1) A "universal" basic income that is NOT UNIVERSAL is a contradiction in terms.

2) Starvation-level UBI is an insult.

It's a nascent rename. My guess is the renaming will be complete when FreeBSD 11.0 is fully released.

Something is rotten at google

EVERYTHING is rotten at Google. They've got a tin ear, a blind eye, and a dead brain.

If you don't limit yourself to stupid brain-dead pared-down models of Xeon, you don't sacrifice the integrated graphics. Those that end in 0 are crap. Those that end in 5 or 8 are good. The e3-1225v5 is cheaper than the e3-1230v5 and has graphics. A much, much better deal. There isn't any 1235 (yet), but there is a 1245 and a 1275.

There are flaws in every sufficiently complex piece of technology, and no, you can't just magically find them all by looking Real Hard.

No. May I repeat myself? HELL NO.

Correctimundo. The one standout brain-dead no-no.

Bzzzzzt. Combining tabs and spaces to make up a single indent level is BRAIN DEAD.

100% of my *.py files use two spaces to indent...

Ah, a man after my own heart. I also use two spaces per indent level. Any more than two is kindergartenish. And yes, I realize that Guido hates me for this, but none of the stuff I write is ever seen by anybody except me, so I don't have to care about catering to those with bad taste. Whenever I used to work on Python code in a team, or made something somebody else might have to maintain later, I did adhere to the dreary prevailing bad taste.

Know what? I use the same 2 spaces in shell scripts, Perl, C, C++, and everything else, too.

The article conveniently ignores Python, a 100% tabbed language.

Bzzzzzzt. Absolutely wrong. Google "PEP8". "Spaces are the preferred indentation method."

Stupid goddam braindead suggestion. Manual syncing, to a crappy slow undercooled USB hard drive, yet - yech.

Assuming the OP's question is a serious one, and assuming the OP's commitment is serious (both of which are questionable assumptions), the answer is to set up a server or NAS with 7 drives in a RAID-Z3 pool; 1TB drives would be plenty, as you would end up with 4TB of triply-redundant storage. FreeNAS, FreeBSD, or Ubuntu 16.04, take your pick, would serve fine. All three of those support ZFS natively.

With a single RAID-Z3 pool, you can lose ANY 1, 2, or 3 drives AT THE SAME TIME without losing any of your data. And with ZFS all your data is AUTOMATICALLY checksummed, and every read operation checks the checksum, and any errors in any read are AUTOMATICALLY corrected before that read returns. You don't have the problem of non-ZFS RAID0, where the system has no possible way of knowing which of your two mirrors is right if for any reason their data differs, and in fact no way of routinely even detecting such an error, let alone notifying the user. And you don't have the problem of the write-hole vulnerability of non-ZFS RAID5.

With ZFS you don't have to take the system off-line to resilver ("rebuild") if you lose a drive and have to replace it. The pool remains perfectly usable during the resilver. In fact for anything less than extremely heavy use, you can barely tell anything is "wrong" with the performance during the resilver.

Obviously this is only a start. A damn good start, but you still need to incorporate off-site backups one way or another.

Both rigid and nonrigid airships in operation are so close to infinite structural rigidity (no significant deformation under operating loads; not infinite strength) that there is no difference in reaction to gusts. As to the difference in cross section, what is the point specifically?

I think there is more similarity to smashing the Staten Island ferry into the dock - something that has been done more than once. They are both crashes. Nothing whatever to do with sinking. On 6 May 1956, the battleship USS Wisconsin collided with the destroyer escort USS Eaton. That was a crash. Neither vessel sank. On 1 February 1944, the battleships USS Washington and USS Indiana collided. That was a crash. Neither vessel sank. A car can't quite stop in time and rear-ends another car. There may be only minor bumper damage to each, and certainly neither one sinks, but you better believe it is a crash.

What is NOT common, what no pilot would ever do, is to keep the elevators in down-ship position all the way until the thing has crashed into the ground. Watch the video. My guess is control system failure.

A design like Airlander 10 is fundamentally a lot more resistant to the common problems that plague blimps during landing, such as susceptability to winds

Highly debatable, and never proved.

It has less inherent lift

Not significantly. At most it has 1.7 times the mass of an equal volume of air. A blimp has close to 1.0 times. An airplane such as the 747 has over 200 times. The Airlander's susceptibility to wind influence during landing is very, very nearly the same as a blimp, and nothing whatever like an airplane. And its flattened shape has the added excitement possibility of dangerous rolling, which is completely absent in a blimp.

a smaller cross section

Wrong. For a given enclosed volume, the blimp shape has a smaller head-on cross section, smaller vertical cross section, and only slightly higher wideways cross section. Even if you reduce the enclosed volume by the ratio of 1.7, the difference made would be very slight.

and more ability to anchor itself down with its fans.

Wrong. The Airlander 10 has nothing more than pneumatic skids for landing gear. The Airlander 50 is supposed to get a full air cushion with the added cpability of suction, but that idea has never been tested under realistic conditions. Gust forces on an airship held stationary on the ground are enormous. Suction pads with enough strength to overcome them could easily induce the hull fabric to tear wide open.

Because an airship is a type of aircraft, ignoramus.