The Anatomy of a Frontal Assault on Apache: Microsoft's Web Server Strategy
Jun 28, 1999, 12:46 (57 Talkback[s])
(Other stories by Conrad Sanderson)
Microsoft's Steve Ballmer has recently made some interesting
noises in Austria. As mentioned in LinuxToday and
Kanzler Klima ... (translated via Babelfish), during a keynote
address (and/or an interview) Ballmer said "Apache is simply
better." And then: "It is our own fault if we do not offer
enough features with our servers to justify their price. When it
comes to hosting several sites on one server version, Apache is
simply better. Windows 2000 will solve this
This is a warning sign 1000 miles high and wide that is easy to
decipher knowing Microsoft's tactics and their history. Everybody
knows what happened to Netscape. Let me dissect what is going to
happen to Apache (and indirectly, to Linux):
- Microsoft admitting that Apache is better ? Let's remember that
Microsoft is more of a marketing company - they would never, ever,
I repeat, ever do that. They are too accustomed to spreading FUD
and proclaiming "our stuff is better than competition" even when it
clearly isn't so. Since they are admitting that the
price/performance/features equation of Apache is better than
anything Microsoft has to offer, it means they have gotten used to
the idea, mulled it over, and came up with a solution. Apache is
not just hurting them financially in the server market - it is also
hurting them in the statistics game, and Microsoft hates not being
on top. They hate not being in control. At last count Apache has
57% of the server market while MS had about 23%.
- The MS solution to the Apache
problem: Windows 2000 will have a simple
(lightweight?) web server - very easy to setup and maintain. This
will be "integrated" into the operating system, just as
Internet Explorer was in Win98.
- This MS web server will also be available as a free download,
for use on Win95 and Win98 machines. This is done purely to spread
it and populate it everywhere - make use of the existing Windows
userbase, since the Windows 2000 upgrade (read: migration) will
take some time. Have you counted how many pirate MP3 sites use
primitive FTP servers running on Windows ? The installed user base
of Windows machines runs into the 100s of millions, while Linux and
other UNIXes that run Apache are only a small percentage of
- This new free MS web server will not be as fast, or feature
rich as their current IIS server. They still want to make some
money out of it. People will be easier to convince to upgrade from
the free server to IIS after being used to a Microsoft product.
Even if the free web server eats into their IIS sales, their
thinking is that it is at least a Microsoft web server and not
- ... which leads us to: Microsoft will add some Internet
Explorer-only features into their new web server (small at first),
thus attempting to control the server and client
side at the same time. Using Netscape to access the web server ?
Too bad it won't do SomeCoolNewFeature - normal people and
Windows geeks won't be impressed...
- The new free web server / Windows 2000 combination will reduce
the appeal of Apache, and hence, Linux (sinceit is the primary
platform on which Apache runs on). The main
attraction of Linux/Apache is the price and functionality.
To most people, its Open SourceNess is a
secondary bonus. Microsoft understands this and is
attempting to weave a self-fullfilling prophecy of "Linux is
only a fad". Stability is of course important, but Microsoft's
Marketing Power (tm) will make sure that this is non-issue when the
advertising (read: hype) campaign starts for Windows 2000.
- Attacking Apache with this kind of tactic is not legally
painful as integrating the browser was. After all, Apache isn't
made by a company. To the legal system and Microsoft, Apache came
out of the ether and its secretary is a website.
- Overall, this strategy isn't much different from what Microsoft
did to Netscape. MS is yet again leveraging the Windows installed
user base to counter a new threat: Open Source Software. But this
threat is of much greater importance and greater stakes are at play
here. They know OSS works. They know it has better stuff than they
make. They know it develops quicker. They are scared of becoming
obsolete. But most of all they are scared of not being on top, not
being able to dictate. Not being able to control.
- Fix the performance problems in the Linux kernel and Apache as
soon as possible, to create yet another viable technical reason for
choosing Linux/Apache over Microsoft products and take the wind out
the sails of Microsoft's PR benchmark department.
- Increase the population of Linux before
Windows 2000 is released by having a good office suite for Linux
(KOffice and Corel Office are good candidates).
Steve Ballmer on Linux: "one of the five
problems, which employ me before falling asleep. But I sleep
nevertheless still quite well."
Let's make him sleep unwell.