Full stack native app development

Hello , as you'all know Unity is growing , since they went Public IPO ,
recently they acquire Meta for vrtualcinematography . which's great news , also they're getting deeper AI technology and tech solutions ----

but it wouldn't it be nice to develop other services for developers ? I mean recently they develop the modified (web based-sorta ) UI Toolkit and they tried to develop mobile UI widget .. but it support only Chinese version of unity ? well it's open source but they should call developer community in a blog or something so more contributes on it .

also it would be awesome future to have if unity have sample integration with blazor with since most SPA frameworks allow custom reusable components it would be awesome if unity adapt ( Declarative HTML) utilizing Entity-Component Architecture on the web via wasm or whatever this allow for bi-di binding without sand-boxed webgl i.e : https://aframe.io/docs/1.2.0/introduction/html-and-primitives.html this meas editing the value inside the browser i.e edit material properties animate commonest value using js libraries

also as they now relying on web technology for UI it would be awesome to develop further tooling like using Unity Tiny project for lightweight 2D physics & animations for GUI .( native vector IMG * lottie animations ) etc. .. Flutter and xamarin uses SkiaSharp- with shader effect it would make sense if unity use their renderer for shader on UI particles etc ,, it should be inherited from native controller .

alot of potential here , I mean look at webflow they creating cool video and documentations and alot of the technologies are probably possible with .net standard 2(mono) if asp.net core is support isn't possible currently due to technical and design issues I would just call for the community gonna boostrap the project . Native Cross App development with unity would be so dope future to have


tryphotino.io is write once deploy everywhere just like MAUI & Electron but using .net just imagine where you could do everything in one framework with easy familiar APIs if unity adapt this project as starting point it would be awesome

 

UI Toolkit isn't web based and this whole thread is nonsense.

 

you're right , at the first glance I thought unity uses html , it has inspired by htmlx ( aka html5) also it's sorta correct :
https://docs.unity3d.com/Manual/UIE-USS.html

this would be even better since most native applications use xaml , plus xaml (shell ) is an easy way for creating menus in xamarin forms .available on chromium as well ( I think chrome has deprecated preview api ) also vs blend has tooling to create animations & layout pages visually ( drag & drop) but i'm not sure if this valid xaml animations

but why this such a cool idea ? and why it does matter ?

WYSIWYG authoring : just like webflow creating responsive UI design across different DPI is cool concept but hard to implement .

CSS & DOM shader : if you don't know this just GOOGLE IT it's dead project from various comanys https://alteredqualia.com/css-shaders/article/

DOM; -to turns HTML DOM elements into interactive textured planes
t's kind of a headache to position your meshes relative to the DOM elements of your web page the right tool to convert dom elements and reusable components into textured planes, allowing you to animate them via SHADERS.

reusable components is the trend of SPA and converting jsx/tsx allow collaborative work for app development i.e the front end could convert figma to usable template .

creating something like this :https://activetheory.net/work/welcome-to-hogwarts would be easier

https://aframe.io/docs/1.2.0/introduction/html-and-primitives.html



similar to Delphi Studio creating apps in one framework with familiar tooling set even if this doesn't make any sense at all .
creating native cross platform app development is really complex I believe unity can simplify it without the need of knowing design pattern like mvvm , mvc , etc , in similar fashion of game development we familiar with, maybe we just need to care about

 

It's not correct at all. What it IS based on is standard markup languages that have been used in UI dev for yonks, which have been developed alongside web software.

Regardless, none of this provides any practical functionality to Unity.

 

I strongly doubt that.

Qt didn't go anywhere, and there's more than one toolkit library about. Plus we have decades of app development before XAML.

 

Aside from the fact that the formatting on these posts is horrific and the poll answers very biased/skewed (there is more answers to this weird question than the 2 polar opposites you added, both of which still make it sound like this idea is sensible if picked ), this makes no sense in any way.

 

I don't understand why it doesn't make sense ? I don't talk about the poll options screw it , for what reasons UnityTeam wouldn't try new developing a POC from this concept?

aside from technical issue / architecture design perspective , some articles mention Unity thought about this before m, again Unity is growing and have other products https://unity.com/solutions

the MAUI & Hot reload caught the eyes of unity dev team but I think there'is other technical issue MAUI is great for many software engineers , but for non professionals ,whom wants to develop apps not games it make sense since Unity have announce Unity as library for XR apps and other stuff .

shouldn't Unity support .net 5 / core 6 ?

 

You have yet to provide a single reason why this would be a meaningful addition to Unity as an engine.

 

HTMLx is NOT HTML5. In fact my searches suggest none of the solutions named "HTMLx" have anything to do with the standard but rather were all named that way because the author thought it would be cool failing to realize there are so many with that name it just drowns them out.

Your posts read very much like a stream of consciousness where you have thoughts and ideas but no real sense of what they would be when combined. Add to that the fact that you don't seem to understand some of the concepts and technologies you're discussing and it just isn't understandable.

https://en.wikipedia.org/wiki/Stream_of_consciousness

 

making beautiful UI -cross platform is really hard( animations and effect ) , It really takes time to implement and complicated / each platform, js library or framework has it's own complexity of doing things , now recently things change a bit , web based approach like PWA & generic permissions API , has it's own limitations and trades but its how things should be ( it's hybrid approach using webview2 to host embedded web apps into native containers ) the web takes unified way of doing things, I don't mean apps should built this way it's just way of porting the logic (business model ) into embedded web app i.e Blazor hybrid & Uno platform

but why Unity should do it instead of other solutions?

-now if you want to use something not mature yet like tryphptino to build Menus Item it's hard where in unity engine extending the Unity Editor is as easy as copy and pasting the documented api. .
app development using a game engine and it's tools is the most relevant thing to do , for same reasons I mentioned
-scene Management instead of (forms /Lauout/views) everything you could like animating using animation timeline & Animator window instead of (XAML Behaviors|vsm ) or whaever
-utilizing shadergraph instead of using SkiaSharp ,
-the web css shaders and Declarative HTML looks so cool .

 

Again, why does this benefit Unity as an engine? Not you looking for another full-stack development library. What good does this bring anyone who is currently working with Unity in its capacity as a real-time rendering platform?

 

if Unity devs release paid prototype , I'm sure not just me many devs would be happy to try .

 

Why would they dedicate resources to making that prototype in the first place when it exists so far from the core functionality and use cases of Unity as an engine? Why is this a good thing for us as developers and Unity as an engine and not just you?

 

Web based approach for UI is awful. Always was, always will be. Doing UI this way is a great way to produce bloatware.

Rendering web page requries you to pull in entire browser with javascript engine, and a single page will take a gigabyte of memory.

The idea is wrong fundamentally.

 

I doubt it. The average online web page right now can suck up a gig of memory, but that's not because of the visible contents or behaviour of the web page itself, it's because of all the often-hidden 3rd party junk which gets carried along for the ride, and all of the stuff your web browser does other than just rendering HTML docs and executing JavaScripts.

These days websites tend to be frivolously wasteful of resources, but that doesn't mean that they have to be.

The approach in general is no more bloaty than many other common software development practices.

 

Some time ago steam decided to switch renderer of their library to web-based approach.
Memory use went up by 700 megabytes to display pretty much the same thing as before.

Web-page based UI pulls in chromium toolkit, because webstandard is huge, and there aren't a lot of toolkits that comply to it. And that thing is definition of bloat. So adding this sort of feature will add a gigabyte of bloat, even if you display "hello world" in there.

While this sort of thing does not have to be wasteful, it currently is. It also offers less options than Qt, which supports layouts and Vector-based rendering which allows rotational elements. It also had fancy shaders.

Speaking of OPs sentiment regarding Xaml, Microsoft consistently produces bad or weak UI toolkits (does anyone remember MFC?) and does not exactly pioneer anything amazing there. Now, they aggressively promote their tools and a lot of people end up using those, but that doesn't mean that those tools are the way to go. Personally I think that the only UI iteration that got it right was Qt. But that's my opinion.

 

My UI system, which has the same functionality between Unity and my custom engine, uses Xaml and it's more fine for game UI work.

 

Enough said.

The fact that poor implementations exist does not invalidate the approach in general.

 

If the only implementation that exists is wasteful, it does.

These days an application that displays a webpage pulls in chromium toolkit, which you can spot because it does this:

Spoiler

Designing your UI with electron so they run on nodeJs fall into the same category.

It is a great way to produce bloatware.

While in a theory an optimal solution could be used by someone, in practice they will not be doing that and will choose path of least resistance, which means producing bloat.

And that's why it is a bad idea.

 

Please read my full quote

like embedded app inside WebWindow , which's is a lightweight cross platform -desktop library now , it's evolved into photino.NET without bundling node like in Electron ."Memory use is especially low on macOS and Linux where the native browser control is far more efficient than Chromium" , but unfortunately the blazor part mobile development is not longer under maintenance , regardless of web performance web stay a platform most things build for to reach wide audience whether it's desktop or mobile . blazor binding (hybird approach ) is cool too but it's not mature and needs tooling also the wasm runtime disrupt bi-di calls .

, I just found new avalonia UI which Ironically working with Unity , it does seems have easier API with unified Acrylic (Gaussian blur) look , however mobile support is not not ready yet


what I really want from Unity spin off for Project Tiny for asp.net core , the Engine has some limitations it can't run two instances , can't mixed with SPA frameworks , (well at least for blazor atm) sure there's js interoperability but wasm couldn't bridge the connection . it would be even cooler if they somehow manged to create similar result to aframe.io , i'm not suggesting to create their own SPA but just try app to create development tools using the engine toolset i.e vector renderer and animations timeline , shadergraph , 2D physics & State manger , and some wrapper for native APIs .

i'm not qualified to say what should be done here , but I think Unity should try , I know this alot of things goes behind full stack development not just the frond & backend like databases , API and Micro-services etc also from engineering architecture , design pattrens , bootstrap entry point DBMS&ORDs etc ,, but there's many SDKs provided for these things for unity, the missing part is rendering & tooling .


Edit : ,

I found Fuse framework which uses

• Uno: A programming language (and a compiler ), which is a dialect of C#. Unlike the traditional C# development tools, Uno generates C++ code that can be compiled with the normal native development tools (Android Studio and Xcode). sounds familiar ?

UX: An XML-based declarative UI language similar to Unity UI element

 

Chromium is far from "the only implementation that exists". Ultralight is designed for light weight, high performance integration in games, is used by big budget studios, and is literally the first hit on Google for "html rendering library".

To be fair, it's 9mb memory footprint may still be more than just writing a UI in whatever you're working in (eg: using Unity's built-in stuff) but then you're losing whatever benefits you were trying to get by using a more common standard. It's a compromise either way.

 

Like already mentioned by other users, your text has issues with formatting and sentence construction.
I will not be deciphering this.

There's also question why people insist on using XML for those sorts of things.

Then again...

Spoiler

https://xkcd.com/927/

 

Again: why?

Unity is a real-time rendering engine, not a full-stack development platform. It makes no sense to pivot to that whatosever.

 

So..you are basically too lazy to learn the Unity UI. You can easily config UI elements as prefabs to act like any framework. I made some that took Firebase React json and laid it out like their widgets. Took a few hundred lines of code total and a couple dozen components to parse the json. But rather than learn this you want Unity to waste money and resources so you can play at full stack dev inside a 3D interactive platform. When using Blazor as WASM, your app is downloading the entire . NET runtime together with application DLL Libraries to the browser. This won't cut it. Learn Unity before attempting to destroy it. You may then know there is no need for hog tying it to such bloat. If yer totally crunk on the idea make a parser and show us yer really a full stack dev and not someone who needs a framework to play at being a full stack dev.