I've been nibbling around the edges of all the Stellar development going on, doing a bit of code and lots of reading. I'm wondering about community interest in a couple of potential things I was considering putting some time into.
1) a "modern" C++ library/SDK focusing on clean, safe, and simple(ish) interface using C++ improvements made up to and including C++17, and perhaps the leading, most solid pieces of C++20.
- I think the existing Qt/C++ library is great, if you are using Qt, but given new standard library and language features added to C++, I hope it would be feasible to create a "pure" C++ version.
- As I've been browsing projects, I note that the javascript and Swift versions look pretty clean. Is there any particular library that other devs feel really nails the interface? I wouldn't want to make C++ code that tries to hard to be something else, but given that, there is still a lot of room to honor the philosophy and style of the other libraries' strengths.
- if no one cares or finds it useful, that's a great thing to know too. My hope was a clean, portable, high level abstraction without unneeded extra cost for that abstraction, which could then be used for everything from desktop/mobile code to webassembly to server side processing. But perhaps C++ isn't in favor enough to bother?
2) I was looking at the vanity address generator and distributed untrusted worker concept discussed in other threads & mentioned in a prior SBC. I started working on building a website to implement it but realized that as written, I think there will be a usability problem with the generated keys.
The writeup suggests that the "miner" that finds the vanity key pair ends up removing their access to the account after adding the buyer's key to the account. This all is fine in terms of security and basic Stellar support, but I think most wallets aren't going to deal with it properly. You'll need the secret key (well, seed actually) in order to import the wallet, in most cases I've seen so far, if it's even possible (those that take only the word mnemonic key backups may just not be an option for these keys?).
That's not a huge deal - the seed can be passed to the buyer too, but when it comes time to use the account, many/most? wallets won't support using another key to do transactions for the imported account, will they? Seems like they all just plan on using the original seed for the account. You can do things manually of course, or semi-manually with a signer app, but still is a lot of extra friction.
I don't really have any proposal to address this - I guess I just want to know if my impression of the situation is correct.
My real interest is in remittance network partnerships and tying them together in order to serve the underbanked/cash oriented workers and businesses around, but I want to absorb more first ?
