RE: HTTP 500-more

> From: Nathan M. Andelin
>
> To use these APIs in a CGI application, you must bind the CGI program to
>  *SRVPGM QZHBCGI in library QHTTPSVR.

A change in one manual isn't enough to remove a functioning API, Nathan, at
least not in my opinion.  If in BPCS we changed the name of a standard
program and shipped a new release to customers, two things would have
happened:

1. The customers would have raised holy hell
2. The person who renamed the program would have been fired

Why number two?  Because the only way they would have gotten such a stupid
change out the door would have been by circumventing procedures.  Making a
WORKING service program STOP WORKING should have set off so many red
flashing lights that IBM should have done something a little more serious
than update a manual.

Even so, I don't think your read of the HTTP server manual is entirely
correct.  Especially since IBM is STILL telling people to use QTMHCGI.  Do a
search on the IBM site on QtmhRdStin and check the first 10 or 20 links.
Whenever a service program is mentioned, it's QTMHCGI.

http://www.as400.ibm.com/developer/java/documents/jcv4r5.html
http://www.as400.ibm.com/developer/ebiz/cgi/tnt_article.html
http://www.as400.ibm.com/developer/ebiz/cgi/workarounds.html
http://www.as400.ibm.com/developer/ebiz/cgi/srv.html

That first link is especially interesting, since the document was released
in November of 2000.  In fact, it brings up an interesting question - do you
use a DIFFERENT QtmhRdStin for CGI programs than for Java programs?  Is IBM
confused?  Aer we confused?  How about you, Nathan, how do you explain the
two different documents, one in 1998 saying (according to you) that QTMHCGI
is dead, while another released in 2000 says to use it?  Oh my aching head.

My contention is that not even IBM bothered to check the ramificatios of
this change, and in fact it is my suspicion that the person who made the
change doesn't even have a clue that it was a bad thing to do.  This last
point is the one that frightens me.

Joe