Re: Correctness vs. speed

From: Jean-David Beyer (j_at_d.b)
Date: 01/21/04 

Date: Wed, 21 Jan 2004 07:57:31 -0500

Roger Leigh wrote:
> Kirk Strauser <kirk@strauser.com> writes:
>
>
>>At 2004-01-20T08:29:08Z, Noi <noi@siam.com> writes:
>>
>>
>>>Clients pay for results. They will only care about what you've written if
>>>it starts causing problems.
>>
>>I'm sorry, but this counterargument is becoming repugnant. Have you no
>>*pride* in your work? Is "good enough for now" really good enough for you?
>>Are you content with your clients and employers seeing your work and saying
>>"well, his code usually doesn't break"?
>
>
> Unfortunately, this attitude is common. "I don't care about how good
> it is as long as it works" is terrible.
>
> I have been criticised for spending too much time designing and
> planning my work. However, in the long run, the rushed jobs and quick
> fixes seem to have an annoying tendency of blowing up in your face and
> costing far more time (and customer complaints!) than doing a
> professional job in the first place would have taken. Meanwhile, the
> work that was done properly and painstakingly works trouble free for
> years, even though it took time in the first place.
>
>

Yup: you will spend the time to do the design at some point anyway.
Either up front before coding, or later when doing unit test,
integration test, responding to customer complaints, etc. The trouble is
that the later in the process you do the design, the more it costs
because usually all work done before that must be done over. And the
costs of a customer complaint can be considerable. A lot of stuff is
really advertized word-of-mouth. The advertizing by a dissatisfied
customer may have more power than anything a vendor can distribute
through the media. 

-- 
   .~.  Jean-David Beyer           Registered Linux User 85642.
   /V\                             Registered Machine    73926.
  /( )\ Shrewsbury, New Jersey     http://counter.li.org
  ^^-^^ 7:55am up 14 days, 19:21, 2 users, load average: 2.12, 2.16, 2.16