Welcome to #DevDiscuss, tonight's topic is:
WebAssembly
You don't have to be an expert to take part in this convo, far from it. Few understand this tech deeply.
Some quick Qs:
- What are your initial thoughts on WebAssembly?
- How will this change the language landscape?
I think WebAssembly will make some fun use cases in all languages possible and probably lead to much more language specialization, but not meaningfully force any major adjustments for the individual developer. #DevDiscuss
Welcome to #DevDiscuss, tonight's topic is:
WebAssembly
You don't have to be an expert to take part in this convo, far from it. Few understand this tech deeply.
Some quick Qs:
- What are your initial thoughts on WebAssembly?
- How will this change the language landscape?
WebAssembly has the potential to be a real game changer. Providing near-native performance for applications could be quite interesting, but I think we are going to have to see if it gets adopted at a high rate. #DevDiscuss
Every language now will be able to target the web. Compilation will make a large comeback and tuning for performance will be closer to our fingertips #DevDiscuss
Welcome to #DevDiscuss, tonight's topic is:
WebAssembly
You don't have to be an expert to take part in this convo, far from it. Few understand this tech deeply.
Some quick Qs:
- What are your initial thoughts on WebAssembly?
- How will this change the language landscape?
Welcome to #DevDiscuss, tonight's topic is:
WebAssembly
You don't have to be an expert to take part in this convo, far from it. Few understand this tech deeply.
Some quick Qs:
- What are your initial thoughts on WebAssembly?
- How will this change the language landscape?
I think languages like Rust could cause this adoption to go up significantly. I could also see a lot of C/C++ native apps getting ported to be "web apps" as the barrier to entry is significantly lowered with WebAssembly #DevDiscuss
But, WebAssembly is a jump for most developers right now, myself included. I don't have fond memories of writing X86 assembly code, but WA isn't trying to make you do that which is good. I think support for garbage collected languages could be interesting as well. #DevDiscuss
The demos and incredible things that developers were able to achieve with ASM.js (like running games in the Unity engine https://t.co/jlK4UzjHpC) give me a lot of hope for the future of the web with WebAssembly. #DevDiscuss
I think the artistic opportunities of Web Assembly could be really cool since it will make frontend code so much faster -- things like 3D graphics and complex generative art may become more of a reality. #DevDiscuss
I am admittedly ignorant about WASM but my biggest concern is that we are, again, allowing code to run in a sandbox. For some of applications that makes sense. We just have to be careful not to sacrifice accessibility, structure & standards. #DevDiscuss
Welcome to #DevDiscuss, tonight's topic is:
WebAssembly
You don't have to be an expert to take part in this convo, far from it. Few understand this tech deeply.
Some quick Qs:
- What are your initial thoughts on WebAssembly?
- How will this change the language landscape?
That being said, I don't think the low-level languages that are normally used with Web Assembly are very accessible to new programmers, so maybe this will lead to an evolution there? 🤞 #DevDiscuss
Aside from compiler backend developers, few people will write webassembly. It already is like this, #mono-wasm is a good example, only the boostrapper shows a little bit of wasm API to load modules. As for the GC... #DevDiscuss
Languages and runtimes have very fine GC semantics that could be difficult to map in a generic way, and the GC may be limited interacting with Javascript objects, or the DOM. #devdiscuss
#DevDiscuss All I know about WASM is the basic premise of it being an assembly language for Browser code, and I feel that such a language is crucial. We're already moving away from javascript and treating it as a compilation target since so many other languages have a lot over it
I honestly can't fully grok why WebAssembly wants to support GC languages. It seems like it would be defeating the purpose of performance in that case and as you call out a GC is not universal across languages. #DevDiscuss
For the first few years, I see it being mostly used for gaming and animation. Eventually that could open up some cool doors for web developers to build some really interesting experiences with JavaScript. #DevDiscuss
Absolutely! This will open opportunity to pick the right language for the right task. Of course, JavaScript will still be the right language a lot of the time on the web. #DevDiscuss
I've been watching #WebAssembly pretty closely and believe it has the potential to be a pretty big game changer. Frameworks like #Blazor have a lot of potential but, need to mature a bit before the they will be adopted in production tier applications.
#DevDiscuss
Straightforward answer there 😄. I've tried to learn Rust once before, but I didn't know what to do with the knowledge. Got any good tutorials or things you could build with Rust that you'd recommend checking out? #DevDiscuss
The point of GC in this context is to be able to make two GC'ed domains interact in a memory-friendly way, and not have to resort to giant hash-tables to map object from one domain to another. Pinning GC instances is easier to reason with. #DevDiscuss
So it is another example of WASM acting as an adaptor between two different languages/runtimes? Allowing them to speak a common language when referring to a GC. #DevDiscuss
I am by no means an expert, but I think looking into Rust would be a good introductory into the concepts that WASM is trying to introduce. @jlaban might have some great resources as well. #DevDiscuss
It could be defined like this indeed, a common way of sharing allocated memory lifetime across wasm modules boundaries, in a way. But that is still in flux, the spec is still being defined. #devdiscuss
Welcome to #DevDiscuss, tonight's topic is:
WebAssembly
You don't have to be an expert to take part in this convo, far from it. Few understand this tech deeply.
Some quick Qs:
- What are your initial thoughts on WebAssembly?
- How will this change the language landscape?
Seems to me like the Rust folks are doing a lot of work to push WebAssembly and integrate in great ways. Lots of people working on both projects.
Would be worthwhile to step into Rust if you want to keep a closer eye on WebAssembly #DevDiscuss
I think the artistic opportunities of Web Assembly could be really cool since it will make frontend code so much faster -- things like 3D graphics and complex generative art may become more doable.
This is so cool:
https://t.co/24awc7GzBvhttps://t.co/q4GYCu9nbJ#DevDiscuss
Welcome to #DevDiscuss, tonight's topic is:
WebAssembly
You don't have to be an expert to take part in this convo, far from it. Few understand this tech deeply.
Some quick Qs:
- What are your initial thoughts on WebAssembly?
- How will this change the language landscape?
Web development lately seems like one hack after another to shoehorn what developers want to use into something the public wants to use. I'm not saying this is bad, but I do think we should focus on more elegant solutions... #devdiscuss
Absolutely. Current technical limitations of JS keep me from getting too wild about the beauty of the web. Jank is sort of a dealbreaker for me. WebAssembly launches us way forward in this regard. #DevDiscuss
That but it also means that all the software investments of the past few decades can now access the web with only a recompilation away. No more awkward Javascript porting of other languages #devdiscuss
In reply to
@y_elkhalili, @kylegalbraith, @ThePracticalDev
I don't think it means JavaScript will be replaced, my interpretation is that it is actually necessary even for WASM. However, it could mean "glue" is written in JavaScript and applications are written in optimized languages. We are a ways away from that though IMHO. #DevDiscuss
I think the focus will remain on compiled languages initially because of the static linking aspect. But, having the .NET engine directly accessible from Node could be very interesting. #DevDiscuss
Agreed. I think the web (for all its warts) really is the only universal content platform at this point, and if we ever want to deliver convincing AR/VR over the web then eventually we’ll need langs more suited to graphics on it #devdiscuss
In reply to
@bendhalpern, @ASpittel, @ThePracticalDev
I'm just hoping it'll compile BASIC so that Wargames text adventure I wrote in 6th grade (and got kicked out of class for) will live to see another day... #devdiscuss
Yeah, I'm definitely not working on a compiler! I've no experience in pretty much any of those languages, so Rust seems most appealing right now. #DevDiscuss
In reply to
@Paul_Bone, @summerscope, @ThePracticalDev, @ComposeMelb
C# is "just" the language, .NET is the whole C# + the whole class library and ecosystem of packages. https://t.co/aXopapBr0J is a specialized framework for creating websites, web apis, etc... and blazor packages the razor rendering engine in a webassembly friendly way #devdiscuss
So true -- AR stuff in the web is such a cool possibility, but its pretty choppy. I built this, https://t.co/1TzFt0nWX0 which is so cool but also could be so much more performant. #DevDiscuss
In reply to
@dangolant, @bendhalpern, @ThePracticalDev
I don't see JS going anywhere anytime soon. I could see some dependencies of frameworks like React and Vue being built in WASM for performance, but the majority of development will still use regular JS. #DevDiscuss
when did native app become a dirty word? web assembly looks cool but I find myself a little considered about the strength of the sandbox that allows for that much power in a browser window. #DevDiscuss
I don't see Web Assembly having a huge impact beyond 3D browser based simulations. It's just overkill for traditional web apps. The performance advantage could have a huge impact on browser based AR and VR though. #DevDiscuss
Why would it be limited ? Think more about what mobile apps are currently doing. Developing larges applications with zero-delay reponses, offline support, using libraries already proven to be working, but using the web as a deployment mecanism instead of appstores. #devdiscuss
I'm working on building a community for developers to mentor and collaborate with others using @code's "Live Share" feature. Check it out and join the conversation https://t.co/NrBL2rlcGp#DevDiscuss
As someone who has to support multiple OSes in an Enterprise environment that used to be windows only, it makes a huge amount of sense. The only thing osx, win10, Ubuntu and redhat have in common is a web browser #DevDiscuss
I've published another post! This one shows you how to use the WhatsApp and Spotify APIs to build a collaborative playlist! https://t.co/JJfMygZRVC#DevDiscuss
Enterprise apps to not be that simple, unfortunately... and if most of the logic can be pushed to the client to leverage client side CPU, this can mean less expenses server side, just API work. As for wasm, it's also a matter of performance and choice of languages #devdiscuss
I agree it's overkill for traditional web apps. The exciting part is all the new applications that can come to the browser. Google showed off AutoCAD as an example. Imagine that! It's much better than trying to run legacy programs in weird compat modes #devdiscuss
If anyone has blog post requests, HMU! Frontend and Python is where I can probably be the most helpful.
Also, check out the blog -> https://t.co/FiJiPqW3zL#DevDiscuss
Imagine the security and access benefits too if we could do something like deliver analytical software anywhere in the world, run in something like a phone browser. Everything client-side? #devdiscuss
In reply to
@ASpittel, @bendhalpern, @ThePracticalDev
Sending out some exciting news to folks who are interested in or have purchased my learn AWS by using it course. 🤫 Spoiler: 2 new bonus chapters are coming very soon to an inbox near you if you have purchased a package. https://t.co/8HBXfPNqkK#DevDiscuss