Certainly, and if most jobs allowed me to ship products I'd like to stay and see them all the way through. A lot of jobs however, end up with the products being stuck in development hell as management keeps changing requirements while being too scared to test their assumptions in the market _and_ don't give raises.
If I'm already going to be stuck building things that never release then I might as well get the most money that I can
I've worked at 3 different companies over the past 13 years (1 large, 1 small, now 1 medium in size) and in all cases a full product development cycle is on the order of 1-3 years at all of them for both hardware and software. I've shipped tons of stuff. If you're finding "a lot of jobs" which don't involve actually shipping product to customers, maybe you're looking at the wrong kinds of companies or the wrong kind of industry?
I am in enterprise shops mostly and the one shop where we were releasing something every week or two was a 5 man engineering team that was left alone by the main company until we started making good money, at which point the company started interfering and we stopped shipping. In the enterprise jobs I have more projects under my belt that were completed and never released versus released products.
Every time it's been the result of either a low level executive losing a political battle and the project being shelved to rearrange resources or a company with little to no competition where executives didn't seem to care much about a million wasted here or there. One project I was on wasted over a million in development costs with the 5 product owners and me as the single dev because the executive who needed to sign off on the completed project didn't like the background color and insisted we started over at the requirements phase because "What else might we have gotten wrong?"
Startups seem more fun/invigorating but the stability and pay haven't been there when I looked so I just stick at the enterprise jobs doing what I can and saving money aggressively to be financially independent and start working at jobs that I want to work out
Some cities are dominated by insurance/banking/healthcare companies. Their IT needs are quite different than the consumer software space.
It's more like a salesguy just landed a big fish and the big fish is making these demands to close the deal, so bolt on some new functionality to your existing platform. Ad nauseum.
Outside of specific industries such as video games, very few software development teams use the waterfall model these days. Viewed in that light, Carmack's comment is difficult to reconcile with standard practice. Much of the software in development today is never really finished, it's under constant revision.
I can definitely see the appeal of finishing a project before moving on to the next one, however.
I think the sentiment still holds, waterfall or no. There are still an identifiable cycle of product development in an agile shop, but they aren't explicitly mapped onto differentiated development phases. It is still possible to see the evolution of a product from inception to maturity, and that is a valuable experience.
The point is that if you leave too early you don't get to feel the weight of your earlier choices, therefore you don't really learn from your mistakes.
There are lots of different product cycles and lots of wisdom to gain by doing only bits of them repeatedly. Which is not to say that Carmack is wrong but that you don’t need to stay in a job you don’t like just to see a project from beginning to end.
[1] https://twitter.com/id_aa_carmack/status/761986843319660548?....