Topic: Microsoft Researchers Suggest Method to Improve Bitcoin Transaction Propagation (Read 17536 times)

But like it or not, if Bitcoin goes viral you are nothing. You are one of 7 billion. The cost to keep your computers on is a few cents; the expected risk for not doing so is your entire bitcoin stash... divided by 7 billion. The economic incentive is outweighed by the disincentive.

maybe i should frame it differently.  i also have an economic incentive to leave my computers on to support the network.  i have a substantial investment in Bitcoin as do many others who compose the network.  its about taking care of one's money.  you don't just leave it out on the front lawn for anyone to take.  you build a safe in a remote area of your house and fortify it to the extent you feel comfortable.  you may buy a few guns also just in case.  it may cost me a few cents to keep those computers on but i want to take care of my money!

very, very interesting insight.  i hadn't thought about that.  if we are all so fascinated by this technology, how could the big boys not be?  i know of several PhD mathematicians, computer geeks, and economists who are fascinated with Bitcoin and these are precisely the type of ppl who'd be employed by these large players.  if they're smart they would be getting ready for the snowball effect.

I don't think of the Bitcoin client as simply a way to spend Bitcoin, even though that's about all it is at the moment. I think of my computers as machines for (sometimes monkey) business. If you want a productive business, you find ways to keep your machinery running 24/7. I think that will be the goal of software designers to craft tools using Bitcoin to stay on making money. I'm not talking about mining, but real world businesses that will depend on a robust Bitcoin network.

[edit]
I wouldn't be surprised if M$, Google, Facebook, Mozilla, Apple, (pardon my bias) etc. doesn't have NDAs regarding major development for BTC based apps. I can't imagine big ballers not creating Open Transaction servers, microtransactions, anti-spam postage based email, deed of title apps, escrow based courier services, and probably hundreds of other unique products and services that take advantage of Bitcoin's unique properties.

Hi, I read some of the posts and I try to summarize it for later readers, and please correct me if I say something wrong:
1. There could be a problem if  out of some reasons there aren't enough people relaying transactions. Example: Bitcoin gets to Visa proportions and only miners have the hardware requirements to receive/relaying transactions. They will (of course) not relay transactions if there is no incentive.

Solutions:
1) Pay all (or statistically some) node that relayed transactions somehow
2) Punish the (greedy) nodes, by cutting connections to them. Then a miner not relaying transactions will receive much less transactions (he could include in his own block)


My Comment:
I think that it it very important to think about it. The solution 2) wouldn't require a change of protocol (I think) and is a quick and easy way to get everyone to relay transactions (even if the network is just composed of miners (due to hardware requirements)).

1) would certainly open a whole lot of new set of problems/attacs, but If REALLY carefully thought through would solve the problem more elegantly than 2).

An other point to 2): Instead of counting how many transactions a node relays (to determine if they are greedy), maybe it should count how much transactions fees they are relaying (just an idea...  but I don't know if that is better or worse, than counting the number)


Ok. It's good that we think about it, thanks to the guys from MS (yes I run Linux)!

Miners are transaction processors.  Mining is how they are verified.

I think this is the most important point. As a concept for a community, bitcoin has been better at making miners than at evangelists and transaction processors. I see this as a game that has made winners out of miners, but is missing players, now. This is a shortcoming of the system of incentives.

Everything costs something!

If bitcoin becomes fully fledged monetary system things like enthusiasm, altruism, philanthropy and philosophy will not be enough to secure its future! After all, it is about money! All tricks will be used both technically and economically to game the system and abuse it. Every expense that is possible to avoid paying will not be paid by participating agents in the network! There is no doubt about that... The big question here is whether all those avoidable expenses are fundamental incentives to keep bicoin network running?

+1

I also do have multiple clients on few servers running currently.

+1

+1

I leave bitcoind running as well, for a selfish non-altruistic reason: I hate waiting for the blockchain to update when I go to use it.  It's slower than hell if you have to wait, and a total non-burden if it stays running.  I selfishly disallow incoming connections, but would quickly reverse that if I heard echoes of people having trouble finding open nodes to connect to.

from the paper:  "We think that if Bitcoin really succeeds the costs of relaying transactions will be very high and that this may cause “regular” non-mining nodes not to relay information (except for sending out transactions that they themselves initiated)."

i disagree.  i leave 4 separate clients open and i don't even mine.  i do it for philosophical reasons to "help" the network.  doesn't cost me anything.

I don't know enough detail to form a proper opinion on this, but I can answer a personal question:
Yup. I always have it open. Never bothered to close it, really.
I do mine in a pool as well, but I believe that's unrelated.

Do people actually leave their Bitcoin clients open? I understand the problem for users who do this, but nobody is forcing people to do that. I'm willing to bet that when this problem is relevant (when we see 1000 transactions per second, compared to the 6000 per day we see now) the network will work FINE even if nobody leaves their clients open when they don't use them. The network would still have thousands of active nodes online every second, even if they're not active for long.

The issue with blockchain size is a real one but we already have lite clients for mobile phones etc.. there are a lot of possible solutions to tackle the problem of too large blockchain. And as long as the computing requirements of verifying transactions isn't too massive for regular PC's to handle, it's fine. If it's more than they can handle, then we have a problem.

You must be living in USA, UK or Australia.

Here, in central europe ALL wired broadband plans are unlimited and this is considered a standard. Of course mobile/wireless ones are always limited, but i guess that's normal & understandable.

It seems that in some things, "developing" countries are more advanced than "developed" countries. Curious.

I think he's talking about Network Neutrality. I am concerned that financial excuses may be used for censorship and exclude a great many people to internet resources. I don't think Bitcoin would use enough bandwidth to concern any ISP, but that wouldn't stop them from trying to block it. Fortunately, there are many ways for Bitcoin to adapt to get around the slow acting ISPs.

There's no incentive to OpenSource, Linux, Bitcoin.   

Not everything is paid to do.

I think we are already there.  Call your ISP and ask what service offerings they have for small office / home office.  You'll probably find that you can get access to any amount of bandwidth at a price.  Sure, you were probably expecting to consume unlimited bandwidth at max speed, pay the low residential rate and offer it to your neighbors, just like some kids go to daycare and assume access to lots of toys means the toys are all theirs.  Even under dialup, your phone company and ISP were free to enforce rules against constant connectivity, even if yours didn't.  Your ISP has got to make money to stay in business too, it's OK for them to set policies on shared resources.

I hope one day we can get back to paying for the bandwidth you want with the right to fully utilize it as well as resell a portion of it.  Dialup sucked for the speed, but the openness enforced by the regulations over phone lines is what allowed the rise of the Internet.  Bandwidth caps and government-enforced ISP level censorship will be it's downfall.