A flaw in Sun Microsystems’s plug-in for running Java on a variety of browsers and operating systems could allow a virus to spread through Microsoft Windows and Linux PCs.
Java flaw could lead to Windows, Linux attacks
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Cool! Cross-platform viruses. Now that’s what I call the march of progress!
Erik
Another reason to open up java: community bugfixing!
The OSS nuts come out to play. I think Java’s track record on security flaws has been good so far, but maybe I’m in a cave. They gave us Solaris, let them rest for a while.
Other than community bugfixing, what reasons are there? Karma on slashdot doesn’t count.
With a carfully designed SELinux policy for Mozilla and the java plugin this problem can be sandboxed, no matter what.
From the article:
The vulnerability, found by Finnish security researcher Jouko Pynnonen in June, was patched last month by Sun, but its details were not made public until Tuesday. Security information provider Secunia posted information about the flaw in an advisory that rated it a “highly critical” threat.
My faith in Sun is waning.
Whoops, June, “highly critical” threat should be bold.
Wierd, it looked fine before I submitted…
“They gave us Solaris, let them rest for a while.”
press releases dont count.show me the license and code
5.0 isn’t affected, only 1.3 and 1.4.
You should visit an insurance company or webservice at some point.
The desktop is a fraction of the market.
“Why would anyone use java?
I just don’t see the point in recompiling your code every time you run it.”
isnt the point of jave “compile once, run everywhere”?
maybe you haven’t read about the benchmarks where java’s JIT/dynamic compilation yields faster results than natively compiled stuff in C++?
read up and find out: http://www.idiom.com/~zilla/Computer/javaCbenchmark.html
“On Intel they found that the Java performance was very reasonable compared to C (e.g, 20% slower)”
20% is not acceptable.
that depends, a game or other app where you rely on speed of computation for stuff like rendering and so on it may have a impact. but for a desktop app like a spreadsheet it will not matter much as most of its time is spent waiting for user input.
http://www.pestpatrol.com/pestinfo/j/java_class_loader_hole_recentl…
I was cruising for porn and java loaded up on a website but my free Antivirus Antivir caught it. Not bad.
That’s it. Guess I am a troll.
“”On Intel they found that the Java performance was very reasonable compared to C (e.g, 20% slower)”
20% is not acceptable.”
Its mostly some scientific functions that are a bit slow. Unless you don’t do some kind of very comutationally intensive science calculations, you should have no problems. If you do that you probably use Fortran anyway. When did you last do a sin calculation in one of your programs?
You should also have in mind that the difference in speed varies much more than 20% depending on the skill of the programmer. A mediocre programmer may easily write 10 times slower code than a skilled one.
I would also say that 20% slower than native languaes like C/C++ is quite acceptable considering that you develop an application in java 3 times as faster than you do in C++.
If you spend the money saved on developer saleries using java instead of C++ on faster hardware the end result will be that java will give you the fastest end result. Not to mention that you may have benefits from being able to hit the market faster.
This is why java is so popular in fortune 500 while native languages are preferred by hobbyists to whome hardware is more expensive than developer time.
Why would anyone use java?
I just don’t see the point in recompiling your code every time you run it.
Well, People use perl, that compile from source code to binary every time it runs. Given perls long time popularity I don’t see this as a problem for Java. At is at least semi precompiled for the java VM. Besides unlike most other languages optimizes itself at runtime to fit the actual load.
Java, with JIT, is quite fast – as long as you don’t divide. Floating point divisions take forever, compared to anything gcc produces on x86 (and gcc is infamously bad at optimizing code).
Also, Swing was unusably slow the last time I looked at it.
So as long as you don’t use Swing and don’t do many divisions, there isn’t much performance difference between c++ and java. In a few cases, the JIT might actually make java perform better…
– I’m tired of hearing of open source java stuff, don’t like it then don’t use it. Let sun persue their plan of having more OPEN community involvement as stated by many execs.. see how things turn out.
– Performance is rather different with Java–it depends on various factors. For instance when loading some games on a freebsd machine it may seem slow.. while using a java based search engine defeats C. If you take a look at the improvements sun has made to java you will relize that is a good investment. Speed in Java has increased rapidly over the years if you have not noticed and I don’t see it decreasing but improving. Besides, java isn’t tied to a specific hardware archetecture so it’s assumed to run slower… but the speed thing is really improving and i dont see a reason to abandond it.
– I thought this flaw was around for a very long time. Was I the only one that noticed this? I assumed it was the nature of java applets. Great to see sun is improving it though, can’t wait for the new community process.
I would also say that 20% slower than native languaes like C/C++ is quite acceptable considering that you develop an application in java 3 times as faster than you do in C++.
for some of the most illogical reasoning ever to cross the pages of osnews. Congrats. Please choose another profession.
It’s funny how you guys always end up calling Java slow – and no reason to ever use it. Fine – but stop comming back to each Java thread and hollering it!
Java works fine for many tasks – so does Perl, ASM, C, VB.NET etc. etc. ad nausea
Java, with JIT, is quite fast – as long as you don’t divide. Floating point divisions take forever, compared to anything gcc produces on x86 (and gcc is infamously bad at optimizing code).
Yeah right. Given that the JIT emits native instructions, there is no reason why divide should be slow on Java. In my experience, that isn’t the case either. Java isn’t particularly slow when it comes to floating-point divide.
Also, Swing was unusably slow the last time I looked at it.
Sure. And how long ago was this, and what do you mean by unusably slow? Swing on my machine which is a Powerbook runs like a charm. Guess YMMV.
this flam question was posted on planetgnome by fernando
errera
“Why Red Hat is not making any Java-GNOME application? If they want to promote java as a platform for the linux desktop and bindings are stable, why not use them for their custom apps instead of pygtk?. Novell (and lot of more people) are doing cool .NET apps for Linux Desktops.”
It is slow because there are differences in how different architectures handle floating point (exceptions, precision etc etc). Java produces identical results in all arhictectures so it has to sacrifice performance to achieve that.
I am not expert in the subject so I could be wrong. Feel free to correct me or add something
It is slow because there are differences in how different architectures handle floating point (exceptions, precision etc etc). Java produces identical results in all arhictectures so it has to sacrifice performance to achieve that.
I am not expert in the subject so I could be wrong. Feel free to correct me or add something
Have you tried it? If you did, you’ll know that what you’ve posted isn’t true. The floating point operations all get JITed into the native FPU code, which is why it’s nearly as fast as native C code.
I’ve tried this on x86 and PPC, running operating systems like Linux, Windows, and OS X. The only time when precision is really upheld (i.e. strict math) is in the trigonometry functions which is why Java gets absolutely trounced by every other language in existence when it comes to trig.
I would also say that 20% slower than native languaes like C/C++ is quite acceptable considering that you develop an application in java 3 times as faster than you do in C++.
This is the most backwards statement I have ever seen in my life. A program only needs to be written once, but ran infinite times. The end result, not creation, is what must be optimised.
If speed were the only factor and time the ability to maintain the software were unimportant then everyone would be programming in assembler or raw machine code.
There are always trade-offs and speed versus development time and maintenance factors will always be considered. In large businesses with complex business rules and constant changes Java is superior in all but the most peformance sensitive areas.
> I would also say that 20% slower than native languaes like C/C++ is quite acceptable considering that you develop an application in java 3 times as faster than you do in C++.
>for some of the most illogical reasoning ever to cross the pages of osnews. Congrats. Please choose another profession.
still in school eh?