alvatar
commented
7 years ago All good, althought some details about number 4.Fees can go to 2 different wallets of which you can have exclusive control of one. However making an interface to make the transient wallets and escrows accessible to you without going through the programming interface is a more involved development undertaking (over 13 points). In any case, if a few users show interest in this we should move to real Escrow anyway, so none of us would have the ability to release any of the funds.
cosmicbro
Does this seem right?
0) Turn on live money & do a test. If it works, celebrate!!!
1) Document all the screens (would love to do this together)
2) Confirm we can reach all the screens (again, doing this together makes sense)
3) Setup some QA outsource (like rainforest?) to do our automated testing every time we make a change. (Note, this will probably take me like 10 - 20ish hours to setup initially and create training / support materials for them, go from step 1 (video, photos, etc) to step 2, etc.) so would be helpful for us to get through steps 1 & 2 together to make sure I don't miss anything.) How many independent tests should we have them run? I assume if they can get to each screen consistently 10+ times we're probably good to test with live users? What are your thoughts here?
4) Create key backups for CT wallets. Both of us should be able to access whatever wallet(s) our service controls and we should have paper / trezor backups stored securely as a redundancy. As we scale our staff we'll probably want to do an M of N signing scheme but for now we're probably ok with both having access to all wallets.
5) Start doing live testing with people who aren't us. I'll start small at first, will try to get through like 10 successful live money transactions before making an announcement. Then the plan will be to progressively do growth experiments.
Huzzah!
Lets launch this thing!!!