I would realy like to see the scoring of the blockchains you tested as much as which one you did test. I am assuming you did only look at permissionned-less blockchains.
Can you explain why you did not try permissionned blockchains (like most companies and institutions are trying to do)?
Your assumption is not correct. First we would be using a permissioned and adapted steem network, not the public net.
Second, we looked at "mainstream" corporate blockchains - everybody we know uses either Ethereum or Hyperledger Fabric. For Ethereum, it's not the public main net but private networks, lately built using the PoA protocol.
The conclusion was that Fabric is more a toolkit that requires considerable time and resources to build a solution than an "out of the box" blockchain. And Ethereum, which has been used for a PoC, is not suitable for storing data.
Had there been more time and resources to scour Github for even better blockchains, I'm still not sure I would have continued looking. If you take a look at the slides I presented in this post, you can see that steem has a unique combination of features (not all of them are on the slides here, there are more) that make it a particularly good bet
On a different topic, I like the steemmessenger project, congrats and keep up the good work !
In a PoA settings I would have gone for the smart contract capabilities over just text. But lately I have been thinking that DPoS is well suited for private blockchain.
I am working on comparing blockchains in my PhD so at some point I will have to look at Hyperledger. As you mention deciding which one to look at and what to look is a pain.
Thanks for steemmessenger. I just recently joined the project and have fun programming around steem and doing some cryptography. By the way I like your posts, it is rare to find valuable discussions n blockchains.
Thanks.
"Smart contracts" are just code executing in a trustless environment. But that is irrelevant in a corporate / institutional settting where trusted execution is the default. Even getting the hierarchy to consider replacing trusted execution with its distributed, trustless version is a pain in the neck. So no, "smart contracts" are definitely the least useful feature of blockchains for corporates.
If you want my opinion, Fabric is an over-engineered architecture that responds to two rather contradictory design goals:
Besides, Fabric is not a product but a toolkit: you need a couple millions to get started and "everything is possible" ... because nothing is built-in. In their tutorials "Alice" can happily double-spend and spend money it doesn't have. When asked, they answer: "oh but you can implement that in business rules" (but forget to add: "provided you pay for X more man-days")
Looking at blockchains is a lot more fun when your horizon is a PhD thesis than when you are expected to deliver with a young and little proven technology and 250K€ a project estimated at 8M€ (32 times more) using a classical centralized architecture.