W3C

Payment Apps Task Force

30 Nov 2016

Agenda

See also: IRC log

Attendees

Present
Rouslan, Jenan, Alan, Ian, Tommy, Adam
Regrets
Conor, Jason, AdrianHB
Chair
Ian
Scribe
Ian

Contents


WebEx phone gateway is down for now

So you need computer audio I think

Next meeting

7 Dec

Tuesdays, 10am ET or 11am ET (for 1 hour)

Should we change to that time?

<rouslan> +1 either

Tommy: -1
... But I can live with it

<adamR> +1 either

https://github.com/w3c/webpayments/wiki/PaymentApp_Notes

Recent spec edits

WebIDL fix

https://github.com/w3c/webpayments-payment-apps-api/pull/71/commits/738e08739133a6e52482db7446b473ad7fe2f64b

Tommy edits regarding cancelation and failure handling:

https://github.com/w3c/webpayments-payment-apps-api/commit/9dcf7d251d1c1f00b13825f96123d5dcc124e105#diff-eacf331f0ffc35d4b482f1d15a887d3b

Tommy: Main thing I did was to say that the rejection of the promise is not necessarily fatal.
... the user agenda can decide what to do next.

Issue 69: What does Icon mean?

https://github.com/w3c/webpayments-payment-apps-api/issues/69

Proposal:

1) We refer to web app manifest rather than roll our own

2) We decide how much of that spec to reference

<Zakim> rouslan, you wanted to talk about icon

rouslan: I think you expressed my view - we should reference web app manifest
... and follow their examples.
... this will benefit implementers

https://www.w3.org/TR/appmanifest/#manifest-and-its-members

IJ: Do we need all the manifest members?

rouslan: That's a hard question to answer here; I think we should continue discussion on GitHub

http://caniuse.com/#search=manifest

AdamR: Marcos may have an opinion on referencing it (re: stability)

IJ: Should I update the spec to say "we are looking at referencing web app manifest"

AdamR: +1 to reusing the two names from that spec, and leaving a note to say we are looking into importing or referencing web app manifest
... please include in examples so people can understand

<scribe> ACTION: Ian to update the manifest information with pointer to this issue and updated member names [recorded in http://www.w3.org/2016/11/30-apps-minutes.html#action01]

Filtering

previous action - AdamR to take a stab at writing up description of payment app function to return boolean if filter matches; see

https://www.w3.org/2016/11/02-apps-minutes.html#item03

AdamR: It was not clear from our last chat was whether we want to define a lambda-based mechanism, or if we want an algorithm for matching capabilities of payment apps with filters described by payment requests

rouslan: On declarative v. procedural....
... data v. function
... I prefer the lambda approach
... I think Adam wanted to be sure that the function could not access the network...I'm not sure how easy that would be to enforce
... what would be good as well is a timeout
... e.g., 400ms

adamR: Will require some time to discuss; find out how sandboxes are described elsewhere

IJ: EME

https://www.w3.org/TR/secure-contexts/

https://www.w3.org/TR/secure-contexts/#monkey-patching-sandbox-flags

https://www.w3.org/TR/html51/browsers.html#parse-the-sandboxing-directive

AdamR: There's a chance that this is unique and will get odd glances

IJ: Due date?

AdamR: Regrets for 7 Dec
... suggest we review this 14 Dec

Issue 12: Passing details on payment options

https://github.com/w3c/webpayments-payment-apps-api/issues/12

IJ: New views on this?

Rouslan: It seems that it's not too difficult to support options
... but we haven't blended the UI of options and payment apps
... we are still undecided on this

IJ: Should we have the spec say "provide option data and display may vary" or wait more to get experience?

adamR: I am ok with current approach for the time being
... I think that if some browsers display options, there will be pressure on browser vendors to display

IJ: The spec could say "browsers and payment apps may exchange option information"

AdamR: I'm ok to wait to see how things land and update the spec
... I prefer that the data structures be defined and to say that display of option is done at the preference of the browser
... if not displayed, then the field could come through undefined
... the issue is "more clicks" from the user to make things happen when they can't choose options directly

rouslan: I am recalling recent conversations on blink list about vagueness of spec
... so let's be more specific and say user agent SHOULD display options
... and UA gets to experiment with exact user experience

AdamR: If option is not displayed, need to send appropriate (null) data to the payment app

[jenan asks UI question]

https://github.com/w3c/webpayments/wiki/PaymentApp_Notes#user-experience-descriptions

IJ: Yes, options might enable user to pick card (e.g.,) straight from mediator UI

AdamR: Goal is to minimize user actions

<jenan> roger, thanks!

Summary:

- include option data in the communications between mediator and payment app

- mediator SHOULD display options

- where user selects option, that information sent to the payment app

(AdamR: Yes, via a unique ID)

- where mediator does not support options, then ID is null to payment app

<TommyT> +1 to MUST

rouslan: chrome shows most frequently used payment option, and user can expand to view all of them
... so does that mean we are not displaying all options (and violate MUST)?

Tommy: It's easier to implement MUST than SHOULD
... since we need to implement support for payment apps that supply options and those that don't
... the MUST here is on the mediator, but there's an option for payment apps (to have options or not)
... are both options and not-options legal?

jenan: What about language that the user must be able to select all options (rather than speak about display)

<rouslan> +1

<rouslan> 1-

<rouslan> +1

rouslan: Yes, +1 to language about user must be able to select all options

<scribe> ACTION: Ian to update the spec to talk about options, user selection of options [recorded in http://www.w3.org/2016/11/30-apps-minutes.html#action02]

Opera / Samsung update

IJ: How is it going? can we help you as a group?

Tommy: I have received a lot of requests from third party payment app implementers
... and they want a platform to test apps
... so I wanted to see if I could implement this in Chromium, and our colleague from Samsung has started implementing the difficult stuff.
... so I'm going to let him!
... I will try to tackle some of the easier problems
... we would like to be able to make some prototype versions of Chromium for app implementers to test
... not sure how long it will take us...we've built some pieces but not put them together yet

IJ: We could organize a hackathon (e.g., Boston)

OR

scribe: in Chicago around our FTF meeting (if we have it in march)

Tommy: We should be able to have something together in Q1

Other issues

https://github.com/w3c/webpayments-payment-apps-api/issues

Regarding issue 23

AdamR: I think that is mostly covered by our shift to service workers
... there are still questions about what permissions we expect the apps to use
... I think we want apps to have access to the web platform (e.g., cameras, etc for authentication or QR codes)
... it is a question we need to address
... I do expect the question will come up later

Next week's call

Ian: AdamR, perhaps you can weigh in by email on this proposal:

https://github.com/w3c/webpayments-payment-apps-api/issues/48#issuecomment-261786242

Summary of Action Items

[NEW] ACTION: Ian to update the manifest information with pointer to this issue and updated member names [recorded in http://www.w3.org/2016/11/30-apps-minutes.html#action01]
[NEW] ACTION: Ian to update the spec to talk about options, user selection of options [recorded in http://www.w3.org/2016/11/30-apps-minutes.html#action02]
 

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.148 (CVS log)
$Date: 2016/11/30 15:02:15 $