The other day I noticed (again) the shadows macOS draws around focused windows, and I thought to myself "if I could turn that off and get .003% battery life back, I would 100% do it".
I would never have thought that if I hadn't used Linux (or another FOSS *NIX), where you can customize things down to modifying the code yourself. Hell I had a window manager for a while where the way to customize it was to edit the code and recompile it.
That Alan Kay quote "...because people don’t understand what computing is about, they think they have it in the iPhone, and that illusion is as bad as the illusion that Guitar Hero is the same as a real guitar" really summarizes the thing best I think, but the whole interview [0] really fills out the sentiment. People may or may not think they're "computing" or whatever, but there's a kind of tug of war between mindlessly shuffling around spreadsheets and slide decks and trying to create something that meaningfully improves peoples' lives.
Blame whatever you want for this:
- we don't factor in pollution externalities into costs, so Electron (and web) apps are commercially viable despite their high energy use
- tech oligopolies do illegal things to squelch competition, so competing organizations are in a race to the bottom re: efficiency, dark patterns, addictive features (building "engagement"), etc.; see Writely (Google Docs) vs. MS Office, etc.
- in capitalism you solve everything with money, and money washes away all sins, so all other concerns (efficiency, civil rights, etc.) are mooted
- there's essentially no useful social safety net in the US, and one of the consequences of this is that we need full employment programs for the middle class, who subconsciously know this and put pressures on the market not to obsolete them (no need for humans to captain Excel around anymore), but to let them burrow more deeply into enterprises (make Excel more and more powerful and complicated such that only humans trained over years can use it). This is analogous to the dynamic medical companies face: treatments make you rich, cures make you bankrupt.
> we don't factor in pollution externalities into costs
Suppose this is solved and it turns out your local electricity price literally doubles from $0.10/kW-hr to $0.20. Suppose your Electron app adds 10W to what you'd otherwise use, for 10 hours/day. Your power bill just went up by 1 cent per day, almost unnoticeable to an individual.
I actually agree that to understand why the software ecosystem is screwed up in the ways it is, we should be looking for the economic/legal/otherwise systemic incentives driving the patterns. But that's different from "I can think of a way the world is bad according to my politics, that must be it."
> But that's different from "I can think of a way the world is bad according to my politics, that must be it."
It's not productive to dismiss someone's argument based on their politics, unless I guess you think I'm an extremist (I'm not). Politics is a core part of how we understand our world, our societies, our institutions, and our cultures. I'm open to having my politics changed and I love to discuss them, but I won't be dismissed because I believe climate change is a huge threat, or for my other economic political views.
---
Continuing in good faith, raw kilowatt hours diminish what a potential pollution cost externality regime might do, and I wouldn't say that a system based on them succeeds in factoring in pollution externalities.
CO2 is more brass tacks. The global CO2 budget is ~40 gigatons of CO2 a year [0]. There's ~8 billion people around, making the per-capita carbon budget 5 tons of carbon/year. You could tax this in different ways:
- Software producers have their software measured for energy use, paying a tax per N users per Wh
- Individuals are allotted 5 tons of carbon/year for free; can pay $2/lb from 5-6 tons, $4/lb from 6-7 tons, etc.
- Industrial sector participants are allotted 50 tons of carbon/year for free, etc. etc.
Don't forget about scale. You need to multiply the numbers by 100s of millions of users, times the amount of time these apps are running (3651010), times the amount of apps (idk, per user like 10, 20, 40?). Plus of course the overhead from extra hops over the internet because apps are being request-happy...
The proposal was that if dirty power were more realistically priced, then people would not use Electron apps.
Code that's a hot spot in a data center already does get more attention to its efficiency, and the data centers already tend to be sited for clean power. The one correct bit here is that pricing the externalities would increase that trend of data-center optimization.
(I guess I should add that I support carbon pricing.)
I would never have thought that if I hadn't used Linux (or another FOSS *NIX), where you can customize things down to modifying the code yourself. Hell I had a window manager for a while where the way to customize it was to edit the code and recompile it.
That Alan Kay quote "...because people don’t understand what computing is about, they think they have it in the iPhone, and that illusion is as bad as the illusion that Guitar Hero is the same as a real guitar" really summarizes the thing best I think, but the whole interview [0] really fills out the sentiment. People may or may not think they're "computing" or whatever, but there's a kind of tug of war between mindlessly shuffling around spreadsheets and slide decks and trying to create something that meaningfully improves peoples' lives.
Blame whatever you want for this:
- we don't factor in pollution externalities into costs, so Electron (and web) apps are commercially viable despite their high energy use
- tech oligopolies do illegal things to squelch competition, so competing organizations are in a race to the bottom re: efficiency, dark patterns, addictive features (building "engagement"), etc.; see Writely (Google Docs) vs. MS Office, etc.
- in capitalism you solve everything with money, and money washes away all sins, so all other concerns (efficiency, civil rights, etc.) are mooted
- there's essentially no useful social safety net in the US, and one of the consequences of this is that we need full employment programs for the middle class, who subconsciously know this and put pressures on the market not to obsolete them (no need for humans to captain Excel around anymore), but to let them burrow more deeply into enterprises (make Excel more and more powerful and complicated such that only humans trained over years can use it). This is analogous to the dynamic medical companies face: treatments make you rich, cures make you bankrupt.
[0]: https://www.fastcompany.com/40435064/what-alan-kay-thinks-ab...