[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference abbott::java

Title:JAVA
Moderator:KOALA::CIOT
Created:Mon Nov 13 1995
Last Modified:Wed Jun 04 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:236
Total number of notes:1251

230.0. "Top Five Java Myths!" by MARVIN::GOODWIN (Pete Goodwin) Fri May 16 1997 12:38

From: http://www5.zdnet.com/anchordesk/story/story_799.html

"Top Five Java Myths"

Behind all this good news is an ugly fact the Java promoters don't want you to
know. Many of Java's key promises are myths. Here are the five fables that
could lead you astray. 

1. Write once, run everywhere. Java's key promise. Fact: It's not true today
and it may never happen. Applets that run fine on Windows 95 may not run at all
on the Mac (or vice versa). And may run so slowly on 16-bit Windows, they
become unusable. I'm still hoping we'll end up with Java virtual machines that
are consistent across platforms, but right now we're still just wandering
around.

2. Businesses will rewrite their applications in Java. Of course not. Fact: We
still have production applications running today in Fortran and Cobol. People
hate to throw stuff away. And why should they?

3. People will do everything new in Java. Every language has its strengths.
Fact: Java will never be optimized for everything. Visual Basic, C++ and other
languages will continue to thrive.

4. Java is secure. Java is more secure than Microsoft's ActiveX. It is also
more restrictive. Using the standard Java "sandbox," you can't cache downloaded
Java applets on your local machine, forcing you to download the same Java
applets over and over again. Fact: As Java is modified to allow local caching,
local storage, and local printing, it will become less secure.

5. Java is an open standard. Yes, Sun Microsystems has turned Java over to a
standards body. But on a glacially slow timetable and in a manner that keeps
all the control with Sun. Fact: Java is less open than Microsoft's ActiveX. 

What's your favorite Java myth? Jump to our Jesse's Berst Alerts forum and join
the discussion. Or send me a TalkBack message

I'll post the best myths right below this story. 

Enjoy this week's announcements, marvel at the amazing progress of Java... but
don't fall for the myths. It's okay to wander through the maze of product
promises and Real-Soon-Now announcements. Unlike Theseus, you won't find a
bull. But you will find plenty of bull$&!#. Just be sure to wend your way back
to reality before you make any serious decisions. 

    Copied without permission
    
    Pete
T.RTitleUserPersonal
Name
DateLines
230.1Gotta love those 'religious' debates...STAR::COOKJust say NO to that AccViO! - Java!!!Tue May 27 1997 19:2036
    
    >>1. Write once, run everywhere. Java's key promise. Fact: It's not true
    >>	today and it may never happen.
    
    	I'd agree that you need to use care when you define 'anywhere'.
    	(Sure, Apple may go bust and never produce a reliable JVM, but
    	 meanwhile, lots of Java code is working on multiple platforms
    	 as intended.  Most reasonable people know that the promise
    	 assumes that JVMs implementations are robust and compatible.)
    
    	Hell, the Win32 API, as implemented on NT vs Windows-95 has
    	it's share of imcompatibilites and platform-specific restrictions.
    
    >>2. Businesses will rewrite their applications in Java. Of course not. 
    
    	That's a 'straw man'.  I agree.  Most reasonable people don't
    	interpret Sun as claiming that any large portion of working
    	applications will be rewritten in Java (just to gain 'Java-unique'
    	advantages).
    
    >>3. People will do everything new in Java. Every language has its 
    >>	strengths.
    
    	That's another 'straw man'.  I agree.  Most reasonable people
    	don't interpret Sun as making this claim.
    
    
    	Etc.  All these statements that you call 'myths' seem slanted
    	just so that someone biased *against* Java can call them myths
    	and spread more half-truths.  Have you programmed anything in Java?
    
    
    My 2-cents worth...
    
    						Dave
    
230.2There's a lot of hype for Java out there...MARVIN::GOODWINPete GoodwinWed May 28 1997 03:4916
    Have I programmed anything in Java? Yes, I'm currently working on a
    project with a Java applet.
    
    I've already been caught by the first myth - though it might be
    Microsoft's implementation at fault. I tried developing with Visual J++
    Trial Edition only to descover it doesn't work with Netscape 3.0. I had
    to go back to JDK 1.0.2 or J++ V1.0 for compatability.
    
    Whilst the underlying language seems fine, its the stone age AWT and
    RTL that amaze me the most. Trying to format numbers with Java seems to
    require re-inventing the wheel - give me printf/scanf anyday. As for
    the weak GUI in Java, soon to be replaced by a myriad set of toolkits
    etc. to program better GUI's why wasn't this considered in the first
    place?!?
    
    Pete
230.3IMHOTLE::JRICHARDWed May 28 1997 10:2217
    
>    Whilst the underlying language seems fine, its the stone age AWT and
>    RTL that amaze me the most. Trying to format numbers with Java seems to
>    require re-inventing the wheel - give me printf/scanf anyday. As for

It's not /that/ different than iostreams.  Except iostreams are much
more flexable.

>    the weak GUI in Java, soon to be replaced by a myriad set of toolkits
>    etc. to program better GUI's why wasn't this considered in the first
>    place?!?

I think it was, but turned down.  If I remember correctly, there were
various reasons; speed and time to market were among them.  Overall,
I have to give Java a bit of slack when it doesn't live up to the
marking hype right away.  It has only be released to the public for 
a little over a year!  It's still a baby.
230.4XSTACY::imladris.ilo.dec.com::grainneGrainne Ni ChoilighThu May 29 1997 13:108
Re: .2

I've had no problem building Java applets with both Visual J++ v1.1
trial edition and Visual J++ v1.1 released edition (it was released
some time ago) and using them with Netscape v3.0 and Netscape v4.0
PR3,4 & 5 

230.5MARVIN::GOODWINPete GoodwinFri May 30 1997 04:524
    I kept getting ScrollBar errors in Netscape 3.01 with J++ 1.1 Trail. I
    started using JDK 1.02, then J++ 1.0 and it all started working fine.
    
    Pete
230.634512::GHEFFDo you feel like swimming?Fri May 30 1997 10:4010
    Regarding point #1, it's been brought home abundantly that VM != VM
    != VM.  My applets (which I develop using Cafe Mac) regularly have
    little (and sometime big) inconsistencies when carried from one
    platform to the next.  Even the differences between the appletviewer
    and the browsers on the same system are enough to make me pull out my
    hair.
    
    Is it enough to make me stop doing it?  No way. 
    
    #Gary