Home Messages Index
[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index

Re: Microsoft Makes OpenOffice 'Dangerous'

Roy Schestowitz wrote:
> OpenOffice.org is a Minefield
>
> ,----[ Quote ]
> | As we all know, Novell's "customers" now also enjoy protection from
> | Microsoft litigation as well.

Microsoft seems very determined to undermine OSS any way it can.
The latest tactic seems to be to offer trojan horse plug-ins.

The cite below reminds us that Sun licensed software from Microsoft,
and that this software could be included in StarOffice, but not in
OpenOffice.  The OpenOffice "core" code is protected by OSS Licenses,
but the OSS license doesn't protect distribution of proprietary
implementations (aka StarOffice).

Keep in mind that StarOffice has Wizards, Themes, Skins, and enhanced
plug-in filters, including files to convert a number of proprietary
formats.  OpenOffice has a content driven filter (no reverse
engineering, but processing based on publicly available documentation).
 StarOffice has a Microsoft Licensed filter to convert MS-Office
documents to Open Office.

Sun and Novell are also getting licenses for OpenXML, which is
Microsoft's proprietary implementation of XML.  It's very similar to
ActiveX and has the ability to include inline binaries, execute them,
and have all of the access rights of the user - without the user having
knowledge of these vulnerabilities.

Microsoft really hates ODF, because ODF is a public standard
implemented entirely in OSS licensed code, which includes patent
enforcement waivers for the OO version.  It is as much of a threat to
Microsoft's Office as TCP/IP was to IBM's LU 6.2/SNA or DECs DECNet.

It's important to note that ODF is designed to provide EXACTLY the same
kind of standard as that of TCP/IP.

Back in the late 1980s, there were many who argued that ISO/OSI was far
superior, and at the lowest level, it was.  Unfortunately, there were
so many proprietary extensions, supersets, and deviation, that it cost
nearly $50,000 just to get all of the documentation for ONE developer.
And even with all of the books, it turned out that the protocol was not
implementable in it's entirety.

TCP/IP on the other hand, did have some flaws.  It didn't have enough
address bits, it didn't have enough length bits (which meant that
packets had to be padded to multiples of 8 bytes).  There were other
concerns as well.  What TCP/IP had going for it is a complete set of
public specifications implemented entirely in OSS code available in
public license source code that was freely redistributable.  These
could be used as reference implementations, or they could be used
directly in operating systems ranging from MVS to Windows to UNIX to
Linux.

As a result, TCP/IP was implemented on nearly every platform, and was
widely used in corporate networks.  Even when other servers such as
Netware were being used, the NetWare servers could communicate with
each other via TCP/IP.  When the Internet went commercial, it was
trivial to use the already widely available TCP/IP, including the
Trumpet Winsock (threadsafe variation of Berkely Sockets), to
interconnect millions of computers in a matter of months.

> | So, now you have your choice - accept Microsoft's license terms,
> | which both Sun and Novell have seen as necessary, or risk Microsoft's
> | wrath. And, having seen the lengths that they are willing to go to
> | stem the adoption of ODF, I am not taking any chances.
> `----

Microsoft was offering technology which made making the transition from
a "Windows Only" solution to a "Linux/Windows hybrid" solution much
easier.  At the same time, by licensing this technology, and assuring
that the commercial vendors offered the Microsoft licensed software
ONLY in the proprietary products, Microsoft is able to cash in on it's
own collapse.

Microsoft may be looking to sabotage Linux and OSS, or they could just
be trying to make sure they have some form of assured revenue stream
when Linux and OSS displace 20-40% of their traditional "Windows Only"
client systems.

Microsoft's track record has been dismal.  They've managed to stab
nearly all of their "partners" in the back.  Nearly every deal and
settlement looks good, but ends up being a poison pill to the partner
and pure cash to Microsoft.

> http://boycottnovell.com/2006/12/05/openofficeorg-is-a-minefield/

Keep in mind that most OSS projects have a core OSS framework, usually
in a "source only" license.  Then there are provisions to add plug-ins.
 The plug-ins can be OSS or proprietary.

> Macros aside, there is C# code being embedded and there are various moves
> which aim to destroy ODF adoption, including a self-commissioned 'study'.

OSS has always been a delicate maze.  Look at Linux.  The kernel is
GPL, the libraries are LGPL, and the X11 toolkits are other versions of
OSS.  Even Qt comes in both OSS and commercial flavors.

> The idea is to preserve control over the format and make OpenOffice software
> that's not free (fork if needed).

My guess is that the Microsoft portion will be offered as a "plug-in" -
in almost exactly the same way that the StarOffice plugins were
offered.  Look at Crossover.  This is an application which uses
Microsoft licensed technology to extend WINE to be compatible with
newer versions of Windows.  Win4Lin is a version of Windows Xen that
allows users to run Windows applications on a "bare bones" Xen client.

Keep in mind that, because Microsoft gets unit volumes and royalties
based on the the sale and distribution records, they have a very good
picture of who well these are spreading.  This may have been
Microsoft's big wake-up call.

Microsoft has figured out that they really have to play nice with
Linux, if they want a "soft landing" in the transition from
Windows/Office to Linux/OSS.

On the other hand, Microsoft has betrayed so many partners, defied so
many court orders, and weaseled out of so many settlements, that it's
hard for anyone, especially the OSS and Linux community (which has
grown to nearly 200 million users world wide), to really trust
Microsoft.


[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index