[5101] in java-interest
Re: java-interest-digest V1 #63
daemon@ATHENA.MIT.EDU (Jonah Benton)
Mon Jan 29 02:15:45 1996
To: java-interest@java.sun.com, 97jgonza@ultrix.uor.edu
From: Jonah Benton <jonah@arcus.net>
Date: Mon, 29 Jan 1996 00:44:44 -0500
At 08:44 PM 1/27/96 -0800, owner-java-interest-digest@java.sun.com wrote:
>From: "Jeff \"Crusader\" Gonzales" <97jgonza@ultrix.uor.edu>
>Date: Fri, 26 Jan 1996 18:12:50 -0800 (PST)
>Subject: Re: VRML
>
>Hey, has anyone read the Steve Jobs interview in the latest WIRED
>magazine? I wanted to hear opinions that object to Steve Jobs' argument
>against Java and other client-side web products. I figured I get a
>response here. :> Thanks.
>
>Jeff Gonzales
>- -> University of Redlands / Johnston Center
Hello,
My recollection is that Jobs feels Java and other client-side technologies
splinter the client platform, giving Microsoft the opportunity to come in
and "own the market."
In response, I'd say first that given that Jobs is spending a lot of time
and energy promoting a server-side solution (WebObjects), his comment can at
least be seen as self-serving.
However, I don't think his argument holds water anyway. It's nice to think
that there's some easy way we can all band together to prevent Microsoft
from achieving dominance somewhere on the Net; unfortunately, I really don't
think that's the case.
I think the mistake in Jobs' reasoning is that he's thinking there's just a
single client platform- the browser. To the contrary, I think that the
browser incorporates at least 3 major client platforms, while at least four
more are coming in the near future. The three are: client-side program
execution, client-side script execution, and basic GUI and data entry
services. The four I see coming are: sophisticated data validation services,
some sort of session management, some form of personal authentication
capability, and low-overhead transaction services. (The first two may
already be on their way to being solved by things like Netscape's cookie
architecture, the client side scripting languages, and Allaire's Cold Fusion
cgi-based data validation facilities.)
Jobs mistake, again, is that he's implicitly giving some kind of precedence
to HTML over the other client technologies- that somehow the other client
technologies are competing with HTML, and that stopping work on the other
technologies will help HTML become more "ubiquitous."
That's just not the case. The technologies listed above are largely
orthogonal and complementary. They each have their uses and limitations, and
the functional overlap between them is not significant. If everyone else
stopped working to provide those facilities, Microsoft would win by default.
Even with the efforts by third parties on the above fronts, Microsoft has a
competitive entry just about everywhere. Just off the top of my head they
have VBScript, Explorer, Blackbird, Excel- and Word-to-HTML tools, the
Internet API, the proposed integration of the web server with the file
system, their agreements with Verity and the credit card and banking
industries...
Jobs is mistaken if he thinks that Microsoft can't own the web client
platform if everyone just sticks with HTML. He's mistaken because all of
these other services are needed, and if third parties don't supply them,
Microsoft will.
I think in fact that the Java VM, though a risky platform for developers and
users for security reasons (yes, I've read the white papers, blah blah blah)
has the best chance in the bunch to succeed as a platform not controlled by
Microsoft. It's full featured and it runs on everything; MS's competing
technology only currently runs on Windows (though they're working on _that_
as well).
Though I wouldn't accuse Microsoft of plotting sabotage, I'd be surprised if
Bill didn't have 2 or 3 people working over the Java VM design and the
various implementations, looking for potential security holes to exploit in
marketing campaigns once their own technology arrives. And I'm sure that
while he's looking for holes in the architecture, he's also got people in
the VC++ group working on Java-language-to-Intel and
Java-language-to-Java-VM compilers, just so that if the Java VM turns out to
be a viable platform, he'll be able to sell developers the tools to create
for it.
--
Jonah Benton
Director of Technology
Arcus Incorporated
http://www.arcus.net/
-
This message was sent to the java-interest mailing list
Info: send 'help' to java-interest-request@java.sun.com
