Ask HN: Is commercial desktop software dead? | Hacker News Reader

Ask HN: Is commercial desktop software dead?

25 pointsnewintown20 comments

I am currently working on a tool for marketers which is a "traditional" software application for Windows, OS X and Linux.

Checking Product Hunt, Hacker News and Reddit I have the feeling everything B2B is web based these days. Is that true?

Shall I be worried I am working on a project which is already dead before even announcing it?

View on HN

As an end-user, I'd rather download and use high quality desktop application than use webapp. First of all, this means no Electron/js/html webapp bundled as desktop, but a real native executable.

Amen to that :)

Absolutely not. As greenyoda said, there are things that web browsers just can't do, such as gain access to the file system or to certain OS APIs that aren't available in a browser. Sure, there may be a trending movement towards SaaS for a lot of apps (SaaS is super hip), but not every app can go that route.

Previous discussions on this topic:

- https://news.ycombinator.com/item?id=11748812

- https://news.ycombinator.com/item?id=4280463

- https://news.ycombinator.com/item?id=7762926

- https://news.ycombinator.com/item?id=5050700

thanks for the links! All very interesting and it's kind of reassuring to know I was not the only one worrying about this (and the general positive attitude)

If your software does something that could be done just as well in a web-based product, customers would probably prefer to have a web-based product, since it can be centrally installed and managed (even if they run it on their own servers).

However, there are still some things that can't be done with web-based products, and for those things people still use desktop software, like Photoshop. A marketing tool doesn't sound like one of these applications.

Why Linux? I wouldn't expect lots of marketing people to be using Linux on the desktop.

Well, there are online only photo editing tools, but yeah, most still go for Photoshop.

The tool I am working on is mostly focused on analytics and data analysis, so even if the main target market is marketers I would not compare it to mailchimp or buffer.

Linux comes for free as the tool is developed using Qt & C++, but yes, probably not a top platform for marketers.

If you say so!

https://www.photopea.com/

I think it depends on the topic a lot.

Desktop apps are good if you are actually making a use out of desktop capabilities.

They can also be good for low maintenance low churn apps if they don't need any servers. Because you can publish a working program and then forget about it until the next version (not babysit/monitor servers 24/7).

But if it's not making much use out of desktop, and needs servers to function then why bother with desktop anyway?

If something can easily be a web app I'd say it's the better choice. Quick safe easy access and you get to control it however you like (like pushing lots of updates every week).

See this post by patio11: http://www.kalzumeus.com/2009/09/05/desktop-aps-versus-web-a...

The point about piracy is important and is a major hidden factor here. The cloud is DRM.

Personally, as a user, I much prefer desktop applications for software of any serious complexity. It's just that much better integrated into the system UI, there's no lag, and best of all: it's running on my computer. (Aside from privacy concerns, I often don't have Internet available...)

For developers, though, I can see why webapps are so much nicer to work with. (See patio11's blog post quoted by drdrey.) So I would posit that the defining metric is complexity: if your software is really complex (and/or doesn't need a server), make it desktop. Otherwise, go for the web.

Even in B2B / enterprise software there are still quite a few desktop applications, some for legacy reasons, some due to specific use cases.

For example, fairly recently I've come across a business application that's been implemented as a Java desktop app and deployed via Java Web Start because its main geographic target areas are - sometimes remote - locations in Southeast Asia where often you have flaky Internet connectivity at best.

Hence, deployment and occasional updates can be done online but daily use for the most part happens entirely offline.

Yup. Try getting something like solidworks running in the browser. JavaScript is not there yet by far.

If you know what you're doing, the UI/UX of a native desktop application can be light-years better than that of a web application. For almost any purpose.

SaaS, by contrast, is typically easier to monetize, distribute, and update; it can also be easier to build, depending on functionality and UI complexity.

I might be wrong, but SaaS is a business model, not a technology or SaaS is not just web-based apps.

You're right; I should have said "web-based SaaS"

Mostly yes. The only exceptions are very high end areas like CAD and high end games that make full use of local hardware.

Everything else is going web. Even native clients are often boxed web UIs due to the cost savings of developing the UI only once.

Make it worth it for the client to install your app on his computer. For example, make it launch fast, work offline and if it makes sense, integrate it with the notification bar.

I work in marketing and if a tool isn't web-based I have no interest.

you can always start with a desktop version and move to a web app if things go well/there is demand.

Reply on news.ycombinator.com