1. 22

Given increasing evidence that organized human life might not be possible by the end of the century and possibly sooner if feedback loops kick in, how does global warming impact your work? What outlook do you have about the future of technology?

  1. 13

    I’ve only recently calculated that the majority of emissions I generate comes from long-haul air travel. Most of this is personal travel, but occasionally business travel as well. I plan to strongly prioritize conferences in Europe in the future (allowing for rail), and will opt to see talks from remote conferences online instead.

    1. 10

      I live in South Korea. After Fukushima, government decided to phase out nuclear energy. Just after two years of this policy, proportion of coal in energy mix, both in absolute and relative term, skyrocketed. This means there is no hope South Korea will meet its Paris Agreement obligation.

      I do nuclear energy evangelism in my spare time. So far it doesn’t seem effective, which makes me depressed. People seem to care more about irrational fear of nuclear energy than global warming.

      1. 10

        I was a pretty big nuclear energy proponent for a lot of reasons, but seeing a lot of stats on solar in particular in terms of pricing makes me feel like the political uphill battle of defending nuclear doesn’t feel worth it relative to advocating for large scale rollouts of renewables.

        If you can find spaces accepting of nuclear (whatever happened to thorium anyways) that’s great, of course. At this point I no longer try to argue against the 100% renewable positions though. Perhaps with enough willpower they can happen

        1. 2

          I think it’s a bit different in South Korea. South Korea has continued to build nuclear power until recently and there was no cessation at all. South Korean coal consumption increased sharply (more than 10%) in 2017, almost entirely due to nuclear phaseout. I am not especially in favor of nuclear energy, I just think nuclear energy is the most feasible option (yes, more than solar) against coal in South Korea.

      2. 21

        One of the biggest pet peeves[1] I have with the software industry, and something I try hard to be conscious of in my work, is that we have a tendency to build software for our hardware, not the hardware our customers will use, and sacrifice performance for our convenience; this is the old maxim of “CPU time is cheaper than developer time”.

        This is problematic partly because we neglect the needs of many users. We buy nice computers, peripherals, data plans, and home internet access plans because they are our tools; we need them. However, this means that something snappy and responsive for us, which we don’t question during our design and testing, could be very slow or even unusable (in the case of highly restrictive timeouts) for people with less money or simply no particular reason to buy an expensive phone or PC.

        It is also problematic because CPU usage generates heat and uses energy. Modern CPUs are very, very good at “spinning down” via P-states and C-states [2]. I suggest you read the linked blog posts, but the very reductive TL;DR is that CPUs can use less power than their maximum TDP when executing sparse instructions (P-states) and can use even less power during periods in which one or more cores is completely idle by shutting off almost all power to those cores (C-states).

        This means that two developer practices can introduce major energy savings: batching and optimization. Optimization is obvious - if your program requires fewer instructions to run, it will run for a shorter time and will allow the CPU to sleep in a C-state for longer.

        Batching, collecting expensive operations to be run together, can have a similar effect. Rather than running one 50ms operation every 100ms, we could run three 50ms operations every 300ms. The duty cycle is the same (50%), but the CPU is likely to be able to enter a deeper sleep for a longer time during that inactive 150ms than during the three 50ms inactive periods, thus saving power.

        Similar techniques can be applied with great success to memory access, storage access, and especially radio access (WiFi and cellular).

        This is one reason I constantly rail against tools like Electron, which trade off developer efficiency (not having to learn a GUI toolkit, maintain multiple front-end codebases) for extremely terrible performance, huge memory usage, and constant CPU wakeups.

        In short, inefficient programs produce heat and use energy, not to mention causing problems for less wealthy or less technical users of one’s software, and the maxim that “CPU time is cheaper than developer time” is less accurate than many seem to think.

        1: In that it’s probably not that big of a deal compared to other things like Amazon exploiting warehouse workers or people writing racist predictive policing systems, but it annoys me a lot.

        2: https://software.intel.com/en-us/articles/power-management-states-p-states-c-states-and-package-c-states

        1. 7

          Semi-shameless plug: my company builds software to help automate clean energy project development, which at this point, given the rapid decline in price for hardware (solar modules, batteries, etc.), is really the last major hurdle in getting the biggest consumers of energy (commercial and industrial facilities) to transition over.

          On the one hand, you could say that my work will benefit from this impending doom. But really, my team sees this more as a ticking countdown to motivate our progress.

          Also, though I can understand how broadly speaking this may be off-topic, I’d love (perhaps selfishly) to see more energy-tech-related topics on the front page :)

          1. 10

            Given increasing evidence that organized human life might not be possible by the end of the century and possibly sooner

            Becoming desensitized to this sort of absurd alarmism is partially why so many people quickly discard legitimate, science-based warnings about the actual risks of climate change.

            1. 6

              I live in a small island nation, that will continue to see drastic changes due to climate change.

              I haven’t thought much about its impact on my work, but I’ve thought a lot about my work’s impact on it. And more generally, I didn’t know if I could stay in tech at all.

              But, I found a job that would have me work on embedded systems to help reduce supple-chain waste on a large scale, so that seems like a positive use of my existing skills.

              1. 6

                Most of the world is ignoring it or doing so little that the damage will happen. So, I’m mostly ignoring it. I do encourage people to take precautions like not living on the coast. Also, push for reusing and not wasting stuff in general to reduce the negatives which go beyond global warming.

                Although I agree with LeoLamba, I will not situation is opposite of you want to develop secure, open hardware. Both power management and microarchitecture are a patent minefield. Anyone wanting to keep risks minimal will have to use 20+ year old constructions or nodes. Think how efficient your boxes were in 1998. You’ll still need speed which will require SMP and cooling. So, addressing secure, open hardware without being sued to death might require doing more environmental damage.

                I’m still hoping one can port 20+ year old methods directly to 28nm or whatever to still reduce some damage plus get a speed boost. Plus use or license cutting-edge, weird stuff out of CompSci in power management.

                1. 1

                  Since I can’t edit, I think I typed “I will add that my situation is opposite.” The “I will not situation is “ was apparently an epic, auto-fill fail on my new phone. Still figuring this thing out (e.g. turning broken features off).

                2. 6

                  I prefer to hope that we will avoid worst case scenarios such as a complete breakdown of human civilisation, so I’m taking actions based on that assumption.

                  I’d be very pleased if I could apply my skills directly towards mitigating climate change. In all likelihood, we will need technologies for negative emissions, ie ways of sequestering carbon from the atmosphere. There isn’t a viable pathway to keeping warming under 2 degrees without it. Additionally, it appears that the energy storage required to even out the intermittent electricity generation from renewable sources is still an unsolved problem. As such, I’m keeping an eye out for opportunities to put my skills to use in those spaces. Also, since my country (New Zealand) is now planning legislation to reach net zero emissions by 2050, I’m also looking for opportunities around the changes that will be triggered as a result.

                  Aside from that, since I haven’t found a way to contribute more directly, I donate some of my earnings to organisations which are working towards systemic change that will drive emissions down (that includes legislative changes and things like divestment from fossil fuel companies that 350.org campaigns for).

                  1. 11

                    Global warming is important, but realistically we can’t address it until we have regained political stability (and significantly improved on the pre-Trump status quo). Goals for the next 10 years are:

                    1. keep my family safe
                    2. avoid civil war, fascism, etc.
                    3. repair the cultural rift that has people at each others’ throats

                    If I can make impacts on longer term issues during that time, great, but it’s hard to think about right now.

                    1. 5

                      So, essentially you’re saying that since Trump was elected we are collectively incapable of doing anything but running in circles shouting about imminent fascism? Any efforts to improve technology wrt. environmental impact cannot realistically be expected to succeed, because politics? Seems like a terrible, self-defeating attitude to me.

                      1. 1

                        Global warming is not a technological problem insofar as you can’t just invent a widget to solve global warming. Even if your widget is something like “planetary scale air filter”, you will not be able to build or operate it without social/political backing. Also:

                        If I can make impacts on longer term issues during that time, great

                        1. 4

                          It’s not a black and white issue, and it’s not going to be ‘solved’ by one major breakthrough. Their point is just that there’s no reason why the current political situation in the USA needs to bring everything to a halt. If you don’t have the time or headspace to deal with it right now, that’s absolutely okay (what matters is you’re aware of it)! Everyone’s circumstances are different, but collectively, we can’t afford to just put it on hold, and it doesn’t have to be at the expense of other important issues. If anything, I’d hope that it might have the power to bring people closer together (if a threat to humanity can’t do that, what can?).

                      2. -1

                        Yes, you’re right that we can’t solve this problem with technical solutions. Other commenters notwithstanding..

                        1. 2

                          What makes you think that? Climate change is in many ways a technical problem, how do you think we are going to solve it if not by adapting our technology?

                          1. 3

                            Did mere technology or lobbying/sales decide what kinds of power plants will be all over many countries? Did technology itself create the disposable culture that adds to waste or did user demand? Is there a technological solution in sight for the methane emissions from cattle whose beef is in high demand? On other side, would we be storing endless amounts of data in these data centers appearing everywhere if technology didn’t make storage and computing so cheap? And is there a technological solution to avoiding them throwing that stuff away on a regular basis when customers want new stuff or manager want metrics to change? Is there a technological solution to getting people who neither care nor are legally required to care to stop doing damaging behaviors?

                            Sounds more like people-oriented decisions are causing most of the problem. Even if you create a beneficial technology, those people might create new practices or legislation that reduce or counter its benefits. Actually, that’s the default thing they do which they’re doing right now on a massive scale. I think we just got lucky with low-power chips/appliances since longer-lasting batteries and cheaper, utility bills are immediate benefits for most people that just happen to benefit the environment on the side.

                            1. 2

                              It is obviously not merely technology that got us here. But these problems are all about technology on a fundamental level and if we want things to change, we need the tech that makes these changes viable. No point lobbying for an alternative that does not exist.

                              Sounds more like people-oriented decisions are causing most of the problem.

                              Always an interplay of technology- and people-oriented decisions. But changing technology is much easier compared to changing people, which has resulted in utter dystopia many times.

                              Even if you create a beneficial technology, those people might create new practices or legislation that reduce or counter its benefits.

                              Same with well-intentioned legislation. But companies have no intrinsic incentive not to use beneficial technology, only to inflate its impact for marketing purposes (like the faked car emissions). They do have an incentive to game legislation, otherwise there would be no point to that legislation (in general; individual cases might profit from being good examples).

                      3. 3

                        If the sea keeps rising I might need a boat to go to work instead of a bicycle. Realistically though we don’t change much at work except the airco is on more often due to the buildings bad dissipation of heat. I would love to have solar panels on it, but we rent the building.

                        1. 3

                          I’m not too worried about me, its my kids I worry about the most. I have some hope for clean energy to become more cost effective than coal/gas and that will just solve the problem but idk.

                          1. 5

                            Zero impact.

                            1. 6

                              This is pretty far off-topic, and most likely to result in a bunch of yelling back and forth between True Believers.



                              OP didn’t even bother to link to the claimed “increasing evidence”. This is a bait thread. Please don’t.

                              1. 17

                                Shrug. I find the complete lack of political awareness at most of the tech companies I’ve worked at to be rather frustrating and I welcome an occasional thread on these topics in this venue.

                                1. 13

                                  It’s possible that many of your coworkers are more politically aware than they let on, and deliberately avoid talking about it in the workplace in order to avoid conflict with people who they need to work with in order to continue earning money.

                                  1. 1

                                    All work is political. “Jesus take the wheel” for your impact on the world through your employment decisions is nihilistic.

                                    1. 8

                                      Not trumpeting all your political views in the workplace does not mean completely ignoring political incentives for employment or other decisions. I’m not sure what made you think GP is advocating that.

                                2. 3

                                  Obviously “off-topic-ness” is subjective, but so far your prediction re: yelling back and forth hasn’t happened. Perhaps your mental model needs updating… maybe your colleagues are better equipped to discuss broad topics politely than you previously imagined?

                                  1. 4

                                    Obviously “off-topic-ness” is subjective, but so far your prediction re: yelling back and forth hasn’t happened.

                                    Probably because everyone on this site is good and right-thinking — or knows well enough to keep his head down and his mouth shut.

                                    (Which has nothing to do with the truth of either side’s beliefs; regardless of truth, why cause trouble for no gain?)

                                    1. 5

                                      To me, the people on this site definitely handle these discussions better. Hard to say how much better given that’s subjective. Let’s try for objective criteria: there’s less flame wars, more people sticking to the facts as they see them vs comments that re pure noise, and moderation techniques usually reduce the worst stuff without censorship of erasing civil dissenters. If those metrics are sound, then Lobsters community are objectively better at political discussions than many sites.

                                    1. 5

                                      These all seem to say one thing: climate change is going to be worse faster than some other prediction said. But that does not even remotely address your claim that “organized human life might not be possible by the end of the century and possibly sooner”. What on earth makes you think you know anything about what conditions humans need to organize?

                                      1. 1

                                        This is a good point. I guess my “evidence” would be past civilization collapse as a result of environmental destruction like what happened on Easter Island.

                                  2. 3

                                    Don’t care or think about it.

                                    1. 3

                                      I see it as the final verdict. The world of 2118 will either be substantially better or catastrophically worse than the one we’re in now. Either we’ll have eliminated most forms of economic scarcity– being “poor” will mean flying to Antarctica or the Moon in the off-season– or we’ll be in some degraded, nearly apocalyptic state. There’s no middle ground.

                                      We have to overcome corporate capitalism– artificial scarcity, racism and sexism, the valuation at zero of animal welfare and the environment, the massive waste of time that is “work” for most people, and political corruption– and we don’t have a lot of time in which to do it. Gradualism seemed fine (ignoring the human costs of inequality) 100 years ago, but now we have to act. We can vote out a bad president; we have much less control over the global climate. It’s actually not temperatures but wet bulb temperatures that’ll kill us: if those get to 35°C– a typical 35°C summer day has WBT of around 25°C– then we are in bad shape. And once that kind of weather becomes possible– right now it virtually never happens (Persian Gulf an exception) because of thunderstorms, but those derive from temperature differentials rather than absolutes– it will become possible in most of the inhabited world and very few locations will be safe from it.

                                      1. 3

                                        Yes, I think of it as humanity’s greatest test. Either we fail to come together on the issue, and fail as a species (because the fact we’re in this situation at all is easily humanity’s greatest failure), or we succeed and overcome the huge challenges in front of us and see incredible progress. It feels like it’s teetering on a knife edge now, but we can’t let ourselves become too cynical about it.

                                        1. 2

                                          I believe 2100 will prove the Great Filter hypothesis of the Fermi Paradox

                                        2. 1

                                          It doesn’t at all. And this is grossly off-topic.

                                          1. 2