Solana is the 5th largest blockchain by market cap in the world, has a ton of investors and ongoing development by a lot of talented developers. Just because you haven't heard of it doesn't mean it's not a legitimate project.
I think a platform like this is as impractical as you are thinking. However, there are potential other things you can do by mixing centralized services with decentralized ones. You could create a forum like hackernews that utilizes traditional infra but restricts access to users that hold NFTs or something similar. Think of something like a Patreon type deal where access to the service is tokenized. I've been kinda thinking of stuff I could do with a NFT project I've been working on and it's a thought I've been bouncing around.
There's no way I'd want to create something that is similar to what you're thinking, because it's pointless as you are thinking. Nobody would pay for this. Also messages such as these have no reason to be on an immutable blockchain. But you can certainly create specialized services for people that support your projects using ideas like this quite seemlessly.
ETH's current scalability problems are beyond terrible, but there are alternatives. You can use chains like Polygon or Avalanche that are EVM, so you get all of the capabilities of Ethereum, but without insane gas fees.
There are non-EVM solutions as well like Solana, which has substantially higher throughput while transactions cost a fraction of a cent.
Try out other chains than ETH before ruling out web3 imo. There's a lot of engineering and incredible products outside of the main ethereum ecosystem which is being masked due to ETH's scalability problems.
There are also standalone computation solutions that are under development such as Truebit, which if successful, could allow for smart contracts to execute complex calculations off-chain, avoiding both the increased gas fees due to complexity, and bypassing the gas limit altogether.
It's pretty simple really. You just seem to have a fundamental misunderstanding of what an NFT actually is. It is simply a single issue token. The current implementation is nearly 100% for art purposes, in that they are tokens that point to cloud storage. As a developer I can create whatever services I want, based off of a determined, finite collection of single issue tokens. The token can do whatever the developer of the token wants it to do, and grant access to whatever services they want it to have access.
If I was to implement this with traditional web infrastructure, it would increase the scope of my project by likely 2-3 times. I am a single developer, and yeah I may have like a decade of experience at this point, but utilizing this instead of integrating traditional payment and authentication services, as well as traditional database services, dramatically decreases the scope of my project. Seriously, it would probably not be a feasible project for me alone if I stuck to traditional infra. It is empowering me, as a solo developer, to create something unique that I hope people find some enjoyment out of.
Let me be very straightforward here - I have spent the past 4 months working full time creating interactive generative art, that I 100% absolutely never would do so, unless there are economic incentives. This is an independent venture, I'm running it as my first business venture outside of the traditional corporate world, and I want to create something unique. I am planning on charging a very fair price for mint, basically only enough that if my full collection is minted and I do nothing else, then I would extend the amount of time I have until I have to go back to the corporate grind from about 6 months from now to about a year. I'm not creating a collection and charging like $500 per piece, I am targetting like $20.
Just because there is a ton of hype in the technology and people are making get quick rich schemes left and right with really low effort collections, please don't rule out the tech entirely.
I think that one of the most exciting facets of web3 stuff is around how easy it makes authentication and authorization scenarios. I've been working on a project that is an interactive NFT art project, where the interactive capabilities of the art are only possible for the owners of the piece. Providing authorization to the service is as simple as verifying ownership of the token. It's been pretty fun, I used an open source game engine as both the art generative tool as well as the interactive HTML5 app that allows for users to interact with their art.
That, and utilizing wallets as proof of identity, is so mindnumbingly simple compared to OAuth+OIDC for authentication and varying strategies for authorization. Granted the project is a small scale project, but with web3 architectural changes, it is empowering me to create a 2 person project (I am the only engineer too), that would be much more daunting of a project in traditional web architecture.
It's very exciting stuff, and I hope that people are able to see the tech for the possibilities it provides, especially when it comes to the NFT space. An NFT is just simply a single issue token, it can do whatever the developer wants, despite the common misperception being it is solely a link to a static image file on arweave / IPFS. Unfortunately the market is completely saturated with low-effort projects so it is very, very difficult to get eyes on innovative projects, but I hope that can change, and hope that I can create a project that allows even a small amount of people to see that there is so much more to the technology than what people have considered in 2021.