Re: OpenDoc [was: Plug-In Spec ]

Paul Prescod (
Thu, 14 Mar 1996 13:45:56 -0500

Date: Thu, 14 Mar 1996 13:45:56 -0500
Message-Id: <>
To: "Daniel W. Connolly" <>
From: Paul Prescod <>
Subject: Re: OpenDoc [was: Plug-In Spec ]
Cc: Paul Prescod <>,

At 12:18 AM 3/14/96 -0500, Daniel W. Connolly wrote:
>"head and shoulders above"? I've heard this claim, but I'm still looking
>for hard evidence to back it. 

I based my statement on two things: my experience with SOM, which, to me is
very simple because it encapuslates tried-and-true OO paradigms such as
inheritance, dynamic binding, etc. vs. COM, which seems (to me) kinda like
the inside of a C++ runtime. 

I've also seen OpenDoc (and SOM) perform really quickly and I've only ever
seen OLE drag my machine to its knees.  

The other standard OpenDoc features are multiple active objects, irregular
frames, well-defined UI guidelines, explicit differentiation between
"viewer" and "editor", source code availability, CORBA compliance, language
independence (due to SOM), multi-page objects and an easy transition to a
distributed object model (perhaps exaggerated). Microsoft has announced
counterparts to some of those features, and might have delivered on them by
now. It's hard to remember what products Microsoft is actually shipping and
what is only a press release.

Here's some references for more comprehensive comparisons. The first two are
from Mac magazines, but they seem quite even handed. No, this isn't an
informational RFC, and I don't think I'm the right person to write one.

An old, but quite comprehensive article:

For the record: I think this article is a little outdated in its view of
SOM. Using SOM a couple of years ago was like C++ 10 years had to
do a lot by hand because your tools didn't know what the hell was going on.
This article was written in that period. SOM and C++ have become a lot more
integrated since this article was written. Even so, the last time I used SOM
was in the era of this article, and I found it flexible, powerful and easy
to use.

Another fairly comprehensive article:

IBM's View:

Biased, of course.

Microsoft's View:

I can't find anything on Microsoft's Web site. Maybe someone else can point
me to information representing their side of the story.

 Paul Prescod