Hacker News Reader

Comment by dstaley | Hacker News Reader

It's still crazy to me that a $16B company can't make the financials of a true native app for Windows and macOS work. Microsoft has some real soul searching to do to fix up Windows development and make it easier and more attractive to companies. Thankfully it seems like both Microsoft and Apple are at least trying to make things easier. Apple will probably do the best with Project Catalyst (I mean, they already got Twitter!), but Microsoft is investing in React Native for Windows which could prove really interesting. Hopefully Microsoft doesn't settle for just making Electron better.

View on HN

I don't even think they need to make a native app; I use plenty of Electron apps day to day, like VSCode, Atom, GitHub Desktop and they're all fine.

Slack however is borderline unusable. Ghost processes, background workspaces silently closing and not delivering notifications, processes pinned at 100% CPU, silent crashes, pauses while typing, and even sometimes character drops.

I've tried it on three laptops from two vendors all with 16gb of memory, 100% SSD and an i7.

I've got Mac user friends that tell me they've never had any of the above issues, and I believe them. I've used it on a Mac, though not as a daily driver, and it's fine.

It just feels to me like Windows is a second tier platform for Slack, where they're fixing the bare minimum to not lose market share.

I'm a Windows user and I never have any of those problems. Our organisation is only about 10 people so I guess we are not pushing it very hard.

When my late model macbook pro fan starts running, it’s always slack. And thats several times a week.

At least it’s not another Box spinning beachball.

Maybe the lesson is that a decent solution to a painful problem and great sales and marketing are a good recipe for success.

It is a recipe for market share. Profitability and a realistic valuation is yet to be seen.

But I don't see anyone claiming a native app is the difference to profitability.

I don't think they do, most "office worker" computer resources are way over provisioned, at least in the developed countries.

Does resource consumption matters to me? yes, but that is because I compile stuff, run VMs, do simulations, like many other developers.

My grandma or dad at his office? they could do just fine with 1/10th of their resources.

This is not an argument for wastefulness, but the reality of business world and the value of computing resources.

If the owners get rich by cashing out on an IPO, that sounds like a "recipe for success" to me.

The pain of forking development into multiple separate teams with completely different technology stacks, bugs, and coordination overhead is just not worth it for most startups. It slows you down when you need to be able to adapt quickly. A lot of unicorns in the past would have addressed this by simply focusing on one platform initially but for an application like this that is not feasible. Even Linux support is not optional for a company like Slack since a lot of the decision makers on stuff like this tend to be in the IT department which in many companies would include at least a few Linux users. By going all in on Electron, they ensured access to all relevant platforms from day one.

The reality is, that most of their user base is completely fine with performance as is, which I agree is not ideal. But it's good enough. My guess is that they will gradually fix issues by benefitting from Electron improvements, swapping out bits of js for more efficient WASM based stuff, etc.

Mostly native app strategies are an extremely bad idea for small start ups these days. You triple the cost (or worse) of your development and inevitably you are going to do better on some platforms than others. I know people obsess about native on platforms like IOS but I find it interesting that there are quite a few unicorns out there succeeding with glorified electron/react native code bases. Slack is basically shipping a web app packaged up as an electron app on Android, IOS, Mac, Windows, and Linux. You can run the thing in a browser pretty much without loss of functionality.

Slack is in no way a "small startup".

They were when they started out and not getting distracted by a multi platform strategy is what allowed them to stop being one.

They’ve had more than a couple of years with plenty of resources to fix their apps.

I just explained you how not doing that was crucial to their growth strategy. Not having to coordinate bugs, features, and releases across multiple teams with different tech stacks is really nice when you are trying to outpace your competitors. Most of these follow a similar strategy. Pure native chat apps are not that common anymore.

if they can't build a couple binary apps with a $16,000,000,000 valuation then they're doing something seriously seriously wrong.

I've worked for companies 1/2000th that with native binaries on all 5 big platforms (linux/win/osx/ios/android).

Slack is a $16 billion dollar company on paper but still not profitable and showing no signs of profitability. It’s not like Slack is printing money.

They will be if they inject ads in between your messages (free version only).

"Rid Belly Fat NOW! Click Here!"

Haha, and that will be the last slack message I ever read.

I think the pure-web play will end up being a better decision in the long run tbh.

I think from an app developer's perspective, web-based apps are definitely the way to go. That's because it's so incredibly difficult to build a good web, Windows, macOS, iOS, and Android app. I'm hopeful that one day making good native apps will be as easy and accessible as web development.

It’s significantly easier to build a good native app than a good web app.

However, it’s significantly harder to build several native apps for each OS/platform with the same level of polish and functionality than a single web app that will work on all of them.

Do Microsoft and Apple really want native apps? I've got the opposite impression: they've tried very hard to herd desktop apps into Apple Store model. Third party vendors, like Adobe, are also forcing clients to adopt subscription schemes.

Those are still native apps. I believe parent is comparing to bloated Electron crap that requires an entire Chromium instance to put a GUI on what is basically IRC and text editors.

What does a subscription have to do with native apps though? All of Adobe's apps are native apps even though they require a subscription license.

Sorry I wasn't clear. The underlying problem is piracy. Big players like Adobe can still use the subscription model. For most vendors, the easiest choice is going the web route.

Deployment is another PITA that makes App Store model desirable for the vendors. For Apple and Microsoft it's even better because they get a cut, they curate their platforms and have the upper hand over vendors lest they become competition.

I doubt that Slack would ever offer subscriptions through the App Store.

With $16B valuation they can afford not to, of course.

Maybe because the real future is web apps and has been for more than a decade now.

In the same way Linux is perpetually the future of desktop?

Not really comparable, the web already has massive adoption, just needs to get better at high-performance client apps. And with WebAssembly + PWAs, and WebGPU in development, it's getting very close.

"Linux already has massive adoption in the server and embedded spaces, just needs to get better at office-related client apps"...

2019, celebrating the eleventh consecutive "This is the Year of Linux on the desktop"...

If it’s been the future for a decade, and that future isn’t here ... is it really the future of applications?

I would replace "the future" with saying that it's much more profitable for vendors and less convenient for users, so it's being adopted as fast as vendors can overcome users's resistance.

It is actually much more convenient for users. They can switch computers/phone/OSes and have the same experience. The same is not true with native apps.

I don’t want the same experience cross platforms. I don’t want the “Mac experience” when I am using Windows (see iTunes) or the “Windows experience” when I am using a Mac (old versions of Office).

No, what you want is the same experience. Which is what the web gives you. What you are describing is native apps trying to replicate the same experience and failing because they are native.

Edit: Native apps require more maintenance which adds to the delta in experience. If a single app is created (web) there is no delta.

Are you seriously trying to tell someone what they want, when they clearly and rationally laid out specifically what they actually want?

It's more convenient for some applications, but it's not the case yet for many applications. Evolution of browsers will probably favour web platform for everything except the more close-to-the-metal ones.

Yup, you used a web app to post your comment. Apple dropped Flash for native web HTML5. Is the future really to develop an app that can only run in 4 places (Windows, mac, ios, and android) dominated by three major companies?

Yes

I keep hearing this but where are all the good web apps?

Of course they can, they just won’t.

Reply on news.ycombinator.com