[CLUG-chat] MS's IDC TCO Study
charles at finebushpeople.co.za
Fri Mar 5 12:43:27 SAST 2004
On Friday 05 March 2004 09:50, Mike Morris wrote:
> IIRC the multiplier for "high tech" spending is more like 21 than 1.4
Got scientific/academic backing for this figure? It looks to me more like the
multiplier of how much longer an African has to work to buy the same
high-tech item. The 21 figure would come from the per capita GDP of the US
(Oceana region) ($13946) vs the per capita GDP of Africa ($652)
The 1.4 comes from Barry Standish at UCT. I comes from this: a company
decides to build a road, the money they spend on materials and labour goes to
people that then spend some of it on food, housing, cars etc. Each of their
little spends gets spent further and further, and increases economic
activity. For every such rand spent in SA, R1.4 of economic activity
registers in the GDP.
So, the 1.4 is the measure of how much benefit we get by spending our money in
SA instead of overseas on licenses.
I just don't know how to use the 21 in the analysis. If I multiply the HW, SW
figures by it, the case for MS as a viable choice gets blown out of the
parny. But then you could argue that this figure applies to every expense on
the chart, as if a farm labourer in the Karoo were trying to set up the
system compared to a labourer in Wisconsin. It would only say that the whole
kaboodle, whether linux OR MS is way out of reach of the average African.
In absolute terms, a Woolworths would need to pay the original rates as set
out in the table, regardless that it _would_ take an average African 21 times
longer to pay it off than the average American. The only point of difference
is that labour is relatively cheaper than in the US, so having a labour
intensive system is cheaper in SA.
<on he rambles>
On the other hand, if getting a linux system into your farm kaya is two
thousand rand cheaper than the equivalent MS system, that is the equivalent
of an American saving 42000 on the purchase of a system!
More information about the Clug-chat