The same thing would be cheaper to operate without the crypto backend. And it wouldn't raise the awareness/price for Sia.

In the long run, it's always going to be higher than what companies dealing with storage services. First, because of effects of scale, second, because of the overhead of dealing with the crypto layer / extra network costs.

Possible but doubtful. Check market price history - the overheads of dealing with crypto are historically tiny compared to the profits of being forced to hold any.

At time of writing storing a TB for a month is ~10 usd. That looks to be cheaper than s3

S3 does a 1000 things on top of data storage. It will always be a more expensive option to compare to. Try wasabi as mentioned or backblaze B2.

Wasabi is 5$ per TB and month.

I don't think you can run a service like Skynet without a crypto backend. It does a _lot_ more than something like IPFS. It can keep files online without needing the original uploader or any user to actively seed them (uses a storage marketplace instead), among other significant features.

I meant you can run skytransfer without it, not Skynet itself. (It's basically Megaupload)

SkyTransfer has the benefit of being "infrastruture-less". Rather than needing someone (the developer or someone else) to run a server, it can sit on top of the same Skynet infrasturcture that runs everything else.

SkyTransfer was built in the absence of any devops. No costs to the developer, no servers that need to be run by the developer, and users can perpetuate the service after the developer leaves without having to do any devops themselves.

SkyTransfer also gets to leverage the identity and authentication infrastructure of the Skynet ecosystem. Building an equivalent service off of Skynet would probably take 5-10x as many developer hours.

The gateway operator/uploader still have to pay for the service and data. Node operators take their fee for the devops and maintenance. Even if the cost is spread thin, it's still there.

I'm not sure why you think building it off Skynet would be much harder. It's basically a web form going to an encryption function to upload to link generator (and a download path handler). On the backend you either buy local storage or pay for an external service which supports pre-signed S3-style uploads. Neither sounds like a lot of work.

The key thing here is that the operator/uploader is different from the app developer. As an app developer, you don't need to build or operate any backend at all, and you have no costs.

Buying local storage and paying for an external service both cost money, and the costs scale as your app gets more users. There's also maintenance overhead, you can't just deploy your app and completely forget those things exist. As a Skynet dev, you have no bills, no maintenance overhead, and it's truly a deploy-and-forget sort of thing. As the dev, you don't need any sort of Skynet account or signup.