1/ *Centralized* apps building atop decentralized protocols may be the winning combination, each furthering the adoption of the other.
2/ As many have pointed out, “decentralize all the things” is a mistake. Decentralize the things that benefit from decentralization.
3/ Starting with centralized apps (think @coinbase @circlepay etc), if you’ve been paying attention, they are utilizing an increasing amount of decentralized protocols & open-source crypto infrastructure, to the benefit of their end users.
4/ The realization this week that @coinbase and @circlepay built their #stablecoin on top of @zeppelin_os was another tick in this trend. github.com/centrehq/centr… 
5/ Centralized #crypto apps that build upon decentralized protocols have:
-Global capabilities from day one
-Lower costs due to the on-demand and perfect-competition nature of protocol services
-More rapid innovation as they build upon others’ open source building blocks.
-Global capabilities from day one
-Lower costs due to the on-demand and perfect-competition nature of protocol services
-More rapid innovation as they build upon others’ open source building blocks.
6/ Long term, centralized #crypto apps that build atop decentralized infrastructure will find it the key to what allows them to scale beyond, and outcompete, incumbents that fear public networks & cling to owning their infrastructure.
7/ Having covered cloud companies for 3 years & seeing how long it took big incumbents to embrace the public cloud (and some still haven’t), I have a hard time seeing these incumbents embrace decentralized protocols before it’s too late. Their loss.
8/ All of this leaves decentralized apps (#dapps) in hot water. As many have been pointing out, pure-play dapps are not getting much use.
9/ Connection points to end users need to be hyper-responsive, and due to the laws of physics, #dapps are fighting an uphill battle against centralized apps in response times.
10/ Projects like @graphprotocol or Picolo Network (picolo.network) are aiming to help #dapps with their response times, but it remains TBD if it will ever be good enough for a human to not discern the lag.
11/ I also think customer service via #dapps is going to be a hard problem to solve, as customer service is high-touch and doesn’t come for free.
13/ For these reasons, if you’re focusing on investing in #cryptoassets (as @placeholdervc is), I recommend focusing on layer 1 protocols, and important middleware protocols atop the layer 1’s.
14/ Look at what protocols the centralized apps are building atop, and getting real utility out of.
Those will have fundamental value, regardless of how much some obsess over “economic abstraction” and “infinite velocity.”
Those will have fundamental value, regardless of how much some obsess over “economic abstraction” and “infinite velocity.”
15/ As always, I’m open to being proved wrong, but I feel reasonably confident that in the short-to-medium term it will be the centralized apps that continue to capture most all of the users — longer term, as people work to make dapps more responsive, we’ll see.
