W3C home > Mailing lists > Public > public-webpayments@w3.org > February 2013

Re: Building A Paid App For Firefox OS

From: Kumar McMillan <kmcmillan@mozilla.com>
Date: Thu, 28 Feb 2013 10:31:56 -0600
Cc: Web Payments <public-webpayments@w3.org>
Message-Id: <C32394DF-43B2-4045-8C37-21EE6CF2CAF8@mozilla.com>
To: Linus Olsson <linus@flattr.com>

On Feb 27, 2013, at 12:06 PM, Linus Olsson <linus@flattr.com> wrote:

> Hello all, my first reply here. I'm CEO & Co-founder of Flattr btw.
> 
> Just wanted to add that this is in reality not a problem. Lets face it, people that don't want to pay will always find a way to get around things. The energy should be put into making it as easy as possible for the people that want to pay instead of using it on a problem that is in reality not possible to solve. Or worth it. Many people also choose free/open/pirated/hacked/etc because it simpler then paying.

Hi Linus.
I definitely agree. The article on the Hacks blog was meant for developers to get a feel for how the app payment API works. It didn't mention the buy flow but that is our #1 priority to get right. It's currently a pretty seamless flow; the user can put a charge on her phone bill with the press of a button. However, the first-time buy flow still needs some optimization since there is a bit of sign-up involved.

Naturally, Mozilla's navigator.mozPay() API will be competing with all other web-based payment flows so the market will force us to make it smooth for users and keep transaction fees low. That is, *any* web based pay flows can take payments on Firefox OS. No one will be blocked from the system. An app can drop in a stripe.js script or whatever they like. The navigator.mozPay() API is just one that we will ship on the device so that developers don't have to think too hard about taking payments if they don't want to. (This applies more to in-app payments which I'll be posting an article on shortly.)

It will be interesting to see which payment flows rise to the top in Firefox OS. I'm curious about Affirm (which was posted to the web-payments list). They are allowing first time buyers to essentially pay on credit without any sign-up which is very clever.

Kumar

> 
> That is one of the reasons we love the way the Mozilla does this.
> 
> (Sure we at Flattr are even promoting the idea of voluntary payments for the things you like and want to support, but that is probably asking to much of many app developers.)
> 
> Linus
> flattr.com
> 
> 
> On Feb 27, 2013, at 18:43, Paul <paul@pogodan.com> wrote:
> 
>> On Feb 27, 2013, at 12:31 PM, Kumar McMillan <kmcmillan@mozilla.com> wrote:
>> 
>>> 
>>> On Feb 27, 2013, at 10:53 AM, Melvin Carvalho <melvincarvalho@gmail.com> wrote:
>>> 
>>>> At first glance the Firefox Marketplace for Firefox OS may look similar to the Apple Store or Google Play Store but there is a key difference: it does not lock you into Mozilla or lock you into your Firefox OS phone. It enables you to sell a web app that will run on any open web device by way of the receipt protocol. Non-Mozilla marketplaces can participate in selling apps on Firefox OS out of the box by implementing the receipt format and users won’t notice anything different when running a paid app from either store.
>>>> 
>>>> When other devices support the receipt protocol then theoretically you could pay for an app once and run it everywhere. There is, of course, a chicken vs. egg problem here so Mozilla hopes to be the egg that helps prove out the decentralized receipt concept and iterate on the protocol. Mozilla invites other vendors to help us work on getting receipts right so that paid apps are as portable and “webby” as possible.
>>>> 
>>>> [Read More...]
>>>> 
>>>> https://hacks.mozilla.org/2013/02/building-a-paid-app-for-firefox-os/
>>>> 
>>> 
>>> Hi Melvin, thanks for posting to the list. I am following the web payments work here and we (Mozilla) are certainly interested in making payments work on the web in a decentralized manner.
>>> 
>>> In comparison to Pay Swarm, I'll point out that our web app receipt format says nothing about how to *pay* for the app (that's way harder!).  However, I think the decentralized receipt format is the most well understood part of Firefox OS' current approach to payments. It is a simple JWT format using standard crypto. The key synchronization is a little complex but it should hopefully be easy enough for other marketplaces to implement. We'll find out how easy once some others try it out. Right now, only Mozilla has implemented the receipt format.
>>> 
>>> As far as I know, neither iOS or Android have attempted to make open payment receipts that third parties can verify. As my article states, we are interested in building an ecosystem where anyone can sell apps and anyone can build a *runtime* that supports any paid app.
>>> 
>>> I'll be posting another article soon about how in-app payments work but IMO those are not as webby and decentralized as app receipts. There is lots of work to do!
>>> 
>>>> I wonder if the firefox marketplace could be a good way to bootstrap the web app eco system …
>>> 
>>> The current open payments concepts came out of our labs group a couple years ago. They are our first take on payments so I think they will evolve once they're in the wild.
>>> 
>>> -Kumar
>> 
>> This is definitely interesting. A couple questions I have are:
>> * would the app fail to boot/go into 'pirated' mode if the receipt server went down/started handing out bad responses?
>> * with a fully open source OS and app, it seems it would always be fairly trivial to bypass at least the client-side part of the equation, or is there something I'm missing that would prevent the user from just removing the receipt validation code? obviously if the app's actual functionality required data from a server you run that would change things
>> 
>> 
>> -- 
>> Paul Meserve
>> Pogoapp | www.pogoapp.com
>> 
>> 
> 
> 
> 
Received on Thursday, 28 February 2013 16:32:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 28 February 2013 16:32:27 GMT