Disclaimer: I'm not an expert. I'm an idiot with an internet connection. I've probably missed a lot, misinterpreted a lot, and failed to account for a lot. This is a fast hot take, the application in question is barely in its infancy, and I'm in no position to predict where it's going. That said, I've read through the TON whitepaper(s) and have some thoughts I'd like to share.
The non-technical "Primer" (marketing material) can be found at https://icorating.com/upload/whitepaper/gNQ7e9z3lCGi519Wz8mmC0Kg8aA0goeZKAQ802vo.pdf
The technical whitepaper can be found (with some digging) at
https://drive.google.com/file/d/1lqVlrgiztnA5dkOHP7-ENDKT1FgZuCUV/view
This is my initial impression.
TON is interesting. There are a TON (ha... haha... chirp chirp) of interesting ideas in the papers.
Too many. Way too many.
This project appears to me to be 7 or 8 separate projects, lumped under one brand-able umbrella, and each and every one of them appears to be unproven, untested, and hyperambitious.
Have you heard of EOS? https://eos.io/
Yeah, that's one of sub-projects, except the TON version is even more ambitious.
It's got DAG-esque Tangle-type routing between infinite shard/workchains
each shard is made up of accounts, and is... isolated... except for the ability to pass messages to other shardchains via what they call Instant Hypercube Routing.
Which seems to be... well, it's not a DAG, it's a UPCG , an undirected potentially cyclic graph (just made that up, my graph theory is rusty, sorry)
There's this concept of self-healing individual accounts within a shardchain, able to write new valid blocks over invalid blocks, which I don't think is very well articulated. Lots of words, very little meaning.
The authors act as though the self-healing is clearly stated, well-planned, well articulated, and repeatedly reference the section where it's explained in both papers, but I don't see the implementation details or answers to the (to my mind) obvious questions of reliability and security in a system that allows invalid blocks to be kept as part of the chain.
The entire platform and its various pieces are planned from the outset, but there are gaps in the implementation details. That makes me wary.
The staking/"mining" mechanisms are interesting, mostly ripped straight from PolkaDot (another interesting, very ambitious project)
They talk reeeaaaally big about the requirement/importance of maintaining decentralization, but their implementation includes a centralized (if autonomous) reserve that artificially maintains prices by "buying back" tokens.
There's an assumption that the system will democratize itself, reducing the total number of Grams (tokens) held by the reserve to the point where the network/community can decide its own direction... but it's just hope.
Direct quote: "Later, when less than half of all Grams remain under control of the TON Foundation, the system will become more democratic. Hopefully it will have become more mature by then, without the need to adjust parameters too frequently."
Developers working on the open-source project (there is only ONE mention of open source, and it's here) will be paid out of some of the unallocated grams held by the reserve.... but with a minimum TWO YEAR vesting period. If you work on the project, don't expect even nominal compensation for a while.
It's a gorgeous whitepaper, from one perspective, and an absolute crock of worthless crap from another, since it's SO big and SO overplanned and SO underarticulated, with so many features that are just assumed to be proven, assumed to exist... when they don't.
If you think projects like OMG and EOS and IOTA are ambitious, perhaps too ambitious, ho ho ho I got news for ya.
Another thing, and this one's a big one.
The project is being funded from the start. As far as I'm concerned, that's a mark of death. That money kills hunger.
Not only is it funded from the outset, it's privately funded. They're doing their second round of private pre-sales. Literally a pre-pre-ICO. And it's private, open only to select Silicon Valley investors.
AND as far as I can tell, there's no way to turn the money they raise into development work, because of the payment plan they've come up with. Unless, of course, they use it to fund a centralized development team. Which they will surely do. And it will surely fail.
Why? Because it has this plan to follow, a plan that's simultaneously very broad in scope, very specific in requirements, and incompletely articulated, and I don't think that's how successful decentralized systems emerge. They emerge; they start with a loose direction, a seed, and incorporate new features as those features become necessary/available/possible.
This project appears to me to require all (or most) of the pieces in place to even begin functioning.
Gotta have the Hypercube routing, the infinite sharding, the self-healing accounts/minichains, the wonky (cool, but wonky and as far as I can tell untested) POS mechanisms...
All of it has to be in place at the outset for any other part to work.
There's no identifiable starting point.
Look, I could be 100% wrong about absolutely everything I've written above. I don't have the experience/knowledge/vocabulary to grok every aspect of this concept, so I'm not qualified to dismiss it. Take my opinion for what you think it's worth, with as many grains of salt as you need.
Maybe TON will evolve naturally into precisely the thing it's trying to be... but I don't see it.
I see a big ole' paper with every cool idea a group of very smart people could think of crammed into it, and very little room for adaptation/evolution.
I'm going to be watching TON as closely as I can, and I'll update my opinion on the project as more information becomes public.
I really, sincerely hope it succeeds beyond my wildest expectations and becomes the magic decentralized super instant hypercube routed awesome thing that it's meant to be...
But I'm skeptical.
I will say one last thing, a sort of silver lining:
VCs and investors in Silicon Valley are buying in like crazy.
That says very little about TON itself, as VCs move in herds and are often proven wrong.
But it says a TON (chirping intensifies) about the future of blockchain technologies, and indicates a shift in opinion from "meh, blockchains" to full-on FOMO among holders of existing capital.
Existing money is TRIPPING over itself to get on board.
This is why the emergence of a novel system rarely results in a "war" or a "takeover" of the old. The pieces of the old want to become part of the next level emerging simplex, because it makes everything better for everyone.
Blockchain is ba-booming, and we all stand to gain from the value this new sector will produce.