T-address support reminds me of the web before Firesheep happened. Most websites supported only HTTP and used HTTPS selectively for payments and password input forms. Arguments were made that that HTTPS’s cryptography was to costly to implement for all connections. Most web users didn’t understand the distinction and Firesheep made it really simple to hack into other people’s accounts when they were on the same open WiFi network.
The situation is similar for Zcash. Most users don’t understand the distinction. Some of my friends have unknowingly put themselves at risk because they assumed “Zcash is private, right?” Those friends are knowledgeable security engineers, who are fully capable of understanding the protocol’s nuances. I have to assume the situation is even worse for everyone else.
Whether shielded adoption is hindered by the fact there are no libraries that make it quick, simple, and secure to set up shielded support in common platforms, or by regulatory fears, these arguments remind me of the ones that were made for only using HTTPS on password forms. We are forced to explain to users that to be secure they need to maintain a shielded balance and spend to t-addrs from there, which is reminiscent of telling people to “look for the lock” whenever they log in or enter their credit card information. Users shouldn’t have to think about it.
If Zcash adoption is contingent upon transparency—whether because of regulatory fears or poor shielded UX—that’s not real adoption. It doesn’t count. If someone is only integrating Zcash because they can re-use their existing bitcoin libraries, one of three things is probably true (a) they don’t value Zcash enough to do the work to integrate shielded, (b) they are afraid to integrate shielded, or (c) integrating shielded is simply too expensive. In cases (a) and (b), they are not really a Zcash adopter—they are not aligned with our mission. In case (c), that’s on us for not making it easy.
I don’t think the solution is to delete t-addrs from the protocol just yet, we need to first fix the usability of shielded, making it as easy to integrate shielded payments into anything as it is to integrate PayPal into anything, and then market Zcash to form a solid ecosystem of shielded adopters. Then we can delete t-addrs because nobody will be using them. IMHO, all future protocol changes should be in service of these goals. Everything else like getting rid of the trusted setup, ZSAs, PoS—while valuable—are opportunity costs we can’t afford when we need to be 100% aligned on making the technology eminently usable and establishing solid niches of adoption.