Re: Human minds on Windows(?)

John Quinley (jquinley@aros.net)
Fri, 9 Jul 1999 12:05:52 -0600

-----Original Message-----
From: mark@unicorn.com <mark@unicorn.com> To: extropians@extropy.com <extropians@extropy.com> Date: Friday, July 09, 1999 11:30 AM
Subject: RE: Human minds on Windows(?)

>Rob Harris Cen-IT [Rob.Harris@bournemouth.gov.uk] wrote:
>>I don't agree with this. If there's one thing that componentware will be
>>it's bloatware. See, it's unlikely that you'll find a set of components
that
>>service your system requirements alone, and no more.
>
>Yep, I was reading an article a few weeks ago about all the wonderful new
>bugs introduced by people using C++ classes without knowing what they
>actually did; users were finding all kinds of unprogrammed 'features' in
>applications which had been built into those classes for some other reason
>and still worked, often breaking the application in the process. The author
>also compared a C program he'd written which compiled to around 30k to a
>one megabyte C++ program he'd downloaded from the Net which did exactly the
>same job; disassembling it showed around 40k of code which actually did
>anything useful, the rest of the program was just unwanted crap from the
>C++ classes the programmer had used.
>
>But hey, it's only RAM, disk space and support calls, so who cares?
>
> Mark
>

I have written and managed programs > 100K LOC written in C++ that compiled down to less that 1 meg. It seems like the program your friend downloaded was not compiled with a compiler that strips out unused code (which is a feature of any good compiler).

John Quinley