Topic: Mobile money slowly turning East Africa into cashless society - page 3. (Read 8146 times)

I agree with you about swapping SIM cards being a pain and also that unlocked phones would be required.   This is far from a perfect solution, and dual SIM phones are rare and can not be assumed.   My goal would be a solution that would work with out a major investment in new phone technology by the user. 

Unlocking phones is a mature cottage industry so I'm not too worried about that, but the dual SIM swapping is unavoidable.  People swap SIMs all the time to save money when traveling by using a local pre-paid GSM card.  It's not elegant, but it could work.


I think you missed my point about why two SIMs would be required.  The first SIM would be from a local cell carrier and provide the SMS infrastructure for communicating with the Bitcoin gateway via SMS.   The second SIM card would not provide any telecommunications at all and would only be programmed to act as a Bitcoin wallet and to facilitate transactions via SMS messages sent using the first SIM card.

So to your question about costs, they would not be zero.  At a minimum the costs would include:

1) cost of an SMS to the Bitcoin gateway to initiate a transaction
2) cost of an SMS from the Bitcoin gateway to notify that the transaction confirmed
3) cost charged by the gateway to perform the Bitcoin gateway functionality

Creating a payment address and checking your bitcoin balance would all be handled locally using the 2nd Bitcoin card so no costs for that.  If you need to check an exchange rate, that would also require an SMS message and reply.

Unless people are using dual SIM phones, this is going to be a pain in the ass.  SIM cards are small and easily lost and swapping out SIM cards is a nuisance in many phones.  A lot of pre-paid phones are also locked to particular networks and will only accept SIM cards from those networks - you'd need to either configure existing cards issued by the network or people would need to get their phones unlocked.

There's a limit to how much trouble people will go to in order to make transactions.  You also need to look at what the effective fees on microtransactions would be.  The infrastructure with which the second SIM communicated would not be free, but it would need to be affordable - how do you make it affordable if you're not using the towers of existing providers to route communication with the Bitcoin SMS gateway?

any NPO, or individuals willing to set one up, looking to move this concept or other possible solutions forward then please to see my thread re potential funding for the research, development, & implementation of this

https://bitcointalk.org/index.php?topic=68162.0;all

all I can say is that I believe it would receive a sympathetic hearing from someone who could offer realistic backing to potentially make it happen

Sounds like they use USSD and the SIM toolkit for the M-Pesa system.  Unfortunately any cell operator based system requires the operator and thus also the the government to support it.  An ideal system would bypass as much as possible any requirement for government or big business support.

I've been thinking of how you could do SMS based Bitcoin transactions with low cost hardware (ie. no smart phones/internet) and still be secure; I think you could do it all with two SIM cards.  One SIM card could be from the operator and thus have the ability to send/receive simple SMS messages.  A second SIM card could be independently produced that acts as a Bitcoin wallet and contain SIM Toolkit software necessary to communicate with a Bitcoin SMS gateway to create transactions and check for confirmations.    This second SIM card would not have any ability to log into the cell providers network, but should be able to check and queue up SMS messages in the phone's memory.

Programming for SIM toolkit is a PITA, but simulators are available so a proof of concept wouldn't be too hard to produce.

The system used in Africa is based on the GSM standard that lets the operator put simple menus on the users screen:

http://en.wikipedia.org/wiki/Unstructured_Supplementary_Service_Data

I think it is as simple as the seller requesting a portion of the buyers prepaid phone amount to be debited and sent to the merchants account.

I can see BTC really thriving in parts of the world that don't have as much infrastructure already set up. You can bypass corrupt regimes ruining your local currency and pay for things easily knowing that you'll still have something of value no matter what is happening locally.

How does Mobile SMS money transfer work in Africa?  There must be some gateway to an account entry system for that to work.  A gateway to initiate and verify Bitcoin transactions should be possible in the same way.  That would enable Bitcoin payment with out local access to the internet.  You would lose some decentralization by using a gateway, but there could be competition among gateways.   I also assume transactions would be denominated in BTC, otherwise there would have to be some way to determine the current exchange rate with local or international fiat currency.

I can image a scenario like the following (assumes an android device):

1) person 1 wants to pay person 2 in BTC
2) person 2 generates a public address on their phone and sends it to person 1 (via bluetooth, camera scan, local wifi, etc.)
3) person 1 generates signed transfer from their wallet to person 2's public address and sends it to person 2 (via bluetooth, camera scan, local wifi, etc.)
4) person 2 verifies terms of the transfer and sends an SMS with the signed transfer to the Bitcoin gateway
5) the Bitcoin gateway posts the signed transfer to the Bitcoin network and waits for some predetermined number of confirms or rejections
6) the Bitcoin gateway sends an SMS to person 2 saying the transfer succeeded or failed

- The BTC gateway operator will incur internet and SMS network fees and will demand some sort of micropayment, perhaps in BTC, for each transaction cleared. 
- The BTC gateway could run on an android phone sitting somewhere with more reliable wifi access.
- An African Bitcoin ecosystem would also need someone willing to exchange Bitcoins for local fiat currency (and/or foreign currency).

I also wonder if the user side could be implemented using SIM toolkit (like M-Pesa) on non-android phones.  That would instantly increase it's penetration.

Meh. Internet is a whole different story. In Ethiopia for example all the telecom is monopolized by the evil gov't, which is doing its best (effectively) that internet do not work at all there.  I wonder what is the point of all those internet cafe in Addis. I suppose just to meet ppl in person or as a cover for some other biz.

Africa is certainly prime for the picking.  Mobile SMS text messaging money transfers are the norm over there.

Bitcoin could certainly make a splash. 

What do you think about using Bitcoin's deflationary properties to push capital into Africa:
http://bitcoinmedia.com/our-next-adoption-phase/?

http://mobilemoneyafrica.com/mobile-money-slowly-turning-east-africa-into-cashless-society/

Mobilemoney Africa summit: Someone from Bitcoin should go there...
http://mobilemoneyafrica.com/mobile-money-event/

"A frequent comment I hear, particularly from international banks, is that the unbanked aren’t for them."
http://www.mobilemoneyconsulting.com/2011/09/21/uncategorized/sibos-have-banks-missed-the-mobile-payments-boat