If C is so great, why isn't the entire PiTiVi project written in C?

Don't get me wrong, I'm a big fan of C, and it certainly has its advantages when writing performance-critical, CPU-intensive code. But it can also be slower language to develop in than higher level languages like Python. That cost in human development hours matters.

For code that doesn't need real-time execution guarantees, which describes a whole lot of projects, Python's performance is already good enough. But where it really shines is that Python drastically improves developer productivity, enabling the delivery of more value to end-users in less time and with greater flexibility as the needs of the software's users change.

The industry migrated from C to Python for the same reasons they migrated from assembly to C. (And I predict it will migrate again in my lifetime.)

C isn't a superior language to Python, it's just a different one. They make different design trade-offs and both have a well-deserved place in programmer's tool bag. As does JavaScript or Lisp or Scala.

Though, as much as you might stan for old-fashioned C and hand-written assembly, benchmarks tend to show that JIT-compiled statically-typed languages like Java can out-perform well-written C and C++ code (if you discount the JVM's startup time, which is completely fair in the case of long-running services.)

C is never going to go away. COBOL, Pascal, and BASIC still haven't gone away despite how much many people wish they would. Python is also going to be around for decades to come.

I don't see this as a problem.

Once routing is established at layer-3, there's no reason you couldn't run a DHT or CAN system (such as Freenet) at a higher OSI layer. As we say in our presentation, the Internet is broken on many levels but we're starting from the lowest ones. Projects like Freenet are starting from the top-down, rather than the bottom-up. Both are necessary and meet in the middle.

To be honest, I'm a bit surprised at all of the negative comments I've seen regarding the "zombie apocalypse" bit. We have tried to make it clear that the zombie thing is a joke, the real inspiration and goal of the project is natural disasters and political activism. On our original wiki, those goals were listed as items 1 & 2, the zombie bit was added as a whimsical third scenario but it seemed to catch on people who were interested in our project.

In reality, we are developing Byzantium to be useful in exactly the kinds of situations you describe. We have been discussing these ideas with people who have experience or expertise in emergency situations, such as political protests (Egypt, Syria, and OWS) and natural disasters (Katrina and Port-au-Prince). We are eager to make Byzantium into a solid tool for first responders who have a need to communicate and organize the relief effort. But also, we want it to be useful to individuals and communities seeking to prepare themselves against outages.

Yes. The routing protocols we have chosen are layer-1/layer-2 agnostic. You can connect to other nodes through any physical link you have available. Dial up, satellite, HAM radio, a VPN tunnel, carrier pigeon, ...

We believe that it can be done, that it can be done securely, and that it can be done securely for iPhone users. Obviously there will always be challenges and vulnerabilities in any system this complex, but we can make an effort to address those or at least make it easier for end users to exercise common sense. We're not there yet, but we think this is a step in the right direction.

We're still having some trouble supporting MacBooks, particularly because none of the core developers own one. We'd love to help you troubleshoot the issue, as we certainly don't want to exclude Macs from being nodes. Best way to reach us is through our mailing list or in #byzantium on Freenode.

I recommend Commotion Wireless.
http://commotionwireless.net/

The great folks at The Serval Project are working on that and we're keeping a close eye on their work.
http://www.servalproject.org/

We have started on this already. https://github.com/Byzantium/Byzantium/wiki/Setting-up-a-mesh-node-without-byzantium. We will have a lot more on this in the coming months. We're heading to an International Summit for Community Wireless Networks in early October to collaborate with other mesh projects and communities on some standard for interoperability. Also, the software we're using to build Byzantium will be packaged and submitted to the top Linux distros for inclusion in their repositories, but for now it's easier to develop and test on a closed platform.