🚀 go-pugleaf

["Followup-To:" header set to comp.lang.java.advocacy.]
On 2004-05-24, David Kastrup <dak@gnu.org> wrote:
> In general, it is a good rule of thumb for good engineering to first
> work on a project until it basically functions, then scrap it and do
> another version from scratch, after having learnt all the necessary
> lessons.

Like that airport building in France perhaps?

Its generally accepted in the Windows world that it won't be expected to
work until version 4 or later, but no other engineering discipline that
I've come across works like that. If things like cars were made as badly
as software, then there'd be an outcry. If Ford sold cars with bad brakes,
then they'd be stomped on by consumer watchdogs and the competition. How
people just accept such poor quality in software I don't know.

> Windows needed about 4 completely different code bases in its history
> in order to learn its lessons.  They are only now getting to a point
> where they are close to being competitive.  Except in the gaming
> compartment which is their stronghold.

People improve on the past - just watch Java and .Net circle around
eachother. Over broad software patents may unfortunately put an and to
this. Steady improvement is better than radical reengineering though. 

Another daft thing is that people complain about the cost of changing to
Linux, but they're having to scrap their old VB apps and learn a
completely new programming paradigm with .Net. Active Directory was
completely new. XP's interface was apparently different enough to
confuse many users (doesn't say much for the average user). 

The attitude in the computer press seams that Microsoft could invent the
Completely Innovative Morse Code Input Device, and it will be praised as a
revolution in computer interaction. You only have to lift one finger now!
The daft thing is that then Linux will be blasted for not having a Morse
Code Input Device. Morse code would become "User Friendly" by definition
and the terrible old keyboard won't. Some Linux distributions would then
lose some lovely usability enhancements to become like Microsoft.
Meanwhile Microsoft will try to take these innovative ideas from Linux
and try to fit them in. 

I can see a possibility where Linux becomes Win95 to try to appeal
to those users, and Windows becomes Linux taking the multi user model
(C:\Documents and Settings anything like /home anyone?) and nice interface
ideas like file entry completion and "reveal desktop", multi desktops
and dock-apps in Longhorn. Its a strange world.

 - Richard

-- 
   _/_/_/  _/_/_/  _/_/_/ Richard dot Corfield    at    ntlworld dot com
  _/  _/    _/    _/      
 _/_/      _/    _/               Time is a one way street,
_/  _/  _/_/    _/_/_/                  Except in the Twilight Zone.