Dennis Ritchie, father of C programming language, dies

CNET News ^ | Oct 13 | Steven Musil

[budj]

"Dennis Ritchie, an internationally renowned computer scientist who created the C programming language, has died at age 70. ..."

[JRandomFreeper]

Are you quoting from the story of Mel?

Real men do program with a soldering iron.

/johnny

[JRandomFreeper]

I was thinking more along the lines of FF.

/johnny

[Ackackadack]

Whoops! The above code was botched since the PRE tag yanked the include directive symbols when I cut and pasted it from my xterm.

I was gonna say, #include... what?

[Erasmus]

I was just remarking on your rwxrwxrwx permissive nature.

[MarkL]

At one time, she was the head of the Software Development Lab at MIT.

Well, I guess that's the reason she hates C. That's where emacs was developed, a text editor built around a programming language! Real men don't need screen editors! We use ed or ex! Real, real men use here documents and hex editors!

Mark

[DaxtonBrown]

[Yeah, with C it will run “really, really fast,” but it will also give you wrong results due to bugs.]

A) real programmers don’t have bugs, they only have features in their code to prevent idiots from misusing their works of genius
B) if you are misusing your pointers or overshooting the ends of your arrays, that’s your problem for being an idiot, not the problem of C.

[JRandomFreeper]

I am like that, sometimes.

I'll own it.

Although in reality, I'm more of a 644 kind of guy.

/johnny

[JRandomFreeper]

if you are misusing your pointers or overshooting the ends of your arrays, that’s your problem for being an idiot, not the problem of C.

Some of us actually did that on purpose, thoughtfully, for a specific reason.

Once upon a time, it was the only way to peek and poke at the real metal.

And dumping a big ole whack-a-doodle right past the end of an array, right onto bare metal works great, should it actually run.

/johnny

[MarkL]

I got my start in computers in 1986 as a repair technician and taught myself programming by reading AT&T System III Unix source and system diagnostics also written in C.

Wow, I came up sort of similarly... I learned C on unix (no source code for me) with the 1st edition white book, the Russian Doll book, and the v7 Bell Labs programming guide and system manual, both of which could have passed as telephone books.

I'm somewhat bummed that I never really kept up with it. I did my last programming back in 1988 and joined the PC networking world.

Mark

[JRandomFreeper]

And I'm more obtuse than angular. I missed your joke compeletely, until I got it.

Then I LOL'ed.

/johnny

[RussP]

“And dumping a big ole whack-a-doodle right past the end of an array, right onto bare metal works great, should it actually run.”

---

Isn’t that “feature” also a classic security hole?

[JRandomFreeper]

It's all actually much easier these days.

You can read the source, and see what the drivers are doing without reverse engineering them.

And there are no overwhelming bookshelves to deal with.

The documentation that you'll never live long enough to read is larger, and online.

/johnny

[MarkL]

I've still got the text for AWK, SED, and grep on the shelf

I think in my last move, I finally lost some stuff I was keeping for sentimental reasons: A number of the original Oreilly "Nutshell Handbooks," back when the covers looked like grocery store bag paper and stapled together, including their termcap and programming in curses books... And a couple of prized wall posters I had picked up from the original Computer Literacy bookstore... And my entire collection of Guru and Unix World magazines... I used to have a yellow t-shirt with Root the Dragon on it...

Mark

[NVDave]

There’s a reason why FMS systems are written in Ada.

When it has to be correct (or else), using C or C++ is like playing Russian roulette with a 1911.

[JRandomFreeper]

Isn’t that “feature” also a classic security hole?

Well, yes, technically. But I wasn't the one that brought it up on usenet, many, many years ago.

My hands are clean (as they ever are).

/johnny

[Gene Eric]

>> const char s[] = “Goodbye World!”;

const char s[] = “Goodbye, World!”;

[MarkL]

Actually... Some of us used that on purpose to write directly to memory.

As did I, using both C and IBM 360 System assembly language. Would occasionally write "self modifying code" to solve issues, but it could be a real pain dealing with memory segments on the Intel architecture. It's really vague to me now... It's been nearly 25 years since I worked as a programmer.

Mark

[JRandomFreeper]

When it has to be correct (or else), everybody on board is fornicated. I don't care if you are writing code for nukes.

Ain't no such animal. Not in C, not in C++, not in c#, or pascal or basic or ADA.

Anything that requires humans has flaws.

Own it. Live it. Revel in it.

But don't think you got something right because it had to be.

It just hasn't failed yet.

/johnny

[NVDave]

C had pointers to pointers because you could do that addressing mode on the PDP-11.

In the old days of V7, I could look at a line of C and tell you exactly which PDP-11 instructions the compiler would emit. The mapping between C and PDP-11 assembly is very obvious. When they started porting C off onto other architectures, C looked much more powerful than originally thought by PDP-11 hackers, because we could do all those things in one line of PDP-11 Macro assembly.

For example, copying strings char-by-char with pointers, with auto-increment. Blindingly obvious that it boils down to one PDP-11 instruction, with two registers, indirect auto-inc.

[Bobalu]

Yup, C is a dangerous sharp thing for a fool to play with.

I use C to program embedded systems.... GCC mostly.
ARM/AVR/TI...etc