__/ [marc_CH] on Tuesday 01 November 2005 22:29 \__
> In article <dk6vaj$13nj$1@xxxxxxxxxxxxxxxxx> newsgroups@xxxxxxxxxxxxxxx
>> >> The news got even worse: Longhorn was irredeemable because Microsoft
>> >> engineers were building it just as they had always built software.
>> >> Throughout its history, Microsoft had let thousands of programmers each
>> >> produce their own piece of computer code, then stitched it together
>> >> into one sprawling program.
>> > That's how *all* non-trivial software is written, dingbat.
>> Watch you mouth please. These were not my words. They came from an
>> article, which I happen to agree with.
> It's your agreement with which I am taking issue. How do you think large
> software projects are written? How do you think large software projects
> *should* be written?
Okay, fair enough.
Software should be written to become more cohesive by using specifica-
tions, use cases and smarter ahead-planning. If an operating system per-
mitted access and full control to any hacker in the world, something had
definitely gone rotten. If even a patch was incomplete, it is then negli-
>> Microsoft often portray the Linux development 'model' as one which
>> involves many 'cowboys' building standalone components. Prior to this
>> revelation, it was assumed that Microsoft used their vast resources to
>> build software in a more principled manner rather than compose 'code
> Where do Microsoft 'often portray' this about Linux?
Public speaking and word-of-mouth can be just as damaging as one would ex-
pect. While it's true that several obscure applications have been coded in
somebody's garage, the core (kernel) is carefully administered and tested
by professionals. That is not what anti-Linux campaigns would have you be-
Yesterday, for the first time in ages, the extent of the zombie attacks
seems to have decreased. I hope it's not merely a one-off.