[vox] [fwd] [NBLUG/talk] Q/A with RMS

[Brian Lavender]

RE http://blog.reddit.com/2010/07/rms-ama.html

The interviewer seems to discount LISP, similar to the way some in
Lugod think that C is a very good language. To me, C is very dangerous
and most programs are difficult to analyze for correctness. There are
too many things a programmer can do wrong. C should be left for simple
small programs or used to implement a higher level language.

I wouldn't discount LISP. While at OSCON I had a realization of one of
the things that computer science is really about.  A program is really
just a list of instructions recongnized by a grammar to which lexical,
syntatical, static, and dynamic analysis are applied. Apply an input
state and out comes an output state. To which languages that implement
pattern matching and tail end recursion are very powerful and will start
to dominate the language frontier. I realized the power of SCALA, ML,
and Erlang. And, they're almost exacly like the syntax of a Context Free
Grammar. While I don't have a lot of experience with LISP, I think most
discount it. I am sure we will see more of it in the future, and not just
for coding emacs, although I understand that emacs is a good example of
a core implementation in C and langauage built around it in LISP.

LISP is probably why Stallman wishes (if he only had one wish from the
magic proprietary software genie) that Autocad was liberated into free
software. It uses LISP type instruction.


On Mon, Aug 02, 2010 at 04:07:34PM -0700, Bill Kendrick wrote:
> Seen on NBLUG's list.  I honestly don't have time to go look right
> now, but thought I'd share it.  Enjoy!
> 
> (For the uninitiated, RMS = Richard Stallman, founder of the GNU
> project and Free Software Foundation, creator of Emacs, GCC, GDB, etc.)
> 
> -bill!
> 
> 
> ----- Forwarded message from Troy Arnold -----
> 
> Date: Thu, 29 Jul 2010 21:50:27 -0700
> From: Troy Arnold
> Subject: [NBLUG/talk] Q/A with RMS
> To: talk at nblug.org
> Reply-To: "General NBLUG chatter about anything Linux, answers to questions, etc." <talk at nblug.org>
> 
> Dunno how many of you frequent Reddit, but there's a forum where people
> volunteer to answer questions that the community throws at them.  Someone
> persuaded RMS to respond to a handful of the top questions (as voted on by
> the community).  Here are his responses:
> 
> http://blog.reddit.com/2010/07/rms-ama.html
> 
> It's the most interesting read I've had in sometime and leaves me with even
> more respect for the guy.
> 
> Not enough to use Emacs though ;-)
> 
> -t
> 
> ----- End forwarded message -----
> 
> -- 
> -bill!
> Sent from my computer
> _______________________________________________
> vox mailing list
> vox at lists.lugod.org
> http://lists.lugod.org/mailman/listinfo/vox

-- 
Brian Lavender
http://www.brie.com/brian/

"There are two ways of constructing a software design. One way is to
make it so simple that there are obviously no deficiencies. And the other
way is to make it so complicated that there are no obvious deficiencies."

Professor C. A. R. Hoare
The 1980 Turing award lecture