Flutter 2 • Hacker News Board

Re: Flutter 2

#471

Post by dmitriid » Wed, Mar 03, 2021, 10:53 PM UTC

eseidelGoogle wrote: post #155

rp1229 wrote: post #127

Earlier quoted context omitted.

The new Google Pay app on iOS is incredibly sluggish -- animations stutter, switching pages has hesitations. I'm not sure what they mean by "quality" in this case.

[Flutter Eng. Dir here] We were also not satisfied with the performance of the initial GPay release. We've been working with the GPay team the last couple months and have made significant improvements within both Flutter and the GPay app. Hopefully the next release of GPay will be out soon and others will be able to see the progress we're continuing to make with the team.

The ad:

> Google Pay switched to Flutter a few months ago for their flagship mobile app, and they already achieved major gains in productivity and quality.

And then, the reality:

> We were also not satisfied with the performance of the initial GPay release.

Ah, the classic false advertisement.

Let me re-write the ad for you:

> Thanks to several-months-long involvement of Flutter's development we could finally fix some of the issues plaguing our app that we re-wrote in Flutter. We're still not close to the release, and we only hope it will be better.

Re: Flutter 2

#472

Post by gagege » Wed, Mar 03, 2021, 10:53 PM UTC

randomchars wrote: post #35

The idea sounds very appealing, especially to an indie developer: being able to ship your app on multiple platforms from a single codebase is kind of the holy grail. Unfortunately, the result is.... I tested this Flutter example[1], on a 16" MBP with and i7, and it is janky as it can get, it feels like I'm using a 15 year old computer. From the code it looks like the whole thing is... rendered on canvas? I'll pass. […

What's up with scrolling on all these demos? It feels insanely slow and jerky.

Re: Flutter 2

#473

Post by kroltan » Wed, Mar 03, 2021, 10:55 PM UTC

vsmenon wrote: post #414

okamiueru wrote: post #368

Earlier quoted context omitted.

Haha. That's alright. That looks super neat though, I must admit. The only equivalent that comes to mind would be using an abstract class. I still might fail to fully understand what that code example does, but, would this be somewhat similar? abstract class AppState {} class AppLoading extends AppState { final int progress; } class AppSelectingLevel extends AppState {} class AppPlaying extends AppState { final Level…

[Disclaimer: I work on the Dart team.] The TS code indeed looks cool. This is an area we're looking at. One point, though: we try to be very careful to not regress performance or developer iteration time (e.g., type checking time) when we introduce new language features. E.g., structural typing can be more expensive in general to type check since we need to recurse.

Fair enough, that is a meritable goal.

Have you considered not going full-on structural-typing but still providing some sort of union? In fact, you could go for one with even stronger guarantees, like the sum types in Rust or F#. (with Rust going as far as to call them enums too)

I'll admit I have the faintest notion on what causes that kind of complexity on a compiler, so my suggestion might be an even worse idea.

Re: Flutter 2

#474

Post by oblio » Wed, Mar 03, 2021, 10:55 PM UTC

Hixie wrote: post #447

oblio wrote: post #277

Earlier quoted context omitted.

Well, good luck maintaining all of that 10 years from now. I can bet on the web being around, well developed and maintained in 2031.

I'm not really sure what you mean by "all of that". What I described _is_ the web.

It's painting stuff in a canvas and not using the standard web controls.

It's the web in the sense that almost every UI toolkit also offers a canvas (for example Tk does this) and you can reimplement every control using it. That doesn't make your new toolkit in a toolkit Tk.

Re: Flutter 2

#475

Post by ksec » Wed, Mar 03, 2021, 10:55 PM UTC

qbasic_forever wrote: post #286

dmitryminkovsky wrote: post #268

Earlier quoted context omitted.

> History has shown us over and over again, always bet on the web. This is a mantra on HN. But is it true? I don’t think it is if you’re trying to build a big product mainstream people will use. People want native apps. No one needs the apps to work in 25 years. The web is just part of the picture, an important part, but still just a part.

The same engine of HTML + JS has powered entire businesses and industries from $0 to billions and billions of dollars over 25 years. Amazon.com would not exist without web browsers, and its entire trillion dollar+ business is an enormous bet on the web. I would say there's zero danger of the web going away when folks like that (and many, many more... Google, Facebook, etc.) depend on the web for every microsecond of…

>Amazon.com would not exist without web browsers, and its entire trillion dollar+ business is an enormous bet on the web.

Except 70% of e-commerce, 90%+ of Social Media are now on Mobile Apps, where they were all previously 100% Web.

The problem with Tech industry is that too much thinking is about centralised, decentralised, Technology with backend front end. etc. When its users or customers dont give a fuss at all.

Re: Flutter 2

#476

Post by ksec » Wed, Mar 03, 2021, 10:57 PM UTC

dmitryminkovsky wrote: post #331

qbasic_forever wrote: post #310

Earlier quoted context omitted.

Clubhouse is targeting the middle-class silicon valley market right now (quite literally, those are the only people who got invites early on), who of course are infatuated with apps and have the money for fancy phones, tablets, etc. It's smart marketing for them to go where their customers are right now. The web is for everyone--I can go to an internet cafe in a slum and browse and buy from the same Amazon.com as I c…

Not sure what personal income has to do with anything. Hand held touch devices have revolutionized Internet access. The deeper down the income scale you go, the less likely you are to encounter a personal computer while you’ll still likely encounter a smart phone. Amazon will work on the web or as an app on that phone, but with the app the user experience will be better. That may be because Google and Apple have inte…

Apart from Google which has a complex answer. Most of the traffic from FAANG are now coming from Apps.

Facebook could have shut down their Website and it wouldn't even have a 10% revenue impact.

Re: Flutter 2

#477

Post by ksec » Wed, Mar 03, 2021, 11:00 PM UTC

peterstensmyr wrote: post #374

qbasic_forever wrote: post #246

Earlier quoted context omitted.

History has shown us over and over again, always bet on the web. Sure you might not like JS as a language, you might find CSS confusing and full of warts, etc. but it's here and will be here 20, 50, and 100 years from now. We're not going to just sit up and throw away 25+ years of progress and history on the web overnight. Tech like java applets, flash, silverlight, etc. come and go like fads. Who knows if Google wil…

I generally agree with you, but I don’t think it’s unequivocally true. Facebook bet on the web/HTML5 for their iOS app, and Zuckerberg later referred to it as the company’s biggest mistake.

That was because he trusted his engineers that believed in ideology rather than objective and rational evidence.

Wanting / Loving something to succeed is one thing. Whether it will actually work or succeed is an entirely different matter.

Re: Flutter 2

#478

Post by avar » Wed, Mar 03, 2021, 11:00 PM UTC

ryandrake wrote: post #436

hanniabu wrote: post #301

Earlier quoted context omitted.

I don't understand why more resources aren't being put into Qt and making that easier to use or building a more "web-developer-friendly" abstraction layer on top.

I don't even fully understand why "cross-platform UI development" is such a holy grail. Is it that expensive to separate your business logic from UI and write the small UI layer in whatever the platform's "best practice" native language is? Is it that hard to find developers who know more than one programming language? With a lot of these frameworks and higher level abstractions, if you go off the toy-app happy path,…

And write your shared business logic in what language? If it's a C library perhaps you could do this on iOS and Android, but how is that going to run in a browser?

I'm aware of things like the LLVM based C to JS compilers, but they're not really viable for anything non-trivial.

That's why a lot of shops are writing the same logic N times in N languages and frameworks. It can actually be easier than having to target something these wildly different platforms share.

Re: Flutter 2

#479

Post by dmitriid » Wed, Mar 03, 2021, 11:00 PM UTC

eseidelGoogle wrote: post #98

sydd wrote: post #87

Earlier quoted context omitted.

I've tested on my phone (1yr old pretty good Android one) and the whole thing is very laggy runs with like 10fps. What about a11y? The whole thing feels like Adobe Flex in 2010. It solves a great deal of problems that the web had but introduced a plethora of others.

[Flutter Eng. Dir. here] Still relatively early days for Flutter Web, so I would not be shocked if it's not buttery everywhere. However, we would certainly love to learn more. fluter.dev/support has links as to how to file an issue if you're interested. https://flutterplasma.dev/ is one demo to try. We expect to be updating flutter.dev/web and flutter.dev/showcase to update more over time.

> However, we would certainly love to learn more.

It's... It's very easy to learn more for yourself.

- Ask your company to provide you with a MacBook Air (not the M1, a regular one) or a mid-tier PC

- Use any browser other than Chrome (and other thank Blink-based browsers)

- Go and use the very same demos you so proudly present

Boom. Learning.

Re: Flutter 2

#480

Post by bsimpson » Wed, Mar 03, 2021, 11:00 PM UTC

eseidelGoogle wrote: post #98

sydd wrote: post #87

Earlier quoted context omitted.

I've tested on my phone (1yr old pretty good Android one) and the whole thing is very laggy runs with like 10fps. What about a11y? The whole thing feels like Adobe Flex in 2010. It solves a great deal of problems that the web had but introduced a plethora of others.

[Flutter Eng. Dir. here] Still relatively early days for Flutter Web, so I would not be shocked if it's not buttery everywhere. However, we would certainly love to learn more. fluter.dev/support has links as to how to file an issue if you're interested. https://flutterplasma.dev/ is one demo to try. We expect to be updating flutter.dev/web and flutter.dev/showcase to update more over time.

Oooff, the edges in that demo look terrible - particularly the grey boxes in perspective. The demo boasts that it isn't a video, but the interpolation looks as bad as one.

I hate to be critical - I'm sure a lot of work has gone into this - but there's still lots of room for improvement too.