Well, look at that, it confirmed. |
|
|
|
|
Ah, an opportunity for a shark. What premium would you be willing to pay to exchange your potentially worthless transaction for one with confirmed sources? Would you take 50%?
A couple months ago I tried a transaction, 74ccfd796cf2cb404a2245f699fd0e5cdf2a3a6b8a7b9cf1169ff2f87c124557, with no fee and it only took 11 days to confirm. I haven't tried it again more recently to be sure they do make it through still eventually. Bitfury was the miner that confirmed my 0-fee transaction.
Perhaps someone could look through the recent blocks for 0-fee transactions that made it through.
|
|
|
|
I suggest compromise or walk away. Prolonged contention is unlikely the best course.
Agreed, compromise is always the best option for both sides. Competition always has consequences for both. However, in this case, I think competition is essential, I think bitcoin should be segwit. And there's the rub; walking away means one "side" gets the legacy/brand and the other is an altcoin. Ok but what kind of a compromise do you propose and why do you think it is a good idea? I used to think the same way as you but I realized that it is easy to say "do a compromise", but when asked what kind of compromise I cannot answer or I realize I do not know much to comment. Totally agree; saying compromise is indeed the easy part; creating it is the hard work. In this particular case it will start with both sides agreeing to something important and then each side giving in enough to be meaningful. For example, if both sides can agree to something like this; "It is important to keep things decentralized." If there is no agreement to something important then walk away. If there is true agreement then proceed to the next step; each side gives up something meaningful, e.g.; 1) All will promote/encourage multiple independent development teams. One dominate/central development team goes against our agreed to guiding principle. 2) All will agree to some small increment in the protocol at first and then measure/observe the impact and move towards results that work best toward decentralization. This is just an example. The actual negotiations would likely end up somewhere else. When both sides feel heard and respected then there is the possibility of compromise. Even in a walk away situation, it should be possible to negotiate inheritance of the legacy/brand; perhaps something along the lines of ETH/ETC. |
|
|
|
Segwit is the compromise. How can you not understand this?
Saying a thing doesn't make it so. We can pretty much tell if it is a compromise when both sides agree to it. |
|
|
|
I suggest compromise or walk away. Prolonged contention is unlikely the best course.
Agreed, compromise is always the best option for both sides. Competition always has consequences for both. However, in this case, I think competition is essential, I think bitcoin should be segwit. And there's the rub; walking away means one "side" gets the legacy/brand and the other is an altcoin. |
|
|
|
|
I suggest compromise or walk away. Prolonged contention is unlikely the best course.
|
|
|
|
|
What's the default? Will my full node prune and/or not store witness data? How can I tell?
Even 3rd parties that store privkeys for folks (which I don't like either but it is what it is) need access to a well-connected full node, right?
|
|
|
|
I run a non-mining full node; it entertains me to do so. I configure it to allow 60 links (8 outbound (default) and 52 inbound). Although it does vary, I do find my node runs along with near the maximum number of links all the time. Well, after a restart, sometimes it can take many hours to build back up.
Although the miners could (and probably do otherwise how do they get transactions?) run a full node (or more than one for redundancy?), there's nothing obliging them to accept many incoming links. Wouldn't that leave users fighting for limited connections without folks like me? Should I increase my link count even more? I configure my mobile phone based wallet to only connect to my full node exclusively. Sometimes I can't get a link so I go to my full node a disconnect a peer (sorry).
So far I have plenty of bandwidth, CPU, and storage. I guess I would be ok with a bigger block but would be very happy with a cap on transaction size. Am I helping at all to keep things decentralized?
If I drop off then it might not be such a big deal but if other folks like me do then what?
most pools connect via things like fibre or supernodes as it used to be called and let the fibre/supernodes propogate the data out to all the other nodes thus taking pressure off the pools from getting huge demand for direct connections from random users. as for the random users and merchants that build up the symbiotic relationship of the diverse decentralise per network that keep pools in line and ach other inline. thats more of a question of the 8 dgree's of separation. for instance if instead of 52 connections there was only 8. if everyone had only 8 connections 8*8*8*8=4096 the data would not propogate to everyone in 4 hops/relays(based on bitnode count ~7000) 8*8*8*8*8=~32k the data would propogate to everyone in 5 hops/relays 10*10*10*10=10k the data would propogate to everyone in 4 hops/relays 20*20*20=8k the data would propogate to everyone in 3 hops/relays 84*84=7056 the data would propogate to everyone in 2 hops/relays so ill leave you to rationalise if you should step up and be more of a super node by going upto 84 as a nice healthy 2 relay number. P.S if your going to want to help the network sync. doont use prunned/no witness features. otherwise people cant grab full data from you. if you want to use prunned/no witness. then just let 8 connections and be at the bottom end cesspit of nodes that cant reliably sync with each other Thank you! I fully intend on hosting the full blockchain, i.e. no pruning. I will watch my bandwidth, CPU, and storage consumption but my guess is I will be ok. The connectedness of the full nodes is one thing; I am wondering about the non-full nodes too. Having a place for them to connect to is important, right? |
|
|
|
|
I run a non-mining full node; it entertains me to do so. I configure it to allow 60 links (8 outbound (default) and 52 inbound). Although it does vary, I do find my node runs along with near the maximum number of links all the time. Well, after a restart, sometimes it can take many hours to build back up.
Although the miners could (and probably do otherwise how do they get transactions?) run a full node (or more than one for redundancy?), there's nothing obliging them to accept many incoming links. Wouldn't that leave users fighting for limited connections without folks like me? Should I increase my link count even more? I configure my mobile phone based wallet to only connect to my full node exclusively. Sometimes I can't get a link so I go to my full node a disconnect a peer (sorry).
So far I have plenty of bandwidth, CPU, and storage. I guess I would be ok with a bigger block but would be very happy with a cap on transaction size. Am I helping at all to keep things decentralized?
If I drop off then it might not be such a big deal but if other folks like me do then what?
|
|
|
|
|
Folks have at least two views of scaling.
The one I am use to is where a function is measured to get work done at some rate and then after some change/improvement the function can then get more work done at a higher rate. For example, a truck might be able to haul so many tons then some bright person gets the idea to hook a second trailer behind the first and double the rate. Scaling like this comes at a price, e.g. miles per gallon, etc. Scaling like this will have some physical limit, e.g. it may become dangerous to keep on attaching more and more trailers, etc.
Another view of scaling involves making a more subtle sort of change/improvement. Instead of just increasing the capacity, the idea is to arrange to do the work with less effort. Using our example, some bright person reengineers the things being hauled to be much smaller and lighter and so many more fit into the first trailer.
Nothing precludes taking advantage of both kinds of scaling.
I have seen occasions where folks dismiss the first sort of scaling although I have yet to find out what they term it.
|
|
|
|
|
Fungible money: When one cannot determine any difference between instances of money. For example, we each have a dollar bill, we swap and no one can tell. But, if the serial numbers were recorded then it can be determined which bill is which. A clean crisp new bill is worth more than an old raggedy bill in some/many places.
Bitcoin is an interesting mixture of really good fungibility and not even close to fungible. If a miner takes care then they can create a portion of Bitcoin pretty much out of thin air with virtually no traceability; without care in theory it can be traced. Each time bitcoins are transacted they tend to become more and more traceable.
I heard once that if you have a US $100 bill in your possession it is very likely to have been used in an illegal drug transaction. In fact, it is quite likely to have trace amounts of cocaine on it. I have not confirmed this but it sounds plausible. If so then arguably you are in possession of something that endangers you. It would behoove you to wash such clean.
Some altcoins are superior to Bitcoin in terms of fungibility, although they might come with other issues.
|
|
|
|
|
Ok, then what shall we do?
|
|
|
|
|
Compromise or watch someone else eat our lunch. Chop Core into bite-sized pieces and find the right subset that appease BU et al.
Miners; sorry, but one way or another you are eventually going to lose your grip.
|
|
|
|
|
One can imagine increasing the block size limit from 1MB to 2MB (or any other value) whilst capping the size of transactions to say 1MB.
|
|
|
|
Fairly recently I got a 0-fee transactions through but it did take 11 days. Your mileage may vary. Past performance is not a reliable indicator of future behavior. Be prepared to recover if your transaction doesn't ever go through.
On my knowledge the mempool times out earlier than 11 days. Assuming you needed to restart your wallet using zapwallettxes=1 and sending same amount again? But @OP "yes" you can send without any fees. It's just a ballot if those reach their destination, at all. Just adding 1 Satoshi to each byte works marvels. I did not have to zap and resend. The original transaction went through without any help. I was comfortable with 11 days; many/most people might not be. |
|
|
|
Fairly recently I got a 0-fee transactions through but it did take 11 days. Your mileage may vary. Past performance is not a reliable indicator of future behavior. Be prepared to recover if your transaction doesn't ever go through. I watch https://bitcoinfees.21.co/ and until very recently there were about 1500 0-fee transactions per day going through. Now it looks like there are none in the memory pool and a declining number going through. Maybe there's another site reporting differently? |
|
|
|
|
Have we got your term yet for improving processing?
|
|
|
|
|
If I were a miner (too expensive for my taste) then I would configure my mining software to mine only 0-fee transactions unless it didn't fill a block in which case then I would include the lowest fee transactions until it filled a block up to the limit allowed. I am not most people.
If the developers would develop a less expensive way to be a miner then I would appreciate it. Perhaps they could let miners like me win ever so often (every tenth block?) if we produce a block full of 0-fee transactions?
|
|
|
|
|
Show Posts
