John writes in regard to progress:

> More telling are the following statistics:
>
>     Time to complete one transaction using an HP 2934 connected via 19.2
>     serial connection to a classic Micro GX with 2 Mb RAM and running the
>     BASIC/3000 version of the program:
>
>           43 seconds (average)
>
>     Time to complete one transaction using Netscape 3.0 on a Pentium 200
>     connected to the same LAN segment as the MPE/iX 937 system, utilizing
>     the freeware NCSA server and native mode SPLash! code:
>
>           54 seconds (average)
>
>     CPU milliseconds per transaction (average)
>
>           311 ms (average) on Micro GX running BASIC/3000 program
>           562 ms (average of all processes involved - NCSA server,
>                  form input/output translator, and application server)
>                  on MPE/iX 937 system
>
>  Wow, what progress!

What John didn't mention in this head-to-head comparison is that the 937 is
about 30 times faster than a Micro GX -- and if the GX's code were moved
(without modification) over to the 937, ancient as it is, it should perform
the same transaction is just one or two seconds.

Recognizing this, it makes "progress" all the more difficult to define.

Wirt Atmar