Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 19 Nov 1999 20:53:00 -0800
From:      "David Schwartz" <>
To:        "Thomas Valentino Crimi" <>
Cc:        <>
Subject:   RE: Marketing vs. technical superiority (was: Judge: "Gates Was Main Culprit")
Message-ID:  <000001bf3313$1ee4ce90$>
In-Reply-To: <>

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help

>> I don't believe that. Why wouldn't we all just stick with the previous
>> generation? Shouldn't we be just as tied to it? If this was true,
>> this be an argument why _no_one_ would upgrade? This argument  fails
under a
>> reasoned analysis and has no empirical evidence to support it.
>   I believe there is very strong logical support, and empirical evidence
> wherever you go:

	Actually, all the empricial evidence points the other way.

> It's one thing for me to tell my friends not to send me Word documents
> as attachments ("Email me another word doc and I'll email you my
> kernel"), but when a potential employer sends me an offer written in the
> file format of their choice, I do my best to find the software to read
> it - it's rather embarrassing to say "I'm sorry, I can't read your
> document, convert it to something leigible and send it again" -
> sometimes you have the leverage to do that, sometimes you don't.  If a
> large part of the 10% of people who get the "wizz-bangy" new computer
> every year installed with the latest software in their office are the
> people many want to communicate with, more people upgrade, and
> eventually you have to match the majority or be the 'special case' in
> everyone's book.  It wastes their time and yours.

	Do you understand what 'empirical evidence' is? What happened to you
personally, or how your Aunt Edna feels is not empirical evidence of a
market trend.

	I can point to just as many such arguments as to why people shouldn't
upgrade. I mean, I think "do I want to be part of the 10% that gets new
WhizBangy features and risk having 90% of the people being unable to open my
documents?" But that isn't empirical evidence either.

	There have actually been several fairly through studies looking for tie in
effects and tipping effects. They've not found any evidence of any in the
comptuer software market. That doesn't mean it doesn't exist, but it does
mean that your anecdotal evidence should be met with suspicion.

> So, yes, compatibility (and therefore having similar systems) is king.

	If this were true, nobody would upgrade unless the benefits were so massive
it was worth losing compatability.

> And so long as someone worth doing buisiness with is upgrading, it may
> very often be your requirement to upgrade as well.

	In other words, if it's worth upgrading, it's worth upgrading. And if it's
not, it's not. If the upgrade offers features people want, they'll upgrade.
If not, not. This means that Microsoft has to keep innovating to keep market

> In a perfect world,
> word processors would use a common file format where features would
> degrade gracefully (similar to HTML ignoring tags it doesn't know).

	If you believe that, why not make one and sell it? If this would allow one
group to upgrade and get the new features and another group to never have to
upgrade, it sounds like it would be a win all around. Microsoft will go out
of its way to help you -- even Microsoft's competitors admit that Microsoft
supports its developers better than pretty much anyone else around.

> Obviously, the best thing for a company to do is to make token
> improvements every year and make the file formats incompatible.

	This would just cause people to stop upgrading. Worse, it would allow a
competitor who made real improvements or committed more to compatability to
steal the market easily.

	In any event, I think this whole issue will become irrelevant in a few
years. It seems like HTML/XML and friends are going to become the
interchange format of choice all around. No thanks to the FreeBSD/Linux
people who are doing everything they can to kill it.


To Unsubscribe: send mail to
with "unsubscribe freebsd-chat" in the body of the message

Want to link to this message? Use this URL: <$1ee4ce90$021d85d1>