In a permissionless blockchain there is nobody, no company, no consortium that controls or governs the blockchain. It is the individual (company) that controls its own resources (e.g. money) within the (legally or by consensus) agreed upon blockchain.
For example in the Bitcoin blockchain it is the holder of a private key that controls the holder’s money. For Bitcoin it is the global consensus (and resulting market ask/bid prices) that gives Bitcoin its value.
There is some romance blooming between the Ethereum Alliance and Hyperledger, maybe they are going to define an interface?
However we think it is important to choose for a global blockchain like Bitcoin. This and similar global side- or blockchains have nice properties like:
Being permissionless, no governance needed;Being PoW secure and data cannot be changed;Having a dirsize that only grows linearly;Being deflationary and green (sic!) because it restrains consumption and compound interest buildup to the few;Removing Silos (single Databases) and making new entrance in the consortium easy.
Hyperledger and to a lesser extent Ethereum do not hold all of these properties. Of course almost all blockchains function well as enabler for consortia to share data.
When, often, a blockchain is not needed, use tested languages like SQL database languages, Python, NodeJS, C++, Java and the like for the purpose they are best, i.e. real world (data) solutions.
Use a blockchain wisely only in cases where it is useful.
1. Like for money bookkeeping or smart contract administration of digital goods
2. or for accounting of real world goods. Not fully tamper-resistant, but the gameability (i.e. making money by cheating) is less, when the real world data is accreditedly validated e.g. by IoT and/orby using oracles blind to the contract and its participants (see work by MIT.)
Only in those cases that censor-resistance (immutability) and/or no-governance is beneficial we would choose for a blockchain, else choose for tested development tools, like SQL, NodeJS etc.