Bitcoin Forum

This discussion is based off a Michael Saylor interview found here: https://youtu.be/-LFofHs0Lsk?t=1504

I am in a mentoring program for an ISP, where I have 10 months to deliver a capstone project to upper leadership.

I think this would be a great opportunity to build a project based on the idea Michael Saylor talked about, to spread bitcoin's influence and apply to a use case.

The idea is about securing areas of large networks against ddos attacks, or single websites against scammers.
Honest people go through free, dishonest people pay a price. This is currently impossible, how do we deter criminals from adversarial countries if we can't punish them?
Anyone that wants their internet traffic to go through to a large network, has to post collateral. The gateway to the network would be able to quickly verify the collateral is valid and redeemable.

Once the cybercrime has been attributed to certain packets, the collateral attached to those packets is redeemed. This would economically deter criminals.

So, open discussion:
What problems with this do you see?
How would you design this system?
How would you make it so ddos packets from botnets can't share the same transaction used in the collateral?
What potential hurdles are there?
Would this technology be used by the ISP for good?

What if an ISP changes the transaction to get users to send them funds?
Who pays the fee for the transaction in this case and if its on the ethereum network, you might be better off charging them btc for the access rights.
What if a company - maybe a news agency - says "now you've shown us how much you have, would you like to donate" what if they selectively do it so people who can collateralise a larger amount only get that or get another message that attempts to get them to sign over a few % before they can see the page.
How is it better than the current hash cash system? I could see it being used as well as that but can't understand how it's better.



You'd need a network of servers scattered around areas the isp serves that could facilitate this.
On second thoughts though, why wouldn't you just get them to sign a message to a crypto address with funds in it, then you only have to do one blockchain query and less algorithms need to be run - you could get the user to sign a randomly generated number with a bitcoin address and then you'd just have to query the balance on the address and the balance on the signature - as long as there's a system thst puts a cool down per address then this shouldn't be another way to ddos a server.



The idea would be to make it so the collateral is high enough to stop a botnet.

Even using a $5 will be enough to stop servers from being overwhelmed afaik (as long as the isp and the Webserver have good congestion management).


This would have a huge privacy tradeoff and might not be used for that reason or there may be considered unsafe or untrustworthy by users.

Cryptocurrency is probably difficult for the average person to understand still and making them hold on to crypto could get them to use it and lose it (eg accidentally sending funds to the wrong address).

Hash cash is a gap in my knowledge, will learn about it!
Want to make sure the design doesn't acrue any cost for honest users.


Confused by some of this, not sure why querying a balance on an address would be needed.
Took part of the MIT open courseware class on cryptocurrency engineering, but still far from understanding contract schemes.

It's the screen you get that comes up with "checking your browser" or something like that with something moving on the screen. It sends your browser a problem to solve in a given amount of time and determines whether you have computational power to be a normal machine or not.


Regardless of the method used, querying the blockchain would have to be done.

You'd always have to query a balance on an address just the way you suggested would cost the use unnecessarily.