From: Lee Daniel Crocker (lee@piclab.com)
Date: Mon May 01 2000 - 15:30:14 MDT
> > ActiveX components are simply Windows executables. Making a runtime for
> > them would basically mean making a runtime to allow Windows programs to
> > run on other OS's. This is *not* a simple task, and not one Microsoft is
> > willing to allow (they refuse to document the APIs, hence everything has
> > to be reverse-engineered).
>
> What alternate universe are we talking about here? I know it can't be the
> real world, because I currently have on my hard drive a complete set of
> Microsoft documentation for pretty much the entire Windows API set. They
> don't quite give this away, but they come pretty close - it's only a couple
> hundred dollars a year for a complete subscription.
>
> Obviously they don't completely document every possible bug, side effect or
> unsupported call in existence, but that isn't because there is some
> conspiracy to keep it secret. Rather, it is because they only have so much
> money to spend on developer outreach. The information they do make available
> is far more complete that the docs I used to work with for Unix and Mac
> APIs, and about as good as I would ever expect to see for a product that
> changes substantially every few years.
Sorry, but his reality is a lot closer than yours. As a former Microsoft
employee involved in O/S work (NT 3.1 in my case), I agree that there was
never any specific intent to hide APIs from application developers.
Indeed, Microsoft donated office space in its own building for various
application developers to work with APIs under development.
But at the same time, it was clear that this benefit was extended only to
application developers that increased the share of Microsoft's OS
business. Free software devlopers need not apply, and developers of
competing APIs like WINE were clearly unwelcome. Word Perfect got
support; the little guys didn't. It was also clear that if one of the
MS application or compiler groups needed a new API for something, adding
it and getting it to work was a much higher priority than telling anyone
else about it (which we'd get around to eventually).
And ActiveX is another ball of wax entirely--it was, from its very
inception, intended to subvert the current de fact standards of the web
to increase the use of less-flexible Microsoft products. I may be on
Mike's side in the Anti-trust issue, but his recent posts about things
like ActiveX are clueless. It was never intended to be secure, or open,
or interoperable, or implementable by anyone else or on any hardware--it
was designed to do Microsoft-like things on Intel-like hardware, nothing
more and nothing less. Comparing ActiveX to Java is just marketing hype
with no relationship to reality. I happen to think that attempting to
undermine public standards is a perfectly legitimate business tactic if
you can pull it off; but I certainly won't argue that MS didn't intend
to do exactly that.
In fact, it disappoints me that the other Microsoft case (v. Sun) gets
no press, because that one is a great example of two things: (1) That a
competitor succeeded in creating a freely-available cross-platform
standard that really worked and that gained popular acceptance at the
expense of Microsoft, and succeeded so completely that Microsoft signed
a contract clearly agreeing to abide by the standard--in short, Sun won,
and demonstrated that monopoly power can be overcome. And (2), it shows
an example of Microsoft actually doing something wrong, by violating
their own signed contract to abide by the standard. This is not like
anti-trust case where they only violated silly rules the government
imposed on them, but a case of them willingly taking on an obligation
and then breaking the deal. For /that/ they should be drawn and
quartered, not for the stupid anti-trust stuff.
-- Lee Daniel Crocker <lee@piclab.com> <http://www.piclab.com/lcrocker.html> "All inventions or works of authorship original to me, herein and past, are placed irrevocably in the public domain, and may be used or modified for any purpose, without permission, attribution, or notification."--LDC
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:28:20 MST