1. In most areas, they eventually will!
2. If there is a permissively-licensed project in that area, at least there's a decent chance that the "reimplementation" will be a fork of that project. At that fork has a decent chance of staying open or even being merged back into the upstream. That benefits the community more than a proprietary from-scratch rewrite (which would follow from a GPL-only open-source scene)
If you're going to talk about theoretical behavior from big companies, you can make stories any way you want.
Let's say a big company selling computers wants to include a PCB editing software by default. If KiCAD was Apache licensed they may be tempted to make a special version for their customers, as a unique selling point. But it's GPL, so they have to choose between rewriting completely (a huge process), or just publishing the changes and being happy to include a good program.
Or a company makes modifications to a GPL program for internal use, and decides they want to share it with partners/customers later.
I have no reason to think these stories are more or less likely than the story of a company completely rejecting the GPL option but still deciding to upstream their changes.
Same with Microsoft and the Windows TCP stack lifted from BSD as well.
Compare with GPL licensed projects, like the Linux Kernel & its license making many projects possible, like the OpenWRT project for example.
Companies that use software like it Free. Companies that develop software like it permissive. There are more users than developers, especially for large more important programs.
It's simple in my experience.
Many big companies have some set "A" of code that they want to keep private, and some set "B" that they don't care about keeping private.
Lawyers are worried that at some point someone will accidentally include GPL code in something from "A" and force it to be made public. So they ban GPL entirely. They could in theory just ban GPL code from "A" and allow it in "B", but they can't trust that among thousands of employees none will make a mistake, so they just ban the GPL entirely.
Because upstreaming a patch once is cheaper than maintaining your own proprietary fork forever. It externalizes the effort of maintaining it in the future. That's the point that the article makes. And it's true in my experience. My employer allows and encourages me to contribute back to our dependencies. Those aren't the core of our business and our competitive advantage
But for applications, if they are willing to invest the costs to rewrite the GPL option I don't see how your argument makes sense.
That's always been the deal. Open source is not a guarantee of free support forever! It's a guarantee that you can always fork and keep using the project (or even developing it further)
There are some projects/people that I donate a small monthly amount. I don't do it because I'm expecting a specific feature to be developed, but simply because I think it's important to have support developers and let them work on something without being concerned about how to make a living out of their work.
To me, it's not. What are the outcomes that we are talking about here?
> You clearly feel when you do the wrong thing.
Do you think that the developers of, e.g, KHTML felt they were doing the wrong thing when they started working on a browser engine for Konqueror under LGPL? Would they continue working on it had they known that Apple would take their work to build what is arguably the most freedom-restricting web browser out there?
Living a life free of guilt and regrets. People make mistakes and can't foresee everything. That's normal. That shouldn't imply any guilt. I was talking specifically about cases where, in advance, you think that you're doing the wrong thing at the cost of short-term convenience.
Regarding KHTML, I don't have any data and can't say for sure. But I'd be surprised if most developers regretted their decision later. They knew in advance that they were getting into an LGPL project that can be used as part of something freedom-restricting. And, in my subjective opinion, Safari isn't so catastrofic and harmful that it defies common expectations of how harmful a freedom-restricting product can be
Sorry, this is not an outcome. Maybe you can say it's an strategy to help you define your priorities and it can help you set your goals in whatever endeavor you're taking on, but it's not something that has a clear goal post.
> Safari isn't so catastrophic and harmful
Taken in isolation, no it's not so bad. But the issues of Safari are the context where its development happens. Safari is intentionally stripped of any feature that could threaten Apple's market dominance in the mobile market.
And that is the problem with too-permissive software licenses.