Amazon's offering is substantially less flexible than that of existing players (shapeways is the name that comes to mind; but there are others), who already accept basically any STL that isn't horribly munged in some way and spit the result out in a number of different materials.

You still have to model the part, or buy a (currently rather expensive) 3d scanner to do it; but if you are willing to put on your CAD hat, you could have the part by next week, just not from Amazon. I wonder if they are just moving slowly, or worried about the copy cops coming after them once people start knocking off action figures or something...

It lacks the sci-fi appeal of pure printing; but there are a variety of techniques that use the 3d printed part as the first step and then subject it to additional treatment steps in order to make up for those sorts of deficiencies.

As long as the subsequent processing steps don't change the dimensions(or change them in predictable ways that you can compensate for) you can get away with whatever tempering, annealing, and so on your application requires.

Depends on what you pay.

A poorly calibrated fused filament unit will produce stringy junk that delaminates if you look at it funny. A well calibrated one will achieve something reasonably close to what the plastic it is using is actually capable of. Outside the cheap seats, you can print all kinds of things(especially if you count parts that require one or more additional processing steps as '3d printed'. Printing wax, for example, is pretty undemanding, and allows you to do lost-wax casts of more or less any shape that will cast properly, without needing a printer that can sinter or melt metals. Some of the techniques for producing ceramics are in the same vein, the printer just needs to tack the ceramic material together long enough for firing, which takes care of the mechanical properties.)

The one thing that is (relatively) easy with injection molding that 3d printing (to my knowledge) isn't so hot for is overmolds. When injection molding you can use insert molding or multi-shot systems to achieve the (enormously common and fairly popular) combination of a rigid plastic structure with an elastomeric surface treatment for grip or aesthetic reasons. For prototyping purposes you can get paint-like coatings that emulate elastomeric overmolds that you can brush on to 3d printed parts; but the quality isn't as good and production takes longer.

Are you honestly equating process separation in multi-user OSes with rootkits?

This EA: you only get StarForce if you pre-order at select retail partners or buy the launch-day DLC...

Given that this is EA we are talking about, I can definitely believe that they'd somehow manage to be paranoid about 'piracy' of a game they are giving away. However, since it's also an older game(pre "Origin" store/client/pox-on-humanity and originally distributed largely on retail disks) and being given away it would be unsurprising if as little effort as possible was put into modifications for the new distribution.

Does anybody know how deeply baked-in SecuROM has to be? Would the developer/publisher have a 'clean' version that is then put through some sort of SecuROM conversion step, or would you have to go further back, and deeper, into the development process to cleanly rip it out?

I'm baffled at why including it would be worth much (especially if the license agreement involves any sort of volume-based payment, which would likely wipe out any minor benefits in audience tracking); but if it is sufficiently difficult to rip out then it would be understandable why EA wouldn't bother doing so(aside from just being evil).

But admitting 'neurological changes' is tantamount to doubting free will, and we just can't have that! Despite any and all evidence to the contrary, it simply must be true that a 'will' or 'self control' exists independent of any squishy brainial biology, yet somehow capable of controlling its function. Never you mind that this makes little sense, or that fiddling with self control through experimental manipulation is practically a psych research hobby, this hypothesis is simply too intuitively attractive to deny!

Outside of DEA flunkies and hardcore suffering enthusiasts, I don't think that there's much support for skipping opiates(indeed, it is commonly held that pain is under-treated); but there is an awareness that prescription opiates are a fairly common introduction to opiate dependency, especially in populations that would otherwise have few introductions to them.

Unfortunately, we barely know how pain works, and really don't have many alternatives to work with. The painkillers that aren't addictive are mostly OTC junk that pain barely notices, and the ones that actually work are typically close relatives of quite addictive compounds. At least the pillheads get their fix manufactured under FDA quality control rules, which makes them safer than the junkies.

I vaguely remember some talk about an emulator at one point; but aside from that the two OSes have essentially zero in common. WebOS was (in my opinion) sadly underrated and died tragically young (I wouldn't be surprised if the situation has improved markedly; but back when 'Android tablet' meant 'Motorola Xoom running 3.0' it wasn't even fair how superior webOS was... Now that LG has it, it's probably gone to shit.); but it had absolutely no relation to palmOS, other than organizational.

The 'conduit' synchronization concept was pretty good as well (in an environment where 'eh, it's a computer, just give it TCP/IP and call it a day.' was not yet practical). The actual sync client, at least for Windows, was a total piece of shit; but conceptually the 'conduits' model was about the nicest flavor of PDA synchronization available before the rise of handhelds with their own data connections. PalmOS never handled those particularly neatly.

If you and everybody else responsible for the code running on the device aren't sloppy programmers, perhaps...

Even if your code is perfect, you still run the risk of having the other guy's program start scribbling over yours unless you feel like re-implementing absolutely everything whose behavior you don't entirely trust.

If it's just a screen connection issue, ADB/fastboot should be enough to wipe any internal storage good and hard.

If it's more broken than that, you'll have to go inside. If something has come loose that you can put back into place, you win. Otherwise, you can either pray for a friendly JTAG connection or physically destroy the flash chips.

I can only assume that 'failed to uncheck the checkbox' is exactly the sort of mutual agreement that contract law enjoys talking about the importance of, if not actually acting on it...

That would be nice. Unfortunately, while others can provide tips, only we can compromise our principles...