Board index » kylix » Re: DElphi/Kylix Vs. Java ??

Re: DElphi/Kylix Vs. Java ??


2003-12-09 05:58:17 AM
kylix1
"pNichols" < XXXX@XXXXX.COM >wrote in message
[..]
Quote
So you are saying NET is open? You still have not answered my question as
to
why MS will not allow anyone to create their own version of NET that
would
be totally compatible with MS using the MS code base. Sun has done this,
MS
will not!!!
Actually you've been repreatedly told this is all your supposition and facts
point in different directions.
Do you have any evidence of MS' intentions of sueing Ximian or Novell? Of
course not!
These are all speculations you're selling as facts or something that will
most certainly happen.
You don't know that as we don't but we all have are elements suggesting the
opposite.
So stop going in circles and just admit you cannot know what MS is thinking.
Mono is a good strategic partner for MS, not the opposite.
 
 

Re:Re: DElphi/Kylix Vs. Java ??

Alessandro Federici wrote:
Quote
"pNichols" < XXXX@XXXXX.COM >wrote in message
news: XXXX@XXXXX.COM ...
[..]
>So you are saying NET is open? You still have not answered my question as
to
>why MS will not allow anyone to create their own version of NET that
would
>be totally compatible with MS using the MS code base. Sun has done this,
MS
>will not!!!

Actually you've been repreatedly told this is all your supposition and
facts point in different directions.
Do you have any evidence of MS' intentions of sueing Ximian or Novell? Of
course not!
These are all speculations you're selling as facts or something that will
most certainly happen.
You don't know that as we don't but we all have are elements suggesting
the opposite.
So stop going in circles and just admit you cannot know what MS is
thinking. Mono is a good strategic partner for MS, not the opposite.
MS reps, Alessandro, have stated that they reserve the right to take legal
action to protect their intellectual property (the question was asked and
the response given in a back issue of Web services and JDJ www.sys-con.com,
concerning a clean room inplementation of NET and Mono in particular). You
are welcome to search this article for yourself, or go and look for back
post where I gave these references and quotes in another post on
jbuilder.nontech. Do you honestly think that if someone were to acquire the
MS specifications and code specs, that MS would not sue?
MS will not hand the MS CLR/CLI to anyone. Sun has. That is the point.
Net is not compatible across implementations (look at the Mono page and
judge for yourself). Java is compatible whether you are using Sun's
JVM/JIT. IBM's, Blackdowns, or JVM implementations from Jeode to JVMs for
embedded devices. Why are they compatible? Because Javasoft has a JCP that
handles the specifications, code specs, APIs, and future direction of Java,
that is not controlled by Sun alone. It is controlled rather by the JCP. I
gave JQP the site where you can see it in legal balck and white for
yourself. If you want to refuse the documents themselves, be my guest, but
they are sufficient enough for IBM, BEA, Oracle, Nokia, CA, Borland, all
of which are a part of the JCP. Does Borland, IBM, Nokia, Oracle, etc.
determine what NET will be? Can they? Can they create their own version of
NET that is 100% compatible with MS NET because MS has given them the APIs,
code specs, source code, etc?
.
 

Re:Re: DElphi/Kylix Vs. Java ??

Quote
>I'm use Java apps only as user and i found following disadvantages of
>Java based apps.
>1. Insufficient performance.
RP>Unfounded claim, Ender. Insufficient for what? Scientific
RP>calculations involving huge matrices? Well, in that case, use
RP>Fortran.
Insufficient GUI speed, this just may be visible by eye on low end and
average desktops. Insufficient speed of any CPU consuming operations, not
necessary scientific calculations. This may be manipulation of large count
of pieces of some data. For example sorting or building some kind of dynamic
data structure. Note, you can see higher than usual CPU load.
Quote
>2. Memory requirements for VM.
RP>Tunable parameter, thank you.
Don't know how it can be tuned, but nearly all Java based apps i encountered
eat much more memory than it's native colleagues. Of course it does not
matter for 512Mb of RAM, but for machine that have 128Mb of memory and
successfully run _other_ apps it may be hurt. I think it is because of VM
overhead but that's the point also.
Quote
>3. Common stability of application depends from implementation of VM.
RP>Goody, is that a disadvantage to Kylix and/or Delphi? I submit that
RP>Delphi,
RP>Kylix as well as Java is subject to the quality of the runtime, be
RP>it DLLs, shared objects or JVM implementations.
Ok. I just very rare encounter fatal errors in Delphi based products because
of error in RTL, and often encounter errors in Java based products because
of JVM unable to work prooperly in certain environment (for example Russian
locale).
Quote
>4. Usage of not native GUI widgets.
RP>That's nonsense. Swing uses native widgets in the end. If you want
RP>to use non-native ones, no problem, if you want native ones, no
RP>problem.
I'm not care what _may_ be done, i'm count what already _was_ done. Remember
once, i'm not Java programmer, i'm user. Or you say that i'm can easily
replace VM or part of VM to use native GUI widgets. And there is absolutely
no problems with results of such operation?
Quote
>Two bright examples:
>1. Try to use oracle Java based stuff... you will wonder how
>corporation that created such beautiful product as Oracle Server also
>created such low quality slow and unstable java-based tools.
RP>Yeah serious. Oracle 8i uses JDK1.1.8, now that is truely ancient.
RP>Oracle 9i uses JDK1.3, more modern, true.
BTW, i'm talking about 9i. ;-) And it is bright example, not 8i.
RP>The question is, is the instability of the
RP>Java stuff in Oracle Java's fault or Oracle's fault.
I don't know. But i has common feeling that Java-based apps tend to be less
bullet-proof than Delphi or C++ based.
RP>FWIW, I agree with the fact that combining Oracle and Java looks like a
RP>plan but isn't. However, that's not because Java is bad, it is because
RP>Java has nothing to do in a database to begin with.
It is question of good libraries, not language itself. I.e question of
_implementation_, not specification.
Quote
>2. Any web site that use Java. If you don't own latest P4 with half
>gigabyte memory just hear sounds of your HDD.
RP>True, applets take a lot of time to load.
And this is still disadvantage.
RP>The advantage is that they are much more secure than let's say ActiveX
RP>controls <shudder>.
I think 99% of sites that currently have any applets does not need nether
ActiveX, nor Java. Usually while surfing i'm disable both and feel perfect
with it.
As user i think that Java (and anything based on it) still not have enough
quality to compare with native apps. Too much problems with Java executing
environment, not with language.
---
Andrew V. Fionik, Papillon Systems, Unix Programmers Group
For reply use "ender" instead of "fionika" in e-mail.
 

{smallsort}

Re:Re: DElphi/Kylix Vs. Java ??

"pNichols" < XXXX@XXXXX.COM >wrote in message
Quote
MS reps, Alessandro, have stated that they reserve the right to take legal
action to protect their intellectual property
So does Borland, Sun, IBM, RemObjects Software and any company that has a
little bit of business sense <G>
Your point being?
Quote
[..] Does Borland, IBM, Nokia, Oracle, etc.
determine what NET will be? Can they? Can they create their own version of
NET that is 100% compatible with MS NET because MS has given them the
APIs,
code specs, source code, etc?
That would be silly business. You don't reinvent the wheel. You improve on
it or make a better one.
Borland's business is not to remake what's ther, but to offer a better
solution. Same goes for IBM and the others.
Do you buy Nike or Mike? Which one do you think is gonna be the market
leader? You create Reebok or Adidas. That's the good business idea.
Still, once again, I dare you find any evidence of threats on Mono or the
other .Net implementations that are around.
It's all nice speculation, but nothing else. Facts talk.
Do you honestly think that if MS wanted to sue Ximian, they would publish
articles about it on the MSDN?
Do you think they would be so stupid to have Java zelots (which still think
they are better off for some magical reason <G>) say "we told you!"? Come
on. You should know better! Suing Mono would only *kill* a lot of .Net
chances.
--
Best regards,
Alessandro Federici
RemObjects Software, Inc.
www.remobjects.com
 

Re:Re: DElphi/Kylix Vs. Java ??

Alessandro Federici wrote:
Quote
You create Reebok or Adidas. That's the good business idea.
except that I'm fairly certain both Reebok and Adidas predate Nike.
--
.. P.
 

Re:Re: DElphi/Kylix Vs. Java ??

Ender wrote:
Quote
Insufficient GUI speed, this just may be visible by eye on low end and
average desktops.
Another example. C++ Builder X. Just installed. Even menus work slow. PC -
Celeron 850Mhz, 384Mb RAM, Linux Mandrake 9.2.
 

Re:Re: DElphi/Kylix Vs. Java ??

Ender wrote:
Quote
Insufficient GUI speed, this just may be visible by eye on low end and
average desktops.
It is a common fact that Swing is large and a little bit overbearing. True.
However, if somebody either buys a Java program or develops a Java program
that must be run on low end systems, he should take that into account, and
if need be, look for alternatives. Other than that, not considering
resource consumption when buying or creating programs in the first place is
a grave error. Blaming Java for the insufficiency of either buyer or
developer is not a good argument. Java has its uses, and in many cases, it
is a better option than native apps. Point in case might be the ease with
which Delphi applications can be ported to Kylix applications. Not exactly
smooth, especially if you're using third party components.
Quote
Insufficient speed of any CPU consuming operations, not
necessary scientific calculations. This may be manipulation of large count
of pieces of some data. For example sorting or building some kind of
dynamic data structure. Note, you can see higher than usual CPU load.
Again, same argument.
Quote
Don't know how it can be tuned, but nearly all Java based apps i
Enlighten yourself and execute:
java -X
in a console and pay attention to the -Xms, -Xmx and -Xss switches.
Quote
encountered eat much more memory than it's native colleagues. Of course it
does not matter for 512Mb of RAM, but for machine that have 128Mb of
memory and successfully run _other_ apps it may be hurt. I think it is
because of VM overhead but that's the point also.
Again, same argument. That the JVM has overhead, yes it has, and it has its
uses. That is what comes with the language. And that is what designers must
take into account. Yes, it might require a bit of tuning, but the
opportunities to do so are there, contratry to the possibilities a native
application has in that respect.
Quote
>>3. Common stability of application depends from implementation of VM.

RP>Goody, is that a disadvantage to Kylix and/or Delphi? I submit that
RP>Delphi,
RP>Kylix as well as Java is subject to the quality of the runtime, be
RP>it DLLs, shared objects or JVM implementations.

Ok. I just very rare encounter fatal errors in Delphi based products
because of error in RTL, and often encounter errors in Java based products
because of JVM unable to work prooperly in certain environment (for
example Russian locale).
Yes, locale specific errors are probably more common, but again. If you want
to maeke a fair comparison between what Delphi provides in terms of
functions and what Java provides in terms of functions, I think that Java
has the upper hand, especially when it comes to networking, threading and
security. Comparing the Delphi RTL to the Java RTL (that is, the standard
packages that come with J2SE plus the JVM) will probably lead you to the
comclusion that the Java RTL so to say is much much larger than the Delphi
RTL.
Quote
>>4. Usage of not native GUI widgets.

RP>That's nonsense. Swing uses native widgets in the end. If you want
RP>to use non-native ones, no problem, if you want native ones, no
RP>problem.

I'm not care what _may_ be done, i'm count what already _was_ done.
Remember once, i'm not Java programmer, i'm user. Or you say that i'm can
easily replace VM or part of VM to use native GUI widgets. And there is
absolutely no problems with results of such operation?
I am saying that Java, or more precisely AWT uses native widgets.
Quote
BTW, i'm talking about 9i. ;-) And it is bright example, not 8i.
True. I once tried to install 9i on this piddly little laptop. No chance.
Problem? Java.
Quote
I don't know. But i has common feeling that Java-based apps tend to be
less bullet-proof than Delphi or C++ based.
That's totally untrue. It's actually quite easy to make bulletproof Java
applications, and there are a host of open-source libraries and tools that
enables a person to create great applications. If and when my company is
able to create an application in three weeks onrelatively unstable specs
that is on time and above customer expectation, what does that tell you?
One, we a bunch of cool engineers, and two, we choose the right tools for
the job. Again, the stability of an application is very much unrelated to
the tools chosen and very much related to the quality of engineering.
Quote
It is question of good libraries, not language itself. I.e question of
_implementation_, not specification.
I disagree here wholeheartedly. It is a question of not combining
conflicting concepts in the first place and proper engineering in the
second place.
Quote
As user i think that Java (and anything based on it) still not have enough
quality to compare with native apps. Too much problems with Java executing
environment, not with language.
No. It is a matter of design design design, engineering engineering
engineering and testing testing testing.
--
Ruurd
 

Re:Re: DElphi/Kylix Vs. Java ??

Ender wrote:
Quote
Another example. C++ Builder X. Just installed. Even menus work slow. PC -
Celeron 850Mhz, 384Mb RAM, Linux Mandrake 9.2.
The specs say PII 500MHz or higer 512MB minimum. Why then run it on a
crippled processor with too little RAM? And if you do, don't expect stellar
performance.
--
Ruurd
 

Re:Re: DElphi/Kylix Vs. Java ??

Quote
>Another example. C++ Builder X. Just installed. Even menus work slow.
>PC -
>Celeron 850Mhz, 384Mb RAM, Linux Mandrake 9.2.
RP>The specs say PII 500MHz or higer 512MB minimum. Why then run it on
RP>a crippled processor with too little RAM?
Hold on. I'm not said that it going to swap. I'm saying that even trivial
operations not related to large memory use is slow.
RP>And if you do, don't expect stellar performance.
Another products, far more complicated and feature rich run much more
faster. This is called "poor performance of java-based app".
---
Andrew V. Fionik, Papillon Systems, Unix Programmers Group
For reply use "ender" instead of "fionika" in e-mail.
 

Re:Re: DElphi/Kylix Vs. Java ??

RP>programs in the first place is a grave error. Blaming Java for the
RP>insufficiency of either buyer or developer is not a good argument.
Please correct me if i'm misunderstood. You saying that when program working
slow because of memory and/or CPU overhead of RTL (comparing to other
programs) then buyer or developer of program is guilty, not developer of
that RTL or entrie concept?
RP>Java has its uses, and in many cases, it is a better option than
RP>native apps. Point in case might be the ease with which Delphi
RP>applications can be ported to Kylix applications. Not exactly
RP>smooth, especially if you're using third party components.
Quote
>Insufficient speed of any CPU consuming operations, not necessary
>scientific calculations. This may be manipulation of large count of
>pieces of some data. For example sorting or building some kind of
>dynamic data structure. Note, you can see higher than usual CPU load.
RP>Again, same argument.
Then role of Java little differ from role of VBA - be constructor for highly
optimized blocks written in native languages. Except for that Java is
cross-platform.
RP>Again, same argument. That the JVM has overhead, yes it has, and it
RP>has its uses. That is what comes with the language. And that is what
RP>designers must take into account. Yes, it might require a bit of
RP>tuning, but the opportunities to do so are there, contratry to the
RP>possibilities a native application has in that respect.
Native apps does not use more memory than they really needed.
Quote
>Ok. I just very rare encounter fatal errors in Delphi based products
>because of error in RTL, and often encounter errors in Java based
>products because of JVM unable to work prooperly in certain
>environment (for example Russian locale).
RP>Yes, locale specific errors are probably more common, but again. If
RP>you want to maeke a fair comparison between what Delphi provides in
RP>terms of functions and what Java provides in terms of functions, I
RP>think that Java has the upper hand, especially when it comes to
RP>networking, threading and security. Comparing the Delphi RTL to the
RP>Java RTL (that is, the standard packages that come with J2SE plus
RP>the JVM) will probably lead you to the comclusion that the Java RTL
RP>so to say is much much larger than the Delphi
RP>RTL.
Quote
>>>4. Usage of not native GUI widgets.
RP>>>That's nonsense. Swing uses native widgets in the end. If you want
RP>>>to use non-native ones, no problem, if you want native ones, no
RP>>>problem.
Quote
>I'm not care what _may_ be done, i'm count what already _was_ done.
>Remember once, i'm not Java programmer, i'm user. Or you say that i'm
>can easily replace VM or part of VM to use native GUI widgets. And
>there is absolutely no problems with results of such operation?
RP>I am saying that Java, or more precisely AWT uses native widgets.
Why C++ Builder X does not use native widgets for my Linux? And main
question what is native widgets for Linux? Qt, GTK? It does not look similar
to any Qt or GTK app i have in my system installed.
Quote
>BTW, i'm talking about 9i. ;-) And it is bright example, not 8i.
RP>True. I once tried to install 9i on this piddly little laptop. No
RP>chance. Problem? Java.
By the way, what window manager you using? I'm noticed fatal problems with
some WMs like fvwm, icewm or windowmaker, but none with KDE or GNOME.
Quote
>I don't know. But i has common feeling that Java-based apps tend to
>be less bullet-proof than Delphi or C++ based.
RP>That's totally untrue.
You cannot say that i'm lie to myself. :-)
RP>It's actually quite easy to make bulletproof Java applications, and
RP>there are a host of open-source libraries and tools that enables a
RP>person to create great applications. If and when my company is able to
RP>create an application in three weeks onrelatively unstable specs that
RP>is on time and above customer expectation, what does that tell you?
RP>One, we a bunch of cool engineers, and two, we choose the right
RP>tools for the job. Again, the stability of an application is very
RP>much unrelated to the tools chosen and very much related to the
RP>quality of engineering.
Do your cool engineers able to create Java program that will run faster, eat
less memory and be stable than native program that doing same things? Of
course we assume that both programs require equal resources for design and
testing and be developed equally qualified programmers.
 

Re:Re: DElphi/Kylix Vs. Java ??

"Phoenix" < XXXX@XXXXX.COM >wrote in message
Quote
>You create Reebok or Adidas. That's the good business idea.
except that I'm fairly certain both Reebok and Adidas predate Nike.
Sigh... It's pointless to make examples...
 

Re:Re: DElphi/Kylix Vs. Java ??

Ender wrote:
Quote
>>Another example. C++ Builder X. Just installed. Even menus work slow.
>>PC -
>>Celeron 850Mhz, 384Mb RAM, Linux Mandrake 9.2.

RP>The specs say PII 500MHz or higer 512MB minimum. Why then run it on
RP>a crippled processor with too little RAM?

Hold on. I'm not said that it going to swap. I'm saying that even trivial
operations not related to large memory use is slow.
No, it is called you are attempting to run a program with less memory than
is recommended. If you need to support old hardware, then Java is not a
good choice. All of the GUIs we have written in Java, however, would run on
your machine just fine. You would notice little difference between a Kylix
compiled app and one of our Java GUI apps using 1.4.2 jre.
Huge apps eat huge amounts of memory. You are confusing JB and BuilderX with
the same amount of options in the Delhi IDE. Truth is, they contain many
more functions and libraries.
98% of Java applications are not going to be at the same level as a full
fledged IDE with huge Libraries, Wizards, Plug-in capabilities, Compiler
options and (in the case of JB and BuilderX), compiler options (as in which
compiler do you want to use), etc. I am sure that as extensive and opne and
BuilderX is (letting you link and compile against X compilers), you are
going to have a lot of overhead involved in the linkers and
compiler/de{*word*81} checks. Sure, Borland could have made it smaller, but
flexibility is much more important.
BuilderX is not complete either. I do not know if it has been optimized yet
for release.
 

Re:Re: DElphi/Kylix Vs. Java ??

Alessandro Federici wrote:
Quote

That would be silly business. You don't reinvent the wheel. You improve on
it or make a better one.
Borland's business is not to remake what's ther, but to offer a better
solution. Same goes for IBM and the others.
Do you buy Nike or Mike? Which one do you think is gonna be the market
leader? You create Reebok or Adidas. That's the good business idea.

Huh? No one can invent a 100% compatible MS.NET period!! It is impossible,
since MS allows NO ONE but MS to determine the direction and future for
NET. That is the point. IBM cannot create a 100% compatible NET for AS400,
AIX, OS390, etc. Neither can Mono, Blackdown, Oracle, Ensignia, Nokia, etc.
That is not true with Java. IBM, Blackdown, Ensignia, etc. have their own
VMs and JITs that are compatible.
Attempting to make the argument that NET is as open as Java is foolish. If
Mono is allowed to make and release a NET, then there implementation can be
compatible and open and extensive, but it will not be MS.NET, it will be
Mono.NET.
But how does this help MS? It doesn't. Instead, it will hurt MS, since there
is no source of revenue for MS. And you really think MS will let this
happen? It would go against every bit of the past history and business
practices of MS.
Of course, the larger argument is "do not write any code today, wait till
whenever and use NET, since I like it better" (of course I am speaking
about non MS.NET). "Don't use Java (which runs on X today), wait till
whenever." Java wasn't invented by MS, so we cannot use it." Thses sound
like valid arguments don't they? <VBG>
Quote
Still, once again, I dare you find any evidence of threats on Mono or the
other .Net implementations that are around.
I gave you where the published information could be found. You can search it
or refuse to believe it. It is in these publications and perhaps others.
 

Re:Re: DElphi/Kylix Vs. Java ??

"pNichols" < XXXX@XXXXX.COM >wrote in message
Quote
Huh? No one can invent a 100% compatible MS.NET period!! It is impossible,
That is not what I said could and (maybe) should be done.
Please reread.
 

Re:Re: DElphi/Kylix Vs. Java ??

:-)))))) Interesting point of view. You trying to say that speed of program
depends from meeting memory requirements instead of actual memory usage? CBX
perfectly fit into 384Mb. It take much less than 384Mb of memory, and it
does not swap. So - memory is not the cause of slowdown.
Did you tried to install CBX? How you feel after speed of installer from
Zero G? Clicking the check box take about second for switching check box
state. PC - P5 with 512Mb of memory, for Windows and for Linux equally slow.
p>If you need to support old hardware, then Java is not a good choice. All
p>of the GUIs we have written in Java, however, would run on your machine
p>just fine. You would notice little difference between a Kylix compiled
p>app and one of our Java GUI apps using 1.4.2 jre.
p>Huge apps eat huge amounts of memory. You are confusing JB and
p>BuilderX with the same amount of options in the Delhi IDE. Truth is,
p>they contain many more functions and libraries.
BuilderX contain many more functions and libraries than Delphi or Kylix?
---
Andrew V. Fionik, Papillon Systems, Unix Programmers Group
For reply use "ender" instead of "fionika" in e-mail.