Re:Java's not exactly pining for the fields just n
on
Java vs .NET
·
· Score: 1
Why would I want to put in Intel assembly code into my Pascal program? I never had a need to drop in 68000, PDP-11, or Cyber assembly language, why would an 8080 be any different?
It is not that you necessarily NORMALLY would, but can, and this is just one example that defeats any type of full cross platform Pascal compiler instantly.
BTW I have done Pascal on PDP, 68000 and many other platforms as well, and trust me - THERE have been times I have had to drop to assembly in professional production.
So don't take the fact that just because cross platform compilers are available for ANY lanuguge that it somehow equates to an architecture consisting of a language running in a VM on multi-platforms like JAVA tries to do.
And this is especially true with languages that try to take advantage of a platform API like Deplhi and Kylix (Pascal) that support more than the antiquated char based applications. Borland did a pretty good job of trying to support both sides of the fence, but there are still SO many things that have to be rewritten and that break when moving from one code base to the other.
JAVA's VM had grand design ideals in its begining, but has failed so far to meet what they promised back in 1996.
It is still considerably slow and still has many problems running applications correctly in a mutli-platform environment. So everyone that believes that by building in JAVA that their software is truly portable at this point are diluting themselves as well.
I have too many developers in my companies that deal with this stuff on a daily basis, and I see the problems they come up against all the time.
I am not a kiddie that is just getting their feet wet in development and trying to post something on slashdot that I have no idea about.
I've been using Mandrake 9.1 here in California, which probably suffers more brownouts and blackouts than any other state. As long as I reboot when it tells me to, I don't lose anything...
Maybe in the news, but then you haven't lived in rural america, especially nevada. A few flashes of lightning in the sky, and cities come to their knees.
California gets the press, but living in So California, I have seen only two power outages in many years. (And a few brownouts) Our companies in Nevada see them almost weekly.
Just an FYI LOL...
Re:Java's not exactly pining for the fields just n
on
Java vs .NET
·
· Score: 0
A portable VM for Pascal... maybe you could call the machine language for that "pcode"...? Wow, I wonder why no one ever thought of doing something like that.
Wow and it still isn't a portable VM... Dropping Intel assembly in your pascal code and then try compiling it on your new G5.
Crappy software is crappy software. On any OS. But if an app is going to crash, let it crash - as long as it doesn't have an adverse affect on the the rest of the processes or usability. Applications should be written around the OS, not the other way around, IMO
In Theory, yes, but in reality, NO... Where have you found this perfect world where every programmer always follows the rules and doesn't write crap software?
(And this applies to the *nix world as much as to this Windows example. How many times have we all come across a piece of *nix code that was coded by a moron so that it crashed on certain distribuitions and not others?)
Additionally, when you are building NEWER versions of OSes that need to maintain vast backwards capatibility with crappy software that somewhat ran on the older OSes, even though it created stability problems, it is much easier to have the NEW OS just look out for the application or the 'application's poor type of coding' and fix the application execution in realtime.
In the END RUN, what does it matter if the Software was properly written and runs perfectly, or the software was half-ass written, but STILL RUNS perfectly because the OS is smart enough to fix it.
The END USER WINS in EACH senerio. This is not about 'punishing users' by making their app fully crash all the time just because the person that wrote it was a moron.
Talk about creating User Friendly OSes, I sure hope you don't develop UI professionally on any OS.
Geesh...
As for your debugging senario, you are talking out your ass. If you are not debugging and trapping errors, then how in hell do you expect your software to be virtually error free? Whether or not you are developing it on Win9X where it will crash or WindowsXP where it won't? (Or on a *nix where it would.)
This whole realtime crash protection in WindowsXP is just an extension of a concept of 'protecting application memory', like MOST modern OSes do already.
If we were to apply your logic, then no one should make OSes protect application memory either, because all third party developers should just know what they are doing and rely on them to keep the programs from crashing or crashing each other.
What a fool...
Re:Isn't that pining for the fjords? Anyway...
on
Java vs .NET
·
· Score: 1
VB is not as much a programming language that became a GUI, but a prototyping tool that became a programming language. The move to VB.NET has taken away all the language simplicity making it a fully fledged language, but made it difficult enough to negate its benefits over other products that fall within its same niche. I do my quick and dirty GUI's in Jbuilder now. If I have to use a fully fledged language, I would rather use a portable one.
I agree, but I also contest that there are better GUI based portable languages. Starting with Delphi and Kylix as just an example.
The speed loss of using tools like JBuilder are not worth it. The end product is severly crippled in terms of performance and compactness.
Additionally, there are even many C++ GUI based development tools that easily produce multi-os products and still run native code.
I think we agree, just not on the methodology of the end product.
GUI apps don't have to be slow, just because they are a quick write.
Re:Java's not exactly pining for the fields just n
on
Java vs .NET
·
· Score: 1
Before the language was called 'Java', it was called 'Oak'. It was a language for building embedded applications on smart consumer electronics. However, Oak was way ahead of its time in terms of product targeting.
The Oak Language and JAVA may have the same syntatical underpinnings, but JAVA was a web based VM techology using the OAK language.
They are STILL two different things.
It would be like writing a portable VM for Pascal and then telling the world that this new technology was just plain old Pascal.
JAVA is a serious of technologies based around the VM technology, it was later adapted back to embedded systems, but STILL with the JAVA VM concepts.
Sun's not a monopoly and don't make it sound like MS gave the entire.NET framework to the ECMA. Nope, just the C# langauge and the CLI, this would be like Sun turning over javac and java to the ECMA but keeping a grip on anything beyond Java primitives, it's a bullshit token PR jesture. The JCP is a far more public process for directing Java than anything we'll ever see from MS for.NET.
No Sun is wonderful funny bunnies and has been a God send to the Open Source community.
Geesh... Do you really know anything about Sun and their continued stranglehold control of JAVA? (Sun - We will open Java up and submit to standards, then next month, we decided to withdraw and not do this, etc etc...)
Or the other business practices that they have embarked on to screw the open source community and users in general?
Just because they have brought Linux on board for some of their systems, does not mean they are out to play nice with open source now.
Do some Sun history, you will find more questionable practices than Microsoft even. Java just being the tip of the ice berg.
Sun was stiff arming vendor and customers long before NT or Windows had even made a dent in computing history.
Even take the Solaris for Intel of 1993 that was pushed to circumvent the upcoming release of OS/2 and NT, as it was nothing but marketing FUD to keep OS/2, NT, and Novell out of their server markets.
Re:Isn't that pining for the fjords? Anyway...
on
Java vs .NET
·
· Score: 1
.Ack! I hated that!! If I wanted to write really simple GUI apps for windows, I would have used VB. I won't even bother with VB.NET. I see no advantage to using.NET for what VB was ment for.
I agree that on the surface VB.NET doesn't look that better than old VB, but with.NET it pulls VB into a new realm that it has only tried to hack into before. A real programming language. Real Object Oriented concepts and even more managed code with the new power of the.NET framework.
Most people have always hated VB for the (VM runtime DLLS) but then you find that these are the same people that love Sun JAVA because of its runtime VM). Strange...
I guess Microsoft shouldn't have stopped the project to port VB's VM to other OSes. Then the Java zealots would all be VB users today.:)
In all honesty, VB was a great product almost 10 years ago for what it did, but with modern IDE's like Delphi and Borland C++, there is almost no ease of use that VB offers that can't be found in other development languages.
Re:Java's not exactly pining for the fields just n
on
Java vs .NET
·
· Score: 1
There are plenty of Java IDEs that simply blow Visual Studio out of the water in terms of functionality at the moment.
That is somewhat debatable as personal preference, but the Borland.NET tools and languages are pretty top notch.
Borland JBuilder as you mention is nice, but not as robust or nice as their new C# support and.NET or the old standbys like Delphi and Kylix.
A language or model is not made on one company's(Microsoft) incarnation of it. Borland has demonstrated that there is some good in the.NET framework and has provided some pretty good tools.
Having spent a lot of time in development, I still prefer Most of Borland's products over MS for development, but that doesn't mean the.NET framework is bad just because VS is a bit hard to dive into - hence why I use Borland products for even.NET work.
Re:Java's not exactly pining for the fields just n
on
Java vs .NET
·
· Score: 1
Which sums up the whole apples vs oranges nature of this "debate" quite nicely. Java was developed for embedded systems,.NET for enterprise/distributed computing
Although somewhat true now, but missing a few key points...
#1) Java was NOT originally developed for embedded systems. It wasn't until the past couple of years that Sun has moved it into that direction. (There is a difference between embedded systems and a cross platform VM)
#2) Microsoft has other embedded system technology that.NET plays a role in, and works with.
So, a developer working with Windows XP might not know of a real bug, because the OS is doing smoke and mirrors behind his back? That's a high crime of software engineering and makes crappy developers look better than they are.
Um, no... This is why they have something called 'debugging tools' - stay with me here people.
Are you a total moron, or never developed software and just commneting on something you know nothing about?
If you hate Microsoft and want to make this an anti-Microsoft spin, just say so, don't make yourself look stupid.
One thing about this "fire fighting" mode of software development is that for each fire fought 100 could have been prevented.
Yeah, Microsoft tells all third parties how to write their software poorly, it is all Microsoft's fault... Geesh...
Other OSes would do good to learn from this OS model, as it protects itself and end users from crap software. Period.
WindowsXP (and 2k to some extent) can catch bad calls by junk software, fix the call and let the application continue to run for the nerd that insists on using it.
Maybe next time they will actually *learn* something about security. Like YOU DON'T PUT A F*CKING WEBSERVER IN THE KERNEL. On the next IIS-exploit, they are going to find out that even LocalSystem is peanuts compared to Ring 0.
And in the meantime you will actually learn about OS architecture.
IIS having kernel level access is that it has a kernel level DRIVER. Nothing more.
And this is mainly for Cache Performance speed increases. So it is a kernel level driver, just like other drivers used throughout the OS.
Even with the Ring 0 (kernel) driver, IIS STILL has to pass through NT's object based security model.
Now, explain how having a ring 0 driver is a big security problem?
The only problem that could happen is if the IIS kernel driver was poorly written and crashed the OS. (Not likely to happen if it hasn't happened in 2 years of testing.)
I don't know what anyone with a better understanding would make of using "send to" causing a protection fault, periodic mouseless periods of up to 10 seconds to name but two. I was disappointed I must say, especially as I hadn't done anything particulary nasty to it. Maybe it is the residue of a million COM experiments or the introduction of a Java development system, but I'd have expected better not worse. Still - the next machine will no doubt have less problems, assuming it's all pre-installed.
As our techs often say, people need to be given a top 10 list of things users do wrong that could cause an OS to be unstable.
This applies to all OSes, and often are simple mistakes people make when upgrading or moving to a different version or different OS.
hummmm..... thats odd. I was not aware that Xp used a nice distributed hylafax. Or that it includes loads of nice tools for X11. Or that it is actually useable out of the box (I always thought that to do anything with MS, you had to add extra applications that were worth a hoot). Now as to the innovative features of XP such as multi logins, we had those in X back in the late 80's, early 90's. As to other innovative features, what are they?
You really don't know much about XP or NT do you?
XP multi Logins? That has been in NT since 1992. (1993 release)
Maybe you meant concurrent Multi User Logins?
If so, the plumbing, again, was in NT 3.1 for this, and there were several third party add-ons that allowed NT3.1 to do this. With NT4 Multi-User sessions was turned into Terminal Services which was bundled into Win2k.
Support for X11 has also been available from third parties and Microsoft, but not widely promoted. Additionally, X11 is a fairly OLD technology (having worked with the development of it myself years and years ago). It was a great Networking GUI protocol, but for a desktop window manager suffers quite a bit in its incarnation. It was designed to offer GUI support from a Server in place of Text based Terminals from Mini and Supercomputers.
It was NOT designed as a desktop window environment and not something that should be touted as the 'must have' feature.
Even Apple knows this, or X11 would be the basis of its desktop environment, and it simply is NOT.
And to this date, Apple itself does not have concurrent multi-user remote desktop technology, with the exception of what its Unix underpinnings support with an XServer.
Interesting. I was running NT4 for around 7 years at 3 different companies with virtually zero problems. My recent XP upgrade has caused things to stop working that previously worked fine.
Our test labs include systems from 200mhz Pentiums with 80mb of RAM to the latest technology. XP was beta tested on all of our platforms.
Even on the 200mhz machines (including a couple of 200mhz laptops) XP was not only 100% stable, but benchmarked and ran applications faster than Win98, Redhat, and NT4.
Sorry your experience was not better. I do understand that our techs have a better understanding of the NT technology and XP than the average user though.
I think they were stung pretty bad after the release of XP which was worse than previous Microsoft OS's beta versions. Maybe, for once, they are just trying to do it right.
We were in the XP and 2003 beta, and you are off base. XP was more solid of a release than 2000 even, there were several updates in the first few months but they were based on 'application compatibility' more than anything. (Because of the errors generated when a poorly written app crashed and sent a 'bug report' to Microsoft)
So with these fixes, Microsoft made XP aware of the bugs in the programs instead of forcing the third party manufacturers to rewrite or rerelease fixes to their broken software.
That is why the error reporting tool in XP works so well, is that the OS can be made stronger by fixing and working around bugs in poorly written third party applications.
Windows Server 2003 took longer to release because of the re-written IIS and.NET layers. Security was also a main concern, but not because XP or 2000 were insecure.
it lists lots of defects in applications adn lists defects in non-standard
This is the definition of an exploit... If the OS is INCAPABLE of protecting itself from programs creating security invulnerabilities, it is the OS that is failing, not the software.
This is why the Windows9x series of OSes are so badly perceived, because even though their core code is quite stable, it has NO inherent security or mechanisms to protect itself from the software that runs on it. Developers could write software that broke other applications, mess with the system itself, and open exploits in the OS for failure. Just like your Mac OS9 software is incapable of protecting itself as well.
Additionally, the 'extras' and 'apps' you dismiss are very MUCH a part of what is the NORMAL usage in a 'Server' environment for Mac OS9.
Where are the MacOS 9 security ratings? There are none. It isn't even capable of a simple C2 level certification because of its inherent lack of login and system file security.
As for the ARMY getting fed up and moving to OS9, you have no idea what you are talking about. Most of the military that are using Mac Servers have already moved to OSX with plans for all Mac Servers to be replaced by other companies technologies or moved to OSX. Apple itself is working with the military in the transition to OSX, and they are PUSHING THEM AWAY from OS9 for SECURITY REASONS.
I am also not going to debate that OSX is the most secure Server OS either. Apple has a maturity period to catch up with many of the other *nix and OSes that have been doing security work for many years. Just keeping 'services' turned off by default and keeping 'ports' closed by default does not mean or make the OS secure. If it was secure, it could leave these services ON with security in place and not have to worry about these services failing. Period.
I am tired of this debate. I am an OS architectural engineer, and do not have time to educate kiddies about OSes.
Stick with your Mac OS9, I am happy that is your choice; however, I do feel sorry that your do not realize that there are 'better' solutions for OSes, especially servers.
If there weren't, Apple themselves would not have completely re-engineered their OS and would have stayed with the Mac System 9 platform. So apparently you think you know better than Apple themselves and the people that wrote Mac OS9. I suggest you write a letter to Jobs and ask him to resign and appoint you leader of the Mac future because of your 'vast' OS architectural understanding.
First off, even though you try to use it, you obviously have no idea what a hyperbole is. So run off to your dictionary...
Secondly, just go to www.securemac.com - there are MANY examples of insecurities in Mac OS 9, including many that are yet unfixed.
If you want an instance of someone using one of these exploits to hack a Mac OS9 System I suggest that you keep pissing people off on Shashdot with your ignorant (100% secure statements), and you will probably get a wave of them.
Securemac.com shows several vulnerabilities and insecurities in Mac OS 9, and that is all the proof a serious IT administrator should need.
- You will find the numerous Appletalk and other various TCP/IP based flaws/insecurites in System 9, 8, and even OSX that you seek and somehow haven't found and don't believe exist.
I used to always ask people who insisted that the classic Mac OS was just as vulnerable as any other OS how they would get around the fact that the classic Mac OS had no command line.
That usually stopped them fast in their tracks.
They other thing that I asked anyone who was questioning the security of the classic Mac OS was to name one instance -- just one -- where a classic Mac OS machine had been cracked.
First off, there was a debug level command line, just because you may not have been bright enough to pop it open, does not mean it was not there. Additionally, what does a command line have to do with security? Telneting into supercomputers was VERY secure even in the 80s, which is a COMMAND LINE interface.
Secondly, there are MANY instances where Classic Macs were hacked into or compromised. Since the majority of the Mac life there was not an 'internet' that it could be hacked into from the outside world does not mean that local systems, and systems on LANs and even via AppleTalk were not compromised.
I worked for a technical consulting company that was a major provider for the 'graphic design' industry in the late 80s, early 90s. All too often people had their work lost or maliciously destroyed because of the inherent lack of ANY security at the console or LAN level on Macs.
Even Windows for Workgroups 3.1 had more security for local and LAN access, and it was also completely insecure as an OS. Just like the Mac System software, it was NEVER designed to be in a major networked system.
However products like NT, Solaris, Novell, many *nixs, and other major OSes during this same time period were built around a 'security model'.
I feel sorry for people that get caught in the 'old way' is the best, line of thinking.
believe the facts speak for themselves NetAvenger. Or can you cite ** just 1 instance where the Mac Pre X/Webstar combo WAS hacked ** (other than the instance above which involved a third party plug-in exploit, 8 mth duration, and a 100,000SK incentive). By all means do identify any other os/server combo that stacks up in this context. I'm sure your learned perspective will prove quite enlightening. Cheers.
I stand by my comments.
This is not about how many times someone can prove a Mac running System 9 or earlier was compromised.
The point is, Mac System 9 and earlier software is inherently insecure, has almost NO modern OS features, and is a poor choice for a server environment.
You would be better off running a Win3.1 box with a TCP/IP stack. It would be just as secure, as it also would not be a major target, and technologically, it is JUST as advanced as the core structure of Mac System Software prior to OSX. Period.
Mac System 9 and under have no file system level security, no kernel level security, and are seriously flawed in terms of features they offer for a server environment, and in many ways, even a desktop environment considering that technically it has a poor memory management system, an aged monolithic messaging cue, and is an application dependant cooperative multi-tasking OS. JUST LIKE WINDOWS 3.1 which was released in 1992.
OSX is light years ahead of Mac System software, just like Linux, FreeBSD, and WindowsNT which have true security built into the core/kernels of the OS. Also take into account that these non-Apple OSes where doing this stuff for over 10 years before Apple finally got around to giving us OSX. I remember the great IBM/Apple announcements and idea of having a real pre-emptive multitasking OS with security on Macs available in 1993 as planned. I was even eager for it to happen. It didn't. Microsoft Built NT, and the other *nix variants for the Intel Platform took off providing the features Apple (the great innovator *cough*) was supposed to have given us over 10 years ago.
And yet you are here trying to tell the world that Mac System 9 is better? I am truly amazed that you are so lost in your hyperbole.
Your logic amazes me. You are saying that just because Mac System 9 was 'seldom' hacked that it is superior. Using your logic, we should be using CPM, or DOS, or Win3.1 as they were never a major target and never hacked into either.
You are out of your league here; try some place other than Slashdot to bloviate about the wonders of Mac System 9 Software. Most people here know the history and the architectures of the major OSes, in addition to the technological shortcomings of Mac OS 9 - especially in a server capacity.
First off, the security you state is way out of whack.
Secondly, who would trust a server that has poor memory management (80s technology), no modern OS features like: Journalled FS, Pre-Emptive Multitasking, Ability to use Large Partitions, Multiple CPUs, More than 2GB of RAM, Etc, Etc, Etc...
Give us a break, Mac OS9 is technically at the same level as Windows 3.1. Hence why Windows3.1 is not a high risk OS for the internet, there is not much in it to hack short of the Winsock add-on.
So we should all go back to DOS, Win3.1, and MacOS9, so that we don't get hacked. Give me a break...
With modern advances, there are security caveats, but limiting yourself to antiquated technology to avoid them is just ridiculous.
Secondly, do you even comprehend the performance differences between an OS like Mac System9 and *nix, NT or other platforms? There is a massive margin, and System9 falls quite short - especially as a Server.
If Mac System9 can't even play an MP3 in the background without skipping while you are working, how do you think it is going to handle 100,000 hits as a Server?
Take a computer class or spend your time spreading your ignorance to a group that doesn't know better.
But this is exactly how 'Automatic Updates' are set in WindowsXP by default unless their computer nerd friends turn them off in 'fear' of Microsoft.
Which has been the case I have found more than anything. Some 'Guru' or other OS Zealot that hates Microsoft screws with their friends computer and turns off the 'Automatic Update' Notification service.
Just leave it alone people and let it work, especially for your novice friends. Even if you think Microsoft is Satan himself and going to eat your children.
Slashdot Mirror
Why would I want to put in Intel assembly code into my Pascal program? I never had a need to drop in 68000, PDP-11, or Cyber assembly language, why would an 8080 be any different?
It is not that you necessarily NORMALLY would, but can, and this is just one example that defeats any type of full cross platform Pascal compiler instantly.
BTW I have done Pascal on PDP, 68000 and many other platforms as well, and trust me - THERE have been times I have had to drop to assembly in professional production.
So don't take the fact that just because cross platform compilers are available for ANY lanuguge that it somehow equates to an architecture consisting of a language running in a VM on multi-platforms like JAVA tries to do.
And this is especially true with languages that try to take advantage of a platform API like Deplhi and Kylix (Pascal) that support more than the antiquated char based applications. Borland did a pretty good job of trying to support both sides of the fence, but there are still SO many things that have to be rewritten and that break when moving from one code base to the other.
JAVA's VM had grand design ideals in its begining, but has failed so far to meet what they promised back in 1996.
It is still considerably slow and still has many problems running applications correctly in a mutli-platform environment. So everyone that believes that by building in JAVA that their software is truly portable at this point are diluting themselves as well.
I have too many developers in my companies that deal with this stuff on a daily basis, and I see the problems they come up against all the time.
I am not a kiddie that is just getting their feet wet in development and trying to post something on slashdot that I have no idea about.
I've been using Mandrake 9.1 here in California, which probably suffers more brownouts and blackouts than any other state. As long as I reboot when it tells me to, I don't lose anything...
Maybe in the news, but then you haven't lived in rural america, especially nevada. A few flashes of lightning in the sky, and cities come to their knees.
California gets the press, but living in So California, I have seen only two power outages in many years. (And a few brownouts) Our companies in Nevada see them almost weekly.
Just an FYI LOL...
A portable VM for Pascal ... maybe you could call the machine language for that "pcode"...? Wow, I wonder why no one ever thought of doing something like that.
Wow and it still isn't a portable VM... Dropping Intel assembly in your pascal code and then try compiling it on your new G5.
Wow, wonder why that wouldn't work?
LOL
Crappy software is crappy software. On any OS. But if an app is going to crash, let it crash - as long as it doesn't have an adverse affect on the the rest of the processes or usability.
Applications should be written around the OS, not the other way around, IMO
In Theory, yes, but in reality, NO... Where have you found this perfect world where every programmer always follows the rules and doesn't write crap software?
(And this applies to the *nix world as much as to this Windows example. How many times have we all come across a piece of *nix code that was coded by a moron so that it crashed on certain distribuitions and not others?)
Additionally, when you are building NEWER versions of OSes that need to maintain vast backwards capatibility with crappy software that somewhat ran on the older OSes, even though it created stability problems, it is much easier to have the NEW OS just look out for the application or the 'application's poor type of coding' and fix the application execution in realtime.
In the END RUN, what does it matter if the Software was properly written and runs perfectly, or the software was half-ass written, but STILL RUNS perfectly because the OS is smart enough to fix it.
The END USER WINS in EACH senerio. This is not about 'punishing users' by making their app fully crash all the time just because the person that wrote it was a moron.
Talk about creating User Friendly OSes, I sure hope you don't develop UI professionally on any OS.
Geesh...
As for your debugging senario, you are talking out your ass. If you are not debugging and trapping errors, then how in hell do you expect your software to be virtually error free? Whether or not you are developing it on Win9X where it will crash or WindowsXP where it won't? (Or on a *nix where it would.)
This whole realtime crash protection in WindowsXP is just an extension of a concept of 'protecting application memory', like MOST modern OSes do already.
If we were to apply your logic, then no one should make OSes protect application memory either, because all third party developers should just know what they are doing and rely on them to keep the programs from crashing or crashing each other.
What a fool...
VB is not as much a programming language that became a GUI, but a prototyping tool that became a programming language. The move to VB.NET has taken away all the language simplicity making it a fully fledged language, but made it difficult enough to negate its benefits over other products that fall within its same niche. I do my quick and dirty GUI's in Jbuilder now. If I have to use a fully fledged language, I would rather use a portable one.
I agree, but I also contest that there are better GUI based portable languages. Starting with Delphi and Kylix as just an example.
The speed loss of using tools like JBuilder are not worth it. The end product is severly crippled in terms of performance and compactness.
Additionally, there are even many C++ GUI based development tools that easily produce multi-os products and still run native code.
I think we agree, just not on the methodology of the end product.
GUI apps don't have to be slow, just because they are a quick write.
Before the language was called 'Java', it was called 'Oak'. It was a language for building embedded applications on smart consumer electronics. However, Oak was way ahead of its time in terms of product targeting.
The Oak Language and JAVA may have the same syntatical underpinnings, but JAVA was a web based VM techology using the OAK language.
They are STILL two different things.
It would be like writing a portable VM for Pascal and then telling the world that this new technology was just plain old Pascal.
JAVA is a serious of technologies based around the VM technology, it was later adapted back to embedded systems, but STILL with the JAVA VM concepts.
Sun's not a monopoly and don't make it sound like MS gave the entire .NET framework to the ECMA. Nope, just the C# langauge and the CLI, this would be like Sun turning over javac and java to the ECMA but keeping a grip on anything beyond Java primitives, it's a bullshit token PR jesture. The JCP is a far more public process for directing Java than anything we'll ever see from MS for .NET.
No Sun is wonderful funny bunnies and has been a God send to the Open Source community.
Geesh... Do you really know anything about Sun and their continued stranglehold control of JAVA? (Sun - We will open Java up and submit to standards, then next month, we decided to withdraw and not do this, etc etc...)
Or the other business practices that they have embarked on to screw the open source community and users in general?
Just because they have brought Linux on board for some of their systems, does not mean they are out to play nice with open source now.
Do some Sun history, you will find more questionable practices than Microsoft even. Java just being the tip of the ice berg.
Sun was stiff arming vendor and customers long before NT or Windows had even made a dent in computing history.
Even take the Solaris for Intel of 1993 that was pushed to circumvent the upcoming release of OS/2 and NT, as it was nothing but marketing FUD to keep OS/2, NT, and Novell out of their server markets.
.Ack! I hated that!! If I wanted to write really simple GUI apps for windows, I would have used VB. I won't even bother with VB.NET. I see no advantage to using .NET for what VB was ment for.
.NET it pulls VB into a new realm that it has only tried to hack into before. A real programming language. Real Object Oriented concepts and even more managed code with the new power of the .NET framework.
:)
I agree that on the surface VB.NET doesn't look that better than old VB, but with
Most people have always hated VB for the (VM runtime DLLS) but then you find that these are the same people that love Sun JAVA because of its runtime VM). Strange...
I guess Microsoft shouldn't have stopped the project to port VB's VM to other OSes. Then the Java zealots would all be VB users today.
In all honesty, VB was a great product almost 10 years ago for what it did, but with modern IDE's like Delphi and Borland C++, there is almost no ease of use that VB offers that can't be found in other development languages.
There are plenty of Java IDEs that simply blow Visual Studio out of the water in terms of functionality at the moment.
.NET tools and languages are pretty top notch.
.NET or the old standbys like Delphi and Kylix.
.NET framework and has provided some pretty good tools.
.NET framework is bad just because VS is a bit hard to dive into - hence why I use Borland products for even .NET work.
That is somewhat debatable as personal preference, but the Borland
Borland JBuilder as you mention is nice, but not as robust or nice as their new C# support and
A language or model is not made on one company's(Microsoft) incarnation of it. Borland has demonstrated that there is some good in the
Having spent a lot of time in development, I still prefer Most of Borland's products over MS for development, but that doesn't mean the
Which sums up the whole apples vs oranges nature of this "debate" quite nicely. Java was developed for embedded systems, .NET for enterprise/distributed computing
.NET plays a role in, and works with.
Although somewhat true now, but missing a few key points...
#1) Java was NOT originally developed for embedded systems. It wasn't until the past couple of years that Sun has moved it into that direction. (There is a difference between embedded systems and a cross platform VM)
#2) Microsoft has other embedded system technology that
So, a developer working with Windows XP might not know of a real bug, because the OS is doing smoke and mirrors behind his back? That's a high crime of software engineering and makes crappy developers look better than they are.
Um, no... This is why they have something called 'debugging tools' - stay with me here people.
Are you a total moron, or never developed software and just commneting on something you know nothing about?
If you hate Microsoft and want to make this an anti-Microsoft spin, just say so, don't make yourself look stupid.
One thing about this "fire fighting" mode of software development is that for each fire fought 100 could have been prevented.
Yeah, Microsoft tells all third parties how to write their software poorly, it is all Microsoft's fault... Geesh...
Other OSes would do good to learn from this OS model, as it protects itself and end users from crap software. Period.
WindowsXP (and 2k to some extent) can catch bad calls by junk software, fix the call and let the application continue to run for the nerd that insists on using it.
Maybe next time they will actually *learn* something about security. Like YOU DON'T PUT A F*CKING WEBSERVER IN THE KERNEL. On the next IIS-exploit, they are going to find out that even LocalSystem is peanuts compared to Ring 0.
And in the meantime you will actually learn about OS architecture.
IIS having kernel level access is that it has a kernel level DRIVER. Nothing more.
And this is mainly for Cache Performance speed increases. So it is a kernel level driver, just like other drivers used throughout the OS.
Even with the Ring 0 (kernel) driver, IIS STILL has to pass through NT's object based security model.
Now, explain how having a ring 0 driver is a big security problem?
The only problem that could happen is if the IIS kernel driver was poorly written and crashed the OS. (Not likely to happen if it hasn't happened in 2 years of testing.)
I don't know what anyone with a better understanding would make of using "send to" causing a protection fault, periodic mouseless periods of up to 10 seconds to name but two. I was disappointed I must say, especially as I hadn't done anything particulary nasty to it. Maybe it is the residue of a million COM experiments or the introduction of a Java development system, but I'd have expected better not worse. Still - the next machine will no doubt have less problems, assuming it's all pre-installed.
As our techs often say, people need to be given a top 10 list of things users do wrong that could cause an OS to be unstable.
This applies to all OSes, and often are simple mistakes people make when upgrading or moving to a different version or different OS.
No offense was meant.
hummmm.....
thats odd. I was not aware that Xp used a nice distributed hylafax. Or that it includes loads of nice tools for X11. Or that it is actually useable out of the box (I always thought that to do anything with MS, you had to add extra applications that were worth a hoot).
Now as to the innovative features of XP such as multi logins, we had those in X back in the late 80's, early 90's. As to other innovative features, what are they?
You really don't know much about XP or NT do you?
XP multi Logins? That has been in NT since 1992. (1993 release)
Maybe you meant concurrent Multi User Logins?
If so, the plumbing, again, was in NT 3.1 for this, and there were several third party add-ons that allowed NT3.1 to do this. With NT4 Multi-User sessions was turned into Terminal Services which was bundled into Win2k.
Support for X11 has also been available from third parties and Microsoft, but not widely promoted. Additionally, X11 is a fairly OLD technology (having worked with the development of it myself years and years ago). It was a great Networking GUI protocol, but for a desktop window manager suffers quite a bit in its incarnation. It was designed to offer GUI support from a Server in place of Text based Terminals from Mini and Supercomputers.
It was NOT designed as a desktop window environment and not something that should be touted as the 'must have' feature.
Even Apple knows this, or X11 would be the basis of its desktop environment, and it simply is NOT.
And to this date, Apple itself does not have concurrent multi-user remote desktop technology, with the exception of what its Unix underpinnings support with an XServer.
So back to your point? Was there one?
And your reply contradicts anything I said where?
Interesting. I was running NT4 for around 7 years at 3 different companies with virtually zero problems. My recent XP upgrade has caused things to stop working that previously worked fine.
Our test labs include systems from 200mhz Pentiums with 80mb of RAM to the latest technology. XP was beta tested on all of our platforms.
Even on the 200mhz machines (including a couple of 200mhz laptops) XP was not only 100% stable, but benchmarked and ran applications faster than Win98, Redhat, and NT4.
Sorry your experience was not better. I do understand that our techs have a better understanding of the NT technology and XP than the average user though.
I think they were stung pretty bad after the release of XP which was worse than previous Microsoft OS's beta versions. Maybe, for once, they are just trying to do it right.
.NET layers. Security was also a main concern, but not because XP or 2000 were insecure.
We were in the XP and 2003 beta, and you are off base. XP was more solid of a release than 2000 even, there were several updates in the first few months but they were based on 'application compatibility' more than anything. (Because of the errors generated when a poorly written app crashed and sent a 'bug report' to Microsoft)
So with these fixes, Microsoft made XP aware of the bugs in the programs instead of forcing the third party manufacturers to rewrite or rerelease fixes to their broken software.
That is why the error reporting tool in XP works so well, is that the OS can be made stronger by fixing and working around bugs in poorly written third party applications.
Windows Server 2003 took longer to release because of the re-written IIS and
it lists lots of defects in applications adn lists defects in non-standard
This is the definition of an exploit... If the OS is INCAPABLE of protecting itself from programs creating security invulnerabilities, it is the OS that is failing, not the software.
This is why the Windows9x series of OSes are so badly perceived, because even though their core code is quite stable, it has NO inherent security or mechanisms to protect itself from the software that runs on it. Developers could write software that broke other applications, mess with the system itself, and open exploits in the OS for failure. Just like your Mac OS9 software is incapable of protecting itself as well.
Additionally, the 'extras' and 'apps' you dismiss are very MUCH a part of what is the NORMAL usage in a 'Server' environment for Mac OS9.
Where are the MacOS 9 security ratings? There are none. It isn't even capable of a simple C2 level certification because of its inherent lack of login and system file security.
As for the ARMY getting fed up and moving to OS9, you have no idea what you are talking about. Most of the military that are using Mac Servers have already moved to OSX with plans for all Mac Servers to be replaced by other companies technologies or moved to OSX. Apple itself is working with the military in the transition to OSX, and they are PUSHING THEM AWAY from OS9 for SECURITY REASONS.
I am also not going to debate that OSX is the most secure Server OS either. Apple has a maturity period to catch up with many of the other *nix and OSes that have been doing security work for many years. Just keeping 'services' turned off by default and keeping 'ports' closed by default does not mean or make the OS secure. If it was secure, it could leave these services ON with security in place and not have to worry about these services failing. Period.
I am tired of this debate. I am an OS architectural engineer, and do not have time to educate kiddies about OSes.
Stick with your Mac OS9, I am happy that is your choice; however, I do feel sorry that your do not realize that there are 'better' solutions for OSes, especially servers.
If there weren't, Apple themselves would not have completely re-engineered their OS and would have stayed with the Mac System 9 platform. So apparently you think you know better than Apple themselves and the people that wrote Mac OS9. I suggest you write a letter to Jobs and ask him to resign and appoint you leader of the Mac future because of your 'vast' OS architectural understanding.
First off, even though you try to use it, you obviously have no idea what a hyperbole is. So run off to your dictionary...
Secondly, just go to www.securemac.com - there are MANY examples of insecurities in Mac OS 9, including many that are yet unfixed.
If you want an instance of someone using one of these exploits to hack a Mac OS9 System I suggest that you keep pissing people off on Shashdot with your ignorant (100% secure statements), and you will probably get a wave of them.
Securemac.com shows several vulnerabilities and insecurities in Mac OS 9, and that is all the proof a serious IT administrator should need.
PS. I forgot to add http://www.securemac.com
- You will find the numerous Appletalk and other various TCP/IP based flaws/insecurites in System 9, 8, and even OSX that you seek and somehow haven't found and don't believe exist.
Cheers...
I used to always ask people who insisted that the classic Mac OS was just as vulnerable as any other OS how they would get around the fact that the classic Mac OS had no command line.
That usually stopped them fast in their tracks.
They other thing that I asked anyone who was questioning the security of the classic Mac OS was to name one instance -- just one -- where a classic Mac OS machine had been cracked.
First off, there was a debug level command line, just because you may not have been bright enough to pop it open, does not mean it was not there. Additionally, what does a command line have to do with security? Telneting into supercomputers was VERY secure even in the 80s, which is a COMMAND LINE interface.
Secondly, there are MANY instances where Classic Macs were hacked into or compromised. Since the majority of the Mac life there was not an 'internet' that it could be hacked into from the outside world does not mean that local systems, and systems on LANs and even via AppleTalk were not compromised.
I worked for a technical consulting company that was a major provider for the 'graphic design' industry in the late 80s, early 90s. All too often people had their work lost or maliciously destroyed because of the inherent lack of ANY security at the console or LAN level on Macs.
Even Windows for Workgroups 3.1 had more security for local and LAN access, and it was also completely insecure as an OS. Just like the Mac System software, it was NEVER designed to be in a major networked system.
However products like NT, Solaris, Novell, many *nixs, and other major OSes during this same time period were built around a 'security model'.
I feel sorry for people that get caught in the 'old way' is the best, line of thinking.
believe the facts speak for themselves NetAvenger. Or can you cite ** just 1 instance where the Mac Pre X/Webstar combo WAS hacked ** (other than the instance above which involved a third party plug-in exploit, 8 mth duration, and a 100,000SK incentive). By all means do identify any other os/server combo that stacks up in this context. I'm sure your learned perspective will prove quite enlightening. Cheers.
I stand by my comments.
This is not about how many times someone can prove a Mac running System 9 or earlier was compromised.
The point is, Mac System 9 and earlier software is inherently insecure, has almost NO modern OS features, and is a poor choice for a server environment.
You would be better off running a Win3.1 box with a TCP/IP stack. It would be just as secure, as it also would not be a major target, and technologically, it is JUST as advanced as the core structure of Mac System Software prior to OSX. Period.
Mac System 9 and under have no file system level security, no kernel level security, and are seriously flawed in terms of features they offer for a server environment, and in many ways, even a desktop environment considering that technically it has a poor memory management system, an aged monolithic messaging cue, and is an application dependant cooperative multi-tasking OS. JUST LIKE WINDOWS 3.1 which was released in 1992.
OSX is light years ahead of Mac System software, just like Linux, FreeBSD, and WindowsNT which have true security built into the core/kernels of the OS. Also take into account that these non-Apple OSes where doing this stuff for over 10 years before Apple finally got around to giving us OSX. I remember the great IBM/Apple announcements and idea of having a real pre-emptive multitasking OS with security on Macs available in 1993 as planned. I was even eager for it to happen. It didn't. Microsoft Built NT, and the other *nix variants for the Intel Platform took off providing the features Apple (the great innovator *cough*) was supposed to have given us over 10 years ago.
And yet you are here trying to tell the world that Mac System 9 is better? I am truly amazed that you are so lost in your hyperbole.
Your logic amazes me. You are saying that just because Mac System 9 was 'seldom' hacked that it is superior. Using your logic, we should be using CPM, or DOS, or Win3.1 as they were never a major target and never hacked into either.
You are out of your league here; try some place other than Slashdot to bloviate about the wonders of Mac System 9 Software. Most people here know the history and the architectures of the major OSes, in addition to the technological shortcomings of Mac OS 9 - especially in a server capacity.
Geesh...
First off, the security you state is way out of whack.
Secondly, who would trust a server that has poor memory management (80s technology), no modern OS features like: Journalled FS, Pre-Emptive Multitasking, Ability to use Large Partitions, Multiple CPUs, More than 2GB of RAM, Etc, Etc, Etc...
Give us a break, Mac OS9 is technically at the same level as Windows 3.1. Hence why Windows3.1 is not a high risk OS for the internet, there is not much in it to hack short of the Winsock add-on.
So we should all go back to DOS, Win3.1, and MacOS9, so that we don't get hacked. Give me a break...
With modern advances, there are security caveats, but limiting yourself to antiquated technology to avoid them is just ridiculous.
Secondly, do you even comprehend the performance differences between an OS like Mac System9 and *nix, NT or other platforms? There is a massive margin, and System9 falls quite short - especially as a Server.
If Mac System9 can't even play an MP3 in the background without skipping while you are working, how do you think it is going to handle 100,000 hits as a Server?
Take a computer class or spend your time spreading your ignorance to a group that doesn't know better.
But this is exactly how 'Automatic Updates' are set in WindowsXP by default unless their computer nerd friends turn them off in 'fear' of Microsoft.
Which has been the case I have found more than anything. Some 'Guru' or other OS Zealot that hates Microsoft screws with their friends computer and turns off the 'Automatic Update' Notification service.
Just leave it alone people and let it work, especially for your novice friends. Even if you think Microsoft is Satan himself and going to eat your children.
Geesh.