Do you have any idea how many FOSS programs from how many authors/projects I use every day? Let me just name a few: Thunderbird and Firefox (Mozilla), LaTeX (TUG), bash/zsh (a community), gfortran/gcc (GNU), vim (another community), ArchLinux (yet another community), GIMP (GNU), Inkscape (yet another community), and the list goes on.
In which case you a productive member of the OSS community. When I contribute to Firefox, I'll have reason to consider your wishes because we're working together - my code and your code need to play nicely together. I won't have any reason to worry about what Apple thinks of my Firefox code, because their Safari code doesn't affect my Firefox code.
So how do you patch a GUI that you consider "counterintutive and confusing" unless you fork it?
The installer GUI is python code. You can patch it the same way you'd patch any other code. Except in this case, one complaint
is inconsistent fonts, so you don't even have to be a programmer. Just search-replace font names.
Alternatively, the GUI is mainly developed using the Glade"IDE", http://glade.gnome.org/ so you can edit the GUI graphically, right in Glade.
Glade generates Python source, from from there run "diff -Nrup" just like any other patch.
The people who you claim "contribute nothing" actually contribute a lot. They are free testers of your product. It in and of itself is a very valuable asset to have.
That is not valuable to me at all. It already works for me, on my hardware. You testing it for your use case, on your hardware, benefits YOU.
It doesn't benefit me one bit, not if you stop there. There is another step or two you can easily take to make a contribution of it, though.
If you stop at using the software, and pretednign that using=testing, it's a giant PITA to be expected to support hardware that I don't even have
access to. If you really think that's beneficial, explain to me how I can eat your test on your hardware for lunch, or how your use case keeps me warm.
It doesn't.
If you at least submit a careful, specific issue report that will probably be useful to YOU. What's useful to me, what fills my belly, is if
you get me a breakfast taco. I write better software when I'm not hungry, so that also benefits you. As far as using/testing, if you take that
"using" and go a step further and write documentation based on how you use it, that benefits the community, including me, because that
saves me the time of typing out answers to questions. Also, if you take the results of actual careful testing (using != testing) and submit a
careful bug report that _might_ be useful to me, if I happen to be affected by the same bug. Having people simply use software I write
does no good for me or anyone else, though.
So when you use software that's poorly documented, either a) write up what you figured out about how to use it or b) get honest with yourself and admit you're useless in that context, not useful. It's okay, just be honest with yourself and others. I'm not useful in regards to Gimp - I just use it. I am useful in the context of the kernel, because I help with development, just a little bit.
you put something out there, it includes an unstated promise, namely that it will work and be useful enough to invest the time in.
That's called "warranty of merchantability" - "merchant" as in "buy and sell". It's an implied promise that if you buy something from me, I'll deliver something worth buying. If you're not buying anything from me, the terms aren't implied, they are clearly stated. I put it out there because it's useful to ME, so it might be useful to you, too. I have no obligation to spend my days making something you'll like - you haven't given me anything. Every GPL package includes a clear statement of these terms:
15. THERE IS NO WARRANTY
Again, we write GPL software because it's useful for US, or for our customers. We decide how to spend our time, how to design a system we create. If you want to be part of the "us" that makes decisions, edit the damn wiki or something.
Newsflash: Not every user of FOSS software knows how to program.
You missed an entire paragraph written just for you:
"But I'm not a programer!" Okay, so translate the documentation into your native language, or help out on the forums, or maybe even consider feeding the programmers lunch while they work with a $20 donation. Otherwise, you're bitching about a gift.
If you want your software to be widely accepted, listen.
We do NOT want our software to be widely used by people who contribute nothing. What good does that do us? You are not a customer. (Unless of course you are a paying customer). You are the recipient of a gift. Freeloaders using our work, while refusing to donate $10, or edit the wiki, or translate something, or run a proper test suite are NOT beneficial to OSS programmers. Quite the opposite. You're just another oddball configuration I have to support, and another piece of idiot-proofing I have to add to the GUI, with no benefit to me. We don't want it to be widely used, we want a wide base of CONTRIBUTORS.
You know what? If Igor thinks can do it better, then he should fork that thing and roll his own distro.
Or, instead of forking, contribute a patch or two to improve things.
I thought I could improve RAID in the Linux kernel, so I did. Patch accepted, so now when I download a new version of Linux, it includes my fix and thousand of improvements others have made. I thought I could improve Apache, so I did. Patch accepted. I thought I could improve Moodle in a half a dozen ways. Half a dozen patches accepted. I thought I could improve Linux:LVM. I'm now the maintainer.
Forking is the last resort, when no reasonable patches are accepted. If you don't like the way something works in OSS, contribute a fix.
"Everything should be made as simple as possible, but not simpler." -- Albert Einstein