Peace, On Mon, Nov 22, 2021 at 11:31 PM Brian E Carpenter <brian.e.carpenter@xxxxxxxxx> wrote: >> draft-petrescu-v6ops-ipv6-power-ipv4? >> The methodology raises questions about whether the results are reliable >> enough to be quoted. > Regardless of that, it's far from clear [..] how [those results] relate to > total power consumption That was never the point of the draft in question. Alexandre Petrescu tried to figure out if IPv6 was using more power precisely on smartphones, as part of the broad IPv6 adoption rate topic, to figure out if end users could be switching off IPv6 on the smartphone because IPv6 empties the battery.[1] >From purely the sustainability point even if we still were to see a 5% difference between IPv4 and IPv6, that wouldn't have really mattered because the core idea of sustainability is not to reduce total consumption to a halt but to: - replace non-renewable energy sources and materials with renewable ones (for a specific definition of "renewable"); - eliminate physical waste (for a specific definition of "physical"). The author of the TechRadar article doesn't seem to be actually aware of the sustainability and climate destabilization topic (which could be explained by his conflict of interest, as pointed before on the thread). And I would argue that draft-petrescu-v6ops-ipv6-power-ipv4 could not be reworked to a sustainability draft at all because it basically states that IPv6 uses more power than IPv4: - on a particular device; - using an OS based on an outdated by now Android version with unknown proprietary modifications and optimizations by a vendor from a country where the IPv6 adoption rate was below 2% at the moment;[2] - using a specific transmission technology (4G radio); - as measured by a particular, presumably proprietary, tool; - under other particular circumstances; - back in 2017. Even if we take a smartphone from the same vendor today the results could be very different because China has recently jumped to no less than 15% v6 adoption rate[3] and has a well-drafted plan to be IPv6 single stack by 2030.[4] This is just an anecdotal evidence, a beginning of a dialogue that didn't seem to continue ever since, and in no way a part of the climate destabilization topic even by the intention of its authors. Starting an Internet sustainability research with this isn't a good idea IMO. Ref. [1] https://mailarchive.ietf.org/arch/msg/v6ops/42MVZYPwzLV4xiDs8rpW54FOyog/ [2] https://blog.apnic.net/2017/07/11/towards-fully-connected-ipv6-network-china/ [3] https://labs.apnic.net/dists/v6dcc.html [4] https://www.theregister.com/2021/07/26/china_single_stack_ipv6_notice/ -- Töma
