- From: Andrew Bransford Brown <andrewbb@gmail.com>
- Date: Tue, 20 Aug 2019 00:51:12 +0700
- To: Melvin Carvalho <melvincarvalho@gmail.com>
- Cc: Michiel de Jong <michiel@unhosted.org>, Web Payments <public-webpayments@w3.org>
- Message-ID: <CAPS+YFLcgEkA+CfgONZFtaRshK1kA8fOpzM8StLiV7VowOHf2A@mail.gmail.com>
Could you use a GUID? Send the GUID to the user (email?). The user can check the balance. The user can spend it. I suppose the GUID could be stolen, but the user would complain. On Mon, Aug 19, 2019 at 1:49 PM Melvin Carvalho <melvincarvalho@gmail.com> wrote: > > > On Mon, 19 Aug 2019 at 08:38, Michiel de Jong <michiel@unhosted.org> > wrote: > >> Hi Melvin, >> >> Great topic! I like how the scheme is very generic, but maybe at the same >> time that's a downside, because how would you dereference >> 'urn:voucher:12345abcd ...'? >> > > urns are typically not derefencable but you could write a spec to > dereference it at /.well-known/voucher/ID if you wanted to > > and then put it here : > https://www.iana.org/assignments/urn-namespaces/urn-namespaces.xhtml > > >> Maybe start by base64-decoding it? >> > > I didnt specify an coding, tho it was perhaps implied. This is something > it could be neat to spec. Right now it's just a string. Should it be > unicode I wondered? > > >> But what would you see then, and how would that refer to a party who is >> willing to "cash" the voucher? >> > > The party cashing is orthogonal, in this case you are overloading the > identifier to be both an ID and a shared secret. A capability URI as such, > but without the HTTP. > > > https://www.slideshare.net/dappelquist/capability-ur-ls-for-london-web-standards > > Bitcoin does this too, but overloading IDs and public key hashes. > > >> There could be some indication of some account identifier at some ledger, >> but for that, you would need some more mechanics than just the opaque URI >> scheme. An interesting approach to that problem is Interledger addresses, >> for instance. >> > > Im starting with just a simple csv or JSON which will be like the "1 star > of payments", easy enough to get up and running. > > >> >> I would say there are generally two types of vouchers, relational (where >> the issuer has some social connection to the redeemer) and anonymous >> > > I think im starting out here with anonymous by implication. > > >> (where the voucher has a more universal value, against some anonymous >> "bubble"). If you're interested in peer-to-peer vouchers rather than >> anonymous ones, then may I take this opportunity to plug the Network Money >> mailing list I started last year, particularly this post in which I >> concluded that maybe peer-to-peer money is in the end not really what >> people want: >> https://groups.google.com/forum/#!topic/network-money/Z2zAyX1R8Xo. >> > > Thanks, already a member > > >> >> Cheers, >> Michiel. >> >> On Sun, Aug 18, 2019 at 5:02 PM Melvin Carvalho <melvincarvalho@gmail.com> >> wrote: >> >>> I have written a payment server that can use arbitrarily many >>> authentication methods on the web. >>> >>> The outcome of that authentication is to return a verified URI. You >>> could think of it as a super set of WebID, DID, user addresses and so on. >>> >>> One thing I'd like to do is have a voucher system. So the idea with a >>> voucher is that it has a special code, say you email it to someone, or have >>> a scratch card or something. >>> >>> Then when that code is shown the back end is able to let the user spend >>> whatever balance it is for. So it's a long the lines of a voucher, a >>> shared secret or a one time password. >>> >>> This may be similar to a bearer token, im not sure, as Im not so >>> familiar with those. >>> >>> My question in all this is, given that I need a URI that is linked to >>> the voucher. Is there something existing I can use. Or, is there some >>> sensible standard we can start experimenting with. >>> >>> The idea I had was to use the URI >>> >>> *urn:voucher:12345abcd ...* >>> >>> And if that appears in the request you know. the user can spend the >>> voucher, and that allows me to build an app. >>> >>> Any thoughts, ideas or previous work that can be reused here? >>> >>
Received on Monday, 19 August 2019 17:52:37 UTC