Isn’t ICP a *clear* evolution of blockchain technology, am I missing something?

First I’d like to say, that yes, this will be somewhat of a shill and I have recently invested a large sum in ICP. Nonetheless, I invested in the first place because this is the first thing in years related to crypto I find truly interesting and novel.

I’ve been in crypto for a long time and I have the same feeling now as when I first saw Bitcoin late 2012, then Ethereum in summer of 2014.
But decentralization hasn’t been perfect, we still are heavily reliant on centralized services such as AWS, wallets, bridges, oracles and hosting for websites.
ICP is a catch all for these centralized choke points.
It essentially allows you to host a decentralized website with in-built smart contract functionality, ICP calls these “decentralized websites + smart contracts”, canisters.
Meaning that you could actually host *real* web3 applications without any centralized hosting for the website. There’s also no need to download separate wallet software to interact with the dApp. A few days ago I created a new account on [dscvr.one]( which runs on ICP, I created the account by simply running my fingerprint on my android phone. That felt like a huge UX improvement.
ICP also allows for trustless bridges which is pretty awesome and makes oracles obsolete since canisters can make https requests.
I know that some of you might say that the hardware requirement to run a node on ICP is too high to achieve decentralization.
This is a fair point, but from what I can tell they have managed to balance decentralization and hardware requirement to unlock a whole new swath of decentralized use cases.
I just find it very interesting and I am becoming more of a believer the more I immerse myself in the tech. This feels and seems much more different and all these “Ethereum killers” or EVM compatibles, like Matic, BSC, Solana, EOS, etc, etc… ICP seems to me as if it just created the next evolution of public ledger technology.

I didn’t just create this post to shill, I’d like to hear criticism from those of you who have looked into ICP but found flaws of things that you don’t like.
Anyways, these are my two cents, am I missing something or is this just as cool as I think it is?

View Source

21 thoughts on “Isn’t ICP a *clear* evolution of blockchain technology, am I missing something?”

  1. ICP makes it possible to use smart contracts in coordination with other web features, and while this makes for beautiful apps, it isn’t flexible enough to run traditional applications. It also doesn’t have the performance required for heavy calculations, and is quite expensive to use (storage alone costing $0.46/mo).

    For flexibility in running almost anything, I can recommend [Akash](https://akash.network/). This is a decentralized marketplace where you can run most Docker images on trusted providers, with the option of load balancing. It’s also cheaper with [the cheapest tracked deployment costing $0.17/mo](https://www.akashlytics.com/price-compare) which would mean **$0.34/mo** for 1 GB + some free RAM & vCPU. Speaking about flexibility, providers can add public attributes to give you details regarding anything you may need, whether it’s for which CPU model they’re using or how much bandwidth you’re getting.

    If you want to run scientific calculations or similar, I highly recommend [Golem](https://golem.network/). Right now, its best applications are ones that can scale by sharding, to use parallel computations. Think doing 100 similar small jobs on 100 computers instead of 1 large job on 1 computer. [One average CPU-month costs $3.17, or you can rent 100 CPU-hours for $0.44](https://stats.golem.network/network/provider/pricing). Notable examples are [blender_cuda](https://github.com/norbibi/yapapi/blob/24881c9d1ecf80be7ad6b8d56f3f72a71403f64a/examples/blender_cuda/blender_cuda.py) which runs on a GPU, and the entirety of [awesome-golem](https://github.com/golemfactory/awesome-golem).

    Since you’re so focused on ICP which is mostly used to run webapps, I can also think of why [Sia](https://sia.tech/)/%5BSkynet%5D(https://skynetlabs.com/) can be useful. Skynet bases itself on the Sia blockchain, where any user can upload a file or a database entry through a portal, self-hosted, private, or public. With cryptographic proofs, it is possible to verify that files are not lost, are re-generated, are legitimate (i.e., not spoofed by the potentially malicious portal) – all while keeping prices low at [$1.44/TB/mo + bandwidth](https://siastats.info/storage_pricing), and reliability at an assumed [99.9999% uptime and a much larger availability](https://blog.sia.tech/cloud-storage-for-2-tb-mo-8a34043e93bb).

    Reply
  2. I think DFinity’s ICP is one of the biggest crypto scams in history.
    Everyone says luna but no one makes a sound about this coin falling from 700-odd dollars to $8

    Reply
  3. ICP is the opposite of what blockchain tech is supposed to be:

    – Permissioned developer system with KYC, your access to the blockchain can be revoked at any time

    – Blockchain runs only on specialized node hardware sold by Dfinity and only in selected data centers

    Thanks but no thanks.

    Reply
  4. it allows you to host centralized “web3.0” apps on a centralized ledger where everything is centralized and probably less secure than AWS, yes.

    Reply
  5. You’re not missing anything. ICP is actually making great strides in smart contract computation efficiency by garbage collecting old state (no need for a replica to verify the entire chain, simply catch up within the current epoch), subnet splitting (replicas only process a subset of the entire networks messages), hybrid permissioned consensus model (stakers determine the topology of the network, rather than completely decentralized wrestling to define the network).

    The network is still able to tolerate Byzantine faults (when a replica attempts to violate the protocol). The main concern is that stakers define the network, and centralization of tokens means total control over the network.

    However, many players are entering the space- VCs yes, but also developers:

    1. https://infinityswap.one/
    2. https://twitter.com/psychedelicdao
    3. https://github.com/demergent-labs

    These developers own tokens to forward their own projects.

    There’s also [ICDevs](https://icdevs.org/) which are separate from Dfinity.

    Dfinity owns tokens, yes, and so do VCs. But many other parties are entering the space and creating projects which were never possible on other blockchains.

    ICP is so completely different than other blockchains that people look at it and say- it MUST be a scam, because no other blockchains exist like this. But they make the error of assuming that blockchains **cannot** exist like this. That is why this is one of the greatest dark horses in crypto.

    Also most people here have no computer science background, and distributed computing is an esoteric topic even in computer science circles. The gains that ICP brings cannot be understood by people on this sub; they can’t even grasp basic economics.

    [I bought at $5.50](https://i.imgur.com/TAp3LB9.jpg). That doesn’t mean I can’t be wrong, of course.

    But when sentiment is at rock bottom and [devs are actively building and contributing to the network, I’m bullish.](https://forum.dfinity.org)

    P.S. Y’all can read how the network works: https://internetcomputer.org/whitepaper.pdf

    Edit: downvoted with no rebuttal. Classic. If you’d like to refute my case, I’d love to hear it.

    Reply
  6. How is this protocol decentralized? I have played with the testnet just a bit several months ago, but it isnt that different from TRX or EOS tbh, another failed ETH killer.

    Reply
  7. Its not always easy to predict the future, but its always obvious in hindsight.

    BTC brought the public ledger. Eth brought smart contracts. ICP brings decentralization.

    Reply

Leave a Comment