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

Re: When Companies Control Newspapers

DFS wrote:
> Rex Ballard wrote:
> > Remember, I used to work for Dow Jones.  I remember at least one
> > incident when Walt Mossberg, the technology columnist, wrote an
> > article singing the praises of Solaris, or Linux, or some competitor
> > to Microsoft.
> >
> > Microsoft called one of the owner/executives, told them that since WSJ
> > liked Solaris so much, they could let Sun buy all the ad space that
> > Microsoft was about to pull.  They not only pulled their own full page
> > ads at $250,000 per page, but they also pulled the ads of  several
> > OEMs, claiming that they didn't want their logo misused.
> >
> > Over less than a week, WSJ had something like $2 million in lost
> > revenue.  That would have been right after Solaris was released.
> > About 1993.
> How do you know about this incident?

As I said above, I worked for Dow Jones.  I was working on the
DowVision project, and working to help 25 companies provide TCP/IP
based services to serve the real-time feeds that were delivered via a
private X.25 feed.  We ported our server software to Solaris, HP_UX,
and VMS, and we had a fully functional SLS Linux server as early as
February of 1993.  Later we switched to Slackware 1.0, which introduced
us to the WAIS search engine, the NCSA web server, and the Opera
browser (I had seen opera earlier, but it was still very "rough").

Given that we were interested in supporting these other operating
systems, and had them on hour desktops, there were numerous discussions
about supporting some kind of "Desktop UNIX", including Linux.  I even
wrote a white paper for management, suggesting in 1993 that it would
take years for NT to become stable enough to be a suitable candidate
for corporate desktops (it took 4 years from when I wrote the article
for NT 4.0 to be released, and for SP3 to make it stable enough for
corporate desktops).   I also suggested that Windows 3.1 would not be
able to compete against OS/2 (once they got the race condition fixed),
or almost any version of Desktop UNIX.

By 1994, Sun had Solaris for PC, SCO had a good workstation solution,
Novell had a good UnixWare Workstation (unfortunately they wanted extra
for TCP/IP), BSD was still missing some critical pieces (eventually
contributed by Novell, when they own AT&T Unix), and Linux was sitting
on my desktop, running almost as well on an 80386/16 as the Solaris on
SPArC machine.

Is it any surprise that when Microsoft started pulling full page ads in
retaliation, that I might be one of those who was informed?  At the
time, they just wanted me to keep a "low profile" for a while.  Keep my
name, and Dow Jones, out of Microsoft's headlights.  This was when I
switched from posting from a Dow Jones account, to a Digex account.

>  What date was the byline?

I joined Dow Jones in January of 1993, I had Linux working in February,
I wrote a paper in March, and I was "Mr TCP/IP" by April (a derogatory
lable at the time).  I was considered slightly nuts for wanting to put
Dow Jones on the Internet.  We met with WAIS Inc, in November, by which
time, we had decided that UNIX and TCP/IP were the wave of the future
in servers, and Web Browsers were the future for Windows/Desktop.

I do have weekly status reports, but this communication was delivered
"off the record", and it was clearly a warning.  Shortly after this,
Microsoft approached Dow Jones about becoming a "Strategic Partner" in
developing a "News Service" based on Exchange, which would use a
Microsoft controlled version of "something like AOL".  DJ would send
feeds to a "master" exchange server, which would "replicate" to other
corporate exchange servers.  The whole thing was very much like Lotus
Notes team rooms.  We were approached some time in May of 1993.  The
other developers, who had been supporting UNIX, were reassigned to NT
duty.  One of the developers, Mike Bird, was hired by Microsoft, then
"rented back" as a "Microsoft Consultant", along with 2 other Microsoft
consultants who regularly "backed up" their systems.

Eventually, Microsoft killed the "Exchange Team Room" concept, Mike
went on to work with Microsoft on other projects, and the other
Microsoft consultants were also pulled out.  Shortly after that,
Microsoft announced MSN, Sidewalk, and Slate.  It seems that Microsoft
had been milking DJ, Rueters, and possibly others, for technology -
then just "rolled their own" from components they had obtained for
"evaluation and support".

Ironically, one of DJs big competitors, Bloomberg, ended up becoming a
heavy UNIX shop, creating custom interfaces directly to Solaris
servers, and even Solaris workstations.  Dow Jones also used Sun
workstations for Copy Prep and assignment of catagory codes, all in
near real time.

> You  remember the exact amount of the ad space, why not the competing techology
> or the date?

I'm pretty sure it was Solaris.  Remember, Dow Jones had been using Sun
for Copy Prep and other "core" functions for quite a while.  Solaris
was released about 3-6 months prior to NT 3.x, and was actually pretty
exciting at the time.  It was fast, simple, easy to use, and Sun even
had Windows 3.1 emulation (WABI).

Microsoft was spending $4 billion per year, pretty much for favorable
coverage of Windows NT 3.1, which wasn't even out in Beta form yet.
Those who were seeing it were having some serious trouble with it.  The
NT development team at DJ was having some serious problems with race
conditions and deadlocks due to threading.  At the time, NT didn't
offer interproccess communication, and the prior method - creating a
DLL with static variables pointing to shared memory buffers, was broken
by NT memory protection.

It was quite amusing, reading some of the "glowing reports" of NT Beta
experiences.  Keep in mind that the user interface was identical to
Windows 3.1, which meant there wasn't that much that was exciting
there.  3rd party applications usually didn't work right, especially if
they used inter-app communication via shared memory in DLLs.  NT would
simply "copy-on-write" and the "recipent" never got the modified
memory.  It was possible to redesign applications to use the NT shared
memory functions, but the redesigned applications would not run on
Windows 3.1 properly.

> I sent an email to Walt Mossberg (I know who he is - I used to read his WSJ
> column) asking if this is true.  If he replies he will say it's not true,
> and I'll post his reply to cola and prove, again, what a liar you are.  The
> last time I asked a 3rd party (James Gosling) about one of your nonsense
> claims, his response also proved you a liar.
All he said was that HE did not remember me.  I suggested that you talk
to Bill Joy, but you didn't.  Or did you notice that there were
exchanges between me and Bill in usenet newsgroups?

You made ONE query into ONE reference, which even I admited was
misworded, and then assumed that EVERYTHING I've ever said was wrong.

I must be doing a pretty good job of Linux advocacy, for you to spend
that much time doing that much homework.  I'd be curious to know how
many of these references you DID verify, and just forgot to mention.

> > In another even more grizzly situation:  While working for McGraw in
> > 1995, one of the magazines, Byte, was giving positive coverage to
> > Microsoft's competitors, and not being as nice to Microsoft as they
> > wanted - especially with reguard to their coverage of NT 3.x and
> > Chicago previews.
> >
> > Microsoft again tried the "squeeze play", but the editor responded in
> > a public declaration, reminding readers and owners alike that Byte was
> > dedicated broad and unbiased coverage of the entire industry, not like
> > some of it's competiitors who had just become publicity rags.
> >
> > Microsoft took more drastic measures.  They began pulling full-page
> > ads from several other of the 172 McGraw-Hill publications.  The
> > pressure was intense, and eventually Terry McGraw solved the problem
> > by selling Byte to C-Net.  C-Net couldn't handle the pressure either,
> > and eventually Byte was taken out of print.
> All lies.  Every statement you made in the last 3 paragraphs is a lie.
> You're a congenital, serial liar.

Perhaps you should talk to Terry McGraw, and ask him why he felt he
needed to sell off Byte.  I was wrong, it was CMP, not C-Net that
bought Byte.

> Byte shut down in '98 due to business problems.  It had nothing to do with
> Microsoft.  Educate yourself via the Byte Senior Editor from '92 to '98:
> http://www.halfhill.com/bytefaq.html#Q3

According to this cite, Byte was sold in 1998.  I left MGH in 1997, and
even by then, it was clear that Byte was under pressure.  There had
been memos circulated.  Furthermore, being a very vocal Linux advocate
and Unix advocate by then, I was frequently the first to get "warnings"
to not be a "trouble maker".  Again, I resolved the issue by using a
"covert" e-mail address and keeping my name and employer out of
Microsoft's radar.

Your site was interesting, I especially liked this part
If BYTE was really so popular, why was it in trouble?
Oddly enough, BYTE might have been too popular. Most of the proposed
turnaround plans would have voluntarily reduced the circulation by
various means, including raising the cover price. The reason for this
is not easy to grasp for those outside the publishing industry. It's
not always desirable to have the largest possible circulation. The
cover price of a magazine often doesn't pay for the enormous costs of
printing, shipping, and mailing the magazine. The profits come mainly
from advertising revenue, which means a magazine has to be attractive
to advertisers as well as to readers. That's where BYTE was hurting.

Quite simply, Byte was very popular with readers, and up to a certain
point, had been very popular with advertizers.  He doesn't come out an
"spell it out", but he hints indirectly that something had very
suddenly happened which had essentially "shut down" revenue coming from
advertizers such as the major OEMs.  Within a very short time, in about
1995, Byte suddenly found that their revenue had dropped by a
substantial percentage.

Check revenue details for other McGraw-Hill publications at that time.
You will find that computer related revenues also dropped for these
other publications as well.  McGraw-Hill used OS/2 1.2 to actually
deliver the S&P feed, including changes in share distribution, several
times during the day.  When someone "accidentally" dropped an alnico
magnet into the magnetic installation media, we were forced to upgrade
to Warp 3.0, and soon after that Warp 4.0.  Warp was such an
improvement, so much faster, so much more reliable, and so much better
than either it's predecessors or NT 3.1 or NT 3.5, that it got some
good press.

Shortly after that, Microsoft's retaliation was swift and intense.
Terry eventually had to choose.  He sent directives to the editors,
essentially saying "play nice with Microsoft", Byte was the lone voice
of defiance.  See Below:

Back to frequently asked questions
Why were BYTE's ad revenues declining?
We'd like to believe that a magazine popular with readers would be
popular with advertisers, too. But it's not necessarily so. Modern
advertisers tend to prefer a very focused audience. It was always
difficult to articulate BYTE's focus.

>From its start in 1975, BYTE was a technology-oriented, multiplatform
magazine. BYTE adhered to that formula for 23 years. But that's exactly
the opposite of almost all other computer magazines, which are
product-oriented and platform-specific. The vast majority of today's
computer magazines will not cover any platform besides "Wintel"
(Microsoft Windows on Intel x86), and they will not cover new
technology until it's packaged as a product for sale in stores. BYTE's
approach was very different.

This pretty much "spells it out".  Since Byte was giving objective and
often positive coverage to technology other than "WinTel", their
advertizing revenue suddenly dissappeared.

Keep in mind that Microsoft had publicly announced prior to this that
it would be "monitoring" the use of  it's trademarks and logos in
advertizing.  My guess is that federal investigators, interrogating the
right people, under oath, would reveal that Microsoft had ordered the
shift of advertizing by OEMs who used microsoft logos - to stop
advertizing in Byte Magazine.

It's clear that the author had received direct communications to this
effect.  Like a good editor, he doesn't have tape recorded
conversations, smoking gun memos, or the other "back-up" materials to
make claims in a nationally branded Magazine stating that these OEMs
were threatened, so he can't publish it.  But he does allude to it.

I also find it interesting how intense the reaction was:
What happened at BYTE on Wednesday, May 27, 1998?
All publication activity at BYTE Magazine ceased on May 27 and almost
all BYTE employees received layoff notices, effective Friday, May 29.
Out of 85 people in all departments, about 80 were laid off. Those who
were retained were mainly in the sales and marketing department. The
entire editorial staff was let go, with the exception of one person who
was offered a job; he declined. We cleaned out our offices and departed
at 5 p.m. Friday, May 29.

Quite simply, they were given no notice, no warning, no chance to back
up important records, and no chance to challenge the decision.  CMP
effectively "executed" 80 out of 85 people, putting the ENTIRE
editorial staff out of work.  However, they kept the marketing

> > Keep in mind that Microsoft's direct advertizing budget is over $4
> > billion per year.  In addition, Microsoft controls placement and
> > content of all ads featuring the Microsoft trademarks and logos,
> > including all ads placed by the big OEMs like HP, Dell, Sony, and
> > Gateway.  That's another $20-40 billion in advertizing that can be
> > pulled almost instantly, or over excruciating months.
> That's another ludicrous statement by a paranoid individual with no
> knowledge of the terms of contracts between MS and OEMs.

What makes you so sure?
Keep in mind that OEM contracts have been submitted as evidence, and
were not sealed until after they were made public.  During the
antitrust case, numerous contracts were "sanitized", but the body of
the contracts, along with their terms, were obtained by subpoena and
made available through the online (pay-per-view) court records service.

I've already stated that I have reviewed most of the exhibits in a
number of these Microsoft cases, including the Antitrust case, as the
information was breaking.  It's a bizarre hobby, but it keeps me out of
the bars :D.

> <snip other bizarro claims>

Keep in mind that I was repeatedly challenged in numerous exchanges in
a public usenet newssgroup in 1997-8, by a mysterious person known only
as "Roger".  After a number of exchanges over about 6 months, Roger
suddenly stopped asking questions.  Shortly after that, the first of a
series of Antitrust lawsuits was filed.  Other states filed their own
lawsuits, and eventually the Clinton administration tried to "rope them
all in" into a single lawsuit.  The one concession he had to make was
that all AGs had to agree unanymously.

The Bush administration decided to scuttle the agreement, cutting out
several states (mostly Democrat controlled), and cutting a deal without
their approval.

Most of the AGs who filed those lawsuits are no longer AGs.  In most
cases, their political careers ended about the same time as the

It could just be a coincidence.

Once coincidence is "God working a miracle anonymously".

A whole series of coincidences all to the benefit of the same person or
generally points to covert conspiracy.

Don't worry DFS, nobody will touch your precious Microsoft.
Microsoft has proven that they are above the law, and that they can
break the law with impunity, and they are willing to pay "hand slap"
settlements to do so.

Ultimately, it will be the backlash of the end-users, the CEO, CFO, and
COO putting pressure on the CIO and CTO, and the formalized policies of
government agencies, corporate IT departments, and the desire for
published standards that will eventually break Microsoft.

As to WHEN that will happen, Bill Gates said, in 1997, that Microsoft
could not survive in it's current form for more than 5 years (hence the
desire to diversify holdings, move into games and entertainment, and
become less dependent on OS revenue).

If Vista "plays nice" with Linux, letting after-market vendors
customize OEM licensed machines so that Linux becomes the primary
operating system and Vista becomes a client under Xen or VMWare, then
Microsoft could last 5-10 years longer in the OS business.

If Vista "gets ugly", and tries to revoke licenses because they don't
like customizations made by VARs and End Users,  then Vista will no
longer be an attractive option.  VARs won't want to be "locked up" by
Vista, and won't be willing to pay extra for after-market Vista

If high-volume VARs and consultants start building millions of Linux
desktop and laptop machines at a time, for example, IBM's GS division
starts doing mass customizations, for corporate customers - then they
are likely to refuse to pay for, or accept deliveray of Vista machines
- since they will have to install "Windows Free" Linux systems.

Microsoft has always been unpredictable, and has been able to rapidly
respond to shifts in the marketplace.  There is a reason that they are
more concerned about Linux than they are about Apple, even though Apple
captured 12% of the market by volume last quarter.

Thus far, Linux deployments have not adversely impacted Microsoft's
revenue.  Most of the 1000 largest corporations, each with over 100,000
workstations or laptops, signed 3 year support contracts in October
2003, and agreed to pay substantial increases for the same support
provided to retail customers using OEM licenses.  This has assured
Microsoft of consistent revenue for the last 3-4 years, and since these
were renewals of contracts signed in 1997 and then in 2000, it's been a
business model that has worked for almost 9 years.

Ironically, it's the same business model used by Linux and OSS vendors
since 1994.  Companies like Red Hat, Caldera, Mandrake, and SUSE
offered the software for free, but offered the support for very
reasonable rates - often less than 1/4 the price of comparable Windows
licenses on similar products.

And this business model was FIRST PROPOSED in 1984, by Richard
Stallman, in his "GNU Manifesto" (I guess those were drafts, the Wiki
dates it as March 1985).

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