To play Devils' advoate (although I think this is an excellent essay) I think the point of the toaster project was to question the amount of effort that goes into producing cheap toasters, which are of relatively narrow utility. There's a lot you can do with a pencil, as with any general purpose tool: the production of pencils (or keyboards, or CPUs) is an example of capitalism at its finest.
A toaster, on the other hand, sacrifices flexibility for efficiency. It's faster and uses less energy than opening up your stove and making toast on the broiler (a Good Thing), but that's all you can do with it - toast bread products. If you want to have egg with your toast, it's no help at all (unlike your stove, which allows you to cook things in many different ways). If you follow the toaster approach, you end up with a kitchen full of one-purpose appliances which are individually excellent but ultimately cater to peoples' inability to cook. At some point it's worth asking the question of whether having multiple specialized appliances is really more efficient than having one or two that can be turned to many different ones. If you look in the kitchen of a diner or restaurant, there are relatively few fancy appliances but the cooks really know what they're doing.
There's room for both approaches. Programming everything from scratch without libraries would be terribly wasteful. On the other hand, you probably wouldn't respect a programmer who relied solely on library functions and was unable to derive any algorithms of their own, no?
That's also solved by the collective intelligence. Is it better to have simple devices which do only one thing each, or something big which does all, or something simple which can do a lot of things but with more labor? The market has tried all (or a lot of) choices, and the winner is still the toaster.
Are you saying this is just an exalted critique of unitaskers (a la Alton Brown)? It seems to me there's more going on here:
"Perhaps the majority of human activity can be reduced to a desire to make life more comfortable for ourselves, and has thus far led to being able to buy a toaster for £3.99 [among other achievements]. But looking at toasters in relation to global industry, at a moment in time when the effects of our industry are no longer trivial compared to the insignificant when our, they seem unreasonable. I think our position is ambiguous - the scale of industry involved in making a toaster [etc.] is ridiculous but at the same time the chain of discoveries and small technological developments that occurred along the way make it entirely reasonable."
As if "global industry" exists only to create "toasters [etc.]". In agreement with the essay, I think there's a whole lot of context that's being dropped by focusing on a device with relatively narrow purpose.
As if "global industry" exists only to create "toasters [etc.]". You mistake me. As with things like CPUs, stoves are fantastic general purpose tools for preparing food and worthy of industrial effort.
Toasters do seem to pass that test as well, because most western households have a toaster or something similar. In that sense the market has spoken. But it is reasonable to ask how much bang for the buck (and also for the environmental buck) the toaster delivers compared to other food preparation tools. I suppose your view of the price:performance ratio is strongly correlated with your personal toast requirements.
A toaster, on the other hand, sacrifices flexibility for efficiency. It's faster and uses less energy than opening up your stove and making toast on the broiler (a Good Thing), but that's all you can do with it - toast bread products. If you want to have egg with your toast, it's no help at all (unlike your stove, which allows you to cook things in many different ways). If you follow the toaster approach, you end up with a kitchen full of one-purpose appliances which are individually excellent but ultimately cater to peoples' inability to cook. At some point it's worth asking the question of whether having multiple specialized appliances is really more efficient than having one or two that can be turned to many different ones. If you look in the kitchen of a diner or restaurant, there are relatively few fancy appliances but the cooks really know what they're doing.
There's room for both approaches. Programming everything from scratch without libraries would be terribly wasteful. On the other hand, you probably wouldn't respect a programmer who relied solely on library functions and was unable to derive any algorithms of their own, no?