From: Jens Thoms Toerring <address@hidden>
To: Development with and of XForms <address@hidden>
Sent: Thursday, October 13, 2011 9:33 PM
Subject: Re: [XForms] Dennis Ritchie is dead
Hi Marcus,
On Thu, Oct 13, 2011 at 08:23:33PM -0400, Marcus D. Leech wrote:
> Yes, if there ever was a giant upon whose shoulders we all stand,
> Dennis Ritchie was surely one of them.
>
> I haven't
seen Dennis in over 20 years, but on the few occasions
> that I met him, and had a couple of beers with him,
> he seemed like a genuinely-nice person, and the most unassuming
> "giant" one could ever possibly meet. The entire
> IT/Software/Technology industry owes him a debt of gratitude so
> insanely large, and yet 99.9% don't even know who
> he is.
>
> The news of his death has affected me personally today. Not only
> because I've actually met the man, but because I learned
> C and Unix "back in the beginning" (1979). My life has been
> utterly shaped by having lucked into a job in 1979 where I got
> to learn C from "The C Tutorial", and program on Unix systems.
>
> Goodnight, curly-braces man. You'll be missed.
Thank you. I would very much have liked to meet him in person
as you did.
On a lighter note:
one of the really funny things I stumbled
upon was an exchange (seems to have been in comp.std.c back
in 1998) with a reply by Dennis Ritchie:
| > You are right. It was nice back in the days when things like
| >
| > #if (sizeof(int) ==
| >
| > actually worked (on some compilers).
|
| Must have been before my time.
|
| Dennis
I guess there wasn't a lot further discussions after that;-)
And another gem I just found:
| From: Dennis Ritchie <address@hidden>
| Newsgroups: comp.std.c
| Subject: Re: Integer Sizes?
| Date: Thu, 02 Dec 1999 03:19:41 +0000
|
| Someone (well, OK, Nick) wrote:
|
| > ... No, that is not the issue. You clearly have not thought of the
| > problem of handling external data formats, which was the rationale
| > for introducing the exact-width data types in the first place.
| >
| > A device driver or external
file data object of 32 bits long will
| > contain those bits in a certain order. There is no requirement for
| > that to be the same order that is used by C...
| >
| > The matter of byte ordering is too well-known to go into details;
| > it is common for the external value 1 when loaded as a uint32_t
| > to be 16777216. I have used systems where both big-endian and
| > little-endian bit orders WITHIN A BYTE were possible....
|
| Maybe by the time of the C0X standard, things will be pinned
| down well enough that it will become possible actually to
| write a TCP/IP stack, or conceivably an entire operating
| system in C. One might even imagine that it runs on several
| different machine architectures. A hope for the new milennium.
|
| Dennis
While never having met him that makes it obvious to me that
he must have been a very unassuming person
with an admirable
sense of humour.
Best regards, Jens
--
\ Jens Thoms Toerring ________
address@hidden \_______________________________ http://toerring.de