- From: Anders Rundgren <anders.rundgren.net@gmail.com>
- Date: Sat, 28 Feb 2015 17:44:53 +0100
- To: Jonathan Rosenne <jr@qsm.co.il>
- CC: "public-webpayments-comments@w3.org" <public-webpayments-comments@w3.org>
On 2015-02-28 17:13, Jonathan Rosenne wrote: > The issuer (the bank) authorizes the token service provider (e.g. Visa or MasterCard) to issue to a cardholder device (e.g. an iPhone) a token that is in effect a unique identifier ("card number" or "PAN"). The specification is publicly available at http://www.emvco.com/specifications.aspx?id=263. > > The authorization request does contain an unspecified cryptogram. > > It is worthwhile to note that the communication between the token and the token service provider is not necessarily trusted. Agreed. This picture shows more exactly what I'm talking about: http://webpki.org/papers/decentralized-payments.pdf I.e. the Amount, PIN, Card image etc. represent a virtual EMV terminal. A merchant web-site wouldn't be trusted for doing this, at least not in my vision. So using current web tech you would have to do this on a trusted web-site rather than locally. Which is why I'm trying to develop new technology that enables this. Best regards, Anders > > Best Regards, > > Jonathan Rosenne > > 054-4246522 > -----Original Message----- > From: Anders Rundgren [mailto:anders.rundgren.net@gmail.com] > Sent: Saturday, February 28, 2015 1:06 PM > To: Jonathan Rosenne > Cc: public-webpayments-comments@w3.org > Subject: Re: Running "Trusted Code" on the Web? > > On 2015-02-28 11:41, Jonathan Rosenne wrote: >> Are you aware of EmvCo Tokenisation and its use with ApplePay? > > Hi Jonathan, > > Exactly how everything works in detail I don't know; it isn't public. > BTW, this seems to be an issue with just about every payment system... > > Anyway, the principle seems to be that the card or the payment terminal creates a unique payment cryptogram which is given to the merchant. > > Whatever method used you depend on "Trusted Code" including PIN input. > > Without this element, your only option is strong authentication to a payment provider which does the EMV tokenization for you. > > Anders > > >> >> יונתן רוזן >> >> 054-4246522 >> >> >> ב-28 בפבר׳ 2015, בשעה 09:12, Anders Rundgren >> <anders.rundgren.net@gmail.com >> <mailto:anders.rundgren.net@gmail.com>> כתב/ה: >> >>> On 2015-02-28 03:30, Manu Sporny wrote: >>> <snip> >>>> What you're suggesting, though, is a demonstrable non-solution. >>> >>> We were addressing different uses of payment instruments. In my >>> example there's no magstrip only "cryptograms". Your example presumed card data. >>> >>> BTW, this actually raises an interesting question: How can you get >>> away from unsecure card data but still be able storing something "card-ish" in places like PayPal? >>> >>> A possibility is creating a cryptogram with the payment credential >>> which gives the specific party rights to withdraw money. Such a >>> system would not be sensitive to theft. >>> >>> >>>> What is >>>> wrong with the current proposal of "don't expose the sensitive >>>> payment data to any system held by the customer or the merchant?" >>>> Isn't this approach simpler than creating a secure execution >>>> environment for Web code (if that's even possible)? >>> >>> Both are needed. IMO. >>> >>> >>>> I think you're right in that we need something like web intents to >>>> transmit non-sensitive payment messages and cryptograms from >>>> browsers to native apps (like wallets) and back. >>> >>> Yes, that's what I work with, currently on a PoC level. An "early version" is available today: >>> http://www.cnet.com/news/google-paves-over-hole-left-by-chrome-plug-i >>> n-ban/ >>> >>> >>>> I'm not so sure we need a secure execution environment for Javascript code. >>> >>> If you are going to expose things like EMV keys to web-code the code >>> must (in order to pass EMV certification) be fixed in some way. I >>> started here but have given up this path not because it is undoable, >>> but because the number of things to be standardized could easily make >>> this a 10 year project and I don't think we need such. >>> >>> Anders >>> >>> >>> >
Received on Saturday, 28 February 2015 16:45:30 UTC