I do the same. Or, if I want to guarantee a quick end to the conversation I wax poetic about how it's applied mathematics where all those algebraic, geometric, and analytic techniques everyone learns in grade school are extended to apply to things like airline reservations and sub-atomic particle modeling. This quickly puts an end to that line of questioning. (Unless, of course, you're talking to a nerd then you can both wax poetic about applied mathematics all night long.)

I've got decades of F'ing adapters cluttering up the place. I could probably produce a handful of SCSI-Parallel adapters and old TokenRing terminators if you wanted them. I don't need another dozen USB adapters in desk drawers and bag pockets.

As a professional software engineer, I have never looked at the side of my MacBook Pro and wished it had fewer ways for me to connect something. Never have I complained about a few millimeters lost to an rarely used SD card slot or Firewire port. I know that eventually a customer is going to hand me an SD card or old external hard drive and I'll be thankful I can just plug in and have it work. "Professional" machines have to be able to do anything when needed. If Apple removes _THE_MOST_USED_ port for connecting miscellaneous peripherals then these machines no longer deserve the moniker "Professional". The only complaint I've ever had about USB ports is that I don't have enough! (And don't get me started on never-there-when-you-need-them dongles and adapters.)

GRC.com is hardly "some random website." Steve Gibson is one of the most well respected computer security gurus around.

Not in my experience. I maintain and continue to develop several apps that are largely indifferent to the JVM version. In most cases there is a minimum necessary version and very rarely does an update change behavior at all. Those times that updates have forced code changes, they've always been minimally intrusive. Furthermore, to safeguard against update screwiness, just package your own favorite version of the JRE with your app; something that can't be done with .NET's system-wide deployment.

As a real honest to god software developer for the past twenty years and having worked with just about every language and environment one might care to name, I can say that the prevailing sentiment amongst myself and my peers is that we're tired of all the platform/language headaches. From our perspective, there's little compelling reason to write in anything other than Java. As requirements; speed, efficiency, frugal resource usage, etc. pale in comparison to needing maintainable, reusable, run-anywhere code. Time and again we've had to rewrite engineering models, data services, etc just because their old code wouldn't fit in to whatever system was being put together. There are certainly specialty problems where absolute efficiency is of paramount importance, but I'd be weary of short-term gains at the cost of long-term maintainability.

(For the similar reasons, this is why we favor open-source tools. I've written millions of lines of code that can no longer be run because some crucial third-party component's license expired a decade ago.)

Wait a minute! You two expect us to believe you were über-1337 using linux on floppies, but both of you have six digit slashdot IDs. I call shenanigans. ;)

Remember that nature doesn't seem to have infinite precision. Rather its precision is bounded by the Plank scales. Pi, for exemple, need only be computed to a precision adequate to describe circular circumfrence down to the Plank length.
Yes, that's fantastically small, but it is a finite limit.

True, but they broke the first rule right out of the gate: never let someone else pack your bags.

As I understand it, the teams were handed a group of fully loaded virtual machines and told to put them on their network and then secure and defend them. No admin in their right mind would ever do this if they're at all concerned with security.

The 'rootkit' mentioned in the article apparently was an executable present on the VMs at the start and not found until its network traffic was sniffed. Kudos to them for finding it, but on a network built for security it would have never been there in the first place.

As a rule, you always start clean and install only what's necessary to provide whatever services are needed. When a security compromise is found, you wipe the system clean and start over.

But, all of this discussion is a bit irrelevant since this was ultimately not an exercise in building secure networks, but rather an exercise in watching how defenders of an insecure network will respond to a series of attacks.