Just out of curiosity, how's Eclipse's C++ support these days. Last I tried it was pretty sucky, but that was a few years ago.

That is the biggest thing. I know less than a hand full of people that need MS Office. That is mostly Excel. One of them would have a hard time without word (she could do it, but it would be a bit painful), and none of them need PowerPoint.

I'm somewhat of an unusual case, but I'm almost the opposite. I almost never use a word processor and rarely use a spreadsheet, but I do sometimes use PowerPoint.

And the problem there is that Impress sucks donkey balls. I don't know of any presentation software that is even in the same ballpark except maybe Keynote -- and as I don't have a Mac, that's a non-starter.

Tie all those to a minor OS (instead of a dominant OS), and they won't be billion dollar businesses.

You definitely have a point.

But at the same time, suppose that it suddenly became more critical to support other platforms. It would be really interesting to see how well they did. For instance, you point out that the Mac version of MSO is behind the Windows one. But what happens if the higher-ups say "hey, this is now a priority!" as opposed to now, where I suspect it's more "hey you guys, work on this and we'll try to get some extra money." I suspect you could see that gap disappear.

I suspect you'd see at least quite a few people still interested in Office (PowerPoint at least is unrivaled except for Keynote which has limited platform issues), and I strongly suspect you'd see at least fairly strong interest in Visual Studio among devs who work on other platforms.

I think it's your middle group (e.g. IIS and Windows Server) that's in the most danger. But still, the good news there is that I'm not sure how much interaction there is between Windows and desktop systems on the decline and servers on the decline.

It seems to me like there'd be a lot of tossups here.

IMO the real problem with Apple's desktop hardware is they don't offer a mid-range tower, so you lose out on a lot of flexibility. They've only got the all-in-one iMac, the ultra-expensive Mac Pro, and the Mini which is basically just a laptop without a monitor.

Even if there was something like the Mac Pro that used consumer hardware (e.g. not Xeons!) but was a bit more expensive than what you'd pay from most other vendors or if you built it itself it would at least be worth a look. But at the moment for me, an Apple desktop is just out of the question.

Which is easier to do in your head? At least for me, multiplying is much more likely to be easy to both estimate and arrive at an actual answer.

(Remember, unit conversions are never hard; this whole story is about frequent but small annoyances.)

Texas, and at some point soon, Nevada. Also apparently there's 80 in at least Utah.

The melting point matters to people who live in regions of the Earth where temperatures goes below 0C.
I leave it as on excercise to the reader why +1C is hugely different compared to -1C when going out.

As someone who is very familiar with such weather, I don't think I agree. For instance, take perhaps the most important reason that freezing/not freezing makes a difference, at least IMO: you're driving, and want to know how much you have to worry about ice. If it was the case that, for instance, at +1C you knew you wouldn't have to deal with ice on the roads or something like that vs -1C you do, I would. But it's already so fuzzy already: hyper-local air temperatures, amount of sun, ground temperatures, etc. already mean that 0C doesn't actually tell you a ton. You have to be worried about ice potentially even several degrees warmer than freezing, and you may not have to deal with it (even if there's water) several degrees below. It makes the nice round number 0C way less useful.

On the other hand, L/100km is very useful for figuring what you need to arrive at a given destination. Driving somewhere that's 250 km away in a car that uses 10 L/100 km? You'll need 25 L of fuel for that. With gas in the US at roughly $1/L, you'll know it costs $25 to drive there. You release about 2.5 kg of CO2 per L, so the trip dumps about 60 some kg of CO2 in the gas powered car.

But what decision are you going to make based on that figure? Are you really going to choose whether to take a trip based on the gas cost? (I suppose you might want to figure drive cost vs air or something like that.)

Whereas "I estimate I have 3 gallons left in my tank, can I make it 90 miles" is something I compute many times each longish car trip.

Interestingly, I prefer cm/mm over inches and feet over meters; have no preference with gallons/qts vs. liters; prefer ml over fluid ounces, prefer grams over ounces but pounds over kg. How is THAT for confused.

Heh, I'm similar except with a different list: inches over cm and feet over meters; liters over gallons/qts/fl oz; indifferent about pounds vs kg; prefer Fahrenheit over Celsius.

redneck detected. mpg isn't clear about efficiency

But who cares? How often do you really compare about the efficiency? About the only time I would was if I was contemplating buying a new car and trying to figure out if it would be worth it. And how often do you do that, vs when driving around figuring out how far you can go on the current tank? Because for me it's a ratio of about 1 time to 100...

Abbreviation from SystÃfme international des unitÃf©s (I may be missing an 'e' or an accent, somewhere)

Don't worry, so is Slashcode.

No, zero Celsius is "too damn cold", zero Fahrenheit is HOLY SHIT HOW CAN A HUMAN BEING LIVE HERE.

I'm biased because of my location (Wisconsin), but part of the reason that I like that quote is because it matches my own temperature opinions so closely. By my scale, down to about 10F it's really easy to dress appropriately and be reasonably comfortable, and single digits Fahrenheit is where I really start wishing it was warmer. On the other end of the spectrum, I can deal reasonably well up to mid-80s, and 90 is where I really start wishing it was cooler. And ~60 is about perfect. :-)

Really though, places that regularly see 100 F, I have the same reaction to your reaction to 0 F. I really don't understand how someone would voluntarily live in Texas or Arizona or whatever. (And if you say "it's a dry heat" I hope to god someone mods you down as literally flamebait. :-))

Are there many different speed limits in US?

Yeah. 25, 30, 35, 45, 55, 65, and 70 are all extremely common in my experience. 15 and 20 are common for fairly special cases (e.g. school zones or shopping mall access roads), you occasionally see 40 or 50, and other parts of the country have interstate speed limits that are 75 or even 85.

We have a weird situation here in the UK. All fuel is sold by the litre - but no one knows what litres per 100km means or how the cost of a litre of gas will affect them. We all refer to MPG and we know that a gallon is about 4.5x the cost of a litre (yes, our gallons have more litres than yours).

So I tend to be pretty ambivalent about metric vs imperial. I would vaguely prefer if everything switched to metric, but I also don't think that it would make much of a difference at all in day-to-day life. Cooking is about the one time I care about the imperial system. (How many tsp are in a tbsp again?)

But I would posit that one reason that L/100km may not have taken off is because that's a stupid way to measure fuel efficiency.

The times I'd care about my fuel efficiency are (1) when I'm choosing whether to buy a new car and what to get and (2) when I'm on the road and want to know whether I have enough fuel to get me to some future waypoint of interest (e.g. "can I get past the Mobil stations on the Indiana turnpike that are run by a bunch of greedy opportunist bastards and across the border to Ohio where prices drop substantially, or should I fill up now?"). Volume-per-distance is the better way to measure efficiency for the first purpose, as if one car's L/100km measure is twice another's then it'll make me pay twice as much for gas.

But for the second purpose you really want distance-per-volume, so you can look at your gas gauge, estimate how many gallons are left, then multiply. Easy peasy. With volume-per-distance it becomes much more awkward. And #2 happens way more than #1. How often do you buy a new car?

It's 200K to Seattle? We'll, we're averaging 100 kph so we'll be there in a couple of hours.

Actually I think driving is one place where miles actually has a tiny advantage.

Suppose you're going 60 mph. That's a mile each minute, which is really easy to work with. Destination 13 miles away? That's 13 minutes. But if you're going 100 km/h and your destination is 20 km away, now you have to do more actual math: divide 100 [km/h] by 20 [km] to get 5 [1/h], then divide 60 [min/hr] by 5 [1/h] to get 12 [min].

It's not hard, but really nothing in this unit conversion stuff is.

Even at 70 or 75 mph, the 1 mile/minute rule can be adjusted a bit. 13 miles away? Well, that's a bit less than 13 minutes. You do still have to do some calculations to get a more accurate answer, but a rough estimate can be arrived at nearly immediately. And even 60 mph is reasonably common -- many non-interstate highways have a 55mph speed limit, as do most interstates in urban areas. Also even on the interstates, if are going 70 mph but stop for an average of 5 minutes each hour, you get a lot closer to 1 mi/min. (That comes out to 64.1 mph on average.)

OK, I am giving a bit of a strawman. 75mph is 120km/h, which is very nicely 2 km/min, so there you'd just divide the destination distance by 2. At 70mph, you'd divide by 2 then add in a fudge factor, just like I subtract a fudge factor.)

This is made possible of course by the coincidence that 60 mph happens to be in the ballpark of highway speeds and there are 60 minutes in an hour, not because of any actual unit conversions or anything.

(Actually know I wonder what you'd get if you took the average speed limit of the entire interstate system weighted by traffic amount.)