While it's easy to agree with Paul on technology being precariously dependent on conditions that vanish with time given the way they're developed, the motivation is to develop technology this way seems mostly practical. The likely alternative is that technology doesn't get developed at all. The technology of developing technology with less vanishing dependencies, presumably decentralized ones, can't yet compete with centralized forms.
There's evidence we no longer need government level centralization at least, but Coase has something to say about why we trend toward centralization to get things done.
I agree centralization gets things done. I guess the main thing I'm trying to convey is that just because a centralized company has achieved great technological heights, it doesn't mean those advancements will spread, or are at all spreadable, to the larger society. So when approaching decentralized projects like bitcoin we should be wary of centralized advancements because those can disappear much more easily than truly decentralized advancements.
reply
We agree. Wary is the right word.
I guess I'm also not sure if decentralization guarantees spread. Bitcoin development for example remains centralized/specialized: one real implementation in use, a relatively small group of devs involved, etc. Even assuming Bitcoin is properly spread, Bitcoin is worth 250 trillion dollars in aggregate to many of us and participants have a fair shot at owning that appreciation/social good. No other technology might ever have these features.
While decentralized advancements are better, I'm less bearish on centralized advancements. If they are net bad, their head will be put on a pike for all to see. Perhaps not fast enough though.
reply