> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:pubsoft-
> [EMAIL PROTECTED] On Behalf Of shogunx
> No.  This knowledge was open and therefore easily attainable through books
> and digital reference material.  I wish I could have gotten more of the
> calculus when I was younger, but my high school only offered it to second
> year, and the class moved too slowly notwithstanding.  The beauracracy of
> the college I attended for a year before dropping out due to lack of
> deriving any benefit prevented any more advanced mathematics from entering
> my head, as they insisted that the next logical step after second year
> calculus was college algebra 1.

Very good! You are demonstrating your environmental advantages as well as
your intrinsic assumptions. Which of them are true and which are false? Do
you know all the assumptions that you make, given that some are perhaps
sub-conscious, or even perfectly sensible, but based on inaccurate
information?

Let's assume that there is a natural incidence of developers amongst the
general population, and further assume that those who are good at
mathematics early on are most likely to become good developers.

Now consider that by the age of 8, some of the brightest kids are
apprenticed to a trade, and hence drop out of school. Consider that the
literacy rate for finishing high school could be as low as 30% (ignoring the
abysmal quality of some of the schools/teachers). 

Consider that the developer output per region per year is directly linked to
the number of developers. Consider that output as the circumference of a
wheel and each year the wheel makes one revolution. The circumference of the
American wheel is 1,250,000; that of the Indian wheel is 750,000; that of
the African wheel is 50,000. How much further behind does that put Africa on
a yearly basis?

What does that do to the pool of potential developers? Those who make it are
marked more by their ability to succeed in spite of the obstacles than
anything else.

The educational system in most African countries is something that you have
to experience in order to understand to the roots. It decimates opportunity
from the earliest days. 

Even the books you speak about are not sold in many places. If you were a
natural, where would you find the books? At a town hundreds of kilometers
away, and at a price you cannot afford, and a practice that your parents or
guardians do not approve of (they want you working for them, not going to
some school, so they have to do their work).

This experience at the age of 12 or so, is only the beginning of the
problems that the would-be African developer is going to face. How many are
going to make it?
 
> > Never talked to a friend
> > about the cool stuff they were doing on the family computer when you
> were
> > growing up?

No friend of yours has a computer. Those who do are a very tiny percentage.
In the university where one would expect more, there were six PCs for three
hundred students at one point. So where are the starting blocks for the
front line runners?
 
> With slim exception, I was unable to access our computers when growing up.
> They were for "business."

At least they were there. It's an impressive trick to get acquainted with
something you never saw before.
 
> > That's an impressive climb!
> >   Training isn't only something that happens in a formal classroom. Both
> formal
> > and informal training are rare in Africa.
> 
> As in many parts of the world.

Not to the same level. I could be wrong here though. However, the point
remains that this drastically affects transformation of people from
'potential' to actual developer status.

> > > Ultimately it is the developer that determines the fate of his code.
> The
> > > client lacks the skills to command the course of development.

He might have something the developer needs. Cash. Money talks, BS walks
goes the adage. It takes a rebel to buck that. Most people are busy doing
their own thing.

> >   Some clients have developer skills, but not the time to use them. Some
> clients
> > have professional or social networks that have developer skills and can
> > therefore control what they purchase even though their own knowledge is
> less
> > than the developer's.
> 
> This is true only if the code is open and available to the client.  If it
> is not, then the developer can write whatever he/she chooses into the
> application, and the client still has no choice in the matter.  As we
> have seen time and time again, a proprietary developer can write a
> backdoor into an application, get caught doing it, and have plausable
> deniability under the argument that "it was a bug.  I will fix it." My
> point is there is nobody standing over the developers shoulder saying "use
> comp
> instead of diff there."  The argument for open source here is that the
> working class (developers in this case) maintains control over the means
> of production, as well as the raw materials.  This provides a social check
> and balance over the distribution of power.

This is a good point. However you have to factor whether it is a buyer's or
seller's market into the weighting. In a buyer's market, the working class
by definition has no influence over the means of production. When the
working class is very small, it is even easier to maintain hegemony as a
buyer, though a few of the working class will be able to take advantage of
scarcity and extract higher prices. That only works if there are high skills
involved and even then it is more risky for the buyer to rely on scarce
labor, so most prefer just to hire more staff (at much cheaper) and make
them make it work.

OSS seems to have the most value as tech transfer and cost reduction. It
appears to also have the potential to implode the profit structure for
developers. The extra work that one gets from being an OSS developer in my
experience does not match the work/value compression that is expected from
the OSS coder - following from the expectation of cost reduction. 

This could be purely my local situation or my own personal failings so I
would love it if you could show me otherwise. (Mark Shuttleworth? Does that
count? Was the company value derived from OSS, or from being in the right
place at the right time?)
 
> > Some markets have few clients, so developers feel that
> > for their long term benefit they need to do as the client wants. Etc,
> etc,
> > etc.
> 
> That is market induced slavery.

Yes. That's what people go through. Read

http://developers.slashdot.org/developers/04/05/07/1618205.shtml

for some more information. I read a draft that had content edited out before
publication. One developer was made to work for free by his government and
the guys above him turned around and sold it to the government. If you are a
developer, you have no power, no status, no money etc. So there is no
recourse. Even getting it published makes life dangerous for you due to
punitive measures and people avoiding working with you. BTW, that is why
people here do not litigate. Even if you win the case, the next guy won't
want to do business with you for fear that you could also sue. That is the
cultural aspect that Yaacov was alluding to.
 
> >
> > > It sounds that with such a small developer community,
> >   Calling it a "community" may well be a misnomer. Many of these people
> are not
> > in contact with each other, and even less with the next generation of
> > developers.
> >
> 
> And there are not mechanisms to remedy this?

Yes. Work together on projects. These projects pay money, good money. The
work becomes the bond and the team gels. It's hard to get it started because
most of these guys don't even know the others exist. 

We had our first ever African developers meeting only this year. Before
then, we didn't know each other. Replicate and extrapolate. However, that
buttresses my point. What is FOSS without the developer community? Please
think about that and elaborate your conclusions.

> > > the developers
> > > already there can use their skills, and FOSSFA projects such as
> > > http://www.ltsp.org, to increase the developer population by educating
> > > others about programming, using freely available programming tools,
> and
> > > internet resources.

FOSSFA (at least some Council members) do not consider developers to be
important. The vast majority of the Council are not developers. Sample quote
is below (anonymized to protect the innocent guilty party :-):

"Our mission is really not about developers� daily struggles � it is about
policy: Government policy, corporate policy and social policy.  If in the
current circumstances developers feel they have to work with clients who
request work be done on proprietary software how is this our problem? We are
trying to change the client�s perceptions. This cannot be done by individual
developers refusing to do work � it will simply be done by others. For some
reason you have interpreted the FOSS campaign as if it were focussed on
spreading FOSS by banning African developers from using proprietary software
� this is ridiculous as the clients would simply go elsewhere. Developers
are simply not important enough in the chain of decision making!"
 
This is quite a reasonable position in itself but it implies Africa would
adopt FOSS without the developers and without the developer community. 

Mandatory FOSS makes it worse because the developers, who are already
oppressed and poor get used even more. They need to retain the right to
release proprietary stuff as well. They should be the ones to choose how to
release their work.

What is the productive engine behind FOSS? The developers! So where does
this mode of adoption lead to? It's not FOSS if it's not about the
developers at the heart of it, they keep the engine of growth and innovation
ticking.

> >   They could, but on what grounds are you going to convince them to?
> They can
> > get jobs more easily if they become MCSEs. And freedom to distribute and
> modify
> > software doesn't sound so grand beside freedom to meet your own and your
> > family's basic needs.
> 
> People still occasionally pay for things like network access and
> education, correct?  In developing areas especially.  As a MCSE, they are
> still at the mercy of their employers.  What grounds?  A network cafe,
> properly situated, and combined with a training center, can provide more
> than enough to support a family, and educate a population at the same
> time.

Actually MCSEs tend to have more options than developers. Remember my first
post to pubsoft about the life of the web African developer? I mentioned two
ponds in town. That was the reality. You work at company A, or company B.
That's it. A MCSE has much more scope.

Creating a network caf� is well beyond the means of the typical developer,
and also beyond the means of those developers who are quite well off
relatively. Capital is scarce and the developer tends not to be the sort
that the investors feel comfortable with. Money here is extremely risk
averse.

Translation: If you don't already have it, you won't get any of it. We're
talking bank loans at 50% interest with collateral being a house or building
past lintel level on land titled to you. If you had that, why would you need
a loan to open a network caf�?

Having said that there are many network cafes. Most are small operations and
most never grow beyond that. Service provision for Internet is pretty close
to perfect competition, since the service is the same thing really. Google
is Google. All that is really different is the speed. However, with
bandwidth at the prices that you get here, very few indeed use this to
differentiate themselves. Doing so would mean death for the business and
explains why there are so many small players who remain at the same size.

> > > This course of action also stimulates the development
> > > of public internet infrastructure, as these same terminal networks can
> be
> > > used concurrently for public access.  Educating the African people
> about
> > > technology will yeild an increase in the rate of technologization.
> And
> > > increasing the programmer base helps to solve point 3 below.  I
> suppose
> > > you could import programmers from developed countries also.

Important. The aim should be to increase the rate of technologization. The
mode is unimportant at this point. My opinion is that risks should be split
and the market and the developer get to decide what's best for them.

> >   Agreed. Internet infrastructure and importing a programmer base are
> very
> > important.

Yes? Import the programmer base and you have destroyed the local income. The
imported programmers get paid maybe 5-10 times the locals. Or they work for
free (we're here to help!), which is just as bad, if not worse. That extra
cost is money that didn't go to someone local who really needs it badly.

> And with such a small technical population, the labor force necessary to
> construct the infrastructure is limited.  Hence the need to develop more
> infrastructure that, directly or indirectly, produces more people capable
> of producing infrastructure.

There is the need for such infrastructure. It is education and capital.
Education is the mainly role of the state or at least heavily influenced by
the state. Capital can come from leveraging scarcity and releasing
proprietary software more than it can come from free software. 

This begs the question. Ideology and social benefit aside, what makes the
most money, FLOSS or PS? What is the best option for someone who needs money
most of all? Does such a person have incentive to participate in the gift
culture or the 'for status' development sweepstakes?

> >
> > > The community you seek is globally distributed.  I also live in an
> area
> > > with a low technical population, comparatively.  I have found that
> > > community over the wires, like the forum in which we now speak.
> 
> >   It is much easier to integrate into the community if you speak
> English, have a
> > reliable, not-too-slow net connection, have time to learn the community,
> are
> > able to understand the distinctly American/European culture of the
> community,
> > etc. I think it's possible, but a greater effort needs to be made to
> help
> > African developers overcome these barriers.
> >

Don't underestimate the net connection bit. As well as the access to
computers bit. Or for that matter, being confused with a spammer. The wires
don't come here. Look at an Internet map and see the connections and traffic
density patterns. It's dark over here.

> Could you expand on that a bit, Yaacov?  What efforts do you see that will
> help?
> 
> > > That is great opportunity for the African developer.  And the African
> > > businessman.
> >   Agreed.
> 
> And, presumably, anyone willing to relocate to Africa, given the
> cooperation and acceptence of the African people.
 
Not for free, should rather be at the same rates of remuneration!!!!!!

You're more than welcome to come help out. Projects as well as developers
are needed. There is vast opportunity but we start with scarce resources.

If we can assume that with scarce resources cooperation is essential, then
FLOSS developers and PS developers need to work hand in hand. This is the
root of my call for a dual strategy. 

-- G.

AfricanIntelligence Axiom 10110010:

�The test of a first rate intelligence is the ability to hold two opposed
ideas in mind at the same time and still retain the ability to function.�

(F. Scott Fitzgerald 1896 � 1940, American writer)



---------------------------------------------
This service is hosted on the Infocom network
http://www.infocom.co.ug

Reply via email to