[QuantumRoar]

I've heard a few times that creating a decent desktop program with a GUI takes far more skill and experience. I can see why that is, since a fully fletched desktop program can be orders of magnitude more complex than most of the mobile and web apps.

I'd love to see more of the bleeding edge desktop software like the Adobe stuff, Blender, all the DAWs for Audio, etc. But it feels to me that not very many programmers have the skill to pull something like that off. Also, there doesn't seem to be much of an incentive left to invest into such things.

I would very much want that the mind set changes a little bit. Great desktop software makes a profit from selling to customers who value quality and awesome features. I wonder why that is not enough incentive for ambitious programmers. I only ever hear about developers for mobile and web apps getting a lot of money for things I seriously don't care about.

I have the feeling that we use desktop computers for the same things we used them ten years ago. I can't even remember when I had the last "I need that!"-moment for desktop software. So why can't we reinvent the desktop with software, like many try to reinvent what phones and web apps can be used for?

It's like the desktop computer in this day and age is slowly reduced to a platform (OS) for a platform (browser) for web apps. That's probably why many can get by with a tiny laptop or even a chromebook, which is probably the culmination of that thought.

However, I've not lost all hope. My bet is on virtual reality. If someone can build a really awesome VR application where everybody's like "I need that!!!", I can see how the desktop could return to its former glory, considering that a decent VR needs a ton of compute. But there's still the problem that VR is hard, GPU programming is hard, 3D is hard, parallelization is hard and optimizing is hard.

So are there any pioneers left with an enormous amount of skill and time who would risk to start such a glorious and uncertain journey without some VC guy dropping a few million dollars on the table? Or are we all going to end up as web developers because that's where the capital seems to be?

[FigBug]

I don't think it's more skill, just different. I currently work on a DAW. I left for a while because I thought my skills were becoming obsolete and I needed to get in on this 'web' thing. I was useless and I lasted 9 months. To just jump in and learn html/js/css + everything else you need to be a web developer is a huge task. And a lot of stuff just doesn't work -- and the only way to know how to get it to work is experience.

By contrast I think C++ and realtime audio programming is simple, but I've slowly picked it up over almost 20 years. I retreated back to desktop apps and firmware. I'll take another stab at learning web technologies, but not at a startup where I need to deliver asap and learn at the same time.

As much as developers hate sandboxes, the lack of one (except Mac OS until recently) is why I think desktop is failing. Users just don't feel safe installing random software like they do going to websites or installing mobile apps. Users have been conditioned to only install software from sources they trust, and they trust no one.

If I post a link to my new web project, most people will click on it. If I post a link to an .exe almost nobody will download it. I think that's the main issue killing the desktop. Big downloads, compatibility issues, slow installs are also an issue, but I think they are secondary.

[QuantumRoar]

It's rarely a good idea to abandon all your experience just because the web stuff seems fancy or fun (I don't think it is). Your skill is rare (much rarer than html/js/css), so you should use it to your advantage.

The goal for the desktop that I was talking about had nothing to do with these webby things. These should stay on the web and do their thing. You can have those small simple programs as a web app and on your phone. I think the downfall of the desktop is that people use them as if they were feeble things with a mail program and a web browser and some office programs on it, when they're actually capable of much more, which isn't leveraged except for some professional programs and games.

The desktop is - in my opinion - not primarily meant for little things. The desktop can and should handle the bleeding edge of high performance, parallel and accelerated computing that can't possibly run on anything else but a desktop. If we would stop thinking about whether a program will run well on a phone and explicitly target real computers, there would be a lot more purpose to desktop computing.

I'm not sure what that might be that doesn't exist, yet (maybe VR). But I'm sure that you can't do this all by yourself in a windowless room over a weekend, it'll probably be a large project involving quite a few programmers.

Then, the question of whether users will download your software is not the same as whether they will click on a website. Then, it's the same question as will they download Photoshop, will they download Reaper or will they download Maya? Will people download the next big thing in desktop computing?