CGI vs. Middleware

Others have written:

> > An interesting perspective, but in reality custom-built CGI will never go
> > away.  Like assembler routines, there are some situations that require
> > low-level design.
> 
> > > My opinion is that CGI/Perl is "old school" and that the future of client/server www functionality is clearly based on in-process middleware 
> > > solutions.

Ingo Macherius writes: 

> Not true: Perl is low level

I  agree  -- I  would  go  further:  the  Korn  shell  is  also a
high-level language, which can be used to assemble pieces of that
venerable set of reusable  modules called  Unix(TM).  And we have
found that the shell makes an excellent 4GL for use with CGI.

> Not true: Perl CGI is slow
>    There are several efforts to speed up the CGI process. ... [snip]
>    ... there is even the possibility to run a daemon[-]like
>    CGI script that holds a persisten connection to your DB ....

In fact, CGI is extremely  *fast*, if you don't have to interface
it to a big, slow DBMS  process  (which we  don't!).  I challenge
*any*  Web-DBMS  implementation  to compete with our  application
(CGI, Korn Shell, agrep, glimpse, and /rdb) on performance!

>  information != knowledge != wisdom != truth != beauty != music == best (FZ)

I agree with that, too ... great .sig!

Cheers, 
--Steve.
                                           oo _\o
                                            \/\ \
                                              /
____________________________________________ oo ____________
"Sometimes you're the windshield; sometimes you're the bug."
- Knopfler

Stephen C. Waterbury         Assurance Technologies Division
Code 310, NASA/GSFC                           CAE Specialist
Greenbelt, MD 20771                            and Webmaster
Phone/FAX: 301-286-7557/1695        
email:                         waterbug@epims1.gsfc.nasa.gov
WWW:        http://arioch.gsfc.nasa.gov/people/waterbug.html
____________________________________________________________

Received on Thursday, 19 September 1996 11:31:27 UTC