W3C home > Mailing lists > Public > public-webpayments-ig@w3.org > September 2015

RE: Organize a chat on account/ledger capabilities?

From: KETELS Kris <Kris.KETELS@swift.com>
Date: Thu, 3 Sep 2015 15:36:49 +0000
To: Adrian Hope-Bailie <adrian@hopebailie.com>, Evert Fekkes <E.R.Fekkes@rn.rabobank.nl>
CC: Adrian Hope-Bailie <adrian@ripple.com>, Kepeng Li <kepeng.lkp@alibaba-inc.com>, "Adler, Patrick" <patrick.adler@chi.frb.org>, VIGNET cyril <Cyril.VIGNET@bpce.fr>, "j.j.spaanderman@dnb.nl" <j.j.spaanderman@dnb.nl>, Web Payments IG <public-webpayments-ig@w3.org>, "Ian Jacobs" <ij@w3.org>
Message-ID: <966D8C268A095048A7619EDB4AB05826792D0E24@BEEXCL25.swift.corp>
Dear Adrian, dear Evert,

Building on this: at this stage, ISO 20022 only supports XML and ASN.1.  The standard however was designed to be extensible for other syntaxes if need be. For this reason,  as we speak, ISO 20022 is looking into extending the ISO 20022 specification to support API’s and as such, also JSON which is becoming the predominant syntax for API’s.
The goal being that from a single logical model, it should be possible to generate for example both an XML message and its JSON equivalent.

Maybe some more things on ISO 20022 itself: it is correct the specification itself is payable (i.e. the metamodel, the transformation rules to XML, the modelling guidelines etc...) but the actual Message Definitions are free of charge and can be downloaded from the ISO 20022 website (http://www.iso20022.org/). There you will also be able to obtain an electronic version of the repository which for example contains the data dictionary describing all of the business and message concepts used in ISO 20022).


Kind regards
Kris


From: Adrian Hope-Bailie [mailto:adrian@hopebailie.com]
Sent: 03 September 2015 14:20
To: Evert Fekkes
Cc: Adrian Hope-Bailie; KETELS Kris; Kepeng Li; Adler, Patrick; VIGNET cyril; j.j.spaanderman@dnb.nl; Web Payments IG; Ian Jacobs
Subject: Re: Organize a chat on account/ledger capabilities?

Hi Evert,
Good points but I think fundamentally the W3C has evolved in a world where developer-centric processes are favoured and ISO favours large organisations and governments.

On 3 September 2015 at 08:49, <E.R.Fekkes@rn.rabobank.nl<mailto:E.R.Fekkes@rn.rabobank.nl>> wrote:
Hello Adrian,

Well worded response, of which I recognize quite a lot.
Some comments, though:


•         ISO standards must be paid for, but do you think the cost is prohibitive?
I think ISO has a different business model as W3C, but for ISO my organization does not have to pay a (substantial) membership fee.
I do. The majority of W3C standards I look at I do so as part of my research. Only quite far into the process would I need them as a definitive reference and be prepared to pay for them.
For a small dev shop paying for every spec is a non-starter. Access to the specs is also not reserved for members. Organisations become members of W3C to support the consortium's work and to have input into the development of standards not for any privileged access to the specs.


•         In my opinion, we should make use of ISO and other work as components. ISO 20022 is surely a given and could be extended with JSON formats next to XML and ASN.1.
Absolutely. My one concern about ISO20022 is that the messaging seems to be very use case specific. i.e. There is a message definition for each use case and these definitions are extensive.
I suspect we will be looking to define our standards at a higher level of abstraction with a minimal set of required fields and an extensible format since that's how the Web works and most successful Web standards have evolved.

There was some effort from Erik and others to get a comprehensive dictionary from ISO20022 as a basis for our vocabulary but I think the result was that this simply doesn't exist.

ISO 8583 is gradually on it’s way out, but we have to keep in mind that many processing systems are firmly based on this standard and migration will take 5-10 years. Other standards such as security, connectivity (NFC) are necessary elements as well.
ISO8383 is dead. Long live ISO8583 :)
The rate of innovation on the Web is certainly unmatched in most other ecosystems. That's probabaly why the "FinTech revoluton" is taking place as technology spills into industries that have not evolved as fast as they could have.
I like our approach of dealing with friction points and interoperability and letting the existing schemes (like card and others based on dinosaur technology) operate within those new paradigms but at the same time allowing new schemes to evolve that are more suited to the Web architecture.


•         The W3C WPAY must define the gaps and fill these with open standards, building on the fundaments of ISO and others. In the cards world, you could say that EMV dit that as well, athough still not as open as is the case with W3C...
+1 - although we shouldn't expect the world to simply patch the holes and ignore the opportunities to embrace the Web as a replacement for these legacy networks that run on ISO-developed protocols.
At the end of the day the ISO protocols can easily be adapted to run on the Internet as opposed to private networks and I expect that this will ultimately be the way things move (unless the ISO protocols prove to be a hindrance to this or some other factor impacts this).
I see our role within the IG as being the facilitators of a process where everybody wins. Ideally the traditional institutions like FIs, SWIFT, ISO embrace the Internet as the infrastructure upon which they will build their next iteration of payments services and will use mature and trusted Web technologies to do it. In that way the innovation on that system is not limited to large organisations with specialized knowledge of the domain.
The incumbents benefit from an ecosystem that is able to innovate and evolve much faster than before and developers are empowered to participate in this process (even if it's just a guy with his laptop and an internet connection).


Comments welcome!

Evert Fekkes
Rabobank

Van: Adrian Hope-Bailie [mailto:adrian@ripple.com<mailto:adrian@ripple.com>]
Verzonden: woensdag 2 september 2015 15:32
Aan: KETELS Kris
CC: Adrian Hope-Bailie; Kepeng Li; Adler, Patrick; Fekkes, ER (Evert); VIGNET cyril; j.j.spaanderman@dnb.nl<mailto:j.j.spaanderman@dnb.nl>; Web Payments IG; Ian Jacobs
Onderwerp: Re: Organize a chat on account/ledger capabilities?

Hi Kris,

I'd classify ISO 20022 as an exception. The development of the standard was done in the traditional ISO manner but the standard is just a methodology that laid a foundation for open development of the messaging standards.

Traditionally ISO standards are developed quite differently to Web standards. They are developed in closed groups under strict confidentiality and the standards themselves are not free and open for anyone to download and implement royalty-free.

The W3C (like the IEEE, ETT Internet Society etc ) are advocates of open standards principals: https://open-stand.org/ which are quite different to the principals of ISO.

That is not to say there are not reasons for there to be different approaches to standardisation for different circumstances but from the perspective of Web technology development I would strongly favour open standards.

Some illustrative anecdotes:

I worked for a long time at for a very large payments software vendor where we spent a lot of our time ensuring our software was compliant with the latest specifications. These were usually variations of ISO8583 from different networks. All of these specs were closely guarded confidential documents that required careful management of their distribution. To this day I have not seen a copy of the original ISO8583 spec from ISO because my company would have had to pay a lot of money for it just for me to have that privilege.

I am part of the W3C liaison with ISO for ISO12812 and have had to install a whole load of document lifecycle management software on my laptop just to be able to receive documents to do my work.

The reality is that the ISO model may work in manufacturing or similar industries where there are layers of abstraction from the specification to the implementation and the cost of getting your hands on the spec is minimal in comparison to all of the other input costs.

On the Web, where some products are developed with almost zero overheads on tiny budgets as they attempt to get traction this model simply doesn't work. The only people that can afford to be bothered with ISO specs are already large companies the majority of whom are struggling to innovate fast enough to defend themselves against the threat of the small start-ups (especially so in the financial technology sector).

So, I can imagine these incumbents being very supportive of ISO specifications, especially in jurisdictions where they have implicit regulatory backing, but that would simply be because it helps them maintain their defense against disruptive innovators that see proprietary and closed specifications as barriers to entry.

On ISO20022 specifically:

I think the approach taken with ISO20022 is excellent. I am certain it will become the language of payments globally and that is attributable in large parts to the availability of information on it's use. It's the only developer friendly ISO specification initiative I am aware of.

I would like it if JSON encoding was supported as a first-class citizen since that is the the most common data encoding format on the Web (I know it's possible just not easy).

I would also be interested to know if ISO20022 had picked a linked-data format (like JSON-LD) for encoding if the current plethora of message types and fields and complexity could have been avoided but who can say.

On Wed, Sep 2, 2015 at 1:40 PM, KETELS Kris <Kris.KETELS@swift.com<mailto:Kris.KETELS@swift.com>> wrote:
Hi Adrian,

Can you explain why you qualify ISO standards as being the opposite of open W3C standards?
I myself haven’t encountered a more open (financial) standard as ISO 20022.


Kind regards
Kris


From: Adrian Hope-Bailie [mailto:adrian@hopebailie.com<mailto:adrian@hopebailie.com>]
Sent: 19 August 2015 18:58
To: Ian Jacobs
Cc: Kepeng Li; Adler, Patrick; Adrian Hope-Bailie; Evert Fekkes; VIGNET cyril; j.j.spaanderman@dnb.nl<mailto:j.j.spaanderman@dnb.nl>; Web Payments IG
Subject: Re: Organize a chat on account/ledger capabilities?

Hi Ian,
Definitely interested. It would be great if the standards that were used for XS2A came out of this group and followed all of the open standards principles of the W3C and used Web technology as opposed to some ISO standard.
We need to get some direct participation from the EPC, EBA etc
Adrian

On 19 August 2015 at 18:45, Ian Jacobs <ij@w3.org<mailto:ij@w3.org>> wrote:
Hi Kepeng, Pat, Adrian, Evert, Cyril, Jurgen, (and others who may be interested),

I was chatting with Kepeng yesterday about capabilities related to account/ledger access. We were discussing
the sorts of useful things one could do if there were standard APIs for account/ledger access, including building
risk monitoring systems that would have an easier time working with diverse types of accounts. Also, I wondered
whether PSD2 regulation in Europe might be leading to a need for open standards to provide access to
accounts from Web applications.

Maybe we can organize a Thursday call around this topic so that we can determine who is interested, and
whether, for example, there is even enough interest to start a task force.

Please let me know if this discussion would interest you. Thanks!

Ian

--
Ian Jacobs <ij@w3.org<mailto:ij@w3.org>>      http://www.w3.org/People/Jacobs

Tel:                       +1 718 260 9447<tel:%2B1%20718%20260%209447>



======================================================
Rabobank disclaimer: http://www.rabobank.nl/disclaimer

Received on Thursday, 3 September 2015 15:37:21 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:08:44 UTC