On Thu, 19 Sep 2002, Muli Ben-Yehuda wrote:
> On Thu, Sep 19, 2002 at 03:23:27AM +0300, guy keren wrote:
> > that's not exactly what i meant. i was trying to (but failed to) make a
> > distinction between two types of clients - those who develope sotware
> > inhouse, and those who don't. for those who do - they will tend to shell
> > out the less interesting stuff - and that is natural, cause their
> > programmers want to handle that stuff. for companies with no in-house
> I take exception to this statement - every time I've been contacted
> for consulting, it's been because the company simply didn't have the
> knowledge of experience to handle whatever task was at hand. I agree
> with you for some fields, but consider something like kernel hacking -
> a company is likely to do the less interesting ( == easier) stuff in
> house, and hire an expert for the really difficult ( == harder)
you've got a slightly 'ping glasses' vision of things, then, i gather. my
expereience differs. i've sometimes been contacted to perform tasks they
didn't know how - and other times, to perform tasks they didn't have time
to do, or even didn't want to perform.
as for kernel hacking - a copany is likely to want to do it in house, if
its the core of what they do - as someone here already mentioned. in such
a case, they will want you do:
1. help them get started, if they don't have experience with that.
2. help them with advice on which things to do - whle they actually do the
coding, so they will learn.
3. help with specific components, that they could later 'take over'
one thing to consider, thought - the number of times you were contacted -
if you took them all, would that suffice for a living for a long duration
(= over 1-2 years)?
> > the other issue is the repetitiveness. after all - an expert is someone
> > who has already done something several times. the more you did this thing
> > (assuming you gave good results) - the more you'll get asked to do the
> > same thing _again_.
> So you do it better/cheaper/faster. As long as you're allowed to
> innovate and learn from past mistakes, I don't see a problem with
> doing the same thing more than once, because it isn't the same...
i didn't say "only do one thing once". i said "don't want to do the same
thing 5 times". there's a difference here. also, 'doing the same thing
several times' has a higher number of 'several' for small things, then for
large things. you can do the same very small task (few hours long) 10
times and not necessarily get bored, but if you do the same larger task
(say, several month) more then 2-3 times, you will get bored. and if you
do the same large task (several years) more then twice, you'll most likely
not like it. heck, even doing this large task twice might be considered
by the way, i take it for granted that every one has different 'bordom
factors' (did i mention GUI programming todya? ;) ), so that is why the
kind of business gilad talks about could work for some, while not for
others (from this perspective specifically).
"For world domination - press 1,
or dial 0, and please hold, for the creator." -- nob o. dy