How can one *possibly* take serious someone who writes about himself, as Eric Raymond does, on his own website, as follows:

"I'm one of the half-dozen or so most influential people in [the open source] movement; in fact, a lot of people would put me among the top three, with Linus Torvalds and Richard M. Stallman.

The community has a tradition of tri-letterizing its heroes - I suppose that began with Stallman, already a hero when I was a fledgling programmer in the early 1980s, who was generally known as RMS even then. Linus Torvalds is just "Linus", perhaps because (unlike "Richard" or "Eric") one can refer to him by simply first name with very little risk of aliasing problems.

I think I started to be routinely triletterized into "ESR" around 1998."

Here's the link, so that you can read this for yourself if you don't believe me: http://www.catb.org/~esr/who-is-ESR.html

I think that you need to take a look at what the Java market share is competing against. Java does not, on the main, compete with PERL or Python. In fact, Java's biggest competitor, and threat, is Visual Basic.

Than's right, visual basic. Right now, as you read this you are thinking "what a joke, this guy doesn't know what he is talking about".

In fact, I do know whatI am talking about. Out here in corporate america, where the IT standards are truly tested, paid for, and set Java is The King, and not in the prince charles way, but in the Elvis way. And VB is the second in the succession.

That is right, everyone is moving from C, not to perl, python, or even Visual Basic (VB seems to be limited to new app development, not replacement). And they are moving to Java. Not for OS or compiler programming. But for industrial strength web apps.

And do you know why? Two reasons.

1) Java makes networking, threading, and database connectivity trivial. Go read O'Reilly's "Learning java" to see the one page of source code it takes to create an HTTP server. With Java, you don't need to have a massive programming crew to use every processor on your Enterprise 15k system. One programmer can do that. Heck, I can do that, and I am not even a programmer.

2) Java is darned close to being open source. YOu get it for free. You get Sun's Netbeans (Ok, who cares if they have the copyright to netbeans or not, they pay for it) for free, you get the best API documentation on the planet, FREE. You can take the base java class and change it in any way you want. Just extend it. All other classes derive from the Object base class, so there is your true open source. You can write a JVM (the specs are avaialable), you can write a Java Compiler (specs available).

So, what really is your beef? Or is Microsoft so impermeable that you have to look for an easier target for your senseless pandering to the OpenSource community. If you really want to help the freeware cause, start hacking.

Otherwise, you are just a "wannabe".

This is a reformat of the same comment above.

# Layton commented on 27 February 2004:

* *

>>As far as ESR's track record on economic predictions goes,
>>need I comment? I can remember interviews he gave 3 or 4
>>years ago in which he predicted the imminent collapse of
>>Microsoft's monopoly, within 6 months or so. In one
>>interview several years ago he also predicted that Linux
>>would become easy enough for Aunt Tilly to use within 6
>>months. Kind of makes you wonder if maybe ESR's judgement
>>is, just a little tad, distorted."

ESR is part schoolgirl-in-love and part True Believer. In
love with the hacker culture, he's mostly blind to its
limitations and dysfunctionality. Not that he's
unintelligent. But ESR chooses to use his intelligence to
fantasize flowery camelot idealizations of that culture, and
concoct elaborate rationalizations of its failings, than see
it for what it is.

You'll find few schoolgirls and ESRs running the world's
most powerful corporations. Rather, you'll find them
writing books, pleading with open letters, and fantasizing.
I'd warn you against heading down the same path.

>Your take on this reminds me of the take I used to see in
>reference to linux in the server room. Yes, those time
>frames are insane. Microsoft has a war chest that would
>keep them going for 2 or 3 years if they lost all
>revenues. But why do you thing they bought controling
>interest in the NBC television network? Why do you think
>they are trying to gain footholds in other non-computer
>related(or at least hidden computer) technology and
>information businesses? They are afraid that their computer
>software business is going to go away.

Microsoft has been extremely cautious for its entire
existence. It has been trying to diversify for many years.
In particular, its partnership with NBC dates from the mid
1990's and has nothing to do with linux. Gates in
particular is well known to be obsessed with circumventing
any "paradigm shift" that would undermine his business.
Microsoft's success getting its entire base of users and
10's of thousands of apps moved intact from DOS to 32 bit
Windows over the course of a decade is due, among other
things, to this extreme caution.

It may have taken you until 2004 to wake up to this fact,
but I recall it already being a ho-hum topic in the computer
press in the early 1990's.

As far as recognizing a threat from linux, Microsoft already
had a department devoted to studying this threat in the
1990's, long before most of the rest of the world, including
the Justice Department, took that threat seriously.

Could linux some day knock out Microsoft's near monopoly on
the desktop? Why not? Linux has poor technology and host
of problems that would make it problematic for wide use by
end users. But everything is possible. However, I think
it'd be much easier to judge what's going to happen if linux
closed some of its current huge technology lag. Remember,
OS/2 Warp shipped in 1994, a full year before Windows 95,
and was backed by IBM, but didn't displace Windows. You
have to ask why linux would succeed today against Windows XP
with inferior technology to what OS/2 had 10 years ago.

>As far as linux being easy/hard for aunt Tilly, my
>observation has been that if aunt Tilly learned to use
>linux first, then she had a hard time with windows. If she
>learned windows first, she had a hard time with
>linux. Further, she will probably have a hard time with
>windows XP if she learned windows 9x first.

I suspect that your "observation" is nothing more than your
imagination and bias. I started life on a TOPS-20 system,
spent most the the 1980's using Unix and living in emacs,
have worked under a variety of operating systems over the
years since then, and have spent the past several years
mostly using linux and FreeBSD at home and at work. I have
very little experience with Win32 systems. Nevertheless, I
generally find it easier to set up Windows 9x, 2000, and XP,
for common tasks, install software, etc., than any unix
variant. I would regard it as stupid and irresponsible to
foist linux on Aunt Tilly because of your or my bias.

I maintain my linux boxes with a combination of patience,
technical expertise, lots of scripting, grappling with
incomplete, poorly written, or non-existent documentation,
lots of google and deja news searches, copious fiddling, and
prayer. My experience is that linux is, on the whole, a
fairly unreliable and problematic system with poor
technology and performance.

By contrast there are 3 Windows boxes in my home for my
family's use. In my experience, Windows 2000 and XP crash
more rarely (in fact, our Win 2000 and XP boxes have never
really crashed to my knowledge, but XP has occasionally gone
flakey) than my linux boxes freeze up, and my wife and young
children, who have no computer expertise, need no help using
them. I find the technology in Windows to be generally much
superior to that in linux, and the software available for
Windows to be much more varied and of much higher quality
than that for linux.

Thus, my observation is quite different from yours.

>XP has taken too much from a unix minimalist approach to
>desktop design for aunt tillie to figure it out easily,
>unless she started with something unixish first.

What you euphemistically call "a unix minimalist approach,"
most of the world sees for what it is, which is
technological impoverishment. Many insults have been
leveled at XP, but associating it with unix's technological
retardation is new to me.

>On the flip side, I agree that there are some usability
>issues that still need to be worked out in linux. But it
>has made tremendous strides, and there are far fewer than
>there were a year ago.

In my opinion, "usability" is a bit misleading, because it
gives the false impression that what is needed is to simply
beautify and polish up the existing software base. In fact,
there are all too many linux people wasting their time on
the beautification, with endless web sites devoted to themes
for emacs, gtk, gnome, xmms, dircolors, ... What linux
primarily lacks is not this nonsense, but good old
technology, intelligence, and programming talent. Windows
users are spoiled with great deal of technology. You
probably can't expect linux to break into that user base
with its limited and inferior technology, even if it's free.

I've used a number of operating systems over the years and
I've never seen any system develop any where near as slowly
and as haltingly as linux. The original development of
gui's for Mac, Amiga, Windows, OS/2, Java, and other
technologies was generally done in small teams producing
good GUIs in relatively short time, typically a year or two.
And that was producing original technology, rather than
simply copying well established paradigms, as linux does.
(In truth, the development of Windows 1.x, 2.x, 3.x was a
bit Unix-like in its sluggishness, but the MS of today has a
much better process.)

And this sluggishness includes the boost that came from the
dot com bubble, where tens of millions of dollars were
dumped on projects like Nautilus (the worst file manager
I've ever used). For whatever reason, linux seems to have a
heck of a time attracting the kind of intelligence and
programming talent needed to produce technology competitive
with Microsoft. Frankly, if the venture capital of the late
90's couldn't do it, I don't see where it's going to come
from. Perhaps if linux catches on in China and India?
Perhaps. Otherwise, from where?

Just sticking to the user interface technologies, it's a
pretty depressing situation. At base is X Window. Better
windowing systems were already developed in the 1980's.
Keith Packard (prominent for his work on extensions to X
Window) related a remarkable historical anecdote explaining
why X Window has no native bezier curve support (or much
else, for that matter). Apparently, nobody on the original
X Window team knew anything about bezier curves (!!!), so
they left that to a future version which never materialized.
Unfortunately, the kind of technical laxity, where people
who don't know about bezier curves would dare to write a
graphical interface system, is all too common and well
tolerated in the unix world. Now, move up a level from X to
Gtk, the most widely used gui library. The people from Gimp
who started Gtk never expected it to become the standard for
linux gui development, and they've admitted to not knowing
much about writing a gui library. Unfortunately, it shows.
Gtk has been under development for about a decade now. Wrap
your head around that -- a DECADE. Still, it's extremely
awkward, inconsistent, has embarrassing limitations, and it
isn't even fully documented. Go up to the top level, Gnome.
It's been, what, 7 years or so now? It was launched by de
Icaza, who wrote a famous essay called "Unix Sucks," in
which he lamented the lack in unix of things like
Microsoft's component technology. Unfortunately, 7 years
later de Icaza's goals are still far off, and Gnome and unix
still suck. How many years does brainlessly copying
Microsoft's gui take? OSS shows me it might take decades.

>by the way, if Sun doesn't cooperate with doing an open
>source java, there are acouple of cooperating projects that
>are building an open source java environment, that may
>eventually pass Sun up. In 6 months? NO. In 5 or 6 years?
>maybe. depending on what Sun does, and how responsive they
>are to their users. Maybe sooner if the nay-sayers prove
>true and Sun goes out of business.

Yes, there are a number of open source attempts around Java.
Java hit the scene in 1995 and they've had about 9 years
now. It's hard to say whether another 5 or 6 years would be
enough for the open source community to catch up to where
Sun is now. That would be great. One can only hope. Being
only 5 or 10 years behind in technology is a reasonable
expectation for open source software. I would give it maybe
50/50 chances.

On the other hand, what about actually pushing forward,
going beyond the current state of Java? Frankly, I don't
even want to see them try. Java certainly has its flaws,
but nothing on the order of the muddle that comes from the
open source world. Let them stick to things like Perl that
are better suited to the open source style.

# Layton commented on 27 February 2004:

* *

"As far as ESR's track record on economic predictions goes,
need I comment? I can remember interviews he gave 3 or
4 years ago in which he predicted the imminent collapse
of Microsoft's monopoly, within 6 months or so. In one
interview several years ago he also predicted that Linux
would become easy enough for Aunt Tilly to use within 6
months. Kind of makes you wonder if maybe ESR's judgement
is, just a little tad, distorted."

ESR is part schoolgirl-in-love and part True Believer. In love with
the hacker culture, he's mostly blind to its limitations and dysfunctionality.
Not that he's unintelligent. But ESR chooses to use his intelligence to
fantasize flowery camelot idealizations of that culture, and concoct
elaborate rationalizations of its failings, than see it for what it is.

You'll find few schoolgirls and ESRs running the world's most powerful
corporations. Rather, you'll find them writing books, pleading with
open letters, and fantasizing. I'd warn you against heading down
the same path.

Your take on this reminds me of the take I used to see in reference to linux in the server room. Yes, those time
frames are insane. Microsoft has a war chest that would keep them going for 2 or 3 years if they lost all
revenues. But why do you thing they bought controling interest in the NBC television network? Why do you think
they are trying to gain footholds in other non-computer related(or at least hidden computer) technology and
information businesses? They are afraid that their computer software business is going to go away.

Microsoft has been extremely cautious for its entire existence. It has been trying to diversify for many years.
In particular, its partnership with NBC dates from the mid 1990's and has nothing to do with linux. Gates in
particular is well known to be obsessed with circumventing any "paradigm shift" that would undermine his business.
Microsoft's success getting its entire base of users and 10's of thousands of apps moved intact from DOS to 32
bit Windows over the course of a decade is due, among other things, to this extreme caution.

It may have taken you until 2004 to wake up to this fact, but I recall it already being a ho-hum topic in the
computer press in the early 1990's.

As far as recognizing a threat from linux, Microsoft already had a department devoted to studying this threat
in the 1990's, long before most of the rest of the world, including the Justice Department, took that threat
seriously.

Could linux some day knock out Microsoft's near monopoly on the desktop? Why not? Linux has poor technology
and host of problems that would make it problematic for wide use by end users. But everything is possible.
However, I think it'd be much easier to judge what's going to happen if linux closed some of its current huge
technology lag. Remember, OS/2 Warp shipped in 1994, a full year before Windows 95, and was backed by IBM,
but didn't displace Windows. You have to ask why linux would succeed today against Windows XP with inferior
technology to what OS/2 had 10 years ago.

As far as linux being easy/hard for aunt Tilly, my observation has been that if aunt Tilly learned to use linux
first, then she had a hard time with windows. If she learned windows first, she had a hard time with
linux. Further, she will probably have a hard time with windows XP if she learned windows 9x first.

I suspect that your "observation" is nothing more than your imagination and bias. I started life on a TOPS-20 system,
spent most the the 1980's using Unix and living in emacs, have worked under a variety of operating systems over the
years since then, and have spent the past several years mostly using linux and FreeBSD at home and at work. I have very
little experience with Win32 systems. Nevertheless, I generally find it easier to set up Windows 9x, 2000, and XP, for
common tasks, install software, etc., than any unix variant. I would regard it as stupid and irresponsible to foist
linux on Aunt Tilly because of your or my bias.

I maintain my linux boxes with a combination of patience, technical expertise, lots of scripting, grappling with
incomplete, poorly written, or non-existent documentation, lots of google and deja news searches, copious fiddling,
and prayer. My experience is that linux is, on the whole, a fairly unreliable and problematic system with poor
technology and performance.

By contrast there are 3 Windows boxes in my home for my family's use. In my experience, Windows 2000 and XP crash more
rarely (in fact, our Win 2000 and XP boxes have never really crashed to my knowledge, but XP has occasionally gone
flakey) than my linux boxes freeze up, and my wife and young children, who have no computer expertise, need no help
using them. I find the technology in Windows to be generally much superior to that in linux, and the software available
for Windows to be much more varied and of much higher quality than that for linux.

Thus, my observation is quite different from yours.

XP has taken too much from a unix minimalist approach to desktop design for aunt tillie to figure it out easily,
unless she started with something unixish first.

What you euphemistically call "a unix minimalist approach," most of the world sees for what it is, which is
technological impoverishment. Many insults have been leveled at XP, but associating it with unix's technological
retardation is new to me.

On the flip side, I agree that there are some usability issues that still need to be worked out in linux. But it
has made tremendous strides, and there are far fewer than there were a year ago.

In my opinion, "usability" is a bit misleading, because it gives the false impression that what is needed is to
simply beautify and polish up the existing software base. In fact, there are all too many linux people wasting
their time on the beautification, with endless web sites devoted to themes for emacs, gtk, gnome, xmms, dircolors,
... What linux primarily lacks is not this nonsense, but good old technology, intelligence, and programming talent.
Windows users are spoiled with great deal of technology. You probably can't expect linux to break into that user base
with its limited and inferior technology, even if it's free.

I've used a number of operating systems over the years and I've never seen any system develop any where near as slowly
and as haltingly as linux. The original development of gui's for Mac, Amiga, Windows, OS/2, Java, and other
technologies was generally done in small teams producing good GUIs in relatively short time, typically a year or two.
And that was producing original technology, rather than simply copying well established paradigms, as linux does.
(In truth, the development of Windows 1.x, 2.x, 3.x was a bit Unix-like in its sluggishness, but the MS of today
has a much better process.)

And this sluggishness includes the boost that came from the dot com bubble, where tens of millions of dollars were
dumped on projects like Nautilus (the worst file manager I've ever used). For whatever reason, linux seems
to have a heck of a time attracting the kind of intelligence and programming talent needed to produce technology
competitive with Microsoft. Frankly, if the venture capital of the late 90's couldn't do it, I don't see where it's
going to come from. Perhaps if linux catches on in China and India? Perhaps. Otherwise, from where?

Just sticking to the user interface technologies, it's a pretty depressing situation. At base is X Window. Better
windowing systems were already developed in the 1980's. Keith Packard (prominent for his work on extensions to X
Window) related a remarkable historical anecdote explaining why X Window has no native bezier curve support (or much
else, for that matter). Apparently, nobody on the original X Window team knew anything about bezier curves (!!!), so
they left that to a future version which never materialized. Unfortunately, the kind of technical laxity, where people
who don't know about bezier curves would dare to write a graphical interface system, is all too common and well
tolerated in the unix world. Now, move up a level from X to Gtk, the most widely used gui library. The people from
Gimp who started Gtk never expected it to become the standard for linux gui development, and they've admitted to not
knowing much about writing a gui library. Unfortunately, it shows. Gtk has been under development for about a decade
now. Wrap your head around that -- a DECADE. Still, it's extremely awkward, inconsistent, has embarrassing
limitations, and it isn't even fully documented. Go up to the top level, Gnome. It's been, what, 7 years or so now?
It was launched by de Icaza, who wrote a famous essay called "Unix Sucks," in which he lamented the lack in unix of
things like Microsoft's component technology. Unfortunately, 7 years later de Icaza's goals are still far off, and
Gnome and unix still suck. How many years does brainlessly copying Microsoft's gui take? OSS shows me it might take
decades.

by the way, if Sun doesn't cooperate with doing an open source java, there are acouple of cooperating projects
that are building an open source java environment, that may eventually pass Sun up. In 6 months? NO. In 5 or 6
years? maybe. depending on what Sun does, and how responsive they are to their users. Maybe sooner if the
nay-sayers prove true and Sun goes out of business.

Yes, there are a number of open source attempts around Java. Java hit the scene in 1995 and they've had about 9
years now. It's hard to say whether another 5 or 6 years would be enough for the open source community to catch up
to where Sun is now. That would be great. One can only hope. Being only 5 or 10 years behind in technology is
a reasonable expectation for open source software. I would give it maybe 50/50 chances.

On the other hand, what about actually pushing forward, going beyond the current state of Java? Frankly, I don't even
want to see them try. Java certainly has its flaws, but nothing on the order of the muddle that comes from the open
source world. Let them stick to things like Perl that are better suited to the open source style.

*

"As far as ESR's track record on economic predictions goes,
need I comment? I can remember interviews he gave 3 or
4 years ago in which he predicted the imminent collapse
of Microsoft's monopoly, within 6 months or so. In one
interview several years ago he also predicted that Linux
would become easy enough for Aunt Tilly to use within 6
months. Kind of makes you wonder if maybe ESR's judgement
is, just a little tad, distorted."

Your take on this reminds me of the take I used to see in reference to linux in the server room. Yes, those time frames are insane. Microsoft has a war chest that would keep them going for 2 or 3 years if they lost all revenues. But why do you thing they bought controling interest in the NBC television network? Why do you think they are trying to gain footholds in other non-computer related(or at least hidden computer) technology and information businesses? They are afraid that their computer software business is going to go away.

As far as linux being easy/hard for aunt Tilly, my observation has been that if aunt Tilly learned to use linux first, then she had a hard time with windows. If she learned windows first, she had a hard time with linux. Further, she will probably have a hard time with windows XP if she learned windows 9x first. XP has taken too much from a unix minimalist approach to desktop design for aunt tillie to figure it out easily, unless she started with something unixish first.

On the flip side, I agree that there are some usability issues that still need to be worked out in linux. But it has made tremendous strides, and there are far fewer than there were a year ago.

by the way, if Sun doesn't cooperate with doing an open source java, there are acouple of cooperating projects that are building an open source java environment, that may eventually pass Sun up. In 6 months? NO. In 5 or 6 years? maybe. depending on what Sun does, and how responsive they are to their users. Maybe sooner if the nay-sayers prove true and Sun goes out of business.

The question of whether Sun should open source Java comes
down to a question of quality and viability. Would
it really improve the quality of Java? I doubt it very
much.

I've been using Linux as my primary OS for several years.
I like open source on a philosophical level, and I like
some social aspects of the distribution and development
model. I've heard all the arguments about why OSS should
be economically viable and technically superior, etc.
However, my experience is otherwise. On a technical level,
Linux (and OSS in general) is not impressive. Linux is the
most problematic OS I've ever used. I've had huge problems
with performance, reliability, documentation, feature
impoverishment, etc.

I've also programmed in both Java and with open source
technologies like gcc, Gtk, ... . Technologically, my
experience is that Java is overwhelmingly superior to
open source technologies -- so much so that a comparison
seems silly to me.

As some others have pointed out, Java may be viewed as a
specification which anyone is free to implement. If the
open source model were so effective, we'd have decent
open source implementations of Java by now. We don't.
As someone who was once forced (as part of a research
project) to use the open source JVM Kaffe, I can tell you
there's a good reason the OSS people want to get their
hands on Sun's source code, rather than go with their
own efforts like Kaffe.

In summary, Java is very high quality and basically
irreplaceable. Linux and OSS are nice in principle,
but in practice are low quality, replaceable, technologies.
I'd be cautious about mixing Java with something that
produces very different kinds of quality and results.

As far as ESR's track record on economic predictions goes,
need I comment? I can remember interviews he gave 3 or
4 years ago in which he predicted the imminent collapse
of Microsoft's monopoly, within 6 months or so. In one
interview several years ago he also predicted that Linux
would become easy enough for Aunt Tilly to use within 6
months. Kind of makes you wonder if maybe ESR's judgement
is, just a little tad, distorted.

The Java language is just a specification. The Java virtual machine is a commodity today. Anyone can build one. We have standardised on using Java/J2EE in our enterprise for development. We are running a non-sun JVM on Linux. I do not see how anyone is locked into Sun with Java. I can decide to use any J2EE complaint App server or any Java virtual. I have never seen such freedom with any other platform.
By letting Java go open source, what are we really talking about here. The source for the Sun - JDK? Who cares!

Business:
Have a central authority that has the resources to test, develop, defend, and maintain consistency is a big seller to the enterprise. As an example, Perl is a great and very powerful programming language, but for an enterprise to write and maintain applications in a more liberally licensed Perl? Sorry, but having an organization such as Sun overseeing the enterprise need helps keep Java in the enterprise, even if it means holding an intellectual fist (license) over the programming language.
Licensing:
I have not looked into the specifics, but if the only concern is if Sun will ''take away java'', maybe the best solution is to open up the important piece - the JVM. Blackdown is not technically ''open source'' because of the ties to Sun''s JVM licensing/core code. Fix that once piece, and anyone can freely use an open-source, certified JVM and a text editor, while the java core and consistency are maintained. They are now certifying open-source J2EE application servers, showing Sun''s commitment to open source. Next logical step to put some people''s fears to rest is to open up development of open-source JVM''s and to certify those open-source JVM''s.

Although not open source, Java is openly specified soliciting input from individuals, commercial interests and open source organizations. However, after having seen how numerous corporations have attempted to hijack or co-opt it for the sake of their own agendas - and this extends beyond Microsoft - it is in Sun''s and Java''s interest not to have a repeat of the OMG specs (which were delayed by years because of internecine strife), and not to have Java splintered in the way Unix had been in the late 80''s/early 90''s.

Although I am a fan of Linux, I am not so much a fan of it''s non-centralized direction. This has led in the past to what I consider arbitrary changes in the Linux platform, which, for example, has made it more difficult to interoperate with other Unix platforms. I would prefer for Java to have central direction much in the way JBoss has central direction - this is key to ensuring stability.

Sun''s actual weakness as Java''s steward is not that it is uncommitted to open standards and software, but that software has never been Sun''s strength. Sun is still first and foremost a hardware vendor and this saps considerable strength from its otherwise talented pool of software professionals. Sun is willing to change it''s core focus, but it needs to find a steady stream of revenue before it can execute on that.

Someday everybody will wake up and realize that the king is parading without clothes!
And that day the big Java hoopla will deflate and people (otherwise smart and talented) will say to themselves: why did I spend so much time in a platform that the only thing is has for it is "being cool"?

Arguably, it seems to be a problem to choose between ubiquity and control. On the other hand, Sun has continued to push the language to a controlled force that drives a lot of businesses and is regarded as a future language. It is possible to view all sources and participate in even hacking the JVM itself, something I would not even dare to think about... Open Source is a great way of accomplishing work and developing software, but there is also a great risk: unless a piece of software is really a nutcracker everyone wishes to use and work with, it moves into too many different directions, thus leading to a disortion of the original intention. I like to think about it as a larger oligopol many can participate, but few take control - basically the same the apache foundation is doing. In keeping tight control over who commits what and the direction a software is moving is a way of keeping it at its bests. Thus, I think Sun is doing the right thing, although there might be some flaws in how they are doing it, and how fast the overall process is.

Greetings,

Dennis

The otherwise articulate and seemingly knowledgeable Eric Raymond says: "Open source is hardly a zero-revenue model; ask Red Hat, which had a share price over triple Sun''s when I just checked."

This is a gaffe of such monumental distortion that it makes me wonder what other pretty simple facts of life he chooses to ignore.

Techies should refrain from using information to support their view from areas they do not understand. Share price is irrelevant. Market cap is a better measure.

Market Cap of Red Hat: $3.26B
Market Cap of Sun Microsystems: $18.9B

Better still measure of the "size" or "worth" of a company is revenue:

Sun: Ranging from $11 BILLION to $18B annually over past 3 years.

Redhat: Ranging from $80 MILLION to $100M annually over same period.

About 1% of Sun''s.

I wish that ESR would shut up.

He uses his high concept, impractical business models to promote himself and his own private agenda, rather than proving that they work by putting his own money and effort into companies that implement them.

ESR reminds me of a union leader that roams around construction sites, complaining and protesting about discrimination and union rule non-compliance, UNTIL the construction manager hands over a bribe to take his trouble somewhere else.

I don''t think that ESR is taking monetary bribes but I do think that he kicks up a stink to promote himself and get his name in the magazines. Most of the time, ESR can convince himself that he is a great prophet but, occasionally and privately, I think that he has moments when he realizes that he is just a huckster and a shill. His main product is ERIC S. RAYMOND IN CAPITAL LETTERS and open source in lower-case.

I lead projects for my company and I push very hard on Java even though we are Microsoft slaves. One of my arguments (among others) to get Java (over .net) projects approved is "you have serious companies like Sun and IBM behind Java". That helps a lot. I do agree from the technology point of view, that opening Java would benefit Java itself a lot, but if that is going to happen, it should be done in a way for which the API, additions and new features are driven like they are right now. One thing that makes Java very reliable is the responsibility in which the inclusion of new APIs, features and changes are done. Very few languages (any other?) have such stable, logical, organized and well-thought APIs. Once you learn Java, you don''t have to re-learn it when a new release comes out (as it happens with M$ languages). I personally think that Sun''s control has helped a lot to keep things that way. If Java is opened, it should be managed the same way so it will keep its reliability.