CFC Stanbic Bank launches till2Bank, enables merchants take Mpesa payments.

CFC Stanbic Bank recently announced a product that would make it easier for businesses to take mobile payments via Mpesa.

The product dubbed till2bank  launched in conjunction with Safaricom’s
Lipa na M-PESA and Kopo Kopo, not only allows merchants to accept payments on M-
PESA but extends banking services to them.

To obtain a Lipa Na Mpesa number merchants are required to register with CFC Stanbic  by calling 0711 068 888 or texting the word ‘LIPA’ to 22208. They may alternatively reach the bank via email on till2bank@stanbic.com

Payments made via
Lipa na M-PESA will be credited into merchant’s CfC
Stanbic Bank account on demand.

Government Directive : No more cash on matts

This year promises to be an interesting one for the local payments industry. The government has already indicated that it will issue a new policy directive to ban the use of currency notes and coin in public transport starting July 1st, 2014.

The government has, in my view, two clear incentives to push through with this directive:

1) Kenyan currency is notoriously shortlived, and public transport is where Kenyan currency go to die. Once a tout habitually folds a brand new KES 1000 note and holds it between his fingers, the note is fast on it’s way to the wear and tear grave.

2) The cash intensive nature of public trnasport in Kenya makes it harder for the taxman to collect his dues. Matatu owners and operators currently might simply opt to underdeclare their earnings or as many do, not declare their earnings to the Kenya Revenue Authority at all. With money from the public transport system going through formal financial institutions and systems, the industry will become more transparent. It will be easier for the taxman to track earnings from the industry and tax them accordingly.

 

The move by the government might just have thrown a lifeline to Google/Equity’s BebaPay – which is yet to see widespread adoption. Safaricom would also benefit greatly, with it’s Lipa na Mpesa product clearly being a key contender for commuter payments.

This might also aide in bringing order to an otherwise chaotic industry as members of the route cartels and rogue policemen, accosted to taking their pound of lesh from the driver’s window might just find it harder to collect. The driver may simply plead that he has no cash on him.

Ha Kairu, OpenStreetMap and the week that was

It’s been an interesting week for me. I have been working on a live system all week. I picked up a couple of tricks such as creating my repo. The system freezes or hangs as we say here, every so often, and I have to restart  the machine. Except when I do, it doesn’t actually restart the machine, it boots a whole new system. Having to set up my dev environment every time was a bit of hustle. I had to install Perl (which for some reason doesn’t come pre-installed on Fedora 19 live), Git and Vim as well as Irssi every single time. On my rather modest connection this took time. So I cached the packages , and created a repo I could install them from every time I went down.

I got down to mapping out Ha Kairu, my pilot location for nimehama.com in earnest last week. Ha Kairu is a neighborhood along the A2 (Thika Highway), that is home to the Ruiru Campus of Kenyatta University. I uploaded several tracks to OpenStreetMap but not as many as I would have liked. I have chalked up the last week as a recon mission.

Still on OpenStreetMap, I wrote a user diary, proposing a way to tag mobile money agents on OSM. I got several responses which was very encouraging. More on that in a separate post.

A friend of mine was moving out of the area and I got my first property listing for nimehama.com. Except Nginx is still being troublesome, and I have not put it up on the website as yet. But still the data is what is most important. So, earlier in the morning I wrote up the data in JSON. I am hoping to have it available on the website as soon as possible.

I got to work on my personal site moshe.me.ke. I cleaned up the CSS quite a bit. It still looks rather hackish, but it’s definitely getting better. Lots more work to do there. I will still need to update it, since the static blog on moshe.me.ke is falling behind this blog.

It was another quiet week on the Wikipedia editing front. I started an article on Ha Kairu in my sandbox. Right now it’s a little more than a stub. I will be fleshing it out real soon with as much detail as I can.

I set up my blog for all sorts of stuff on Tumblr. That will be the home of The Daily Hash until I have time and opportunity to develop a site for it. I made a couple of posts on The Daily Hash. You can find them at thehashhq.com.

Here is how my rather shoddy week shaped up:

  • Posts, Essays and Articles published: 2
  • Github commits: 18
  • Words translated: 0
  • Miles traveled : 0
  • Gpx tracks uploaded to OpenStreetMap: 6
  • Wikipedia edits: 4
  • Books finished: 0
  • Movies watched: 0
  • Series completed: 0
  • Lines of code written: ~ 40
  • Projects completed: 0

My weekly scorecard – four days late!

Once again my weekly scorecard is late, by four days this time. It was a rather eventful weekend. I attended the November Nairobi Linux User Group meet-up on Saturday, then came to visit my friends Ibrahim and Fred  in Ruiru. In fact I am posting this from Ruiru.
At the beginning of last week I finally got round to working on a web app I had been meaning to write for years now while now. By the end of the week I had a working prototype of the app. More on that here.
Activity on the Wikipedia front was rather light. I took part in a survey on mobile editing. I hope my feedback will help the folks at Wikimedia improve an already fantastic mobile editing experience.
I discovered that the Wikipedia page for my old high school Dagoretti High is fraught with errors and is generally pathetic.
This is all the more unfortunate since a lot of my old school fellows work in tech and ostensibly have access to the internet and the means of improving the page. I made several edits, looked at some of the issues on the page, but there is still a lot more to do.
I hope to get Fred and Ibrahim on board, and write up a couple of Wikipedia articles by the end of the week. On Sunday I finally signed up for microca.st, am looking to test pump.io, and maybe eventually host my own instance.
All in all, a slightly better scorecard than last week’s I think:

  • Posts, Essays and Articles published: 1
  • Github commits: 5
  • Words translated: 0
  • Miles traveled : 0
  • Gpx tracks uploaded to OpenStreetMap: 4
  • Wikipedia edits: 5
  • Books finished: 1
  • Movies watched: 0
  • Series completed: 0
  • Lines of code written: ~ 100
  • Projects completed: 0

mngapi: A Mpesa tarriff calculator built entirely in JS/HTML 5

For a couple of years now I have been meaning to write an application that would calculate mobile money transfer transaction fees, specifically how much I would need to send someone an amount of money inclusive of the transfer and withdrawal fees. I intended to do this entirely in JS.
I could not find such an application then and having to look up the rates on a flyer I always seemed to lose, and mentally calculating the sums is rather tedious.
I discovered such an app exists on the Google playstore the other day. With over 10,000 downloads it is perhaps one of the most popular Kenyan apps on the Playstore.
I am unconvinced of the need for a native app for such a simple rate calculator.
The app does not seem to make any use of native bindings,( has no notifications etc) and it seemed to me my original idea of a HTML 5/CSS/JS app that is capable of running entirely in the browser, and that is easily portable between different OSes, was far superior.
I revived the idea with Ibrahim the other day.
So Ibrahim went out and prototyped the app in PHP to prove he is right or something.
I am not interested in this approach going forward.  Not least because it needs a web server backend. As I stated earlier I am looking to build one that runs entirely in the browser. I was only interested in the logic/algorithmic attitude he would assume.
Decoupling the data from the rest of the app does simplify matters. You know what would have been really useful? A SQL dump of the tarriffs data.
If you’re going to host a database with the tarriffs you ought to consider  building an API to allow folks to access the data  programatically . It is the only aspect of having a backend I find compelling.
Now on to matters js, I have formatted the mpesa tarriff data as json. I intend to push this onto github.
Think anyone else looking to build an application around this will find that valuable.
I hope to have the scaffolding for the app up on github by the end of the week.
In the meantime here is a while loop I built on my phone in under fifteen minutes as a prototype for the application.



var userInput = prompt ( "How many Kes do you wish to send?");

var amount = userInput/1;

var upperLimit = new Array(49,100,500,5000,7500,10000,15000,20000,25000,30000,35000,40000,45000,50000,70000);

tier = 0;

var i = 0;

while (upperLimit[i] < amount ) {      (upperLimit[i++]);

tier++;



var sendingFees = new Array (3, 5, 27, 33, 55, 50, 60, 70);

var withdrawalFees = new Array (0,10, 27, 27, 33, 49, 60, 70);

var fee = sendingFees[tier] + withdrawalFees[tier];

var total = fee + amount;

var display = " transfer fee: " + sendingFees[tier] + " agent  withdrawal fee: " + withdrawalFees[tier] + " total fees: " + fee + " You need: " + total;

alert(display);

document.writeln (display);

I have a working prototype of the app. It ran on my mobile phone browser, it should run on any browser.
With a bit more work it might even end up in the Firefox OS marketplace. My first Firefox OS app is already long overdue.

Last Week’s Scorecard

This scorecard is a whole day behind schedule. I had to give myself an extra day yesterday to accomplish something. It would have looked embarrassingly shabby if I had put it up Saturday morning as I have done past couple of weekends.
I spent much of the week reading, mostly technical books. I have been reading up on Python, JavaScript and NodeJS especially.
I attended the October OpenStack Nairobi meetup at the mLab on Thursday, and it was very informative going over Havana , the latest OpenStack release. The OpenStack project has come a long way. I will be writing a post on that soon.
I finally got round to downloading and installing Locus yesterday morning, but it came with an EULA. Clicking accept on EULAs still makes me very uncomfortable. I prefer Opensource apps. Plus the free version of Locus comes with ads.
I downloaded and attempted to install OSMAnd instead. It came in a 15 MB zip archive, a rather beefy download for an apk. The Zip archive extracts  to a 21 MB application that I could not install on my phone for some reason.
After a couple of failed attempts installing OSMAnd, I finally downloaded and installed OSM Tracker for Android. OSM auth failed when I attempted to upload gpx tracks.
I am yet to find a way to export pois and gpx tracks to OSM from my Android phone.
I finally retrieved my Wikipedia password. Actually, I finally remembered it. Mobile Wikipedia has improved tremendously. I managed to login on Opera Mini and it looks great. I immediately rushed in hoping to make a flurry of edits to save myself the shame of an otherwise shabby week. I went through  my watch list and straight away noticed an error in an article summary box (Mwatate is the capital of Taita Taveta not Voi ). I wanted to make the edit but couldn’t. Even after enabling beta and experimental features in the settings, the pencil icon for editing an article was not visible on Opera Mini.
I then installed the Wikipedia for Android app, a waste of 1.2 MB, as the app is read only, and doesn’t offer any editing capabilities.
On the default Android browser, the pencil icon is visible, and the editing  processing went smoother than expected.
My Wikipedia stats : 26 edits in 1505 days – shameful.
I was pleasantly surprised to find out that Wikipedia is free on my network operator, Airtel Kenya. Between that and mobile edits being enabled on Wikipedia,  I have pretty much ran out of excuses for not making edits.
Here is how this rather slow week shaped up:

    • Posts, Essays and Articles published: 0

    • Github commits: 4
    • Words translated: 0
    • Miles travelled : 0
    • Gpx tracks uploaded to OpenStreetMap: 0
    • Wikipedia edits: 6
    • Books finished: 0
    • Movies watched: 0
    • Series completed: 0
    • Lines of code written: 0
    • Projects completed: 0

    Weekly Scorecard

    “Funny what seven days can change”

    ~ Shawn ‘Jay Z’ Carter, A Week Ago

    I was hoping to upload a lot of gpx tracks this week. I am yet to install an application for making tracks. I had Locus on my previous Android phone, and it did a fine job of generating GPX tracks and uploading them to OpenStreetMaps. I attempted to download the Locus apk, but discovered I had ran out of space on my SD card.
    I had installed OSMAnd on my old phone as well , for creating POIs. OSMAnd can create gpx tracks too, but I never did seem to get the uploading to OSM feature working.
    Wikipedia enabled editing on their mobile site so I have pretty much ran out of excuses. I do not necessarily have to install any app to edit articles or write them up from scratch.
    Except for the fact that I lost my password and have to retrieve it. I hope to retrieve my password over the weekend. I am looking to make  around 5 edits a week minimum. In addition, I hope to write a minimum of one new article from scratch.
    I am a member of the Kenya project on Wikipedia. Information on Kenya on Wikipedia is disappointingly scanty.
    I intend to start out with places, especially places I have lived or places I know well.
    I worked on fleshing out my idea for nimehama.com for most of the week. nimehama.com is still in stealth mode,  there will be more in a separate blogpost at a later date.
    A lot of the week was spent on researching Ghost, a new minimalist blogging platform based on NodeJs. I like Ghost,I have never been much of a WordPress – and by extension PHP – fan. I am exploring the possibility of offering Ghost hosting at netivity.co. If things go according plan I could be the world’s first commercial hosting platform for Ghost, but I am getting ahead of myself here.
    I made a couple of commits to my personal website repo . There are 3 issues around styling that I ought to work on in the coming week.
    I made my first commit to my list of things greater than other things repo , part of a project I am working on to design a fun, creative way for learning to work with git and Github. I encourage you to head over to Github and fork the repo, for lulz.
    So, here is my scorecard for the week.

    • Posts, Essays and Articles published: 1
    • Github commits: 4
    • Words translated: 0
    • Miles travelled : 0
    • Gpx tracks uploaded to OpenStreetMap: 0
    • Wikipedia edits
    • Books finished: 0
    • Movies watched: 0
    • Series completed: 0
    • Lines of code written: 0
    • Projects completed: 0
    Follow

    Get every new post delivered to your Inbox.