Linux is the product of Standards - Re: Linuxes are all proprietary.

From: Rex Ballard (r.e.ballard_at_usa.net)
Date: 09/18/04 

Date: 18 Sep 2004 12:13:06 -0700

Jeff Relf <Usenet_2_@JeffRelf.Cotse.NET> wrote in message news:<_Jeff_Relf_2004_Sep_18_VWSy@Cotse.NET>...
> Hi Rex Ballard,
>
> Re: Your claim that Microsoft's products reduce productivity
> because they are not what you call " standards ",

Keep in mind that throughout the rest of the computer industry,
standards are established by groups representing industry vendors,
corporate consumers, and others who invest in long-term management of
information. Most of these decision makers and representatives are
thinking in terms of standards that will help them maintain access to
legally manddated archives for 7, 10, 20, 30, even 100 years. I still
have to keep a copy of Windows 95 on one of my hard drives so that I
can access quicken files and turbotax files from 1992-1999.
I also need it to access archives stored on Colorado Memory tape
drives.

The key is that all of the factions must agree on common baseline
technologies, and even when optional extensions are added, the core
systems function independent of the extensions. Backward
compatibility is critical as well.

We had standards bodies such as the IETF, ANSII, ISO, CITT, and IEEE
which publish proposed standards, collect review and feedback, make
sure that the standards are complete, and make sure that they can be
implemented based exclusively on the reccomendation. In many cases,
these standards are used as guidlines such as the Federal Information
Processing Standards (FIPS) and other standards used for purchasing.

The most critical part is that an established standard cannot be
altered or obsoleted by some new enhancement. Even when new revisions
of established standards are proposed, the review process assures that
there are negotiation procedures to make sure that an HTTP 1.0 browser
can read from an HTTP 1.1 host, and vice versa.

Microsoft calls it's "brute force" technology "de facto standards",
but that term is actually fraudulently misleading. There are no
processes which prevent Microsoft from making revisions which are not
backward compatible, or implementeting systems that break forward
compatibility. In fact, breaking backward/forward compatibility is a
critical element of their revenue stream. If Windows 95 users using
Office-95 could read all of the documents generated by Office 2003
(saved in "office 95 format") without losing critical formatting
information (the "saved" version ends up looking completely terrible
when viewed by the Windows 95/Office 95 software), then it's not
really a standard.

Standards also assure interoperability between vendors, without loss
of core functionality. If Word 6.0 was a true standard format, then I
could view it with WordPerfect, Lotus WordPro, Applix Office,
OpenOffice, StarOffice, or any version of word, and it would look
exactly the same regardless of which application saved it, and which
application was used to view it.

Consider your Web Browser. When industry standard HTML is used, you
can't tell from the display on the browser whether it was generated
using Netscape Communicator, FrontPage, Word using "Save As HTML", or
using vi or emacs.

You can't tell which server is serving it either. When you look at
the content using a text editor, you can see differences in style, but
you could even strip out things like specified fonts, and still view
the content in a useful format.

> You wrote: <<
>
> ...some of the biggest productivity " losses "
> were the result of the fancy proprietary technology.

I also pointed out how these proprietary technologies, lacking the
benefits of public reviews for security, interoperability,
completeness, and possibilities of "harvesting" information from one
format to another, have resulted in productivity hits such as:
   Viruses
   Worms
   Trojans
   Back-door breach servers.
   Manual reentry of data.
   Manual reformatting of cut/paste data.
   Manual searching of large archives.
   Manually managed revision controls.
   Loss of work previously done (loss of "Good State" products).
   Loss of new work (due to reluctance to back-up).
   Loss of all work (due to ineffective backup resources).
   Loss of all functionalities (reimage of hard drive to recover).
   Loss of key functionalities (lost or nonexistent installation
media/IDs).
   Loss of critical data (because personal files are
           stored intermingled with application code in
           the same directory trees.
   Loss of critical data - due to inability to back up open files.
   Loss of archives - due to incompatibility with previous versions.
   Manual transformation of archives - opening word documents in
      one version so they can be saved in the newer version.
   Loss of systems - due to Manual configuration procedures via GUI.
   Loss of systems - due to lack of fully functional scripting.
   Loss of systems - due to lack of automated maintenance.
   Extended development time - due to GUI-only development techniques.
   Extended development time - due to lack of automated testing.
   Extended development time - due to focus on GUI while
     ignoring business rules.
   Extendend development time - due to frequently changing APIs
     without warning.
   Extended procurement related costs (not including actual purchase)
due to
     lack of "test drive" capabilties.
   Extended procurement related costs (due to wrong choice -
switching).
   Extended procurement costs (due to vendor "dog and pony shows").
   Extended procurement costs (due to vendor incompatibilites).
   Extended procurement costs (due to Microsoft's "revolutionary"
changes).
   Extended integration costs (due to vendor incompatibilities).
   Extended integration costs (due to OS incompatibilites).
   Extended Integration costs (due to API incompatibilities).
   Extended Integration costs (due to API changes by Microsoft).
   Extended Integration costs (due to API incompatibility between
platforms).
   Extended Integration costs (due to Version syncronization
problems).

Is it any surprise that many of those who have COMPLETED Linux
transitions have been reporting long-term savings as high as 80% (cut
costs by 1/5th) over the long term?

> Proprietary plug-ins and applications
> ( MS-Office Macroviruses, OLE transformation ( lack of it ),
> ActiveX controls ( Virus factory and worm garden ),
> and time-wasters ( Video Games on corporate PCs,
> Flash Macromedia pop-ups, general pop-ups,
> spyware, and SPAM ). >>

> You're a smart guy Rex, but you got this one ass-backwards.
>
> Why is Linux continually being forked ? !

Actually, Linux isn't really forked. The Linux distribution is based
on the Linux kernel, which has very much AVOIDED forks. This means
that there is a common baseline. Furthermore, there is a great deal
of effort to maintain backward compatibility. Most applications
written for a Linux 2.2 kernel will run on a Linux 2.4 kernel or even
a 2.6 kernel. This means that less time is spent trying to re-port -
look at the way Windows NT 4 server applications were expected to be
reported to take advantage of Windows 2000 features, and Windows 2000
applications were supposed to be rewritten to use Windows 2003
functionality. Those who failed to do so found that their
applications ran SLOWER on the newer platoforms. Even SQL Server 7
ran slower on Windows 2000 than it did on NT 4. Microsoft even went
to court to try and prevent that information from being made public -
claiming that the benchmark violated the EULA agreement. The judge
ruled against Microsoft, noting that Microsoft was trying to use the
EULA to perpetrate fraud by preventing the publication of results they
didn't contest. Microsoft knew that NT4 apps were slower on Windows
2000. They just didn't want that fact made public.

Many companies did have to commit extensive - and expensive -
resources to reporting the same application to the same core platform,
simply because Microsoft decided to "change the rules".

Linux DISTRIBUTIONS consist of numerous components. Many of which
COMPETE with each other. For example, KDE vs GNOME. But at the same
time, these components are required to COOPERATE with each other as
well. All applications must comply with a number of established
compatibility standards including the ICCCM, IETF standards, glibc
standards, ANSII C standards (with consistent extensions available on
all platforms), POSIX-1, POSIX-2, and POSIX-3, X11R6, and so on.

The advantage - to the consumer is that they can use the GNOME desktop
with the Kommander browser. They can use GIMP on KDE, or they can use
both on FVWM. This means that there is less need to alter underlying
infrastructure and more resources can be focused on creating new
functionality.

Also, because these are modular components, there is less dependency
on a single massive package. I can use OpenOffice for word
processing, or KOffice. Both will run equally well. I can publish a
document using OpenOffice which will look good on OpenOffice,
StarOffice, MS-Office, Lotuse WordPro, WordPerfect, or KOffice -
without manual modifications. If I publish a similar document using
MS-Office 2000, the columns will be screwed up on WordPerfect, the
watermark will try and turn the documents into postage stamps. On
Lotes WordPro the columns will be completely ragged, the tables will
be the wrong column widths, and the images can even end up on the
wrong pages.

> Can you say: " Linuxes are all proprietary "

Actually, they are not proprietary in the sense that there are
nondisclosures agreements that prevent you from looking at code,
publishing comparisons between products, or indentifying potential
problems. In fact, those who get Linux code are encouraged to review
it for security risks, patent issues, copyright issues, or any other
potential legal entanglements. They are encouraged to review it early
and often, so that these issues can be managed as early as possible.
Debian might opt to not distribute something because there is a patent
issue. Red Hat and SuSE might work out a very reasonable royalty
payment for copies sold commercially. Others might even remarket
these core products - paying them a royalty to cover the range of
intellectual property rights.

In many cases, companies like IBM, HP, and SUN even contribute their
own intellectual property, waving patent rights on the basis that
others will follow suit.

> And that's even true when no money is involved.

Actually, it's true when money is involved too. Look at how much UNIX
HASN'T forked. I can quickly move from Linux to AIX to Solaris to
HPUX to UnixWare to BSD and be fully productive in a matter of
minutes. Not only can I quickly adjust to the minor differences
between the systems, I can even "personalize" my environment so that I
get a consistent interface regardless of which platform I'm working on
(And I work on all of them). When I switch to Windows, I can spend
hours getting a newly reimaged system to work the way I want it to
work. When my laptop Windows box get reimaged, it can take 2-3 weeks
to get it back to even the BASE functionality.

> " Proprietary " can mean simply controlling a version,
> even if no money changes hands.

> See Dictionary.COM, Proprietary: <<
>
> 3. ( Eccl. ) A monk who had
> reserved goods and effects to himself,
> notwithstanding his renunciation of all
> at the time of profession. >>

You went directly to the biblical definition and ignored the more
relevant and common meanings of the word.

   1. Of, relating to, or suggestive of a proprietor or to proprietors
as a group: had proprietary rights; behaved with a proprietary air in
his friend's house.
   2. Exclusively owned; private: a proprietary hospital.
   3. Owned by a private individual or corporation under a trademark
or patent: a proprietary drug.

Put simply, when you claim that you own intellectual property, and
that no one but you can control it, and that you can do anything you
want to it (including deface or destroy it), then it's proprietary.
Microsoft has even shown their willingness to destroy their own
property. Look at how they tried to destroy Windows 3.1 in order to
get everyone to switch to Windows 9x and NT.

Linux stanardards are more like national forests. The land is
available to anyone for certain types of use, but this use is
restricted, to prevent destruction of the resources and make it
useless to others.

> You say Microsoft's products are not standards...
> Hundreds of billions of dollars says you're wrong.

Actually, hundreds of billions of dollars says I am RIGHT!

Think of the hundreds of billions of dollars spent to replace
perfectly functional computers with new ones simply because Microsoft
didn't want to support them any more. Think of the billions of
dollars invested reengineering perfectly functional products simply
because Microsoft decided they wanted to change their APIs. Think of
the billions of dollars spent trying to fix preventable problems
simply because there were now public reviews of the "standards"
Microsoft just shoveled into Windows. There were now public reviews
for security, reliability, performance, maintainability, or even
iteroperability. In fact, Microsoft very tightly restricts all such
efforts to perform even the most superficial reviews - maintaining
that any comparisons between Windows and anything else must be
approved - in writing - by Microsoft.

When an independent lab does attempt to publish such comparisons,
Microsoft has the authority to force them to perform their own
carefully rigged comparisons, to exclude any references to the absurd
assumptions made, and publish the results as prominantly as the
original article published by the same company. Microsoft then
includes these comparisons in their "Fast Facts" (Fast with the Facts)
web pages. In most cases, the lab is even required to remove all
references to the prior report. Occaisionally you can find references
to the prior benchmark. The classic example was the Mindcraft
benchmarks, which, when conducted in the industry standard way, showed
Linux 2.2 to be about 20-50% faster than Windows NT4. Microsoft gave
them a new benchmark, using carefully doctored test conditions,
atypical configurations of Windows against "default" conditions for
Linux, to show that Windows was 10-20% faster than Linux.

The funny thing is that Linux 2.4 was optimized to eliminated some of
the bottlenecks discovered during this peculiar benchmark. The net
result was that there have been no subsequent published benchmarks -
since Microsoft has been unable to cook up anything like a standard
benchmark to make a comparison that was favorable to Microsoft.

> Windows is the de facto standard.

Yes. In 1960, segregation was the de-facto standard. Did that make
it right?

The means used to maintain this segregation was illegal, and in 1963
laws were passed to eliminate some of these legal aids to segregation.
 Was the fact that realtors wouldn't show homes to middle class blacks
that were located in predominantly white neighborhoods - to keep black
kids out of the middle-class white schools, right? Was telling a
black doctor or lawyer who had overcome extraordinary obsticles to
excel in his profession - that he had to limit his housing search to
slums and neighborhoods saturated with drug dealers, pimps,
prostitutes, and gangs - right?

Windows is the de facto standards because of contracts that have been
consistently ruled to be illegal. They are the result if
anticompetitive strategies that have been ruled to be illegal.
Microsoft has been under investigation for it's business practices
since 1987, and probably before that. In 1987, as part of a
settlement (which turned out to be worthless) the FTC voted 4-3
against prosecuting Microsoft for fraud and illegal marketing
practices. In 1990, Microsoft was again under investigation and in
1993 it's per-processor licenses and bundling practices were ruled to
be illegal. In 1999, a whole series of bundling practices were ruled
to be violations of the Sherman Antitrust act. During the course of
the case, including the testimony of Microsoft executives, numerous
counts of fraud, extortion, blackmail, obstruction of justice, and
collusion were recited in the courtroom. Unfortunately the scope of
the case was limited to the Sherman Act violations.

Judge Jackson's decision was remanded - because he conducted an
interview with a reporter while his clerks were revising his final
rulings on the remedy. During that interview, he pointed out many of
these criminal activities, and admitted that he was no longer
interested in giving the Microsoft Executives the opportunity to
further perjure themselves and flaunt their appearant immunity from
the law, as they tried to justify criminal acts they freely admitted.

> It is the network, irregardless of the connection employed.

Actually, no. That Web browser is 98% *nix. The web sites you visit
- are 98% *nix. The infrastructure between them - 98% *nix. What
Microsoft "added" and the "proprietary" elements - were the very
features that made it possible to spread the worms and viruses and
spam.

Ironically, if there had been a public review of all of these
"innovations" introduced by Microsoft, they would have been
indentified as nearly identical in principle to the holes that allowed
the "Morris Worm" to collapse the Internet back in 1987. They might
have been willing to accept the concept of certificates, but they
might have insisted that certificates be regulated by the regulatory
agencies that regulate communications. This would have made it a
federal crime to apply for a certificate under false identity, it
would have made disclosing a certificate to anyone without a search
warrant a crime, and it would have forced certificate authorities to
submit to audits and regulation similar to those imposed on telephone
carriers.

If AT&T had declared - back when they were a Monopoly, that they were
going to record and monitor any telephone number and make all of those
recordings available to anyone willing to pay the price of admission,
nearly everyone would have been screaming for the immediate
disintegration of the telephone companies.

Ironically, this is exactly what Microsoft has done.

Since the Introduction of IE4, the first really "proprietary" version
of their web browser, Microsoft has functioned as an unregulated
communications carrier. They have the capability to examine any file,
on any machine, and any part of the memory, on any machine. They have
the capability to obtain any certificate issued to any recipient of
certificats from nearly all of the certificate authorities (Verisign,
Thawte, Equifax...), and they can provide that information to any one
they please, for any purpose. They can provide this information to
law enforcement agencies without a search warrant (functioning only as
an "anonymous informant"). They can even access the computers of law
enforcment agents, political officials, or anyone in their staff, and
make that information available to their opponents. They even
participate in regulated industries, maintaining a substantial share
in companies that provide everything from real estate and banking to
cable and sattellite carriers and services. They directly control a
number of television networks, and can easily influence the coverage
of several others.

Watch what happens when Microsoft gets hit with something really
nasty. Watch the MSNBC and MSN sites early the following day. There
will almost always be a really good "scoop" to provide diversionary
coverage.

Microsoft has made $Billions (roughly 400 billion over the last 15
years), but 3rd party vendors, customers, and competitors have wasted
nearly $1,000 for every dollar Microsoft earned.

Rex Ballard
http://www.open4success.org

Relevant Pages

  • Re: French Gov handing out Linux in schools
    ... >> Which happens to be Windows. ... Many of Linux/BSD is coded to the standards which are designed ... > Microsoft on a project and I was a team member. ...
    (alt.computer.security)
  • Linux is the product of Standards - Re: Linuxes are all proprietary.
    ... thinking in terms of standards that will help them maintain access to ... have to keep a copy of Windows 95 on one of my hard drives so that I ... Extended integration costs. ... Extended Integration costs (due to API changes by Microsoft). ...
    (comp.os.linux.development.apps)
  • Re: Need help on modifying and assembly of a small program!
    ... "broken"...every other "main" browser handles the situation correctly, ... per the W3C standards and RFCs that they should... ... Microsoft won't listen to "put a human on the end of the postmaster ... have two hard drives in it (one 15GB, the other 4GB...yes, not too great by ...
    (alt.lang.asm)
  • Re: Microsoft Says Recovery from Malware Becoming Impossible
    ... Windows malware scourge, a Microsoft security official said businesses ... You say MS takes standards and alters them so they do not interop, ...
    (microsoft.public.security)
  • Re: Why Lisp is too hard for me to use
    ... Windows benefited from the standardization and contributed ... Microsoft had for such a long time. ... There have been a few standards ... but for a different reason; if they were not so dominant and so ...
    (comp.lang.lisp)