I have done this and it is a nightmare. The software that supports the CrystalHD card is buggy as hell, the XBMC support for it is still not great, and the performance is still spotty (running the latest XBMC SVN builds and the latest CrystalHD code from svn). I can't get it to stream The Dark Knight without dropping frames, pausing/resuming causes all sorts of problems, and exiting out of a movie crashes XBMC half the time for me.

ION hardware is absolutely the way to go, no question about it.

Maybe while they're at it, they can add in actually-useful error messages. See http://blog.llvm.org/2010/04/amazing-feats-of-clang-error-recovery.html for some examples. It's shocking how user-hostile GCC is in comparison.

You jumped right on the LLVM part, but apparently don't know what clang is. You should read about clang. http://clang.llvm.org/

Nah, NVIDIA gutted their QA department about a year and a half ago and the people they still had afterwards have been overworked. It's not like the Linux driver team was the only group that completely failed to catch the fan control problem, so it doesn't make any kind of sense that this decision would be tied to that fiasco. The more disturbing part (to me) is, this isn't even the first time that fan control regressions have made it into shipping drivers. But this time the regression caused the fans *not* to spin up properly, as opposed to spinning up too much.

Realistically speaking? Nobody cares about SPARC, least of all NVIDIA. They've got enough shit to worry about, with Intel squeezing them out of the northbridge chipset market (No, you can't have a DMI license!) and ATI/AMD kicking their ass in the general enthusiast market (have you seen the Radeon 5xxx series?).

It's a damn good thing for them they had the foresight to get involved in the ARM business, because that may be the thing that keeps them alive for the next few years.

The only thing that I found surprising about this announcement is how long it took for them to finally decide to kill it. The 'nv' driver wasn't doing them any good in the first place.

A fair point, but aren't there are a larger percentage of mailservers running !sendmail than there are DNS servers running !BIND at this point?

I hope I'm wrong.

Why would they even release it if their ground-up rewrite is so pathetic? Were they worried that BIND might be losing its rich reputation as the worst piece of widely-used network software ever made? If so, bravo, guys.

Make sure you get either an Intel X25-M (though the biggest one they offer is 160GB) or something with an Indilinx controller (OCZ Vertex, for example, up to 256GB). Stay away from anything with a JMicron controller - those drives might be cheaper for bigger sizes, but the performance is crap.

If you'd ever used an iPhone or iPod touch, you would know that the iPod app keeps playing in the background when you switch away. So yes, you CAN browse the web while listening to music. Unless you meant "you can't browse the web while streaming Pandora", which is true. But that's not what you said. :)

So unless Apple has broken this functionality without stating that they have, your first complaint is completely invalid. The screen is higher resolution than DVD, and DVD-quality video is clearly the most popular option in the US right now, so the screen is capable of showing better-than-DVD-quality video... so, no, it's not 720p, or 1080p, but it's still going to look pretty damn good.

Of the issues you raised, the only really valid one is that you can't use a USB thumb drive with it, but that's hardly a dealbreaker.

That said, Slashdot approval has been pretty much inversely related to Apple product success in the past, so it looks like the iPad will do extremely well based on the comments here...

Oh really?

Joystiq and Kotaku seem to disagree with you.

I don't know about the having-to-be-online-to-play thing. I'm usually online. But there has in fact already been a PC patch.

Actually Chrome (well, Chromium) has been rock-solid for me on Linux. It replaced Firefox for me about a month ago. The only major feature that I'm missing at the moment is printing support, which they're working on, but it's not like I print things that often anyway.

It's so much better than Firefox on Linux that it almost makes me feel bad for the Mozilla devs, except that they don't really care that much about Linux users.

If any Google employee ever attempted to look up personal data in the fashions you just described, they would be fired so fast they wouldn't even know what happened. From my understanding, it is currently impossible to look up user data in these manners without setting off a whole ton of red flags, and all of those red flags mean "immediate termination."

What are those cases, exactly?

I was playing around with the latest msysgit (1.6, 12/27) and it didn't seem to be much slower than the Linux git for the things I was doing. (I cloned the xserver repo from freedesktop.org, made a couple of branches from arbitrary points in history, used 'git blame', etc.)

msysgit was definitely slightly slower - it took 47 seconds compared to 27 seconds to clone the repo, for example - but nothing that I'd consider to be "extremely slow". And the xserver git repo is much larger than most typical projects, so I'm wondering what it is that I'm missing.