[interesting part of discussion moved to the top -- bet and flames at bottom]
El wrote:
> Nemo wrote:
> >
> > Average cost of getting a  PC with modem: $1000+tax+duty
> > Average cost of getting an MD:  $100,000+transportation
> > The nurse has to be there in both cases.
> >
> > So, if having the PC allows the nurse to reduce the number of times he
> > has to consult the doctor by say 20%, then that PC has paid back its
> > cost 20 times over.
> Salary of a Professor of Medicine in Nigeria is less then $3000 per
> year and he doesn't get transportation. Kenya pays a little bit more I
> hear, and places like the DRCongo much much less if anything at all. I
> can try and dig it up for Ethiopia.
> Before you ask, a Senior Medical Officer in Namibia gets a maximum of
> $NA35000 (but has to work serious overtime for that) and some
> transportation is provided.
> Salaries are however in local money (which the government of the day
> can print itself), whereas hardware and software requires foreign
> exchange. Which is one reason why there are so few computers. Which
> you of course, again, know full well.

I was talking about the true cost of producing a doctor for the
overall economy, not salaries.  So the $100k referred to the cost of
training a doctor (just the order of magnitude). A reasonable measure
of that is the tuition and other fees for a *private* medical
school. I'm not saying anything about the relative merits of private
or public universities, or African vs other doctors or whatever. Also
we all know that African doctors are underpaid, and that various
subsidies, regulations, even visa restrictions and currency factors
make their salaries or their fees completely unrelated to their true
cost, and even less to their true value.

The point of that number was simply this: the complete training of a
medical doctor costs 100 times more than the buying of a computer.  If
you say it's 80 or 50 or 150 times, that's fine too, you get my point.
Let's call that number R.

> But nevermind the salaries, you simply can not deduct as arbitrarily
> as you do, because you have to provide coverage.
> There are WHO recommended ratios of Medical Officers per population,
> but you need at least three in a district even if it is a small one
> warranting only one under the ratio.
> You have to be able to do operations such as Cesarean Sections and
> Apendicectomies around the clock, 7 days a week, which need
> anaesthetics and assistance (at least two individuals (surgeon and
> anaesthetist), legally speaking at least 3 (one assistant), for a
> Cesarean in fact 4 one for the baby). Even if you sat around during
> the day and twiddled thumbs, it's just not possible to provide cover
> safely with only two people over anything more then a week or so.

Precisely! I never said there should be fewer doctors. I'm saying that
given a certain number of doctors, who I assume are already
overwhelmed with serious cases, can we use technology (the web) to
allow them to make better use of their time? In economic terms, can we
make better use of the $100k that was invested in their education. The
answer, I claim is yes. If we can eliminate some fratcion F (plug in
your number here) of the doctor's load consisting of cases which a
qualified nurse with an uptodate, complete reference at his/her
fingertips could have handled alone, then that frees up the doctor to
do that much more apendectomies etc. than he would have been able to
before. So for every 1$ spent on the computer, you get an efficiecny
gain of F*R$ ; in my example, I put R=100 and F=0.2, so you get 1900%
return on your investment in the web.  What would you estimate for R
and F?

This is an example of one of the advantages of the web that are simply
not there with email or ftp alone.  It enables society to utilize
limited resources more efficiently. In this example, the resource is
the doctor, or rather her time, and by spending some money on the
computer and the web access, you can get an increase in efficiency
which more than pays for the cost of the computer. This is nothing
radical nor is it new. The same argument can be made about
books. Except that no library can ever be as complete, as up-to-date,
as inter-active and as cheap as the worldwide web.  And the computer
itself of course has many other uses (email, patient records, etc.)
which reduce its net cost for this example.

> > But let's settle this with the following: you post some symptoms
> > here, as a patient would describe to a nurse.  I, with my
> > non-existent medical training, will come back with a web-based
> > semi-diagnosis.
> And you will tell us how long it took you to find it.
> > If I fail,then  for one week after, I will replace my homepage with the
> > following four words: "The web is useless".
> >
> > If I succeed, you will have to dig up this messages again using your
> > wonderful glimpse search engine, and write a long self-criticism  (in
> > HTML, no less, illustrated with plenty of JPEGs and nice fonts).
> >
> > Do we have a deal?
> No.
> If you manage (and in less then one hour on line time), then for one
> week after, I will replace my homepage with the following five words:
> "The web is not useless".

Ok, but let's say 2 hours. Do we have a bet now?

> > We'll need to define what is a reasonable enough symptom, something that
> > a poor old bush nurse would not know, but that a normal general
> > practioner would, and also what constitutes a good enough diagnosis.
> You are mixing up symptom with diagnosis by the way.

I don't think so... I was saying you post a not too difficult symptom,
and I post a not too wrong diagnosis.

> > > > Well, if she had the funds, telephone, electricity, a computer, modem,
> > > > computer literacy, English, and the time she might actually try this
> > > > once. However, my village nurses read their treatment manual and if
> > > > that doesn't cover it to their satisfaction they phone me.
> > > >
> >
> > Oh so she has no electricity but she has a phone? Hmm...
> Now I could make some sharp comments about the quality of instruction
> in electrical engineering you must have received, but you know
> perfectly well how telephones receive their power, over the wire.

Yes you could but those sharp comments would be irrelevant and
baseless.   I wasn't saying a person must necessarily have electric
service to have a phone. I was implying that electricity has a deeper
penetration than phones, and therefore it is unlikely that a village
will have phones but not have electricity. I have seen many places in
Africa with electricity but no phones, but I have yet to see the
opposite.  So your claim that it is unrealistic for the nurse to have
a computer with internet access, but can have a phone to call you, is
not convincing to me.

> I share your resentment 100%. But I have never ever said or even
> implied anything like this.  I have been saying the exact opposite
> since at least 5 years again and again.
> I have been raving about the virtues of inexpensive Unix for
> inexpensive hardware (linux on PCs) for years with emphasis on state
> of the art.

Great! The state of the art evolves in network services as well as
CPUs and OSes. The web is one such evolution.  After our bet, when you
recognize the its value, then we will agree on everything!