Its not so much double dipping as in the ads aren't worth much online and advertisers know that very quickly. The TV and print media have been very good at convincing their customers that ads worked and they worked well. Why do you think that being an advertising executives was shown in such a good light in 1960s tv shows?

5 km should be trivial. Less than 10 meters should be doable using consumer grade hardware.

Pulsar emission areas have been mapped to about 2 meters according to some research out of Australia using radio telescopes. If they can map an emission source to 2 meters several light years away, then I'm thinking they should be able to get positions better than 10 meters when combining things. Another way to look at is that 2 m is about 6 ft which is about 6 nano-light-seconds. The trick is to adjust a clock tick to a known pulsar spin but a similar problem was solved with psedoranges with NavStar.

Many modern GPS receivers already have special circuits or software to help remove the pulsar noise from the RF signals.

The big problem with using pulsars is the scary sums that have to be calculated. Most modern GPS devices use a Kalman filter that take into account a high order polar polynomial that has factors such as sun's gravity change, as well as orbital wobbles involving Juniper and Saturn. For pulsars you have to have all of those combined with things that are near each pulsars and your local neighborhood. For our uses I think a 70 terms should about cover it.

How many deaths has it cause by helping spread such fun things like Methicillin-resistant Staphylococcus aureus (MRSA)?

Foxtel has about 1.6 million customers (two cable boxes in a house count as 2) after 17 years of building their network. I don't think they did 1.5 million in a year. Be careful about "pass" rates and connections. Also don't forget to count the ones using satellite as well. There might not even be 1.5 million premisses hooked to HFC in Australia right now.

How do you get that many people with the proper licensees to do the work? In Victoria they will 48 months of training. What your proposing is sort of like getting 9 people to make a baby in a month.

Some of the complaints are technical in nature.
Lots of stuff has been promised that simply can't happen with the current stuff that is being installed. For example the ONTs didn't have the ability to have different VPNs for data or voice traffic. The solution was to layer a VPN on a VPN and the last time I heard, it cost the same for a data+voice channel as just a data channel or a voice channel.

Some of the equipment being installed had American telco inventory tags on it. If the original equipment owner can't offer gigabit (and are currently advertising 15 mb as fast), how will the rolled out system provide gigabit?

It is also looking like nearly all the traffic in an area will go through 2 choke points. I don't think that having a system were two people with bolt cutters can take out most of an entire state is a good idea. If you calculate how many bits have be going through that system compared say a large peering point like AMS-IX, you will find the numbers are orders of magnitude larger which implies there will be some very interesting problems that have to be solved.

Then there is the single shared fiber problem that NT&T has been working on for decades and have finally given up (they now use a name implying two stars for their solution which effectively uses two PON networks). It turns out that single fiber have the same problem that radio does when you want to talk fast, your transmitter blinds your own receiver for a short time and that introduces latency. You also have to coordinate all the transmissions. An ethernet packet on this network takes up about an inch and some of the links will be 70 km long so some how something has to magically get 32 optical devices to talk at just the right time without stepping on each other. To make that even more fun, they are hinging the fiber from poles so the optical length of the signal changes every time the wind blows.

As far as attracting global users to the network, they look at the power bill first and walk away.

I posted this just above, but as this is a thread purely about linux on SB, I'll place it here also:

Be careful. You may get bitten by this bug. The tl;dr version: If your apps use dynamic loading on Sandy Bridge, you may get segmentation faults cause by a bug in glibc.

RHEL should have this fixed by release 6.3. Other clones of EL will get the fix via the update to 6.3 after RH has released it.

Be careful. You may get bitten by this bug. The tl;dr version: If your apps use dynamic loading on Sandy Bridge, you may get segmentation faults cause by a bug in glibc.

RHEL should have this fixed by release 6.3. Other clones of EL will get the fix via the update to 6.3 after RH has released it.

Awesome. Does that mean other countries can extradite US politicians and business men for screwing over companies and in some cases entire countries?

Oh right, what was I thinking... :\

If you're doing a new UI (like say the iPod touch one before it was around), you need a language that allows you to collect events that most languages don't. In fact most languages don't deal with the events at all, they just let the OS cope with some of it and massive libraries of poorly written software for the rest.

What language allows you to hook into the condition where a mouse down event happened here and then there were motion events along this path... It gets even stranger when you start to deal with gestures or eye movements or whatever else you need to do to invent a new UI.

The result of all that leaves C, assembly and forth with top layers in whatever you want.