Difference between flashy and being effective

[jfm2@club-internet.fr]

It is very frustrating for most people including me having Indy being
little more than a RedHat clone woth some packages added like Indy 0.1
was.  It is of course much more glamorous to start a grandiose
programming plan who will end in an installation able to detect
hardware and read in user mind even in case there is little to read
due to ignorance and confusion.

But you have to consider some factors like time, real usefulness and
resources.


First: Writing an ambitious programs takes time.  Sometimes several
years.  That means that during this time you helped nobody to come to
Linux.  And sometimes you never finish because people become
disheartened, and sometimes you find that while you were writing your
program the world has changed in the form of the install in other
distribs being about as good as yours or being possible to ask to most
manufacturers that Linux comes preinstalled so your install is
useless.

Second: Your user has installed Linux thanks to your wonderful install
and then what?  Do you drop him in front of the command line?  Do you
expect that he will find by himself than the help command is 'man'?
Did you pay any attention to allowing him to ask for help or do you
let him to solve his problems by himself?  Do you assume than GCC is
the most exciting app for attratting common people to Linux?  You
have to consider all these points and many more before you decide a
superb install is the most useful thing you can do both for the user
and for Linux.  Because you have to remember that resources are not
infinite so what you assign to the install is not available for
improving what happens _after_ the install, and this (the _after_
install part) will not be solved the day every computer manufacturer
ships Linux preinstalled.

Third: When you look at the forms filled by people willing to
participate a striking fact appears there are plenty of volunteers for
web designing but programmers are scarce.  That is logical it is the
people who had had painful experiences with Linux who are the most
enthousiastic and programmers are not between them. Then you have to
consider that only a third or so of the people who fill the form will
answer the welcome teext and that between the people who answer either
because they didn't measure the effort it takes or because I am a poor
negotiator most of them will drop silently after a couple mails.  End
result is _one_ programmer and even for simple concrete tasks you get
no volunteers.  So even in the case an install would be high priority
how would it be possible to write one?


Therefore the strategy is shipping a distribution with adequate
software, make this software as ready to use as possible, remove
absurd assumptions inherited from proprietary Unix (who were used more
in mainframe like roles than as microcomputers) and learn to put the
infounder the user's eyes.

-- 
			Jean Francois Martinez

Project Independence: Linux for the Masses
http://www.independence.seul.org