Aims of OID4VP payment support

[jogu]

Is there a nice introduction to the goals of EU wallet based payment system somewhere?

I'm really struggling to get my head around the details of the problem it is trying to solve.

In particular what are the advantages of something like https://github.com/digitallabor-berlin/eudiw-sca/blob/ea5db12b59f583f79e3c866896565fa6c93ae2e4/openbanking-r2s.md#payment compared to just letting wallets get standard OAuth2 based tokens that let them initiate payments via OpenBanking APIs? e.g. Is it just that it shifts some liability onto the payee?

[jogu]

The WG call on 8th August will have some payments people presenting their use cases which will hopefully cover this.

[cyberphone]

@jogu There are tons of issues here, but liability shift is not one of them. A non-exhausting list:

• Considerably simpler wallet.
• Improved reliability through stateless operation.
• Improved UX. User conversation is between the Payee, leaving out the ASPSP and PISP.
• Potentially enabling NFC payments without requiring mobile data connectivity (like payment cards).
• Payment network neutral. Supporting VISA/MasterCard/EPI may be important.
• Enables the Payee to chose PSP.
• Deals with multi-phase payments in a more natural way than VRP.
• Opens the path to P2P payments.
• Enables intrinsic support for account balance data
• Competitive with Apple/Google pay.

Last but not least, potentially enables a more decentralized authorization concept which also removes the current single point of failure:

The primary drawback is that you need new Open Banking APIs.

[cyberphone]

An even bigger problem than the wallet and associated flows, is the monolithic nature of current Open Banking architectures, making new or revised APIs daunting, expensive, and time-consuming. It also makes testing and development using "sandboxes" impossible. The Berlin Group's "Signed Payment Request", intended to be used with DigitalLabor's wallet is an empty shell. The need for unique and proprietary implementations in every bank, is a guarantee for extended lead-times and limited "enthusiasm". In addition, the EU banks have already poured over € 1BN into EPI. @sakimura

[Sakurann]

@jogu could we close this issue with the presentation on the payments use-case that seemed to convince the WG of the need for this use-case? https://lists.openid.net/pipermail/openid-specs-digital-credentials-protocols/Week-of-Mon-20240812/000423.html

[jogu]

Yes, as the reporter I'm happy for this to be closed so will close it now.