No you don't have to have exact denoms. If you want to pay 19.9 you will just send 100 and get 80.1 back as a change. What would happen next is that you'd denominate that 80.1 to 8x10 + 0.1, and then start anonymizing that. The process could be even streamlined so that all those transactions would be made at the same time, i.e. 19.9 + 8x10 + 0.1Denomination to 0.1 might be a good temporary solution actually. And I would suggest to pay everything smaller than 0.1 as a tx fee - this will be the cost of the bloat. We can always shift it when price goes up. And we were talking about eliminating 500 and 100 denoms anyway...
The only problem is user have to have enough of these denoms to combine into right amount. If you have 100 DRK anonymized and you want to pay 19.9 you'll need 1x10, 9x1 and 9x0.1. And every time you pay you need to have full set of smaller denoms. So for 5 payments by 19.9 you'll need 5x10, 45x1, 45x0.1. Now imagine you want to pay by 9.9 instead... When you do it ahead-of-time you never know how many of each denoms you'll actually need.
Hmmm... We'd better ask Evan eduffield I guess I'm not sure how this could be presented in UI though.Is it possible to allow users some control over this? If most of what I buy costs a few DRK, I'd rather have a lot of 1's, not so much of the 10's and 100's...
Choose amount of DRK to anonymize: 10 1 0.1 [ 4 ] [ 14 ] [ 6 ] You will anonymize 54.6 DRK
I agree with Crouton that a bandaid can be placed on this issue byI have no idea what else Evan is currently working on, but I'd rather get on with IX and other stuff than spend weeks overanylising this issue. Denominate down to 0.1 or 0.01, bloat be damned, we have time to deal with it if it ever gets to be a problem. What merchants are going to be selling anything in units of less than 0.01 anyway?
Stick a warning popup in the client if a user tries to spend linkable change. Maybe make the auto input choosing a little bit smarter. Move on.
edit: Also, what moli said - make the coin control thing more obvious, make it a routine part of the process. You open your guvpaper wallet, you choose which notes and coins to take out, same thing.
No I wasn't referring to the example. I was talking about using the coin control in spending my real money.No you haven't, you spent inputs that were not anon. They were inputs that could be linked to earlier transactions. If you're referring to the example you posted earlier that is.
Few corrections:I agree with Crouton that a bandaid can be placed on this issue by
However, even if this is not Even's most pressing item, it is worth a discussion to look for a solid long term solution. Definitely worth discussing! If we come up with something that sounds like a solid solution, we can present it for Even to consider on a future revision. No?
- lowering the denominate a little
- Setting aside "dead change" so that the user can have a clear understanding that it should be kept separate unless they choose to compromise anonymity.
I agree - strong privacy feature is a must here. But at the same time it looks like we want to find free solution while we can solve it at max price of $2.37 * 0.1 = 24c (at current price) per anonymous transaction. And thinking in that way... It makes me wonder - does it really even worth do discuss?... Is there any other solution that can give you comparable price-feature ratio?As far as I'm concerned, this is more of an issue and pressing than testing instantx. The core drive of this coin is privacy and right now, change back under 1 and reused harms us. Make sure the core feature works first, then build on it. I realize instantx is done but we aren't just marketing a fast confirm coin... we are marketing a privacy-centric coin. Don't ever forget that. The 0.10 denomination cleans up a lot of this problem but it still leaves an issue with what amounts to being around 5c or less in value (sub 0.10).
Exactly, and that isn't to say we can't re-evaluate denominations, perhaps to 0.025 or 0.01 when DRK is substantially higher so costs are always kept low. Sending a bunch of money with up to 0.10 worth of "fees" privately is hardly worth getting panties in a bunch over. This seems to make the most sense given the situation, risk:reward looks favorable and if we suck it up and change under 0.10 goes to the network, we never have to worry about change linking tx's again. Yes, the blockchain will bloat a bit but there are a few viable options for pruning and it's a far easier problem to solve IMHO than comprising privacy. The blockchain, especially dated blocks containing really old tx's, serves no purpose when the majority are going to unlinkable transactions of similar denominations. It's all fog.I agree - strong privacy feature is a must here. But at the same time it looks like we want to find free solution while we can solve it at max price of $2.37 * 0.1 = 24c (at current price) per anonymous transaction. And thinking in that way... It makes me wonder - does it really even worth do discuss?... Is there any other solution that can give you comparable price-feature ratio?
I understand the sentiment, but I worry about going down the slippery slope of mandatory donations even if the cause is worthy.Dropping it to 0.01 would be a 10x easier sell, but...
Miners and MN ops already have income, devs deserve pay too, send it to the Foundation... 1000 tx's a day @ 0.1 is $240... lets put some worms in Evan's beak (+flare, + UdjinM6, + others) for a change*... :tongue:
Coins to the network, anything else will be spun in a negative light.I understand the sentiment, but I worry about going down the slippery slope of mandatory donations even if the cause is worthy.
I don't think most will mind the transaction fee idea as there is a way to get their coins back, but if we do go with mandatory donations it might be best to do a poll or something to really be sure there will not be a community "revolt" somewhere down the line.
This. Most concisely stated. Temporary, but immediate solution. I support this idea!The fact mixing is free now means the anon tech is cheap. If we lowered the minimum denom amount to .01 and gave the leftover of anon transactions to the miners as tx fee, it would be like incurring a tx fee for darksent coins, the maximum of which would be .00999999~ drk (about $.023). We could test around to see how much bloat it would create and if it was manageable come back to the whole thing later.
I had (more or less) the same idea yesterday and created a new thread for this because it's not about change contracts anymore...Allow merchants to price at ~1 cent granularity, i.e. add 0.1 AND 0.01 denominations, and then trucate any excess and send it to the miners/MNs as a transaction fee automatically. As the price increases, then the denomination levels can be adjusted to maintain ~1 cent granularity.
As far as I'm concerned, this is more of an issue and pressing than testing instantx. The core drive of this coin is privacy and right now, change back under 1 and reused harms us. Make sure the core feature works first, then build on it. I realize instantx is done but we aren't just marketing a fast confirm coin... we are marketing a privacy-centric coin. Don't ever forget that. The 0.10 denomination cleans up a lot of this problem but it still leaves an issue with what amounts to being around 5c or less in value (sub 0.10). Do I dare say rid the random ds fee of 0.10 and use change 0.0999999999 and less as the "fee". It's still "cheap" for privacy.
The order should be (agree or disagree):
Why would you try to bring merchant adoption on a privacy-centric coin that currently is "broken" due to a flaw with change. Further, the road ahead for any sort of mainstream adoption is going to long and hard, simply due to the fact that the name has negative connotations to the average joe. Bitcoin's immediate success was built in the darknet markets for years before it really started getting exposure elsewhere. Even then, bitcoin's name was in a far better positioning than darkcoin's. What will most likely need to happen is a specialized, rebranded wallet that basically serves as a mask for the underlying darkcoin functionality.+1 on that too. But Evan never said it was a priority. Just in the roadmap. I propose a different scenario.
At first, MN were supposed to be the holy grail of DRK rise with the positive feedback loop philosophy. That ship has sailed. To gain momentum, I believe DRK must capitalise itself in the crypto markets ASAP. Darksend works well, and gets better with every iteration. IP obfuscation is good, but not entirely necessary as of this time. If a VPS has an issue, its too simple and fast to setup a server elsewhere. Legal implications? Too soon for that IMHO, but in any case, set it off-shore and done deal.
InstantTX to me far exceeds the anonymity part right now. Anonymity is niche, despite the humungous capital out there, be it micro or macro transactions. But InstantTX is on a different level. If it proves to be totally stable and reliable, it completely nukes every remittance/merchant platform out there, as well as fast paced financial instruments and all.
D+ will for long be associated with darknet markets and illegal activities (media FUD et al.). Gambling and the sex industry. Who would ever buy a product off Amazon, say, needing anonymity?
Widespread merchant adoption is in serious need. And recent polls favout BTC over Apple Pay. But the +/- 1h BTC confirmation time totally devastates the micro-payment cosmos. DRK could seriously make a dent here.
So I would say;
- IP obfuscation
Bingo, the push towards saying DRK is anonymous must live up to those expectations. It will fall flat on its face to still have problematic areas surrounding that feature. The official stance of the team should be against illegal activity but lets face it, everyone would like to see DRK being the standard currency on Agora or Evolution. It only further solidifies its utility. Quite frankly, the currency that becomes the new standard on those sites and any other large darkmarket website will ultimately be the one that wins the privacy-centric coin space. That is the real testing ground for utility. Show that it really can serve its first purpose and then push for further adoption.I almost agree with yidakee because I'd love to see InstantX in action, too. But thinking more about it, right now Darkcoin is actually much faster than Bitcoin, but why are so many bitcoiners still loyal to bitcoin? Darkcoin has been advertised as the best anonymous coin... and then the flaws are discovered... We need to fix our problems ASAP... Plus Evan is going to appear on the online crypto Expo in a few days, he needs to get things in order before being quizzed by his adversaries...
Yidakee, you've been away from DRK too long. This is the current issue:oblox, I totally agree with you. When I mentioned merchant adoption, I wasnt referring to change contracts, but DRK in general. When in comes to money, no one gives a damn what its called as long as they hear *catching* in their wallets. Sure, Darkcoin will have a bad connotation for long, but names gain a new meaning with time, intent and use. Like Obama was thrashed as an infiltrated muslim terrorist that would NEVER win because of the name. There is a tobacco brand called Death, and people buy and smoke it (no joke). Money talks, bullshit walks they say..
There is no flaw in D+, it is totally anonymous. The process of anonymising them, that is a big buggy yes. The coin is not even 1 year old. I suspect InstantTX is a much simpler feature to implement that the complexity of denomination. We absolutely have nailed the spot as the nº1 anonymous coin already, why not nail the nº1 for instant TX?
Oh my... yes I was away during that period. Wooops. I had no idea. Need to read that carefully. Thanks moli.Yidakee, you've been away from DRK too long. This is the current issue:
and Evan also fixed another issue that Aswan pointed out, but there's still something to do with the large denoms.
LOL.. No problem.. As of now, if you blink and miss a few pages on DCT... you really have no idea what's being discussed, like what happened to me... I thought the change issue was if you use your change left from a previous DS transaction, the change would lead right back to your origin address.. So yesterday I did some testing and found that wasn't so... I posted here and people were like "what are you doing moli, what are you talking about, why did you do this..." hahahah.. Found out that wasn't exactly the issue, just because I didn't read everything that Aswan and everybody posted... tl,dr... lolOh my... yes I was away during that period. Wooops. I had no idea. Need to read that carefully. Thanks moli.
No, it isn't. The dead change issue removes anonymity linking future tx's to past tx's. IT NEEDS TO BE FIXED. I see instantx testing is out now but this issue needs to be fixed immediately.There is no flaw in D+, it is totally anonymous. The process of anonymising them, that is a big buggy yes. The coin is not even 1 year old. I suspect InstantTX is a much simpler feature to implement that the complexity of denomination. We absolutely have nailed the spot as the nº1 anonymous coin already, why not nail the nº1 for instant TX?