Maintenance often takes a backseat to innovation. But in the long run, we care about the total cost of ownership (TCO). Not only upfront costs, but all maintenance costs, hidden costs, and everything else that goes into owning an asset.
Cloud infrastructure is well understood enough to make the TCO case against writing, running, and maintaining services yourself. It's the reason why even the largest and most software-centric companies still outsource and buy third-party tools. Why does Google use Workday instead of building its own HR management solution? Why does Google still use an Oracle database somewhere deep in the company's accounting department? TCO.
On the other hand, the TCO of crypto and web3 technologies isn't well understood (yet). But observers can start to piece together the ownership costs. Why do most decentralized web3 applications use Alchemy or Infura instead of running their own Ethereum or Bitcoin nodes to get transaction data? Why do so many users have custodial wallets with Coinbase instead of managing their own? Why do NFT buyers and sellers prefer a centralized service like OpenSea to direct transactions?
But high TCO isn't an argument against web3. In fact, the total cost of ownership is almost always an opportunity. Look at electric cars vs. gas cars. The TCO of EVs was extremely high – bespoke engines to fix and batteries to replace, few charging stations, and low gas prices. Now, it might be smarter than ever to own an EV – longer range, more charging stations, less engine maintenance, and low electricity costs.