Comment by mikeash | Hacker News Reader

Comment by mikeash | Hacker News Reader

I thought it was a bit silly until I saw that he's talking about C++.

The smart-everything, RAII-everything approach that most C++ uses these days tends to make everything a little bit slow. In other languages, there's usually a small number of places that are slow, and everything else is inconsequential. In C++ you end up spending a lot of time spread among a million different bits of code twiddling smart pointers and RAIIing things that don't really need it.

I completely lost it at this paragraph, though:

"If you tell yourself, it’s only a malloc, it’s nothing, and you do this often enough, you will end up with 25000 temporary allocs for a single keystroke. They may only take 50ns each, but I type 529 characters per minute."

That works out to 1.1% CPU utilization when typing at full bore. (50ns * 25000 * 529/minute = 0.011.) In the abstract, that's high. But in a practical sense, it's completely irrelevant.

Shaving CPU cycles off the keystroke handler of an app that only uses 1% CPU in the hands of a fast typist is a massive waste of the programmer's time. This paragraph is followed by:

"When these sorts of things are pointed out, people aught to respond by fixing the problem, so that they can deliver a better product. Instead, they typically start arguing with you about why it’s not that big a deal."

Well yes, because it's not that big of a deal. There are better places to spend your time. A product that uses 0.01% CPU while typing at full bore is not noticeably better than one that uses 1% CPU. You won't "deliver a better product" by addressing this, you'll waste a bunch of time you could have spent building a product that's actually better.

If my engineering estimates and mental math are correct (which is always suspect) that software bug you're describing that wastes 1% CPU on my 95 watt TDP desktop times 5 million users working full time jobs (some kind of keyboard driver I suppose) means 5 tons of CO2 emissions per year just from one little software bug.

Obviously most keyboard drivers have a lot more than 5 million full time users but as a crude engineering estimate (again assuming I didn't screw up the math) 1% of wasted CPU for a million users is a ton of CO2 emissions per year.

Little things add up... This is what leads to things like banning transformer based wall warts, each may only waste 10 watts but you eliminate them from the world and you can like, close an entire coal mine due to reduced electrical demand.

A ton of carbon sounds like a whole lot, but according to the EPA (http://www.epa.gov/cleanenergy/energy-resources/refs.html), a typical coal-fueled power plant puts out 3.8 million metric tons of carbon per year. Kinda puts it in perspective.

That's one of the major complaints about consumer-focused environmentalism. Sure, it helps to recycle your trash. But if you don't deal with industrial waste, you're beating a dead horse on the margins and letting all the live ones run by on a giant racetrack.

The wall warts thing mentioned above is another good example of this. You'll save a few watts per person... but the US averages 10,000W/person of energy consumption, so it will barely be noticeable. Moving to more efficient designs is a good thing, to be sure, but it's barely relevant in terms of making things better. You'll never find enough initiatives of that size to make a dent. You need to attack major energy wasters, like cars and air conditioners if you want to accomplish something.

It's why I stopped recycling. The amount I contribute to the waste stream is insignificant. In fact I would wager that the fuel I use to drive to the recycling center more than offsets the environmental benefit. To say nothing of the wasted time, the hassle and low level stress caused by having bags of trash sitting around until there's enough to haul away.

Simpsons did it, of course:

"A half-ton of newspaper and all we get is 75 cents? That won't even cover the gas I used to go to the store to buy the twine to tie up the bundles."

I recycle a lot of stuff, but it's convenient for me, because my trash service picks it up weekly. I think we actually recycle more than we throw away at this point. But without people coming by regularly to pick it up for me, it's hard to see how it would be worthwhile.

The average American emits about 17 tons of CO2 per year, so you're saving less than one month's worth of emissions for the programmer, and that's assuming their program is used by millions of users on incredibly inefficient computers. With more realistic assumptions, the CO2 emitted by the programmer running his computer working on these optimizations, and getting to/from the office, probably greatly outweighs the benefit gained from reducing others' emissions.

This conversation happens a lot. I point out that some optimization isn't worthwhile, and I get a response along the lines of "every little bit counts." But you're kind of missing the point. Every little bit counts, but there are opportunity costs. It's not a choice between optimizing your keydown routines or doing nothing. It's a choice between optimizing your keydown routines, or spending that same time optimizing something that actually matters.

I'm not arguing for slower, less efficient programs. I'm arguing that you'll make faster, more efficient programs if you avoid optimizing things that don't make a big difference.

Reply on news.ycombinator.com