RE: Webfacing

> From: Buck Calabro
>
> Also, The WebFacing tool is a slam dunk.
(...)
> I get response time in the 2.7 second range after the first touch on my
> model 820 with 3 gig of RAM.

I have a slight problem with the terminology here.  A web application with
nearly 3-second response time is hardly a slam dunk, in my opinion.  There
is no way that a web application should require 2.7 seconds to bring up a
page.  On my little bitty model 270 (370CPW, zero interactive), my average
response time is 130 milliseconds - that's TWENTY TIMES FASTER than
WebFacing.

And it works under load, too.  With 50 users banging on the same machine,
with a 3-second delay between enter keys (that's 10-15 requests a second), I
still get subsecond response (950ms with 50 users on a zero CPW machine!).
Try simulating 50 users on your machine with WebFacing and see what the
results are.

I just don't understand why these horrendous response times are acceptable,
especially when there is no reason for it!  And it's not just my tool - I'm
sure Brad's e-RPG approach or Nathan's web development environment both
yield better results.  (PSC/400 has other benefits <grin>).

I think it's imperative that we nip this particular bit of bloat in the bud.
Two and a half seconds for a web page is entirely unacceptable.   On a
dedicated machine, anything over a couple of hundred milliseconds should be
considered SLOW.

Joe