A 'git push' every now and then may help* in these cases... ;-)

* (both you and the other guys working on your project)

Although this may be OT, I do ask design patterns to my interviewees. However it's not like "tell me what a Visitor is", rather "you have this problem, how would you solve it?". The good ones come up with several solutions, some of which resemble (or exactly overlap with) a design pattern, and that's enough. If they are also aware of this, it's a bonus.

Said that, the value of patterns is not just in using them (they originate from common sense after all) but also in creating a common ground for communication and practice: if you see a class named "Observer" or "Factory" you should immediately imagine what it is about. Presentation patterns could have similar benefits, e.g. an organisation may decide to deliver all presentations in a certain format.

There's nothing new in this, just a few examples: the 10/20/30 rule and pecha-kucha (pronounced like this).

The article is correct, the DSN has just a couple of letters in common with DTN, and nothing to do with the Bundle Protocol.

Delay/Disruption-Tolerant-Networks have been researched and developed by the DTN Research Group and the Bundle Protocol has been an RFC since 2007. It's possible to download an open-source reference implementation from SourceForge.

Actually NASA also use their own protocol, called ION (Interplanetary Overlay Network).

1) Gingerbread was the first version to give a fairly complete user environment (compared to iPhone).
Ergo: Most software has been developed for 2.3.

2) Most contracts with mobile operators in UK (and I guess in other countries as well) last for 24 months, and devices cannot be upgraded in the meanwhile.
Ergo: I'm not going to pay for a new device before the current contract end.

3) Honeycomb is mainly for tablets, which have quite a narrow user base compared to mobile phones.
Ergo: Gingerbread devices > Honeycomb devices.

4) Android versions are not easily backwards compatible.
Ergo: I'm not going to port my current software, unless there's a significant user base (if it ain't broke, don't fix it)

...it's where my unpaired socks disappear.

Does it run Linux?

Suetonius reports Caesar words were spoken in Greek (kai sou teknon - no greek letters in Slashdot comments apparently). A more correct translation in Latin would be "tu quoque fili". Sometimes the whole phrase is reported as "tu quoque Brute, fili mi" (you too Brutus, my son).

If you fancy some DIY, there are several good chipsets for media players: Intel CE4200 (i386) and several Marvell ones (ARM) just to name a few. They normally handle 720/1080p, stereo/5.1 audio and yes, they all run Linux :-)
I personally like Marvell Armada chipsets (have been working on them in the past) and they're also used in some plug computers, which is something you may like to try (double-check the specs though).

Once you've got your HW decoder, you can generally run gstreamer on it. You may need to fine-tune gstreamer for the particular chipset though, and possibly write your own player around it, as using gst-launch is quite hard-core. There are several other open-source libraries to add nice features, e.g. lirc for IR remote controls, DirectFB for GUI/OSD and WebKit for web browsing. Qt may be ported as well, which would be a good plus to me. You may also consider live555, which is what VLC is using I think. Of course XBMC is still a great solution, if you want things working (almost) out of the box.

The best thing would be to double-check the SDK documentation (whenever accessible) and see what's supported by each particular chipset. Whenever a core feature is there (e.g. the codecs you want, input support, etc.) then it's possible to write some SW to use it. Unfortunately, in my experience, Python is seldom considered, but if you fancy C/C++, sky's the limit.

I really hope not!

Then why not making medicine a compulsory subject too? Otherwise people will depend on a selected few to repair their bodies. And we should also add some industrial mechanics ts as well, we don't want to depend upon some elected guys when our [add any mechanical device here] breaks. And what about energy making? That's far more important than programming, and we don't want to depend on another set of few companies' know-how.

I could continue, but the bottom line is: we have to depend upon other people, so let each one be free to choose who shall her/him rely on. I'd rather improve the quality of current compulsory subjects, in particular humanities: Our kids really need to get a broad perspective of human evolution in order to contribute to it, and no CS course will ever teach you how to think out of the box in the same way.

Also, I'm glad that there are people who couldn't care less about programming: many of them are artists and they often enrich our lives with alternative perspectives of the world. And they also mean more job for me.

Always take your umbrella with you.

Many of you compared BitCoins to national currencies, but they are not. This is quite an important factor: at the moment people can afford to "stash" bitcoins because there's no real need to spend them: we still rely on national currencies for buying most of what we need.

However, the fact that it's getting more and more difficult to mine them also means that the 21M limit is getting closer. Once reached, the impossibility to mine new bitcoins will inevitable affect their value and I'm quite confident that bitcoin trades will rapidly increase, as that will be the main way to "earn" bitcoins (vs exchanging them for other currencies).

Honestly I think that having an exchange market has been a bad idea, as well as making bitcoins to last forever. If they'd been granted a maximum lifetime, people would have been forced to trade them as quickly as possible. What happened instead is that many guys mined bitcoins just to have a try, and will never use their stash, so those bitcoins are lost forever.