Decentralising online transactions

Or “online cash” – since I don’t have to hand my physical cash to Visa so they can give it to you for that coffee I just bought. Of course they’d like it that way, hence the push of “mobile wallets”, owned by guess who! (If not owned, reliant on the infrastructure of – this is not a step forward)

Flattr just announced they’re becoming a normal* payment gateway. With some caveats, and differences, that make it more interesting, but the important thing is – we get it now :)

Anyway, one of the comments reminded me I’d done a lot of thinking about how you decentralise the transaction of cash, in a simple sense all that’s happened is that a number in one row of a database is going up and another is going down, right? This should be easy. And I suspect most of it is.

Here’s the brain-dump: (was going to be a comment on that TC post)

Gratz to the team :)

Although this might appear to be a step backwards, towards traditional payment mechanics, the current players need to be disrupted by replicating current behaviours, with better practice, and with less barriers to mass adoption. And more options are always better, the current choices of PayPal or Google Checkout are frankly appalling. We use them because we have to. This slows innovation because we’re locked in. There are also lots of horror stories, do some Googling.

Now of course true p2p is the holy grail, as it were.. maybe build something out on bitcoin? I hear they’re doing well, it just needs some UX work, what doesn’t!

(We’re heading into a time when everything is distributed (like the work to distribute DNS) and trust will be king, services will compete on it, and where choice will be free, doesn’t that sound like a utopian future for humanity?)

Truly distributed services will always have that problem, trust. So there needs to be a system of nodes and providers (anyone and everyone is a node, and anyone or everyone can be a provider), you trust a provider based on other existing proofs. i.e. Twitter could be a provider, you trust them, right? Or your bank (yeah i know, ack), or the guy who runs your local residents group.

You then enter your provider at the point of payment with your unique identifier in a oauth/openid style interaction (Email address? More trust proof), that then gets looked up in the network, and payment is authorised.

Balance and transaction info is stored encrypted and distributed in the network, no node can be locally read. Each unit of currency holds a transaction history against itself, who last owned it, who owns it now, and where it was last transfered. Each unit is also unique with this triplet of info. that uniqueness can be looked up too, records are stored and replicated DNS-style. PGP cash?

And so it goes. money is democratised.



This has been a hard thing to articulate, and I’ve rushed through it here, I’ll try to elaborate more if I work out how to :)

* Normal in that you can make a payment of any amount, previously it was a percentage of your monthly pot based on the ratio of pot size to number of payments. They explain it better, I’m sure.


Are these pocket-sized devices different?

A post of two parts, here’s part one..

There are a lot of very similar looking boxes-you-plug-into-stuff in the market today. The hardware is basically the same..

Ubiquisys femtocell

Logitech Revue for Google TV – engadget article


OnLive micro console

  • the same array of connection ports (USB, ethernet, A/V)
  • similarly powerful and featured chipsets
  • support for local storage for files (SD, USB)
  • wireless comms (3G, Wifi, WiMax)
  • video-out capabilities (HDMI)
  • better on-screen interfaces (e.g. Android)
  • user installable apps
  • they’re all pocket-sized

(While I’ve purposely missed out devices with screens, we can assume they’re part of this wider discussion)

They take data in, they push data out, they let people do stuff on the internet. They are so similar that it’s hard to see the differences. Except they all have very different uses, mainly dictated by the manufacturer and what they’ve allowed the software to let you do. This is changing.

The main difference is that they either create access to high speed data in some form, or they require access to high speed data in some form. So why isn’t that where the convergence is happening? Well it is, kind of, but why aren’t those devices designed to do anything else but that?

So why not have a femtocell running android, that you carry in your pocket, has HDMI, a touch screen, an SD slot. You would be able to play Xbox games in the middle of a field, and then plug into your TV when you get home, and share it’s internet connection with your iPad.

The OnLive micro console is especially interesting, as it moves processing, and thus cost, to the cloud. It streams the video of what you’re playing to you. With that move, you also get a marketplace for games, with Apple app store-like one-click purchase and play. So why doesn’t my wifi router do that? Why isn’t OnLive just an app?

The limits are fairly obvious, and we’re not quite there yet, I reckon a year. Battery efficiency is a problem, SIM cards are a problem (I can’t work out what they’re for, I mean I know what we’re told they’re for, security, but that’s not true. Web-style user auth is more secure, Android will fix this).

The other limit is that some of these devices are meant to be shared. You can’t take your home wifi out with you, what would the other people in the house do while you were out? (Well obviously we’d all have one, so this also will change. Notice that very quickly, quicker than with laptops, households are multiple iPad owners).

I think there’s a huge opportunity here for device collaboration, a standardisation of compatible modules.. or a huge opportunity for one device that does it all. It’ll be the business model that dictates that path, and will be helped along by adoption of open standards.

Now time for part two of this post..

As some of you will know, I’m a bit of a geek about radio spectrum, the politics, tech, licensing issues, and toys. As Churchill said, radio spectrum is one of our most valuable resource a nation has (citation needed). This is because it is finite. We have to work out how to split it up for a large number of uses, only a small portion of which are for public use (TV, Wifi, 3G, Bluetooth, CB, RC, AM/FM, etc.. it’s a long list, of very narrow bit of radio). A large part also goes to private or military use.

This PDF shows the entire range (and US allocation). There is no more. You can’t squeeze bits in-between, or add more on the end. That’s all we’ve got.

It’s this that has held back the next generation on wireless tech, different countries have different parts of the spectrum available for different things, making settling on common bands for WiMax or LTE (both classed as 4G, and firmware interchangeable) very hard for chip manufacturers, and thus handset/device manufacturers too. For example, in the UK, the licenses to operate 4G commercially to end users simply don’t exist yet. If we could all agree on the radio tech, where in the spectrum to put 4G, you’d have these toys today.

Oh, and the cost of the license. That royally screwed everyone having toys like this years ago.. in my opinion :)

I’ve been waiting to write this post for a long time. This weekend I got to chat to Jacob Hsu about device convergence, and heard Martin Varsavsky talking about ideas they’d had for pushing the Fonera, and Chris Gilbert asking, “how can I make my daughter want my device?”.. so it was time to put some of my thoughts out there.

In conclusion..

If all these devices are basically the same, with the same holes, the same connectivity, and the same market, then why are they being designed to do such different things? Why can’t I decide what the device does, use it anywhere, plug it into anything, and share it with anyone? What really needs to change is the concept of the device being the thing you own. What we own is what we do with them, they’re just boxes with interfaces and data flowing through them.