I think we should also keep in mind programmatic usage of memos, like the āSavings circleā app Iām writing right now that needs to download all memos without user interaction because they are programmatic.
Also! Kudos again to Nighthawk! Maybe we shouldnāt nerd out on memos in this thread because it takes away from their accomplishment?
Are you sure about this? I mean I agree you canāt just scan everything scalably, once you hit some number to txs per second, but how about adding a field to block headers that would let you know if a tx to you was in the last x number of blocks? I think this could be done in a way that still preserves your privacy about your tx being in those blocks.
Thinking about this some more, Iām beginning to see why itās a hard problemā¦
Zcash came out of what ended up being my PhD thesis. I started on that in part because anonymous messaging seemed hard. It had been around a long time and people hadnāt solved it. Private decentralized payments were hard too (they foiled Satoshi ), but ā¦ easier than that. Its been nearly 10 years since I made that choice and the best practical deployed metadata resistant (which is weaker than anonymous) system anyone has is Signalās and their approach is basically have a server and donāt keep logs.
Great thread, keep it going.
@gmale thanks for sharing the background. It is hard to debug & find root cause without logging & crash reports.
Yes! I didnāt want folks making a big deal out of it. If the open source ECC wallet doesnāt use those analytics libraries, then it is worth removing those dependencies.
Iām confident that we could deploy (or reuse) a suitable mixnet if we decided that we needed one. And I do take the point youāve made elsewhere about needing all associated communication to also be anonymous.
It took 2+ years to ship a wallet, over a year to do sapling. And those are things where the academic underpinnings were settled years ago. So even if there was an existing solution that met our needs, had been designed, fully vetted, and worked, it would be a major challenge when most of our attention is focused elsewhere. In reality, almost none of those things hold. Iām not joking when I said I started the whole line of research that underpins Zcash because I thought it was easier than anonymous communication and ~10 years later I still do.
Thereās a reason Iām arguing strenuously for fixing the problem now by either 1) not fetching by default or 2) adding that as an option and defaulting to fetching all memos.
Can you elaborate a bit on option 2? Because, to me, that sounds about the same as putting it back into the compact block, which is considerable bandwidth but maybe worth it for some users.
Personally, Iām really a fan of āoption 3ā that you suggested elsewhere of iterating on payment detection. As a wallet dev, pulling down all these blocks and doing all this scanning and managing the associated state is overly cumbersome.
There ā¦ has to . . . be . . . a better way!
Oh yeah, that is the same thing, mostly. Though you might want to let someone who never gets memos turn it off for bandwidth reasons. And if we get much more z2z usage, bandwidth on mobile would be an issue. Sticking it in the compact block format could make that tricky, I think? If not, totally the same.
And Iām all for iterating after we have a stop gap fix. The possible stop gaps are 1) download all memos (either in compact block or separately) 2) default to not fetching memos. Oh, 3) change the wallet protocol to accept singed flat files for blocks . Then you can fetch from services that plausibly donāt log like CDNs or do it p2p.
The current situation is not acceptable. And as I keep pointing out, having been working on this for 10 years, it takes longer than people think to do anything fancy. So we shouldnāt wait
Good idea. paywithz.cash now has a filter carousel at the top so users can select from stores/non-profits that have a z-address. Thereās also a dedicated page for this feature at https://paywithz.cash/z2z
Perhaps I agree with you, if you look at the current situation, it turns out that time is playing in the wrong direction, the longer you wait, the more chances that everything will change.
Iām happy to see this great work and the discussion around it.
However - itās absolutely wrong to use a killing machine as a symbol.
@NighthawkApps Is the āShielded Userā screen with the big yellow spaceship logo an āAbout usā screen? It appears so, with the yellow at Nighthawk button. Have you considered adding āsendā and āreceiveā buttons? Itās not clear to me how to send and receive Zcash.
(Itās not a spaceship itās an F-117 Stealth fighter)
I am personally not seeing a ākilling machineā as a symbol of this wallet software. I am seeing a ādefence machineā instead. To me it is bringing an image of something that is protecting me.
. . . and, perhaps most importantly, the concept of being āoff the radar!ā
Iām with @mika here, and Iād like to thank her for pointing it out. Thatās a machine for bombing civilians.
Lets not do this. Names aside, 80s stealth tech is a cool aesthetic for a maximum privacy wallet and better than the green cyberpunk generics someone might have done.
The attribute the F117 is being used for here is stealth, not its ability to drop bombs. Itās a tool.
And finally, it was designed to hit heavily defended military targets, primarily in a conventional ground war in Europe in which the USSR invaded. Itās not meant to bomb civilians nor does it have the payload or range. It was not nuclear capable. So it was used to take out command and control bunkers in e.g. the intervention in Serbia to stop genocide.
Of course, thereās ambiguity in this. In the first gulf war the US accidentally bombed a civilian air raid bunker they thought was a command and control site. But I wouldnāt be so eager to set the standard of āworst possible useā defining something, seeing as we all work on a privacy preserving cryptocurrency and, however rare, it will be used for evil.
You can disagree with the name and are free not to use it, but lets not pile on other people for choosing it or liking it.
Personally I think name is synonymous with āstealthā or āskunk worksā moreso than it being a military weapon.
