Re: Fool.com Rule Maker Portfolio

["Robert G. Brown" <bob@xxxxxxx>]

Adam Williams <adam@morrison-ind.com> wrote:
>Warning:  This message contains a high rant factor.
I've written comments on the original posting. I will respond to Adam here,
and I'll try not to rant, but no guarantees! :) Lets see how successful I am!

>Projecting ANYTHING in the computer industry THREE YEARS is absurd.
>Java? Linux? Network Computers? The Internet?  Who saw these coming?
Three years isn't impossible, it's "only" hard. Reasonable forecasts are 
usually exceeded. Of the above 4 technologies named, Gartner failed to spot
only one (Linux), and fairly early. The other three have been on their radar
screens since the early 90's, and they have been surprisingly accurate with
those.

Three years used to be two technology generations, now I believe its more
like three. This is going to lower batting averages, but in Baseball, you're
a Hall-of-fame candidate if you only succeed three times out of ten. By this
standard, Gartner's got a place in Cooperstown fer sure!

>Personally I can't refuse stating my opinion that the Gartner Group is a
>troop of high minded fools.  I've read several of their "studies" and think
>I have yet to agree with them, and according to my recollection their
>previous conditions about the "future" of the computer industry have always
>been dead wrong.
I've been reading Gartner Group studies since mid-1984, which makes me a 15-year
audience member. I don't know what articles you've read, Adam, and I have not 
sampled much of their publicly available stuff. I'm sure that there is some
variation in quality, and I pick up on the idea that Gartner has a bit of a 
blind spot when it comes to Linux. I happen to like what they grind out, but
my views are based on different things, and they have a learning curve to climb
in areas that have only recently popped up on their radar. It appears to me
that this is the case with Linux.

Gartner Group provides computer industry related information, surveys, analysis
and commentary to paying customers. These customers tend to be large corporates
with a stake in the business. I encountered them at one of the investment banks
I worked for in the '80's, when I was appointed to a committee charged with set-
ting the direction we'd take in terms of computing, communications, software,
all that "stuff". We got Gartner group studies of the IBM mainframe (and plug-
compatible) market, since we were big IBM customers, and we got their stuff on 
data communications, distributed processing, and office automation. These were 
provided on a subscription basis, and they were NOT publicly available. Sub-
scriptions ran in the several thousand per year area, but they were worth every 
dime. I won't say the research was "brilliant", but it was very, very good, and
showed an insight into the mind-set of each area they covered. They rarely wrote
anything unsound, and if you invested in the companies they covered, you could 
do very well indeed!

I don't know how much truth (if any) there is to some of the accusations I've
read on various web sites about Gartner being "bought off" by Microsoft. I do
know that I was reading the same stuff that IBM was getting, and never saw any
particularly "pro-IBM" bias in what Gartner issued. Microsoft does subscribe
to Gartner Group analyses, and I've had access to some of the subscription
based stuff about PC's, Microsoft, LAN's, and so on. In those reports, they do
no kow-towing to Redmond.

Having written all that, I make no apologies for them.. they're grown-ups, and 
they know what they're doing. While I feel that some of the more gross accusa-
tions leveled at Gartner by Linux advocates are excessive, Gartner has not 
helped things by reacting ambiguously to comments about their relationship 
with Microsoft, especially when it comes to public pronouncements about Linux.
Gartner characteristicly adopts the world view that most of its clientele have,
which is corporate multinational in nature. Should we therefore be surprised if
they don't handle Linux too well or seriously, at least initially?

One mark of the "arrival" of Linux is that Gartner appears to be taking it 
seriously now, and we'll see how true this is over the next couple of months.

>And what is the  "network effects"?
Suppose you're the only one on earth who owns a telephone. It's worthless, you
have no one to call. Telephones are more valuable to you if all your friends
have one, more so if every store you want to call has one, and so on. The
value of a "network effect" product is proportional to the number of users.
Twenty two years ago, when I got my first access to what would someday be 
called "the Internet", almost no one knew what I was talking about. Now, we
converse with people in that medium (like this list) knowing ONLY their e-mail
addresses. E-mail is more valuable now, since more people use it. That's the 
"network effect". 

In my response to the initial posting, I stated that network effects will not 
help Linux, since Linux allows free choice and relies on open protocols, while 
products that require proprietary protocols generate a network effect of their 
own. Telephones produce a network effect, but now that we all have one (yes, 
I'm ignoring the poor, nomads in Afghanistan, etc.) network effects don't help 
any particular long-distance company. They compete on a (more or less) level
playing field based on price, better service, reliability, quality, and so on.
That how I visualize a computing world dominated by Linux, et al.

>.... Everyone talks about "right to innovate" etc... concerning the
>Microsoft Trial.  The real point is that the broke the law,  the same law
>that applies to everyone.
Actually, this has been a great rhetorical red-herring. There is no "right to 
innovate", and the anti-trust laws are NOT there (primarily) to protect the
consumer.

Microsoft protests about an infringement on the "right to innovate", which seems
a clever code word for "We claim the ability to decide what the term 'Operating
System' means". Well folks, I got some news for ya... they don't have that
right! The term "Operating System" has an accepted meaning throughout the 
computer industry, and it doesn't change if and when one company decides to 
include something in one of its products. Other vendors (IBM or Red Hat) have
never had this problem. You can tell when IBM is talking about an OS or about 
a product based on the OS as a platform (perhaps after training, and being 
blessed in a ceremony that enlists you in the arcane discipline of IBM catalog
reading, but he distinctions are there). The same holds true for EVERY vendor
I've ever encountered -- except one!

Innovation is a natural activity that results from competition, but it is not 
a right that explicitly exists anywhere. Patent and copyright law are there to
help product innovative activity (whether these laws or the government is
working correctly with regard to them is another discussion), and no one wants
to stop innovation... most people want to nurture innovation, and that means 
broadening it; allowing as many innovators as possible into the market.

How Orwellian of Microsoft to compose and co-opt the term "Right to Innovate"!
I'd like to see a list of other organizations for whom Microsoft spoke out in
protection of their "right to innovate". Sorry, in light of the Finding of Fact
and their own behavior, the "Right to innovate" rhetoric, coming from Micro-
soft rings awfully hollow. Sounds like they want to reserve the "Right to 
Innovate" for themselves, on their terms.

Now, the anti-trust laws are NOT there to protect "the consumer", at least not
as a first priority. The people who's rights are trod upon first and most by a
monopolist gone wild are the other market players... aka "the competition".
While one may make the argument they "my" rights have been violated by Micro-
soft, the main focus of anti-trust litigation is in protecting the rights of
Netscape or other competitors in the broadest market.

As a minor note, once an organization has been found guilty of violating the
anti-trust laws (things have NOT reached this stage in USA v. Microsoft), it is
much easier for me (as a consumer of MS products) to show that I am also 
eligible to make damage claims, as well.

"Consumer Benefit" or "protection" is another canard. The people handling the
public face of this case for Microsoft knows that there are more consumers than 
there are Suns and Netscapes, and they clearly want to be seen as "protecting
the consumer", especially when coupled with the "Right to innovate". Similar
suspicion would be well applied, in my opinion, to any politician who comments
on these things; there are more voters than there are competitors.

Hogwash vaccine is excellent medicine.

>I'm not certain is hasn't forced the industry into stagnation, Linux hasn't
>made it that far.  
I can't claim it HAS, but there are a lot of signs that this is true. Here's 
and experiment for everybody: Close your eye right after reading the following
phrase:
                                 Word Processor

Did you see a Microsoft word panel?

Think of this as the degree to which Microsoft has managed to set up peoples
expectations of what interfaces "should" look like. I don't think of this as 
stagnation per se, but it's a powerful influence on the market, even on deve-
lopers who want to break out of that mode.

>Without gaining the desktop Linux will die,  proprietary
>extentions to HTTP, WinY2k, and such will make it a non-viable server.  We
>already experience much pain from vendors and customers with IE only web
>sites, and the pervasiveness of ActiveX makes the situation even worse.  It's
>impossible for us to simply "not use" those sites.
This is an excellent point! It's been made elsewhere in different ways, but not
as compactly. By doing this, Microsoft provokes confrontation, since anyone 
(not only Linux developers, but ANY software developer in that space). Some of 
the energy of the Linux community must be "soaked off" in continuing to be, to
some degree "Microsoft compatible". However, the economics of Linux lend itself
to this a LOT better than any corporate competitor Microsoft has ever encoun-
tered. Linux developers seem to be emerging pretty much as needed, and this may
be something that, in the medium term at least, the community can sustain. It
is also no wonder why all the OTHER computer corporates are jumping on the Linux
bus, since without duplication of effort (or provoking antitrust actions against
themselves!) they can, in effect, pool resources and support a platform that 
will keep pace with Microsoft in this area.

If anyone is puzzled by all of this, I can go into more detail on this point. It
is very important to understand WHY the rest of the computer industry sees the
merit of Linux, and part of that merit is (legitimately) as a foil for Micro-
soft.

>WinY2K's domain structure is diffrent than NT 4.0,  which poses problems for 
>Samba.  Thier directory and kerberos implementation is non-standard so they 
>can't interoperate with other systems, etc...
It will be interesting to see what degree of difficulty this poses for network
managers who want to evolve "gracefully' from NT 4.0 to Windows 2000. Conversely
it will be a test of the "adhesive" qualities of Linux by observing the resi-
stance to Windows 2000 if they are not compatible with the installed base of 
Linux servers.

>Alot of this has to do with "stupid" developers who use quick-and-dirty tools 
>like VB and ActiveX without regard to performance/scalability/portability 
>issues,...
OK, this is a PERFECT example of the "network effect". Asking these guys why 
they use this stuff, the responses run something like "VBX gives me what my
customers want, and since they're all on Windows, it's just great!" This is 
exactly why I'm not in that business. Most of the development work I've done 
over the last 5 years has been to a great extent platform independent, and it
ports to any platform (including Microsoft platforms) very nicely, thank you.
When Microsoft changes its tune, I have very little to unlearn. I allow other
folks to write the platform dependent stuff.

>I think the rise of Linux on the college campuses, etc... is going to be the 
>real savior, as that is where the next wave of developers comes from.
One of the principle sources, anyway. One of the reasons that Linux has been
successful and other (perhaps technically superior) systems such as FreeBSD or
even BeOS is the licensing arrangements. It's simply much, much tougher to get
changes into FreeBSD (as an example, I'm not picking on FreeBSD in particular)
than it is into Linux... the process takes more time, and the effort required
to become a player is a lot higher. In some ways, The Linux community selected
itself for the role of it now has. You could also look at some of the other
projects as having, in effect, DISQUALIFIED themselves from that role. This is
what motivates a lot of developers these days, even guys who are out of school
for a generation or me (like me!).

>That belief is one of several reasons I am involved with KLUG.
In closing then, I pose this question to you (and everyone, actually):
  How can we help to promote this?

                                                             Regards,
                                                              ---> RGB <---