AFAIK the only plugins marked as broken are actually broken. I believe the python API is now generated from C++ code directly rather than being hand coded, so some breakage is possible due to that, however, the vast majority will be broken due to website changes, rather than XBMC-side changes, which have been reasonably static for the last few months (i.e. plenty of time for add-on devs to work with).

There is no marking as broken as a matter of course on version update as far as I can discern. If anything, the add-on would simply not be moved into the new versions repository - i.e. current users won't be affected, but it won't (initially) be available to new users until the author has tested and approved it for the new repository.

Considered purchasing a cheap Android box? There's a bunch of them now that work nicely with XBMC for under $100. I'm sure you can get some info on xbmc.org as to which ones work well and which ones are best avoided. Alternatives are getting a UPnP server that will transcode - XBMC won't do that, but there's many other apps out there that will. Ofcourse, there may be other deficiencies with your TVs UPnP client to concern yourself with - thus the suggestion of a cheap client.

There are touch based skins that might operate a little better with the mouse - Confluence works, but sure, it's not designed for it. The official Android remote should have an update for Frodo in the play store shortly (it's been submitted afaik). There are other remotes out there and apk's are available elsewhere in the meantime.

Is your computer plugged into your TV? If so, then ofcourse. If not, does your computer support UPnP, and does it support the types of files on your computer? If so, then ofcourse (subject to the TV having a half decent UPnP client). To fully apprecieate XBMC, you want the former, not the latter.

Are you selling axes? Do they come in other shapes than the one you're grinding now?

Except the trackpad sucks.

It's 24dB/octave in a THX system, at 80Hz. This is typically done with a 24dB/octave Linkwitz-Riley lowpass filter at 80Hz in the receiver, matched with a 12dB/octave highpass butterworth at 80Hz set to match the natural (i.e. 12dB/octave butterworth) rolloff of a sealed enclosure speakers for the mains.

Your sub(s) won't be producing much of anything over 160Hz or so - and if there is anything the directional cue comes so much more strongly from the mains at that point it makes no difference where the sub is placed, other than overcoming room nodes.

So many people know absolutely nothing about the audio engineering behind good quality surround sound.

A possible improvement to the algorithm would be to check all new edges during a merge of two components rather than just testing for whether the constraint of planarity is violated. This would avoid errors in merges - many of the sample images show components that join well on one or two edges, but badly on others.

Don't blame Kim DotCom for the idiots that are the NZ police, or NZ politicians. They all need taken down a notch or two for not only illegal behaviour, but for idiotic behaviour. That we get to enjoy the show while it all goes down just adds to the amusement. Kim Dotcom will no doubt milk it for all it's worth, and so should he!

I'm assuming you subscribe to the idiocy that in order to be in parliament, a member has to be directly elected by voters, rather than being high enough up a party heirarchy to be elected through the popular vote. If so, you're wrong, as _by_definition_ they were directly elected through the popular vote.

In the case you purchased outright from Apple, Apple is the retailer and you'd get exactly the same coverage under the law, as the same devices are being sold by others (under Apples blessing) on 2 year contracts.

You can quite easily keep increasing energy consumption without exceeding some predefined level. Just the amount of each increase becomes arbitrarily small (eg 1 - e^{-x})

citation needed.

If you're genuinely interested in improving XBMC (by pointing out why it didn't work for you) please go to the XBMC forums and let them know of ideas to make it better. "It didn't work" is useless both ways as you quite rightly point out. It's not going to get any better unless you give folk reasons to improve it! XBMC is 100% open. Boxee is not - both the hardware and (parts of) the software are closed. Which is an incredible shame, given it's origins. Just imagine what the XBMC guys could do with even 1/10th of that investment.