Re: We know what you need, and we'll push it down your throat.

[prev] [thread] [next] [lurker] [Date index for 2006/12/28]

From: peter (Peter da Silva)
Subject: Re: We know what you need, and we'll push it down your throat.
Date: 17:08 on 28 Dec 2006
> Why isn't any system with a language (such as x86-compatible hardware, or any 
> operating system exposing system calls, or any editor with it's set of
> commands and menus) "a part of a lock-in strategy"?

When it's an open system.

That is, when the interfaces and protocols are publicly defined, are not under
the control of a vendor of said system, and the definition is complete enough
to make a production quality independant implementation feasible in a
realistic time-frame.

The last part can often only be determined by attempting an independant
implementation. :)

> So MS Office is surely a part of a lock-in strategy, since the formats are
> not documented, and it's hard to make a compatible program. But C# and .NET
> are documented, so why are they a part of a lock-in strategy more than C
> and Unix, or any program for that matter?

Let's see.

		Publicly	Publicly	Easily
  System	Defined?	Controlled?	Implemented?

C# and .NET	Yes		No		No
C and UNIX	Yes		Yes		Yes

Now let me back up these points.

While the UNIX Programmer's Manual was Copyright AT&T, the majority of
the API was published in Software Tools, including a Fortran implementation
of a C-like language and all the system calls that were practical to
implement in portable Fortran. Not only was AT&T not restricting the
use of the API, they promoted independant implementations of both UNIX and
C and only maintained trade secret protection of the source code itself.

The UNIX API contained one patented concept, which was explicitly released
to the public domain.

The UNIX API was not really stable until the 5th or 6th edition of the
UPM, and it really didn't get widespread attention until the 6th edition
was published, along with several important papers on the philosophy and
design of the operating system in the Bell Systems' Technical Journal, and
of course "Software Tools".

The idea of a portable OS and a portable systems programming language weren't
new... C borrowed from BCPL and CPL, UNIX borrowed from Multics... but the
fact that someone had actually done it and ported an OS to a completely
different platform was amazing.

Within a few years there were several independant implementations of UNIX
of various kinds: complete systems like Idris and Uniflex and Regulus, ports
on top of other operating systems like Cromix and Eunice and Phoenix, shells
like the Software Tools VOS, and so on.

Really, those books and papers from 1976 to 1978 pretty much created the
whole open systems movement. How are UNIX and C open? They *epitomise* open!
How can you get "locked in" to UNIX when you can get UNIX from Cromemco
or Whitesmiths or AT&T or Wollongong or Onyx or Microsoft?

Microsoft?

Microsoft was once a big noise in the movement, with Microsoft C and later
Microsoft Xenix... but some time in the '80s they came up with a strategy
that was more directly profitable for them, dumped Xenix since they couldn't
control UNIX, and started deliberately changing the systems they *did*
control faster than they could be copied.

This is not fantasy. It's documented in material from the trials and
independently by people like Marlin Eller (ex-Microsoft exec). People are
concerned about Microsoft's control of C# and .NET not because of what
Microsoft could *theoretically* do, but because lock-in has been Microsoft's
business model for over 20 years.

There's stuff above here

Generated at 03:02 on 01 Jan 2007 by mariachi 0.52