Roy Schestowitz wrote:
> I don't agree. I think this is the more instinctive interpretation, which
> says that cost is being fixed at 0. In reality, you benefit because:
> * The wheel is rare, if ever, re-invented
True, a good programmer writes in such a way that he/she doesn't need to
rewrite the same functions/classes for each simmilar project. As a
programmer you learn to do that early in your carreer, even if it is just
from a lazy point of view, i.e. not wanting to type the same lines for
every project that you do.
> * There is incentive and resources (exiting code) for people to
> _improve_ that wheel
True, it is much easier for new budding programmers to get onto the
programming ladder in Linux, partly through cost, but also because of the
disciplined approach that tends to be taken in UNIX/Linux he/she can build
in a good stable predictable environment.
> * Projects are merging more naturally, which makes development
> cohesive (communal)
Partly true. You tend to get the projects in sourceforge launched to quite
an extent from the programmers personal interest. Understandable of cause,
who is going to give their own time to something that they have no interest
in. But because of this we were left, for example, for quite a time in
early linux with few drivers, because drivers arn't much fun to program.
So a lot of the dull side or programming was done by professional
programmers volunteered by companies. IBM did a lot of work in those areas
that were being ignored by the programming volunteers, the main distros
also put in work in these areas.
> Don't worry, Bear. People's skills will always be needed. If certain tasks
> can be automated, new opportunities will emerge which are less
> 'mechanical' in nature.
In practice it doesn't really work that way, what you tend to get from the
automation of factories, is less need for skilled workers, you can just get
the cheaper unskilled workers and less of them. Take the mines away and
they is no need for your 40 years of mining experience, build in car
diagnostics and you can get rid of the skilled mechanic and replace him
with kids capable of reading the print out. Automate the fault disgnostics
on telephony systems and you can have one engineer to cover the whole of
yorkshire, with many cheaper workers who are only trained on how to wire a
socket. Less skills make lesser people, less satisfaction, no future to
grow into just a life time of 'this'.
So people retrain as their skilled jobs are taken away, office fodder or
parking lot attendants, or people who can wire a phone socket, they train
to be less than they were.
Take away the need for professional companies to take on professional
programmers because they is no value in an end product, then who is going
to pay for the training of new programmers. Like many here I was self
taught, mainly because in 1979ish they wasn't anyone around to teach you,
but at the same time because I was lucky enough to be programming through
the build up of the systems and languages to what they are now, it gives me
an advantage over those poor soles who have it to learn from scratch now.
I agree that Linux gives them a better base to learn on than MS (which is
messy from a programmers point of view), but it is by no means an easy task
to self teach programming in Linux, even with the likes of Eric and the
IDEs. I don't think we could rely on a self teach approach in coming
generations because eventually that means that they are less skills
involved in the programming of our system. So who is going to pay the uni
bills? What student will do a course knowing that the number of jobs are
very limited at the end, because the companies who develop new ideas can't
carry on because they can't make an income out of it.
I know it isn't like this now, but can you really see where GPL is heading?
They are clues around, maybe we should start noting them.
Consider the applications on your current system, look in the menu for
'Office', even if you loaded the whole office section off your distro's
dvd, how long is it since that list changed? Same office applications,
newer versions, but where is That competition. Planner, that came from
UNIX, where is the competition for planner, where has it been added to or
built on since you first saw it, is they even a menu item change since the
first time you saw it?
Move to the graphics menu, that hasn't changed for many many years, again
newer versions, but not new applications.
I am not saying that change is necessary, but this compatition and 'building
on whats already there', it is only really happening in sourceforge in
those areas that peak a programmers interest. As the interest drains
because the product or idea is nearer to a complate application, then it
seems that we are happy for inovation to stop. But what is they that will
stop us going 'stale'. Because when that happens people will drift to
another OS for no other reason than there are no more toys to play with on
I am not sure what we should do about all this, if we ought to do anything,
a flower grows, has sex then dies and a new flower grows from it's seed.
Maybe that is where Linux is going, to have sex :)