IPFS and Filecoin Token – a p2p decentralised replacement for HTTP | Hacker News Reader

IPFS and Filecoin Token – a p2p decentralised replacement for HTTP

avc.com

93 pointsOsiris3019 comments

I find IPFS very interesting, and I actually just created an IPFS pinning service with a friend (https://www.eternum.io/). I hope Filecoin will be as well-run as the IPFS project is, it certainly looks very interesting.

Very interesting indeed. I can see an era where the bandwidth used is paid for by the end user (low cost).

Data between seeds could also be chargeable but at a lower level than the end user - allowing the node to 'seed' the content and collect the bandwidth. This incentives the network to sustain itself and get rewarded for increasing availability.

ISPs and/or even access points like Cafes and Hotels can seed for themselves getting the bandwidth $$

My understanding is Eth blockchain could also allow part of the 'coins' be given back to content creators? though I also understand there may be scaling issues.

> My understanding is Eth blockchain could also allow part of the 'coins' be given back to content creators?

I doubt it. How do you prove (to a smart contract) that you're the creator of the content and not just the first person to put the content on IPFS?

You can have original creators do an out of band verification, e.g. by signing their hash. You would then need a community to vote / decide on the right hash. Given anyone can upload to IPFS, 99% of the data will be junk so you will always need an incentive based quality control system to categorise and curate content.

Looks like it's not anonymous (requires my email address), and costs $0.15 per GB per month.

What advantage does this have over S3 (0.0245 per GB/month) or OVH ($0.0112 per GB/month)?

The advantage is that it pins stuff on IPFS while the others don't? I'm not sure I understand the question, I'm afraid.

Why would anyone care about IPFS? Presumably someone pays for hosting to get their content shared. But if other forms are cheaper, what does IPFS bring to the table such that someone would pay for your pinning service if its not cheaper or anonymous? And why wouldn't someone just use AWS and run their own node to pin it?

> But if other forms are cheaper, what does IPFS bring to the table such that someone would pay for your pinning service if its not cheaper or anonymous?

It's much harder to censor, more permanent (content doesn't go away if your server goes down), etc.

> And why wouldn't someone just use AWS and run their own node to pin it?

Because it's more expensive and more hassle than just using a third party service, of course.

From economics perspective this is this is terrific idea in general.

Instead of wasting electricity and resources to generate 'consensus based artificial scarcity' fiat money like BTC, monetize digital assets available in the net with cryptoderivatives.

---

Terminology question: Are all 'cryptosecurities' going to be called coins or currency? Technically Filecoin is commodity derivative or asset-backed security.

> From economics perspective this is this is terrific idea in general.

> Instead of wasting electricity and resources to generate 'consensus fiat artificial scarcity' based money like BTC. Monetize digital assets available in the net with derivatives.

... or you could just not do that, and save even more resources.

This is BitTorrent with cryptocurrency micropayments bolted on the side. It's literally an "x but with Bitcoin" idea.

Taking illiquid assets and transforming them into a security to be sold in markets is the mark of mature industry.

Internet has created large pool of digital commodities to be sold: file storage, computing power, network bandwidth, regional caching and content delivery. Wrapping them up into financial instruments under common protocols is the way to go.

Has anyone tried https://storj.io/ "Decentralized storage based on the blockchain" ?

It's an interesting idea to have a coin that is backed by something that has real tangible value in the industry (storage space)... I'd like to know whether or not the intrinsic value behind Filecoin (in terms of $ worth of storage space) will be worth the cost of operating the network (in $ terms).

As a buyer of storage space, if you need storage, isn't it always going to be cheaper, faster and more convenient to use your own hard disks or a specialised service like S3?

Will people actually have an incentive to use Filecoin as a storage service instead of S3? If they don't, then the coin has no advantages over Bitcoin.

One thing that immediately comes to mind it: storage locality. S3 will by design always be at fixed (known) distances to the retriever. On the other hand distributed storage may end up having better locality (depending on where in the world you are accessing the data from).

Owning hard disks is the only way to get cheap storage today. Specialized services like S3 are pretty expensive in comparison, even the cheapest ones, like Backblaze or OVH.

As I see it peer-to-peer storage could force everyone to compete globally on price and compete directly for the data, not the customers. Kind of like automatically choosing during upload the cheapest offering available on the global market at any given time, except that the players themselves are aware of the mechanism and know they are competing solely on efficiency, making them pursue efficiency. Eventually this could even put significant amount of storage into the countries with cheap electricity, labor and internet, making storage costs lower than otherwise possible.

efficiency is not the only metric, reliability is a huge factor in any kind of digital resource.

Reliability is what software is for.

It's more for public data than private. It's like p2p filesharing where you can pay for reliable storage.

Judging by the title, it's also for web pages. That's also part of the vision for Ethereum's Swarm.

How is this different from Siacoin http://sia.tech/ Great idea anyhow and backed by Protocol Labs.

Reply on news.ycombinator.com