[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
Re: Something to do While Waiting for Compiles
- To: <tribble>
- Subject: Re: Something to do While Waiting for Compiles
- From: Michael McClary <michael>
- Date: Tue, 14 Aug 90 10:14:52 PDT
- Cc: <vlad!mark>, <xtech>
> From: tribble (Eric Dean Tribble)
> I'll happily argue that clean code is a
> wonderful thing, however it is secondary to working code.
I'll happily agree. But I'll also point out that, though it is
in principle possible for dirty code to work, examples of dirty
code that DOES work - especially in multi-programmer projects -
are vanishingly few.
Far more common is dirty code that works most of the time, with
confused programmers wondering why their mirror doesn't focus or
their NFS occasionally eats files. I don't want to add "their
hypertext backend occasionally burns the world library".
I have no objection to using protected rather than private variables.
What I object to is a handfull of half-implemented coding standards,
left for "when we've got time" to finish them, and things like a macro
named "PRIVATE" that expands to "...protected". One is an itch.
A dozen are a disaster - and amount to a guarantee that things will
never work well enough that you can "get around to" cleaning them up.
Clean code is for when the schedules are tight, not when you have
leisure. The code needs to be clean, not because it's beautiful
that way, but because clenliness is a programming tool.
In this particular case the problem may be minor, but (thanks to
markm's tool) the solution should also be trivial.
I agree, to varying degrees, with many of your other points as well.
But I have other things to do, so I won't answer them until I have