You don't have to cull it down, you just need to organize it into logically distinct groups and assign them priorities. Hoarding isn't the problem, the problem is assigning too high a priority to the hoarded pr0n as compared to the really important stuff.

• Group #0:

Contents: Documents, source control repositories, user preferences, email archives
Maximum Size: 10GB
Protection: 3-way Mirror + Snapshots + Offsite
Total Space Required (way upper bound): 150GB
Total Cost: $3 a month for Crashplan

• Group #1:

Contents: Personal photos, music collection, other people's #0 backups, /home
Maximum Size: 1TB
Protection: 2-way Mirror + Offsite
Total Space Required: 2TB
Total cost $3 a month for Crashplan

• Group #2

Contents: Everything else, media, pr0n,
Maximum Size: âz
Protection: Diskpool, maybe integrity if you like zfs/btrfs

(1) Password managers. One strong password, infinite storage. Maybe one for work and one for home.

(2) DropBox. One file, infinite number of distributed copies. Also available to sync on your mobile device of choice :-)

(3) Discipline. Every password is in the vault, no exceptions. Every change is updated and synced to DB.

Problem Solved.

Again, that sentence is unparseable by people in the real world.

Maybe tech workers have spent too much time in the tech world and forgot that terms like "SIP" and "POTS" are not really part of the language. Moreover, the mental model of understanding used to conceptualize the various bits of the phone system and how they interact is itself part of a specialized skill set. You can't get very far when you are, quite literally, thinking about it wrong.

And it's not just the idiots (although they certainly qualify), I've met tons of research academics, medical professionals, authors, historians whose make fundamental errors in how information systems work and are organized. They ask questions that I just have to answer with, "the fact that you ask this question means you have the wrong picture in your head already".

As an example, I once had a surgeon explain to me that his computer was not likely to be hacked because he always put it to sleep when he wasn't using it. I asked him what about the times it was awake, and he said that "well it can't be hacked because I'm using it at the time and only one person can be on the computer at once". Suffice it to say, I had to go down a few levels to explain things.

Or think of it this way -- he knows as much about computers and information systems as you know about orthopaedic surgery and the organization of the connective tissue in the body. :-)

If you're doing wifi-only, get yourself a static IP, run asterisk, use any old cellphone with SIP support and wifi and skip AT&T, as they are fuckers of the highest degree. Their prices are beyond fucking ridiculous. They want $50/mo for a land line. I got SIP for about ten bucks a month. My Xperia Play is now our cordless phone, and it's also a neato clock. My server is a $20 pogoplug, but in fairness I bought two of them so I could do HA.

You forgot to account for the time of a competent admin that can set up asterisk and is around to troubleshoot it.

$50/mo for a landline is stupid, but make a fair comparison -- you can't rate a system that requires a /.er to design and set up against one that a person with an IQ of room temperature can use.

For many (but certainly not all) applications, especially when it comes to UI, what matters is 95% worst performance, not peak throughput. From the Anandtech review, that's where this drive really shines.

Different tradeoffs have to be made for different workloads -- it can't be boiled down to a single (or even a set of) number(s). Some applications are far more tolerant of worst-case performance than others.

You are right to be sarcastic but you are dead wrong in conflating volatility risk with counterparty risk. The two are actually completely orthogonal -- you can have very little risk of volatility but high counterparty risk, or high/low (and high/high low/low for that matter).

The key is to distinguish from the risk inherent in the fulfillment of the contract and the risk that the contract will not be carried out.

Look, it's nice when you are well versed enough in a language to not have to lookup method/function names, nor their arguments. But let's face it, it's hardly the mark of an amazing programmer to have a photographic memory.

On the other hand, a guy that says "oh yeah, I should use one of those STL things that let's you look up values by keys" and has to go fishing for std::map doesn't inspire one with confidence. Repeated exposure to (and use) of a language (and/or framework API) naturally causes considerable memorization to happen even without trying to memorize everything. It's an indicia of experience.

Of course, experience isn't skill and so forth. But having a decent working recall of how to get around seems to me a necessary but not sufficient condition of being an "amazing programmer".

[ Full disclosure: I do, in fact, use an IDE for most C/C++ development. ]

FTFY.

Until the boffins at Intel or ARM create a processor whose machine code is JavaScript, you need bullshit quotes around that 'native' claim.

If you want to make the argument that you don't need native code, that's your prerogative. Depending on the use case and requirements, you will no doubt be correct in a large number of cases -- I don't need a native slashdot app, the HTML version is quite sufficient.

But why in God's name do you need to make a preposterous claim like that? What does that buy you?

Bluetooth keyboard? Almost all platforms support it.

Also, I have this vague terror about any sort of important scripting being done from a device with a sub-5" screen. No offense, but I kind of hope that I don't rely on those services ...

"They don't want to endanger the space-farers or the public, and they can't let the industry get started and then have a Titanic-like scenario that puts an end to it all in the eyes of the public."

Puts and end to all of what? Did we stop ocean-faring after Titanic sunk? What is this guy talking about?

Won't help -- the time it takes to try even thousands or millions of decoy keys is trivial (10^6) complexity.

Security through obscurity of the key storage is papering over the fundamental new assumption that RAM Is accessible to a hostile attacker.

No affiliation, but this sounds like a good reason to move to TRESOR-like implementation in which the AES key is kept in hardware registers that are cleared when you go to S3 and on each reset. It's still vulnerable to anyone that gets root access to your OS, but a cold-reboot attack or a DMA attack on the RAM are not going to work -- so that's some forward progress.

Anyone want to take a stab at porting it to TrueCrypt?

Yeah, why should we trust judges when they explain the law? And what's with these pesky biologists trying to explain biology. And don't get me started on mathematicians telling me that I'm doing matrix multiplication wrong! And historians! Talking about history!

Anti-intellectualism at its finest.

You should really start referring to them as "my opinions of how the 4th amendment works". As far as I can tell, not a single practicing or academic legal authority has ever endorsed this construction.

Now, of course, it's a free country -- you can represent your views however you want. But you don't get to pick your facts and you definitely don't get to reinterpret the law just because you don't like it (hellooo segregationists).

The problem is that many of us rent or otherwise occupy a dwelling for a much shorter period of time than the lifetime of these bulbs. So either you've got to convince my landlord (or the next tenant) to pay me for the remaining lifetime on the LED bulb (how can they measure that, we don't know) or else I've got to swallow it when I move somewhere else.

The time horizon of these things is just way too long to make sense except for those already fairly stable in where they are.