I think this is as much about training as it is about a technical solution, but here goes.
My wife clicks on anything, so I set her up a Linux Mint machine. I removed everything from the Desktop, and I took the "menu" button off the bottom bar. She could restore it but she doesn't know how and doesn't care about it.
Then I put one, and only one, icon on the Desktop --- for Firefox. I made sure everything was set up, installed AdBlock and Ghostery, and that's it. Done. She can do one thing, which is launch the browser.
In the Poster's case, I would have three icons: Google Chrome (fixed up with AdBlock and Ghostery), Thunderbird (so mail can be read offline), and one icon tied to a script to launch the dialup. (Or even better, use that old dialup-on-demand software; I don't remember the name but back in the day it worked perfectly.) I wouldn't necessarily put a file manager on the Desktop. Too much chance of messing something up.
Set up Thunderbird to automatically fetch mail when started. Definitely switch to a gmail account to get around all the size issues.
I think this would give what was requested, except for remote troubleshooting. Someone suggested dynamic DNS and ssh, and that might work well enough.
For the most part I would also want the machine locked down enough so that an incompetent but well-meaning do-gooder can't make a hash of things.
I would have started dumping on Apple but when Google put the "feature" in Kitkat to disallow general write permission to the SD card, effectively making an SD card almost worthless unless you root your Android phone, I had to admit that there is enough guilt to go around.
Oh, wait, what's that you say? Apple phones don't have a provision for an SD card?
"In the end, yeah, Windows, yuck, but deal-able, and it's really disingenuous to pretend that because they have dumb downed the initial install package to Windows levels, that the actual ongoing user experience of Linux is nearly that plug and play for most folks, so to speak."
I don't disagree, and the average Jane won't be able to maintain Linux (although I wonder if the average Jane can maintain Windows, either, if something goes wrong).
But consider this situation: I set up a Linux machine for my wife to use (she likes to click on, well, everything, and I figured it would be safer). I do updates every so often. I support it for her. But it's not like it needs much support.
She doesn't even know it's not Windows. She has no idea what she's running and doesn't care, as long as she can browse, do email and Facebook, etc.
There is no problem. I could care less if other people run Windows as long as I don't have to. If somebody buys a computer with Windows pre-installed and then decides to run Windows because it's an appliance mentality, no skin off my nose. If they're happy, great.
The primary problem for me, though, is in being forced to pay for Windows. I build my desktop machines from components so it's not an issue there. Laptops
The secondary problem is friends and relatives asking me for help with their Windows computers, but I can happily tell them I don't run Windows and sorry, can't help; however I'd be happy to install Linux for them.
"It's news because it illustrates that, as much as Linux users like to throw stones at Windows, they too are vulnerable. Anyone can pick through the source and find security holes what can be exploited - perhaps even much more subtle ones than anyone would ever find on Windows."
I find this fascinating. Some Windows fans will grab onto something like this, an exploitable bug in Linux, and use that to "prove" that Windows is better. "Look here, Linux has an exploitable bug, obviously it's no good. I told you how much better Windows is, now it's proven science!" And the additional comment about finding bugs more subtle than those on Windows? Where's the evidence for that statement, other than perhaps in the fact that Linux source is readily available while Windows source is not?
Any OS has exploitable bugs. Failure to patch is, as noted by many other posters, the real issue. Don't necessarily condemn one system or the other because there are bugs. Instead, maybe we might look at the track record, in which case some conclusions should emerge.
It is pretty good in places that never developed a Windows culture..
I think this poster has identified the real issue. I doubt that Microsoft bribed people to complain, though I'm sure they provided subtle encouragement. I'm sure they also worked at a high executive level, not with outright bribes, but in the way that sort of thing is always done, the old FUD method.
But really, it comes down to people who are used to Windows wanting Windows, and they'll do that (mostly) even in the face of a mess like Windows 8. "What's this weird Linux thing they're making me use? I never had to use that anywhere else! Other organizations aren't converting, why are we?" And so on.
I don't buy that Windows is inherently more "office ready" than Linux for the vast majority of office users, all else being equal. The thing is, all else isn't equal. I do buy the idea that Windows is heavily entrenched and has a huge "incumbent" advantage, one that is going to persist for a long, long time, whether we like that idea or not.
I've never, ever had the severe kinds of problems you mention, and I've been on Ubuntu or its derivatives (most recently Mint) for years and years. And this is across maybe a dozen machines of all descriptions, and with all sorts of graphics cards, including the dreaded nvidia, which works just fine and only required a little patience.
That is not to say I take the position that Linux is completely golden and Windows is purely trash. There are always bugs and problems. However, given what I paid for my Linux distros, I think I got a really good deal which far exceeds expectations.
The problem is that some vocal Windows people will jump on Linux bugs as "proof" that Linux is not "ready" and so on. But Microsoft's latest antic convinces me that Windows, after how many years, is less ready.
We can cut through all the nonsense and PC vs. science stuff pretty easily.
Are some things determined by genetics? As another poster mention, white people tend to have white kids. How far does this go? I don't know. Objective, non-political science should be the way to answer this question.
But, just for argument's sake, let's say that people with green skin have a tendency to be more violent than people with blue skin. I don't know if that's possible or not, but let's just say it.
The real issue is not that it is or isn't politically correct to state a scientific fact. The issue is pre-judging individuals and even worse, acting on that pre-judgment.
You're green? Oh, you're a violent criminal. You're blue? Oh, you're higher class.
Are there differences between races? Your eyes tell you that there are in a split second. Are there other differences? I don't know for sure but it seems likely. But do I have the right to pre-judge you, to decide you have a lesser value as a human being? No, I do not.