I'm not a user of Windows Phone, but I did just port an Android app I've written to WP7, and in doing so, I learned quite a bit about it... From my point of view (been an Android developer before the first phones were released), it seems like WP8 will be very nice, but WP7 is still lacking in a lot of ways. A few things I noticed:
    - there's not a whole lot of useful multitasking you can do right now, so complex apps that use background services are right out.
    - you can't disable the on-screen keyboard from activating when a text box is focused, so if you have a box that the user can select text from or position the cursor in, you always get the OSK covering half of your UI
    - the screen layout designer is difficult to work with, and doesn't seem like it has many features for supporting different resolutions, MS sure does love their absolute-positioning grid layouts
    - there doesn't seem to be a debug log viewer available in the development tools... or maybe the OS has no logging at all?

I suspect an end user won't really notice a lot of my complaints, but they're there, and the whole experience was a bit disappointing to me, despite my preference for C# over Java.

Nope, I agree, I really didn't enjoy reading the book, despite some of the concepts being interesting.

Right, you're ok with Apple spying on you but not AT&T or Verizon? Fascinating.

Enjoy the rest of your life :)

Yes, but when you have to deal with an *existing database* that has composite keys, you're out of luck.

> If you want to get a job programming, but have never written any software that you've published, then you are probably not worth hiring.

No, sorry, that's not really true. There are a lot of exceptional people in the industry who have never published a thing, because all of their work has been on internal projects for their employer. I don't see why publishing software is very relevant - a lot of published software is complete crap.

But it DOES remember SSH keys, which can have a null password.

I know nothing of Ubuntu, but on my laptop, I just use xrandr bound to a hotkey to do this, and it Just Works(TM) perfectly. I really don't see what the problem is, this isn't 1999 anymore.

They do. It's called the Android Development Phone. Buy one of those, and you have root right out of the box.

What's the preferred 1-wire interface on Linux these days? I used to use owfs, but it's been a few years since I did anything like this.

Years ago, I'd have argued with you here. I've read (and own) all of Niven's "Known Space" works, I started reading and collecting them when I was 14-15.

I just went back and read a few, and read the new "Draco Tavern" collection. Still haven't read any of Niven's work-with-others, but now that I'm older... yeah. There's a ton of FAR better work out there than Niven-alone has ever put out.

Sorta like a childhood memory that just isn't so great to relive when you're an adult. Oh well.

As if looking at the physical packaging tells you everything. You have to remember that "one CPU" does not mean "one chip" - it could mean there are many physical chips, DSP, radio bits, etc. And some of the "multiple-chip" phones do the baseband processor and application processor on a single chip.

The point is, does the processor that runs the user interface also implement the GSM protocol? You'll find that most phones (again, other than the batch of modern smartphones (iPhone, Android, etc)) do this in a single processor.

I imagine using the same die would be cheaper than using a second chip, but surely you don't think it'd have no cost at all?

Incorrect, there are many phones on the market that you are able to reflash the baseband code. Believing that anything like this is impervious to hacking is just naïeve.

That $99 is still something I shouldn't have to pay to work on apps for *my own device*.