Concerns about cost EXISTENCE, not cost EFFECTIVENESS


I recently learned about Ethereum and have read the whitepaper, and I'm excited by the potential for Turing-complete decentralized apps. I have many questions but earlier today I was engaged in a Twitter discussion[0] in which the fact of a "cost per operation" came up. To summarize:

* In order to accomplish any operation, a cost is levied for that operation (in ETH).
* While a DApp may refund a user the cost of the ETH after the operation is accomplished, this means at a minimum that users must first provide a "deposit" to pay for the operation in full.

Given these facts about the inherent nature of computation in Ethereum, what are the (side-)effects on poorly-resourced individuals?

To clarify, I understand *why* there is a cost (GASPRICE, etc.) and in fact I began looking into Ethereum precisely because I was looking for an operating environment that would be resilient against spam-like attacks.[1] Ethereum promises that and I'm excited about its promise!

What concerns me is the fact that, since my DApp has nothing to do with economics—there are no assets, there is no money involved—and the target audience for the DApp is specifically poorly-resourced people, forcing them to effectively "pay" to use the DApp is antithetical to the purpose of the DApp. In other words, to put it crudely, I am concerned that the very same mechanism that will deter spam will also deter my user base because of a technologically-enforced Digital Divide.[2]

From Twitter, I am told that the Ethereum Foundation may provide ETH grants to non-profit public good DApps.[3] That's great.

But what I want to know is what, if any, other thought has been given to the fact that there is the potential for poorly-resourced people to suffer from even more radically restricted access to computation in general if cryptocurrency is required to perform such computation (on top of electricity and physical-world access to the CPU, etc.)?

Thanks for your time.





  • anoneth01anoneth01 Member Posts: 1
    I gave this some thought and I'm thinking this is a case of optimism vs. pessimism on where the costs lie.

    For example, if you want to run your own instance of Predator Alert Tool, you have to host it. In addition, you need some technical know-how to get it running - not a great deal, necessarily, but it's not something you can sweep under the rug. Moreover, maintaining the app once it's running creates an ongoing cost in time. You can beg or borrow your way to get it going, or use an existing open instance, but then you are indebted to whomever was gracious enough to run it for you. And so a simple idea becomes a complex web of obligation, with control going to the people who can find that particular combination of time, money, and know-how.

    As you point out, your instance of PAT is being attacked with spam, and that's a consequence of having that control, even if you are disinterested in exploiting it. It's this "public use of private property" aspect that makes you a target - all your computational resources are up for grabs, so when you provide a visible, public commons and it's attacked, the only way to survive is to be bigger than the attacker. And that encourages working your way up the power chain until you're in control of way more than you actually need.

    On Ethereum, though, PAT would require access to "just enough" computing. You can earn ETH automatically by exploiting your personal computing resources, and then all the obligations around maintaining a server are gone. The cost, in human time-and-effort terms, is going to be much smaller because you don't have to provision a server. But that does leave the aspect of users having to pay for their share.

    So, narrowing the questions(to ones that I don't have answers to): How much does it cost to do things with ETH, relative to the status quo? And will earning be "lumped" towards the high end, or does it downscale to the lightest computing devices?

    If we have a high cost and unequal earning scalability - enough so that most users are burdened - the original argument holds merit. But this is a question related to how cheap and efficient the underlying technology is. If more people have computing, and that access is at a higher average baseline, the cost of running the app relative to the rent earnings of the user's personal computing device should plummet; ETH should be "too cheap to meter" for everyday uses. And if you think Moore's Law will hold out, that's a future worth betting on.
Sign In or Register to comment.