why novice programmers should be encouraged to re-invent the wheel
--------------------------------------
i read a lot of programming books that tell people not to do
the wheel reinvention.
in my opinion.
sometimes beginning programmers should be allowed to 'reinvent the wheel'.
actually the analogy doesnt follow.
the wheel ,as a circle,is perfect,u can't make a circle "more rounder
than it is!!" but of course there is no "perfect software" yet.
if there were, microsoft wouldnt be in business.
xp was better than or an improvement over windows 98 which was a big leap forward from windows 3.e or so (correct me about this version here)
and so on.
they even rewrote their old api ,apple did the same thing too,
so reinventing the wheel is a part of the software world.

don't write software u can download . why not.
writing a software u love is a way of proving ur programming mettle to urself and maybe others.

e.g
some novice programmer may decide to try measure his muscle by rewriting some popular software he's heard about - if he succeeds,
even though he could have easily downloaded it from the net,he'll feel
good about himself.
this wouldnt have been so easy if he had to dream up the challenge first before tackling it.by pitting himself against something popular he'll be able to prove his worth to others.
so ,sometimes reinventing the wheel is good.
besides if one has fun writing software that he could have downloaded freely whats wrong with that-having fun doing sth - the harder way.

then,sometimes by trying to reinvent the wheel the programmer comes to
appreciate the amount of thoughtfulness that went into some of the applications he formerly took for granted and after he survives the ordeal, after many months of having stubborn bugs haunt him,he won't need somebody telling him to desist from re-inventing the wheel that can be downloaded from the net.

without this experience his mindset might not have changed for the better . it still boils down to being given the liberty to - do it again,his own way.

@ Fallguy,
You are rite.
Something its harder to rewrite your own software than to write a new one.

@fallguy. Reinventing the well when need be is a must. I've seen myself move from the band of a visionaire where i'ld just sit back and work my mind through designing some software system to being a techie type where i disdane use of other peoples work and jump into writing code for a similar type of library to being a tinkerer where i'm over joyed to have/use someones code for my new system and would have it no other way. It takes time,intelligence and experience to be able to weigh what u should self write and what u should just take handed to u. For one, i don't think i'll ever want to get involved in writing/developing a 3d modelling package, that will be murder.

reinventing the wheel is a waste of time and resources 99% of the time.

you are better off focusing your energy on solving new problems using an existing wheel and let others worry about improving the wheel

am glad some are beginning to reason independently of articles out there.
like i always preach, read the articles and digest the part that applies to you.

because Steve Jobs writes on his blog that 2+2=5 does not make 2+2=5

@all. Almost always is the case of never re-event the wheel stressed to new comers of a system that we who stress it fail to realise that an incomprehensible library or one that has a large memory footprint or slow or faulted in someway that can't be overlooked will get re-implemented by someone who can conceive of a better way to get it done. Take for instance the now 'extinct' Networking middle-ware from Microsoft called DirectPlay; professionals got stressed in understanding the internals of the system and implemented other more useful systems like RakNET and ReplicaNET. Does that make them go against all that's 'right'? No way. If a system doesn't look right for u and u can't find a replacement for it elsewhere, then go ahead, re-implement something more 'suitable' if u can. No one will 'hang' u for it.