's designed just to keep a large development team busy. I'm not joking, we have a 42 line file that's only purpose is to flip booleans for XAML views.
We're a C# shop so it was nice to share our common C# with our desktop application, but I don't think it was worth it in the end. Sure JS has its problems, but I'll take those problems any day over MAUI.
I hope Avalonia can fix .NET MAUI - it'd be a massive kudos to them if they can smooth it over, but I can't say I'd willingly rely on this project long term.
1 - https://github.com/dotnet/maui/pull/15612 2 - https://github.com/dotnet/maui/issues/8191 3 - https://github.com/dotnet/maui/pull/15655 4 - https://learn.microsoft.com/en-us/dotnet/communitytoolkit/ma... 5 - https://www.joelonsoftware.com/2000/04/06/things-you-should-... https://github.com/dotnet/maui/pull/16965
Well, I guess Avalonia can solve 4 at least as you can negate a binding[1]. Good news for me as I recently started an Avalonia project, and thought you did still need an InvertedBoolConverter.
[1]: https://docs.avaloniaui.net/docs/reference/built-in-data-bin...
Why does everything need to have a soulless mascot now? It’s offputting.
This is awesome… but: On the Web, is Avalonia using Skia to render inside a Skia Canvas?
**insert inception meme here**
Joking aside: this points to MSFT moving away from the whole Mono/Maui investments and into Aspire or whatever they call it. Without MSFT backing this I am not sure if there is much more future left for MAUI (or dotnet on mobile in general).
Avalonia is great though.
> On the Web, is Avalonia using Skia to render inside a Skia Canvas?
Yes. They're also looking at offering Impeller as a render option:
https://avaloniaui.net/blog/avalonia-partners-with-google-s-...
If it runs on Avalonia anyway why choose MAUI? If you have the option just use Avalonia. MAUI is unfinished and its docs is awful.
Do you want to build an app using MAUI? Unless you build an app that barely deviates from the template, expect to desperately search through decade old Xamarin documentation and figure out the details through painful trial and error.
Good luck.
They addressed this.
It is a way to get people choosing the "official" path (ie. choosing MAUI) to experience Avalonia. They are hoping you come for the MAUI and stay for the Avalonia and become an Avalonia developer.
As for, why choose the Avalonia version of MAUI, there are three reasons: - Linux support (the big one I think) - Drawn framework (same renderer on all platforms) - WASM support (probably useful sometimes but not the real draw)
They are making a big deal of WASM here because it is easy to demo. We can all go into it and run it. But do we want to use it for our apps?
As for, why not Avalonia directly? To loop back to the beginning, it is because you do not yet know Avalonia and trust it. The Avalonia team is hoping this helps with that.
Thank you for explaining this. I didn't even realize this was an Avalonia project until you pointed it out. I like Avalonia (and love .NET in general), but I think the messaging on this needs a lot of work. Avalonia is creating unnecessary cognitive dissonance by emphasizing MAUI, which is a competing project after all.
This is what mystifies me about this announcement. Avalonia already works fine on Linux, allowing anyone to build a cross-platform .NET GUI application.
MAUI is supposed to be a wrapper around native widgets. The fact that they had to use Avalonia under the covers to get it to work on Linux seems to defeat the point. (Avalonia is a complete UI toolkit, like Qt or Flutter, that owns the entire stack from XAML to pixels.)
https://avaloniaui.net/maui-compare
Either it takes more than a minute to load or it does not work on iOS with Lock Down mode enabled.
It's using WASM, which Lockdown Mode disables (among other things).
A toolkit announcing “Linux support” is pretty ambiguous. Does it mean Xorg support? Wayland support? Framebuffer support? The announcement provides no clue about this.
They are not announcing Linux support for Avalonia. It has long supported Linux.
What they are announcing is being able to run MAUI on Avalonia such that you can run your MAUI apps on any Avalonia supported target. This includes both Linux and WASM, neither of which is supported by MAUI directly.
Since it's using Avalonia, I'd say it's just X support at the moment. They've announced that they intend to support Wayland, but that was a couple months ago[1], so I doubt that's ready.
[1]: https://avaloniaui.net/blog/bringing-wayland-support-to-aval...
Not to be confused with the MauiKit[0], the cross platform UI toolkit that had the name first[1]
[0] https://mauikit.org/ [1] https://github.com/dotnet/maui/issues/35
Honestly, I feel that everything UI related has gone backwards to the stone age.
I wonder how hard would it be to go back to visual designers like we had with Delphi or VB6. There were flexible layout container components which helped a lot when adapting forms to varying screen resolutions.
Excited about it, but I don't believe it will be good given the previous fiasco with the MAUI initial release.
I hope Microsoft delivers though, we need more alternatives to React + RN or Flutter.
Avalonia is also working towards using the new Flutter rendering backend Impeller which is being used to replace Skia, which is used by Chrome for rendering, turtles all the way down:
https://avaloniaui.net/blog/avalonia-partners-with-google-s-...
Ok, finally catching up to Kotlin Multiplatform Compose.
This is a huge win for people using MAUI. The only thing I worry is some performance loss from using Avalonia backend instead of using directly the target platform.
It seems like alot of people in this thread dont undestand the dotnet stack. MAUI, Native UIs, and WASM are all interchangable. Your cross platform app could literly be built out of everyone of of these components at once. MAUI is not required for use-cases generally when you have a webUI, its really targetting at the simple UI, write-once, run anywhere crowd. Once you outgrow it, you can move to native integrations using dotnet linux/mac/windows. Or you can just integrate with webviews, and have C# backend's or WASM backends. The combination is limitless.
Xamarin turned into dotnet, Xamarin.forms turned into turned into Maui. Name another large tech company that has embraced community projects like this and pushed provided enterprise support for community driven stuff over the same period.
I was writing cross-platform apps before Flutter, angular, or whatever other language you choose for 11yrs now. Find another framework thats done this before the scare tactics of "MS will abandon this" rhetroic. dotnet is the premier cross-platform enterprise ready framework full stop. No other framwork has the backwards compatiblty while maintaining paritity with the latest OS APIs. No other framework can serve millions of pages per/sec while supporting pixel perfect UIs and code-reuse. They have invested so much money in building automation that means as soon as the OS releases an API, your getting access to it. In real world terms, this is what counts if you want to build cross-platform stuff that your clients cant tell isn't native.
I run the same code from 2014, today, in apps in all the stores. Over the years all i have had to change is various namespaces to take advantage of the latest enhancements. Code that ran on dotnet4, silverlight, xamarin, still runs today on dotnet10. I share 90% of code across all platforms yet clients cant tell they are not native apps. Thats what i call return on investment.
I agree that, despite all the complaining, code re-use and longevity on .NET is second to none.
Where there is intense confusion in these comments is in thinking that MAUI is a web application framework. MAUI is "cross-platform" as you say but really mobile first with desktop brought along for the ride. It is in no-way intended to be a web application tech.
Avalonia has created an Avalonia back-end for MAUI which means running MAUI apps wherever Avalonia is supported, and that includes WASM. So now you can run MAUI apps in a browser. You can but that does not make MAUI web tech.
I can run Windows 95 in a browser. That does not make Win32 a web API.
Uggh, here we go again...
Good job!
Pathetic that MS doesn’t manage to do this themselves.
Even more pathetic that MS doesn’t actually use it for any of their products, so forgive me if I have very little faith in the future of MAUI with all its bugs.
Too little, too late. Desktop apps are mostly dead.
Not in the industry I work in - AAA gamedev. Art folks typically would have two, or even three monitors - so good solutions for docking across them (and still working) are required.
A chrome browser by itself can't work that - it's great for many things, but not for Creative Tools.
MAUI is first and foremost a mobile UI framework. It was built for iOS and Android. The desktop is added on as a bonus so that you can target macOS and Windows as well.
If you are building a Windows desktop application though, Microsoft does not want you to use MAUI. You use MAUI because iOS and Android are your top platforms and you want to target macOS and Windows without writing dedicated applications.
Linux has always been missing. This Avalonia port fills that gap.
You would not target the web with MAUI either. I guess "you can" now because WASM is one of the platforms that Avalonia supports. Again, I guess you might if you already have a MAUI app and do not want to create one for the web. But you would never set out to create a MAUI app for the web.
Desktop UI toolkits are dead or stagnant but the desktop ui is still king IMO
That's why desktop machines and monitors can't be given away. Nobody wants them /s
More and more of MAUI is abandoned by microsoft. I wouldn't touch this framework with a ten-foot pole. https://www.youtube.com/watch?v=yil7UVeOx4Q
Three apprecitions:
1) In today's American political climate I think it's appropriate to express appreciation for immigrants to like Miguel de Icaza who dragged Microsoft kicking and screaming into cross platform .NET and is the godfather of .NET Maui
2) As someone who came up developing desktop apps for Windows and Mac, I never liked developing web applications. There was so much lacking, but now developing web apps is becoming like developing desktop apps. Now you get/put your data from HTTP calls instead of file system and database calls or with Blazor and SignalR you don't even have to think about those. This may seem obvious to younger programmers today, but it would have seemed like magic back in 2004 when Dymanic HTML and Ajax (both Microsoft) were being invented.
3) I'm grateful Microsoft has changed their old ways to be a forward thinking company. They still have problems that any GIANT, Inc. organization has, but let's not forget how far they've come.
It wasn’t magic circa 2008-2010 — see silverlight
Agreed, Silverlight was an unfortunate path to choose, a Flash competitor when Flash was dying.
Really... You just had to bring up politics in this discussion?
Out of respect for Miguel de Icaza, yes