Re: report on Leveraging ICT for the Base of the Pyramid

On 05/10/11 16:34, CE Whitehead wrote:

>  > Indeed. I just did a small test texting: "1F332 FISH CAKE WITH SWIRL
>  > DESIGN [??] and 1F35A COOKED RICE [??]" (from Skype) to various devices:
> Do you mean 1F365 instead of 1F332? (The latter is an evergreen tree; in
> any case I cannot make sense of the symbols 1F35A and 1F365, and your
> symbol for "fish" below is at least as useful.)
> http://www.unicode.org/charts/PDF/U1F300.pdf

Yes, sorry. I got my codepoints wrong.

>  > - iPhone: displays 1F35A but not 1F332
>  > - Samsung Galaxy Tab (Android): displays neither
>  > - Motorola V220 (from around 2003): neither
>  > - Nokia xpressmusic (from around 2006): neither - doesn't even recognise
>  > UTF-8 and displays 2 blocks for each character.
>  >
>  > So I think we can discard fancy unicode symbols for a while. Probably
>  > not a problem for currency symbols though, which are (I suspect)
>  > optional in most cases.
>  >
> Yes the symbols seem to be the wrong way to go with cell phones for the
> present; I hope you are right about the currency symbols. In which case
> these could be used to form graphs:
>
> Rice (in tenths of dollars? but $= $.10 would be confusing though
> something might be worked for some other currency symbols; I don't have
> the symbol for cents on my mini keyboard or would have used that)
>
> 1. Town1 $$$$$
> 2. Town2 $$$$$$
> 3. Town3 $$$$$

That's a clever idea. Its usefulness depends on how widely the prices 
differ, of course, and what's the granularity needed to make decisions 
about which town provides the best distance/profit ratio.

>  > I suspect that, to people with low reading skills, "F" is just as good a
>  > symbol for fish than <º)))><.
> You are probably right; that's a nice fish but, as it's not formed using
> a continuous line in a single character space, it's not that useful
> perhaps. I do agree that "F" may be just as recognizeable initially, and
> actually the whole word, "Fish" might even be very recognizable to those
> who deal in it.
>
> What about: <º)))>< Fish ? (Now can someone do rice?)
>
> As far as images go, I was hoping only for the bar graphs showing cost
> differences (described above) plus visual rough maps showing the
> approximate distance and direction of villages from the cell phone
> location. I've noticed variable width characters too and yes those can
> be a problem; of course a repeated "." or "-: showing the distance
> between towns in miles would be "monospaced," and if each town were
> represented by a number then that would be all that could vary (and, for
> the digits 1-9, mine are monospaced on this mini; but those are European
> digits of course; I'd have to try others . . . ).

1-9 digits are clearly variable-width on the old phones I have here. But 
indeed using a single character 'horizontal histogram' alleviates that. 
You still have to be careful about line wrapping, though. So essentially 
you have to assume the smallest number of characters per line among the 
phones your users own.

>  > Opinions from a specialist in low-literacy most welcome.
>  >
> (I'm not that; I have a linguistics degree and have studied literacy,
> mainly writing/reading and bilingual writing/reading but am hardly a
> specialist in this area I don't think. I could not comment on 3rd world
> farmers and what symbols they are familiar with; from travel not study I
> know that many people worldwide, at least those who work away from home,
> are familiar with symbols indicating prices though a graph of some sort
> makes comparison easier. Also of course, whatever symbols/etc are used
> these should be tested on local farmers to see if they are "intuitive" I
> would guess.)

That's good enough expertise for me! There isn't much research on this 
that I can find, unfortunately. Lots of papers on how to tackle 
illiteracy through the use of SMS, but not much on how to design SMS 
services for illiterate people (who have more important things on their 
minds than to start learning to read). It seems the solution lies in 
using IVR for now, and perhaps Web access on the phone, when that 
becomes available and usable.

Cheers,
Max.

Received on Thursday, 6 October 2011 10:10:43 UTC