So I feel that I have to respond here because a lot of what you're saying is simply factually inaccurate and very misleading. I have no interest in defending Engine and there are many things that could be done better, but I do think it's important that people understand what is actually going on.

Transactions on Engine are not distributed across a network

This is incorrect, Engine transactions are Hive blockchain transactions, so they are very much distributed across a network.

Engine is not tamperproof, at any time anyone on the Engine team with appropriate access can modify balances and transactions.

This is also incorrect. You're confusing tampering with a node with tampering with the system as a whole. I can run my own Hive node and modify balances and transactions all I want, but that doesn't mean it's correct based on the protocol of the system. Engine is no different. Sure the Engine team can modify things on their node, but others can validate the info on their node to see if they are reporting correct information or not just like with any other blockchain platform. More on this later.

All verification is dependent on one single third party entity (Engine)

False. The code is all open source for the Engine platform and anyone can independently verify all of the data.

While Engine may in fact as an append only mode, the possibility to edit/delete from the postgres/mongo database is technically possible and relatively easy to do with the correct access.

This is incorrect for the same reason as the "tamperproof" claim above.

I don't know enough about the inner workings of Engine to determine if it is in fact protected by cryptography, I suspect it is not and is just stored in postgres but I don't have enough information to confirm or deny this.

What is meant by "protected by cryptography" is that the transactions are cryptographically signed using a public/private key pair, which Engine transactions most certainly are as a result of being transactions on the Hive blockchain.

Everything @cryptomancer said in his quoted comment is correct and accurate. If I were part of the Hive Engine team, I would encourage third parties to run their own Hive Engine nodes and their own copies of the hive-engine.com website that point to them (or just to allow hive-engine.com to point to other nodes just like peakd.com and other Hive apps can).

If more people were running nodes, then if the Engine team wanted to change the code then they would need the consensus of everyone running Hive Engine nodes to adopt the same changes, or else their node would fork off from the others, at which point it will be up to the users to decide which fork they want to use. This is exactly the same as most any other blockchain platform, and is only not apparent in this case because no one else is running a node - a problem that should be relatively easy to fix, IMO.

I think that relatively simple thing would make it much more clear that Hive Engine most definitely is a blockchain. It doesn't mean it's necessarily a good one, or one that projects might want to use - that's up to each person or project to decide, but just because you may not like it or the people behind it doesn't mean that it's not a blockchain.

Pair that with native smart contracts and we will have something no one else has, free three second transactions with the ability to create your own contracts and tokens.

Also, what?? Numerous other blockchains have this nowadays - EOS, WAX, TRON, and many others including Ethereum sidechains which seem be popping up left and right. Hive simply cannot compete in the generic smart contract platform arena.