I'm guessing here, but I think what you are really asking is, can you make money selling software as-is without any support contract or obligation. It's not that you don't want to support your software, fix bugs and release incremental versions. But what you don't want to be is legally responsible for fixing those bugs and releasing fixes in a very timely manner because your customers demand that. If what I just said aligns with your question, then absolutely you can sell your software and make money.
To give you an example, just look at most games. It's a billion dollar industry, but each customer has very little legal power. You spend $50 on a software package, you install it and run it. It could be awesome or it could be total crap but once you bought it, you bought it, manufacturer will release fixes if/when they feel like it and the buyer can only wait and hope.
"I don't want to make junk" -- I wonder how many developers sit down to write some code and say to themselves, today I'm looking forward to writing junk and nothing but junk? The point is that significant pieces of software WILL have bugs, that just a nature of things. And all of us wanted to produce nothing but gold. If you are a small software shop (or just one person), I think it would be absolutely silly not to listen to your customers and provide support. When you are that small, your customers are essentially your QA and if they are willing to pay you up front AND THEN test your software (because there's only so much testing you can do as one person), why would you not release fixes for them?? That's not to say you have to promise solid turn around times in 2 days or less, but still do your best to keep people that give you money happy or they'll stop giving you money.
It also depends on what kind of software you are producing and your target audience. Do the math. I'm a software engineer and we have 9 developers on the team. Let's say our engineering costs for salaries, benefits, equipment are around $1M per year. Would we spend $100 on a critical or even somewhat important piece of software when if something doesn't work, one of our engineers would have to spend 2 days digging through your code? Those 2 days just cost my team $600, and that's only assuming that other developer's jobs were not hindered by the bug. Otherwise, start multiplying all of their time lost by 9. On the other hand paying $1000 increases our teams operating expenses by 0.1%.
So if it's anything important or semi-important, my personal inclination would be to stay away from your product. On the other hand, if you are producing something non-critical and I try your software with 30-day trial license and absolutely love it, of course I'll spend $100 and buy it.
So IMO these are the factors:
- Small target audience --> support more important
- How critical your product to operations --> support more important
- How polished is version 1.0 (i.e. usable right out of the box) --> support less important. In this point I'm talking even about little things. If I launch your app, and it works but it's a bit quirky or glitchy, that gives me overall impressions that there are potentially other much larger issues that I haven't stumbled on yet.
And don't get hung up on #3. As I said above, all software has bugs and all software could use polish and improvements. But if you aim for 0 defect release, you'll just spend years coding your software without ever selling it. Instead, my advice would be to do the opposite:
- Find people who are interested in what you have to offer
- Release version 0.9 (i.e. Beta) to 2 or 3 of those people with full understanding that it is buggy.
- Use their feedback to...
- add features they want but you don't have yet
- remove features they don't care about and you don't want to support
- fix bugs and bring up the quality
By artificially keeping your initial customer base small, you won't get swamped with support calls/requests and you will be able to work and communicate with them to get your product to a shipping quality. By releasing beta, you could give some of your customers a working copy in 10%-30% of the time than what it would take you to finish every single feature and fix every single bug.