(Excerpts from a message dated 10-10-99, Bob Wright to Murray Lesser)

Hi Bob--

 ML>     IBN made what it thought was a valid marketing attempt to push the
 ML> first Warp version (OS/2 3.0) to the SOHO market when that version was
 ML> first released (1994).  There have been many theories expressed on this
 ML> echo (including a grand conspiracy theory) as to why that effort was
 ML> abandoned.  I know little of the real story, but my own pet opinion is
 ML> that IBM finally learned that it doesn't know how to deal with a
 ML> nonprofessional individual user :-(.  I base this opinion, in part, on
 ML> those terrible 1994/5 TV ads for OS/2.

BW>"Terrible".... being a bit charitable are we??  :)

    Actually, I have war stories, some dating back almost 50 years,
illustrating IBM's characteristic inability to deal with individual
users that might be considered to be above-average "consumers," but I
felt that those tales were off-topic :-(

 ML>     An early version of Star Office (called "IBM Works" at the time) was
 ML> distributed (free) with the Warp 3 Bonus Pak.

BW>Not exactly...

BW>The suite that was released with Warp as IBM Works was the same
  >package as was on the Warp 4 bonus pak IIRC... Mine (from Connect)
  >has copyrights from Footprint Software (ISTR that it was originally
  >referred to as "Footprint Works", in fact), as well as IBM,
  >MasterSoft, Halcyon and Houghton- Miffin some of which are dated as
  >far back as 1983...

    Thanks (thanks also to John Thompson) for the correction.  This is
what I get for relying on hearsay instead of checking for myself.

 ML>  Where else can you find a PL/I compiler for OS/2?  (US
 ML> list price $2,999, although I got the current v 2.1 from IBM for much
 ML> less!)

BW>Of course, you first have to WANT one... ;>

    As does any programmer using OS/2 who has ever programmed in PL/I
and is still writing programs in anything other than REXX :-).

 ML>  I, too, use several shareware application programs (all
 ML> registered).  But very few large commercial concerns will use shareware,
 ML> largely because they want to be as certain as possible that the provider
 ML> of the software will still be around to fix it when.

BW>Although just being high-priced "commercial" stuff doesn't always
  >guarantee that the provider will be around -- or, in some cases,
  >interested in providing support..

    You are so right.  The worst supporter of high-priced software I
ever did business with was Microsoft with regard to a bug in its BASIC
PDS compiler that I reported in 1992.  The only response to my
well-crafted bug report (description of the bug, an example that always
displayed the error, and a workaround) was a form letter saying that my
report was being forwarded to the development group!  (Earlier bug
reports to Microsoft with regard to its BASIC compilers had always
produced a correction on a floppy, by return mail.)  The second-worst
was the late (and unlamented) Borland, whose response was to the effect
that since there was a workaround, and they had so many other bugs in
the product (C/C++ compiler v 1.0 for OS/2), they weren't going to fix
the one I reported :-(.  The best bug-report responses I have received
in the 1990's were from IBM, with regards to a bug I found in one of the
programs included in the first edition of the Warp 4 Developer's Toolkit
(repaired in the next release of the DevCon CD-ROMs), and to one in an
early version of the PL/I for OS/2 compiler (repair went out in a CSD
the following month).  I guess IBM doesn't know that I am an individual
user!

    Regards,

        --Murray

___
 * MR/2 2.25 #120 * You rarely get what pay for; never what you don't pay for.

--- Maximus/2 2.02