Very Computer

  To continue along the lines of this thread.  There is not much Microsoft
can do to compete with Java.  It has released Visual J++, and it's own Java
SDK, but that does not stop the threat to Windows if the rest of the world
becomes "cross-platform" thus having not a need for the Win32 API.

  Microsoft has one ace in the hole, which turns out to be a curse.  If they
could get Visual Basic to run cross-platform by building a virtual-machine
in place of it's run-time DLL, bingo there goes God-knows-how-many VB coders
suddenly going cross-platform.  Not to mention it is an easier to pick up
language than Java.  Microsoft cares not about OOP and proper coding, those
attributes of Java become irrelevant in Microsoft and VB coders eyes.  The
curse is it kill the need for Windows.  Therefore Microsoft will never
implement it.

  Java is the first real threat to Windows/Microsoft.  The momentum of Java
is unmatched by any IS trent over the last 20 years.  A lot of people have a
problem picturing a world without Microsoft rule.  I just tell them to
picture themselves back in 1983 saying IBM will lose it's grasp over the PC
industry over the next 5 years.  

  Things happen.  I can't help but wonder how many MS coders would be happy
if MS lost it's * hole over the industry.  If that were the case they
would not fear competing with MS if they were to go about their own
ventures.  People like to knock MS coders, but the truth is some of the
brightest hacker talent is up in Redmond.  Keeping them stagnant in
Microsoft-land may be keeping the next "killer app" from coming to life.

  Only time will tell where things are heading.  I hope for the sake of
progress Microsoft's shackles on the industry are broken by Java.

  Here's a nickel, got three pennies change?

So how do they make it worthwhile?  There are a wide variety of Win32
capabilities that have no Java counterparts.  A lot of these
capabilities are required to make an application commercially viable
on Windows platforms (ie 80+% of the market).  Thus if Microsoft were
to release a set of Java classes that encapsulate Win32 functionality
they'll get a lot of people to use them.

Quote:>  Java is the first real threat to Windows/Microsoft.

I'm serious.  I write Java applications for a living.  I used to write
portable C and C++ applications for a living.  Java cuts down on the
need for me to run a compile on every machine I wish to target, but it
doesn't cut down on the need to test on every machine I wish to target
or the need to write specific code for every machine I wish to target.
Anybody that tells you differently is either naive or lying.

Java isn't going to do anything to Microsoft.

jim frost

--
http://world.std.com/~jimf

Quote:> >  Java is the first real threat to Windows/Microsoft.

> Why?  What does Java give us in terms of Microsoft independence that
> we didn't already have with C/C++ and a portable class library?

Quote:> I'm serious.  I write Java applications for a living.  I used to write
> portable C and C++ applications for a living.

There IS a difference between "Code for this library, and compile
for any platform you like" and "code with this tool, compile on
your favorite platform and run on any platform you like".

Neither Java nor C/C++ with portable class libraries always live
up to the promise, but Java has a somewhat higher potential than
the C/C++ solution, since it removes some problems with porting
that C/C++ class lib's have problems with...

Quote:> Java cuts down on the
> need for me to run a compile on every machine I wish to target, but it
> doesn't cut down on the need to test on every machine I wish to target
> or the need to write specific code for every machine I wish to target.
> Anybody that tells you differently is either naive or lying.

> Java isn't going to do anything to Microsoft.

--
____________________________________________________________________

---
--- <!--#include virtual="/docs/std/disclaimer.txt" -->

: >  To continue along the lines of this thread.  There is not much Microsoft
: >can do to compete with Java.  It has released Visual J++, and it's own Java
: >SDK, but that does not stop the threat to Windows if the rest of the world
: >becomes "cross-platform" thus having not a need for the Win32 API.
:
: Microsoft doesn't *have* to compete with Java, all they have to do is
: make it worthwhile to write Java applications that are
: Windows-specific.  This strategy is known as "embrace and extend" and
: is something Microsoft is really, really good at.

Far be it from me to be able to predict the future but I think your
assessment is just a little off. It does little good for Microsoft to
write Windows specific Java libraries in an attempt to keep their
* hold on the desktop. I simply need to point to all the hype
around the NC as well as Java. It's not that businesses don't want
really good apps or lots of functionality but they don't want it at the
price of having to support thousands of expensive PC's. So they are
seriously evaluating the NC and Java. If Microsoft targets apps or libraries
at Windows only PCs they will only alienate businesses even more than they
are now.

Yes I have seen the NetPC announcements and I consider them a joke, maybe
their not, I can't be sure but it's not in Microsoft's best interests to
make it cheap to run a PC. Their whole fiscal policy is based on upgrade,
upgrade,upgrade. This is not a cheap philosophy and businesses are getting
sick of it.

:
: So how do they make it worthwhile?  There are a wide variety of Win32
: capabilities that have no Java counterparts.  A lot of these
: capabilities are required to make an application commercially viable
: on Windows platforms (ie 80+% of the market).  Thus if Microsoft were
: to release a set of Java classes that encapsulate Win32 functionality
: they'll get a lot of people to use them.

That 80% of the market may drop very quickly if the NC turns out to be
much less expensive to support then the current PCs. Now it doesn't have
to be Java code running on those NCs but that IS what will be running on
them. If Microsoft releases libraries that don't run on those NCs they will
only feel the rather of companies that expect to run "any" Java app on
a Java "compatibile" platform. Yes Microsoft could try to make it look like
it's the NC makers fault but Sun,IBM,Oracle etc. have a vested interest
in this as well and it is unlikely they would not counter every piece of
Microsoft FUD with their own.

: Java isn't going to do anything to Microsoft.

I cut the other stuff because I don't disagree with it. But that means
nothing in the face of the fact that a large number of major vendors and
businesses have fallen for the Java == cross platform line. This is being
heavily pushed as one of it's big "advantages" and if Microsoft tries to
subvert this I think they will be in for a rude awakening. Needless to
say I disagree with the line I left in.

On the other hand these are just my perceptions and thus I could be wrong,
hell it's happened before(well once in 1970 or so but I know I've been
wrong about something :-) ).

--
Gerry Gryschuk       "If it can't be expressed in figures, it is not science;

Go Habs!!!            "Never underestimate the power of human stupidity"
                         - Lazarus Long(R.A.Heinlein)  

[snip]

Quote:

>Yes I have seen the NetPC announcements and I consider them a joke,

Does this sound a bit like "deja vu all over again" to anybody?
(Read on if not.)

Quote:>                                                                    maybe
>their not, I can't be sure but it's not in Microsoft's best interests to
>make it cheap to run a PC.

Sigh...  It's all so depressingly predictable.

I am not sanguine that they will not succeed with any or all of
their efforts.

Quote:>                           Their whole fiscal policy is based on upgrade,
>upgrade,upgrade. This is not a cheap philosophy and businesses are getting
>sick of it.

Many of them *say* they are, *but*...

"Nobody ever got fired for buying IBM."  Remember that?  Well all
you need do is substitute "Wintel" for "IBM" and you have today's
CYA IS mantra.  And Wintel is doing all that they can to "enhance"
that thought's presence.  IS executives and managers are only
human, after all.  They have lives to live, mouths to feed (if
only their own), etc.  So in spite of the fact that they may feel,
in many (most?) cases justifyably, that Wintel is not the best
solution for a given requirement, they will sometimes operate in
CYA mode.

Regards,
Jim
--
Jim Seymour                     | "With [Microsoft's] ActiveX, you just need a

Systems & Network Administrator |  OLE server running.  But you get what you
Medar, Inc., 38700 Grand River  |  pay for." -- Albert Eng, Dir. of Financial
Ave., Farmington Hills, MI.     |  Technology, Canadian Imp. Bank of Commerce.

  Are you thinking instead of the java p-machine, which is a clever
  idea, even though it's been done before? (the USCD P-machine, for
  example, which is only about 22 years old.)

                ____
  david parsons \bi/ waiting for the p-code backend to gcc.
                 \/

Alexander Lamb

Excerpts from netnews.comp.sys.next.advocacy: 24-Nov-96 Re: Was: SUN'S
JAVA: THE TH.. by david parsons)

Quote:>   Are you thinking instead of the java p-machine, which is a clever
>   idea, even though it's been done before? (the USCD P-machine, for
>   example, which is only about 22 years old.)

Yes, p-code or bytecode systems were first invented many years ago.
However, there is a vast difference in terms of functionality,
modularity, encapsulation, dynamic capabilities, security, networking,
and probably a few other areas between Java and the USCD P-code stystem.

-Chuck

        ----------------+---------------------+---------------------
           I know you're an optimist if you think I'm a pessimist.

What it doesn't give you is perfect portability.  Far from it: the
runtime has myriad differences not only between platforms but even
between JVM implementations on a single platform.  There's no win
there.  In fact, you lose because Java's class libraries are terribly
primitive so you get to work around lots of limitations using native
classes if you're writing any substantial, commercial-grade program.

Quote:>There IS a difference between "Code for this library, and compile
>for any platform you like" and "code with this tool, compile on
>your favorite platform and run on any platform you like".

Quote:>Neither Java nor C/C++ with portable class libraries always live
>up to the promise, but Java has a somewhat higher potential than
>the C/C++ solution, since it removes some problems with porting
>that C/C++ class lib's have problems with...

Two things I can think of:

1. More consistent compilation environment.  In my experience this
eliminates a day or so worth of work by an experienced engineer on a
half-million-line project (the kind of thing I used to do all the
time), which is down in the noise for most projects.

2. No linker to deal with.  That's a help too, but it won't save more
than a day or so worth of work per platform so it's also lost in the
noise.

I do this stuff for a living in *all three* languages.  I know very
well where the relative costs and savings are, and I won't lie about
it to fulfill some hidden agenda.  The only agenda I have is to write
good software as fast as I can, using whatever tools are available to
do it.

Quote:>> Java isn't going to do anything to Microsoft.
>Depends on how you look at it... I agree that Java will not
>make MS have red numbers by the end of any fiscal year in the
>future, but it might force MS to think in quite different
>terms. I say might, because MS engineers seems terribly hung
>up on the old ways of thinking still, even after taking the
>internet into their hearst, they have obviously not understood
>anything about what internet is. (Thinkin for example of the
>ridiculous TCP/IP limitation in NT/W).

Microsoft already beat the Java threat.  They won't be behind anybody
else from this point forward in terms of Java technology, and you're
going to see them extend the runtime in all kinds of useful ways to
leverage Microsoft OS technologies in order to tie Java programmers to
Microsoft's stuff.

jim frost

--
http://world.std.com/~jimf

Quote:>[Microsoft's] whole fiscal policy is based on upgrade,
>upgrade,upgrade. This is not a cheap philosophy and businesses are getting
>sick of it.

Quote:>: So how do they make it worthwhile?  There are a wide variety of Win32
>: capabilities that have no Java counterparts.  A lot of these
>: capabilities are required to make an application commercially viable
>: on Windows platforms (ie 80+% of the market).  Thus if Microsoft were
>: to release a set of Java classes that encapsulate Win32 functionality
>: they'll get a lot of people to use them.
>That 80% of the market may drop very quickly if the NC turns out to be
>much less expensive to support then the current PCs.

Quote:>: Java isn't going to do anything to Microsoft.
>I cut the other stuff because I don't disagree with it. But that means
>nothing in the face of the fact that a large number of major vendors and
>businesses have fallen for the Java == cross platform line. This is being
>heavily pushed as one of it's big "advantages" and if Microsoft tries to
>subvert this I think they will be in for a rude awakening. Needless to
>say I disagree with the line I left in.

What's not a myth is that Java improves software quality and time to
market dramatically compared to C and C++, yet is much more scalable
and portable than something like Visual Basic.  That's where Java's
strengths lie, and why I recommend it.  Java's principle weakness is
that the portable class libraries are functionally weak compared to
native implementations.  That's correctable through native class
libraries without making the code less portable than other languages
and without impacting Java's other advantages.

jim frost

--
http://www.veryComputer.com/~jimf

Gryschuk)24 Nov 1996 06:26:06 GMT writes:

:>: Microsoft doesn't *have* to compete with Java, all they have to do is
:>: make it worthwhile to write Java applications that are
:>: Windows-specific.  This strategy is known as "embrace and extend" and
:>: is something Microsoft is really, really good at.

Read PCWeek. They're going to try just this, a windows-specific JAVA. A
contradictio in terminis, really.
And as far as I can help it they're not going to succeed!!

Groeten,

-=-Reinout

[Team OS/2 NL]

Quote:>Excerpts from netnews.comp.sys.next.advocacy: 24-Nov-96 Re: Was: SUN'S
>JAVA: THE TH.. by david parsons)
>>   Are you thinking instead of the java p-machine, which is a clever
>>   idea, even though it's been done before? (the USCD P-machine, for
>>   example, which is only about 22 years old.)
>Didn't we go over this before?
>Yes, p-code or bytecode systems were first invented many years ago.
>However, there is a vast difference in terms of functionality,
>modularity, encapsulation, dynamic capabilities, security, networking,
>and probably a few other areas between Java and the USCD P-code stystem.

In any case once you build the interpretive environment adding the
rest of these things is pretty easy.

jim frost

--
http://world.std.com/~jimf

Quote:>Excerpts from netnews.comp.sys.next.advocacy: 24-Nov-96 Re: Was: SUN'S
>JAVA: THE TH.. by david parsons)
>>   Are you thinking instead of the java p-machine, which is a clever
>>   idea, even though it's been done before? (the USCD P-machine, for
>>   example, which is only about 22 years old.)

>Didn't we go over this before?

                 ____

                  \/

I do agree that most of the people who buy PC's now will not turn
to NC's anytime soon.  The NC may create it's own marketplace,
though.  And if it can get money from that, it might eventually
add enough features to compete against "Lower-end PC usage".

Quote:> I hear a lot of "may"s and "could"s, but nobody has definitive
> numbers as to what NC cost of ownership will be, nor has anyone
> addressed the software issue satisfactorily.

Quote:> Companies may well believe it, but I use Java every day and the
> fact of the matter is that Java is nowhere near transparently
> cross platform right now.  The portability argument is largely
> a myth, ask anybody who's tried it.  Corporations that buy into
> that particular myth are in for a rude awakening independent of
> Microsoft.

> What's not a myth is that Java improves software quality and time
> to market dramatically compared to C and C++, yet is much more
> scalable and portable than something like Visual Basic.  That's
> where Java's strengths lie, and why I recommend it.  Java's
> principle weakness is that the portable class libraries are
> functionally weak compared to native implementations.  That's
> correctable through native class libraries without making the
> code less portable than other languages and without impacting
> Java's other advantages.

I just wanted to agree with these comments...   :-)

---

Senior Systems Programmer        (MIME & NeXTmail capable)
Rensselaer Polytechnic Institute;           Troy NY    USA

We've also had quite a few years of people thinking about the
issues.  Certainly the developments in hardware performance also
help a lot.  The same developments help Windows95.  Does that mean
Windows95 is certain to fail, simply because the hardware to run
it would have been too expensive 22 years ago?

Still, developments such as "Juice" indicate that Java isn't
necessarily the perfect answer either, and that other bytecode
strategies would work better.  Time will tell, I guess.

---

Senior Systems Programmer        (MIME & NeXTmail capable)
Rensselaer Polytechnic Institute;           Troy NY    USA