Love your explanation of why engineering problems are hard (everything is a compromise of something else, nothing is as simple as it first seems), although I disagree with you about nuclear power. Why should "how much money it makes" be the ruling metric? That's extremely foolish. Despite the high-profile cases, nuclear power is actually one of (if not the) safest forms of power generation. We are ruining people's health and the environment by using things like coal, so we need an alternative. So, nuclear power doesn't make lots of money - so what? If that's all we are measuring things by, then it explains why so many things are screwed up. Apply the same engineering thinking you explained to the performance metrics question: any single-metric performance measurement will be wrong ("good" overall is measured by a number of competing and sometimes conflicting factors; so, in your case "profitability" is a poor reason to say "nuclear isn't a solution").

The prevention and clean-up do need to be factored into the use of nuclear power, but we also need to drag the technology forward to safer designs, not keep limiting it to unsafe, inefficient forms that haven't changed in half a century.

Not necessarily. Time won't solve basic physical limits. Chemical batteries, as most will know, have very limited lifetimes. The RTG on Voyager 1 has been going for more than 37 years. If you rule out radioactivity or nuclear power, then your only options in space are chemical or something like solar. Solar has problems, as Philae has demonstrated. The issue with chemical is that there are hard limits on how much energy you can store in the bonds between atoms - even if we invent a wonderful new rocket fuel or battery type, the maximum limits can still be worked out and they will never exceed that (there's a reason why we use ion engines for space probes, and it has to do with "mass you have to carry" and "how much it can change your speed"). "More technology" will never overcome these problems, unless you come up with something really exotic (like zero-point energy). One that is easy to understand is solar on Earth: we can make it more and more efficient, but we can never exceed 1kW/m^2, as this is the total amount of solar radiation reaching the surface (and, I don't think we've got better than about 30% efficiency). It doesn't matter how wonderful your technology gets, it can never beat basic physics.

The only "high-yield, low launch risk" technology I could think of would be fusion (as deuterium isn't radioactive), but we are yet to get that viable. Apart from that, you're dreaming of magic, no matter how much time you wait.

So, back in the early 90s when we Amiga fans were thinking that the Amiga was the machine of the future, this wasn't really what we would have expected - that one of the last serious uses of one would be controlling an AC system... Well, at least while it controls their AC, they'd be able to still properly multitask and play MOD music and have several layers of side-scrolling beauty, haha. Times have changed.

HP's warranty product support line was so bad that the guys in the shop I used to work at would come up with creative ways to not be the poor sod who had to call them (it was perhaps the worst punishment that could be dealt out). Let's just say that 30-40 minutes and then getting an actual Tier 2 person was surprisingly quick. My only thought on how their process could be so bad is that they did it deliberately so customers would give up instead of getting a warranty replacement - no need to send a new replacement? Profit! (Excepting the fact that I will probably never buy an HP product ever again, and won't recommend them to anyone...)

I disagree. Yes, follow the script as far as checking that you have tried those things (e.g. "Yes, I have already tried rebooting"), but if their script starts wasting time and isn't relevant, it's time to go off-script. If they are not good enough to recognise when the script isn't relevant, then get it escalated (aside from the fact that IMHO such a person should never have a tech support job in the first place). If by follow the script, it's a simple checklist of "Have you tried...? Yes", then fine - that won't take long and doesn't waste either of our time, really. But, if it's going to take 20 minutes to go through something that I know isn't going to solve my problem, or do things I have already done, then the process needs derailing (because, ultimately, the process is broken if it can't deal with recognising what is and is not relevant). I have dealt with both kinds of Tier 1 support, and if they are the "I am going to force you to waste both of our time to do things that have nothing to do with this problem" type, then you politely ask to escalate (or less politely if they refuse).

ISPs that require you to use a particular router? Wow, glad I'm not in America. Seriously, if someone requires that you use specific hardware, and that hardware is rubbish, you request that they let you use something better. If they decline, take your business elsewhere. Thankfully, where I am, the (often) cheapest ISP is also one of the few who has a local support desk. This is a nice combination, but you might not get so lucky. What kind of ISP would have those insane rules though?

This is a pretty significant point: you do get some truly awful callers (this can be because they are stressed and have a deadline they can't meet because of this problem), so a caller being nice is a refreshing change. But, I have a pretty low tolerance for a support desk tech who doesn't listen and insists on following a script when it's obviously not relevant, so there is a point where they can wear down that "social contract" of being polite. If I pick up that someone is out of their depth, I'll usually ask to be escalated or similar (politely, as long as they actually do).

Also, if a call is taking an hour, then either it's a hairy awful problem, or the tech isn't competent. Most tech issues shouldn't go beyond about 20 minutes unless it's a really difficult one. After that, you are probably better served by sending someone out onsite than trying to stay on the phone while the tech struggles for ideas (if this is possible - I did once have one of these where the customer was in the next country, and that was nasty).

So, it's a case of the stupid people ruining it for the rest of us? I'm guessing you were not in a position where you ever talked to anyone who had even the slightest clue, otherwise you wouldn't be saying this. I have had to be a part of both ends, and I can tell you that I pick up very quickly whether the other person (caller or tech) is a moron or knows what they are doing. Two actually tech-savvy people can usually pick each other out, and I've had some great calls talking to someone who has a clue, but one or both of us is a bit stumped by this particular problem. And, it's a whole lot faster than if you get the usual T1 monkey with a script. If you don't have the people and/or technical skills to pick up the different types of users, then you are probably in the wrong job (sorry for the insult). As someone who really does know what I'm doing, I don't mind if the first question is "Have you tried rebooting it?", seeing as, just as you noted, sometimes we all forget something obvious. But, if I answer "Yes, have rebooted it twice, and the problem is obviously not being caused by that, as it's actually xyz", and they reply "Can you please reboot it again?", my reaction is usually not a particularly happy one (unless of course it's a "I've just made a change, can you try now?").

TL;DR: a good tech support person can tell the difference between someone who thinks they know what they are doing and one who actually does, and responds accordingly. Yes, they might cover the basics just to check, but they're not going to infuriate that person with a dumb script.

So my advice is to be polite but be clear exactly what you have done and exactly what the problems/symptoms are (being honest with yourself on your own level of knowledge, as we tech-savvy people tend to somewhat over-estimate our abilities in areas that are not our primary domain), and listen to what the tech has to say. If they say something stupid, politely make it clear why that isn't relevant. If they keep trying to push a script that clearly isn't anything to do with the problem, then ask to get escalated. Make it clear why you need to be escalated without being insulting or rude. At this point, they should go off-script to actually help, or escalate you. If they don't, repeat the "Please escalate me" (or "May I please speak to your manager") line, with increasing force until you get heard. As far as I'm concerned, if a tech can't tell that they are genuinely outdone by the caller and refuses to pass you on to someone who might actually be able to help, then the requirement to stay calm and polite fades. I wouldn't recommend outright insults and swearing at someone, but there is a point where I've had techs be stubborn enough that they are just wasting my time and theirs, and that's not acceptable behaviour for someone paid to help the customers.

But again, to temper this, don't be so arrogant as a caller that you won't listen to actual advice. Remember, you called them for help.

I love tech support calls (both receiving and making) where the person on the other end of the line picks up that they don't need to dumb-down the conversation, and can move more quickly through the steps, or completely skip ones that aren't relevant. "Can you do a traceroute to 1.2.3.4? Ok, that drops after the second step?" is so much nicer a conversation than having someone tell me "Please click on your start button, then go to Run. Now type see-emm-dee and press Enter. Have you done that?" "Yes, of course I've done that; let me guess, you want me to try pinging the gateway's IP address? I told you, I already tried that before I called you, and while I was waiting for you I just tried a traceroute. What do you mean you don't understand what I just said?"

The thing is, and this is my biggest gripe with tech support, that the tech support person at a minimum should be able to tell the difference between a problem on the computer and a problem with the router - simple "identify and isolate the problem" stuff - especially when the person they are talking to can give them intelligent feedback (or, tell them clearly where the problem is or is not). If someone in a call centre is unable to recognise troubleshooting ability in the person they are talking to - and have the humility to pick up "This person is actually better than me, I'm out of my depth and need to escalate" - then they shouldn't be doing that kind of work. Just because there was an obscure case where clearing the recycle bin fixed it doesn't mean that it's applicable in every case.

Being able to tell when the customer is so frustrated that you are about to lose said customer, and then escalating them appropriately really is needed - and isn't that hard! Just don't have some moronic rule about "You lose points if you escalate" or "You shouldn't bother the T2 techs". If it's needed, make it happen.

I once had to deal with HP's faulty hardware returns line, and that was brutally painful. First, you wait maybe an hour, only to be talking to someone with a distinctly middling command of English and a script that they will not ever deviate from ("No, I haven't tried restarting it, because the power supply is actually dead. No, I can't turn it on, I have already tried that. No, I'm a computer tech, I know when something is dead. No, argh!"). The second time I called them, I started it with "I am a professional computer technician, and before you start your script, please be aware that I have already tried all the basic troubleshooting on your script. I just need this part replaced." to which they responded "Have you tried restarting it?" (going back into their script), to which I started having to go to "Shut up and escalate me now!" over and over until they finally did (had to play the angry customer, a lot more rudely than I would have liked, but there was nothing else I could do). Tier 2 there was at least sufficiently competent that we got the issue solved, but still what I would call "barely competent". Having been on the receiving end of tech support calls, I don't like having to be nasty, but if that's the only way to get service, they are doing it wrong...

Or, that there is an entirely different option: we've been getting Genesis wrong due to approaching it with the wrong worldview (The Lost World of Genesis One, John H Walton). I don't hear this being talked about much, even though it really does seem to be a good way to link "believing the Bible" and accepting science (i.e. it makes most of the points of contention disappear), and I think being able to find some common ground between the two sides of the debate would be good. I am a little concerned though that my hope of "let's consider all ideas, even if they disagree with what we previously thought" is too courageous a position to expect, so this very thoughtful book might get missed by those who most need to read it (defenders of Creationism)...

Actually, yes, Starcraft would probably the first of the RTSes to make the list (mostly thanks to South Korea), even though it was far from the first. It has had quite a significant cultural impact (which the others can't really boast). Maybe I'm just a bit nostalgic...

Hmm. That list seems to make sense, though WoW is perhaps the outsider.

But, what about Asteroids, Space Invaders, even Street Fighter? Civilization or Sim City? No love (yet) for strategy games at all, which should probably start with either those or Dune II, or perhaps something from the Command & Conquer series (either C&C1 or Red Alert 1). Although, maybe I'm a bit too much of a geek - quite likely, Age of Empires would beat some of those to the list.

However, these, and most of the other suggestions being made, are more about iconic or revolutionary (started a genre) games, as opposed to simply famous. And, they put the requirement of it influencing outside culture (so recognisable by more than gamers). The Sims, Minecraft, Sonic, and yes, as much as we geeks don't want to admit it, Angry Birds (as derivative as that is - i.e. there's no Scorched Earth, a much earlier artillery game) are all legitimate nominations I think. Perhaps they might have had less ire if they kept to solely classic games for the first round, although I do remember hearing something about an unofficial WoW theme park in China... Interesting that WoW gets in before Warcraft though.

I can see how difficult it would be to be objective on something like this, so I applaud their attempting it, even if I don't entirely agree with the choices. Maybe they included WoW to try and have something relevant to a modern audience? (Although, arguably Minecraft is more recognisable today, if perhaps still a little new. I expect it will get there soon enough though).

Oh, and do the wiring yourself, then get an electrician to just check it. That way, you have control on quality, and you save a lot of money.

Several people have commented on going passive thermal, good insulation, etc, and really, these are the things that you will value long-term (as well as, by the sounds of the type of person you are, the "hackability" - which is where the often-reiterated conduit comments come in, as that really is your best option - well, any easily-accessible cable run system; I suggested false ceilings further up as another alternative).

If going concrete slab foundation (probable nowadays), put a layer of expanded polystyrene under it. This thermally isolates the house from the ground, and means you can do things like passive solar, using that concrete block as your heat storage.

Of course, if you care about the environment, you are better off minimising soil disturbance, so instead of cut-and-fill as is common now, build around the existing landforms and use piling instead of a concrete slab (a piled house has the advantage that it's potentially more flood-resistant, and it gives you a crawl space that will be very useful for running cables in future).

Some of the suggestions are very region-specific; do you need to worry about pipes freezing? Is it dry or wet where you live? How much is heat or cold going to be an issue? Snow? Tornadoes? Rain? Drought? How much does the local climate affect your design decisions? (e.g. Southern California is going to be quite different to Seattle, which will be quite different to Florida - and these are assuming, like others here, that you are in the USA; something like Norway or Fiji or Egypt would be very different again). Design for the local conditions (e.g. don't make a grass lawn if you live in a desert - that's just dumb, and irresponsible). Local building codes, water take laws, height restrictions, material availability, etc matter a lot, as do things like whether your specific plot of land has covenants on it.

Build responsibly: have a think about the ecological and economic impacts of what you are doing. Look at life-cycle analyses for everything, and if you can, try to get the best score on whatever local "green star" certification system is available (not only will this mean you have a more energy-efficient home, you also minimise environmental impact, and usually - somewhat counter-intuitively - your build cost as well). Recycled materials are great (shipping containers, crushed glass and waste paint in concrete - also can look amazing when polished - recycled wood, recycled aggregate/crushed concrete). Depending on where you are, you might have to do some work to find local suppliers and contractors who will work with (or know how to work with) such things, but it is worth it.

As long as it's not a desert, make use of your stormwater. At the very least, go for a low-impact build (green roof - added insulation as well - or raingardens, or any number of other neat solutions), or simply run it to a tank and use it for grey water (watering the garden, flushing toilets) - or even as your primary water source. A water tank above ground level is harder to build, but means it doesn't need a pump (be creative).

If you are going really crazy with your tech ideas, consider crawlspaces. You probably don't need these in every wall (though the fun you could have with secret passageways should not be understated!), but even just under the house or in the ceiling would make accessing conduits etc easier (also, think about the roof structural design, and whether or not you can actually negotiate it without needing to be a contortionist or a midget). A totally bonkers but awesome idea would be a full-height tunnel that runs under the structure as a "backbone" for all your services - if you're feeling really crazy, this could even go to the property boundary, making service relocations a breeze - and, even the option to check the mailbox in the rain! (I once got a tour of a large auditorium that had one of these running under the carpark, and it was pretty awesome - would be especially so if you have kids). You won't need to care about the walls there lined with pipes - in fact, it would be a good setting if you ever choose to make sci-fi fan films!

Solar passive thermal design and good insulation cannot be understated. Think also about humidity control though - if you have a sealed-up house, consider how you are going to avoid it getting damp (unless in a desert). Use ground-loop heat pumps as they're the most efficient (called "geothermal" by some here, though that term makes me think of this kind of thing).

LED lighting, of course, and the suggestions of a separate low-voltage DC circuit for these. One idea I had a while ago was having all the lighting on a (probably ethernet with PoE) system, whereby each individual light can be controlled centrally. Even better is do that with LED clusters that let you control the brightness and colour (mood lighting, disco, or whatever you want it to be). You'd probably need an Arduino or Raspberry Pi or something at each light fitting (or cluster), so it could get expensive. With that, I figured you have touch panels where you would normally put light switches that turn off and on (to a pre-set default) with a single touch, fade brightness with a sliding touch from the bottom or top, or go to a full-house control view on a long touch that gives more specific control (probably a top-down view of the house allowing control of each room). This would be amazing, but probably too much cost and work to be realistic... LED ropes recessed into the ceiling in an extended architrave is a nice way to add ambient light control.

Go solar. I actually wouldn't advise going off-grid though: feed back into the grid instead - this way, you have a backup if your systems aren't cutting it, and any over-provisioning still nets you money. One suggestion I heard was to have the panels facing West not South (or North) - this is less efficient overall, but gets more energy in the high-demand evening (i.e. shifts the supply curve to more closely match the demand curve).

For water heating, I'm not a fan of on-demand gas systems (you get full pressure and good heat, but they tend not to be useful for small amounts of hot water, or any on/off cycling). Heat pump hot water cylinders are the most efficient I know of (short of solar thermal, or maybe a wet-back log burner if you have a free supply of logs, and the space to build it as a proper big boiler). Presumably you could have a ground-loop heat pump hot water system (GLHPHWS?!?), though I haven't seen these myself (and will not be attempting to pronounce that acronym). Actually, a true geothermal heating (i.e. you live on top of an active volcanic field) is probably the most efficient, but that is very location-specific. On that note, if you have that sort of heat source, then you don't need any other energy source; you just have to hope the volcano risk doesn't eventuate in your lifetime.

Design and build something to outlast your lifetime. Most houses now are built with a 50-year lifespan max. I reckon that (especially considering the life-cycle impacts of construction materials), we should be building for 200 to 1000 year lifespans - however, this only makes sense if your structure is somewhere that isn't likely to get bulldozed in the next few decades. Building to reduce maintenance costs is good though (but it does usually push up your capital outlay in the short term).

One thing you could consider is a modular design: either internal walls that can be easily configured, or interconnected "pods" of some kind (shipping containers are good for this kind of thing) - build what you need now, and expand later if you want. I think one of the keys to longevity is having it able to adapt without too much difficulty.

Soil is a great insulator. Use your imagination (also, having your rooftop as a usable space seems to be a much better idea than the pointy, water-impervious space-wasters of the urban sprawl).

Bamboo is an interesting material I have only recently come across: strong (tension and compression), lightweight, grows really fast, environmentally sustainable. Using bamboo laminate, you can build all sorts of things (google it).

You don't know what technologies are going to go extinct and which will fly, so don't spend too much money on cabling etc - but do make it easy to change. I would suggest you install what you need now, plus likely expansion (e.g. put 2 Cat6 where you think you need 1, but don't run fibre and coax and HDMI and string telephone everywhere unless you can see it might be useful - put it in later if and when it becomes a need).

Some have joked about Faraday cages to keep out the NSA, however the reverse is probably more important to consider: anything that blocks RF is going to kill cellphone and wireless signal, which can be a real pain (we live somewhere with patchy cell reception, and it's annoying). Steel framing, reinforcing, or even joinery can do this, as can anything thick enough. If, like us, you don't bother with a landline any more, this is especially worth considering. It's also a lot easier when one router is sufficient (I once helped someone install wireless networking on a house with a concrete-on-corrugated-steel floor between the levels, and nothing got through that (of course, not even cabling, because of the impenetrability of it, and there not being any convenient ducting).

Fruit trees are great: fairly low maintenance, and you get free, fresh food.

In general, think ahead, and ask around (which is exactly what is happening here). Build for maximising future utility. Exceed minimum compliance and go for well-thought-out, simple-yet-easily-modifiable designs (note that the first half of this sentence is almost a direct quote from an engineering review of buildings in the Christchurch, New Zealand earthquake, so it's about more than insulation). Tech dates quickly, so don't spend too much tying yourself down to one thing.

An idea I had a while back was an open-plan ground level with sliding/folding partitions (reasonably heat-proof) that allow you to close down the open area into smaller, more cozy sections. e.g. you can have the lovely big open-plan, light, spacious area in summer (maybe even with a water feature beneath a glass floor or something fun like that), but then can easily close up to have separate rooms that are easily heated in the winter (or when you want some privacy from noisy children).