Posts by jwenting

Worse than pointless. Active support for your nation's enemies is treason. Treason in times of war is typically dealt with rather quickly and decisively. The fact that the US isn't doing that at this time shows they've become weak and are an easy target (at least that's how the terrorists and their symphatisers like this woman will argue).

Her actions also sully the memory of her son in the extreme. That's not the act of a mother, it's the act of a traitor.

Why should we help someone who doesn't even take the time to write out his sentences?
If you're that lazy you don't deserve help.

crossposted in 2 other forums.

crosspost.

hmm, seems some school has assigned a class full of kids this assignment. I see it popping up on several forums all over the net, and all at about the same time.
All the same question, and pretty much the same attitude ("do my work for me").

yes, that's serialization.
It works if you take some basic precautions (like making sure every field in your class is itself Serializable or transient).

Of course you should mark things that rely on external data as being transient so they don't get serialized as those external resources likely will not exist or be different when you deserialize it again.

Some more ideas (or rather the reverse, how to turn a Mac into something useful...) http://www.applefritter.com/image/tid/114
http://www.jagshouse.com/macquarium.html

Nice photoshopping.

Now without Photoshop: http://www.usefilm.com/image/894242.html

One thing you should have learned by now is how to think for yourself, and that includes coming up with ideas for projects.

Richard, could it be the XML serialiser isn't correctly serialising the graphical (thus binary) data, leading to your document being corrupted?
I've not used the system, but seeing as XML was never meant for binary data that is a likely scenario.

Or better, use RandomAccessFile and just write the data you want at the place you want it selectively.
Takes some getting used to as you're no longer dealing with Writers but Streams but well worth is.
Typical sequence:
- move the file pointer
- create a buffer of data to write to the file
- write the buffer

Maybe so, but you can't rely on anything that's not from an official statement to be correct.
There's too many rumours out there, and most of them are based either on wishful thinking, misinterpretation (deliberate or not) of older official statements, or just made up out of malice.
Remember the "Microsoft is changing DOS to make it imcompatible with Lotus 1-2-3" myths from the 1980s and early 1990s? Many people believed that, even after they got that new DOS version and 1-2-3 still worked (they just thought Microsoft had failed to do what they wanted to). In fact Microsoft went out of their way to make sure 1-2-3 WOULD work, as it was the main application used by many of their own customers and thus having it work was vital for their sales...

uh, not just distributed systems and in distributed systems too it depends on the design of the system :)

The Interface defines your public interface. Handy when you have multiple implementations. Typically you don't want any class using your classes (instead of creating instances of them) to have to care about what actual implementation type they're dealing with so you write those classes against the Interface instead.
You can then send them something else completely as long as it implements that Interface and it'll work.

Except when you think like that and think that you can do the same things you did in C++ in Java and always get the same result (or even get them to compile)...

you're the one who's not paying attention. You might be a little less agressive when you're not getting an answer you understand, your behaviour is not making you any friends.

Your failure to get even the basic commandline syntax correct is your problem, not ours. People give you hints as to what to do, even spell it out for you, and still you don't get it.
Apparently you're not even trying, so why should we try to help you any further?

Hello,

It has been my experience in the business world that the nVidia drivers are quirky and unreliable, whereas the ATI materials have been stable and trustworthy. I have seen a lot of nVidia cards cause some monitor problems, and unstable graphics. The business environment that I am referring to comes from publishing and CAD applications

My personal experience with these brands is almost diametrically opposed to yours :)
Unstable (to the point of rendering the card inoperative until the driver is replaced) ATI drivers and rock solid NVidia drivers.
Of course if you go out and install leaked and tweaked beta drivers (which many people seem to do because they promise higher framerates) you're on your own, but I've experienced situations in which there was NO release driver for an ATI card that worked.

Worst situation: An ATI card for which the release driver failed to put the card into any graphics mode (do they test those things?).
The only driver that did work was a beta driver for a completely different chipset (beats me, but it worked).

Probably failed to activate the AGP slot after deactivating the onboard chipset.

who is "u"?
who is "i"?

Use real English in writing and you might understand what others write as well.

try your favourite search engine.
wikipedia also has an excellent article on prime numbers complete with explanations of the most popular algorithms.
None are particularly difficult to implement and I'm pretty sure your teacher wants you to do your own homework rather than submit something written by someone else.
You're there to learn something after all.

Woooow, a BIT of patience may be appropriate.
You must be the most impatient poster I've ever seen, complaining about not getting an answer after less than a minute.

We're not here to proofread your programs. What is the output you're getting, and what were you expecting to get?

Use code tags, I (and many others) won't read any code that's not formated in a way that makes it readable.

Or try the card in another PCI slot if the machine doesn't even see it.
Could be you have a bad slot, happened to me once on an old machine.

Seems the charging circuit's fried, not the polarity protection on the mains inlet.
Either the charging circuit has its own protection (likely) or it's a goner and needs replacing (which in a laptop could mean replacing the mobo depending on the design).

She'd hardly notice if she's a good mum and does real cooking :cheesy:

Site doesn't exist. It's a domain squatter. Not tried any of their "suggestions", the URLs look suspiciously like sites that try to do naughty things like install CWS.

I usually take a 1 Tesla magnet, sure to get rid of anything on that drive :)

More seriously, just boot from another medium and remove the Windows or WinNT directory and your OS is effectively hosed.
Disk might still try to boot but will do you little good.

More generally, repartitioning the disk will hose anything on it just as does formating.

No, no "major corporation" is behind this.
It starts as a joke and then someone takes it seriously and starts sending it to several hundred people.
When that happens a few times it gains critical mass.

There is NO way for anyone to know who is getting that message unless they themselves sent it to you personally (and even then there's no way to know if you got it for sure).

it means

if (x > rect.x) {
  left = x;
} else {
  left = rect.x;
}

Yea, people would simply change to ICQ, AIM, yahoo instant messenger if MSN charged for there messenger.

The same hoax has been doing the rounds about those as well periodically :mrgreen:

Elsewhere (Sun's forums in fact) someone came to the conclusion that this whole thing only affects the (SAL) software accelleration layer and not the HAL (hardware layer).
Sounds perfectly reasonable to me, and it's what I conclude as well on rereading the info Microsoft puts forward.
Avalon will mainly replace GDI+, which is something most people programming DX or OpenGL avoid anyway because it's slow.
On top of that Avalon seems to offer the capability of being used as a layer on top of DX AND (and that's the thing people are ranting about) of routing DX and OpenGL calls through Avalon (which will then translate them to something else for specialised hardware I guess).
Microsoft states clearly that doing that will incur a performance penalty (which is hardly unreasonable to expect as you're moving operations from the videohardware into the CPU which not only isn't optimised for those operations but also has other things to do).

I'm not talking university grade math here, but the nitty gritty details of the language specification.

If you can memorise the JLS and have a logical mind you're quite a ways towards being able to pass the exam.
The 5.0 exam is designed to make that harder by introducing more tests of practical abilities (which though existent in the 1.4 exam aren't large enough in volume to seriously degrade the performance of those who are just cramming).

The main thing degrading the 1.4 exam though is the low passing score which Sun introduced to make the exam more appealing to relatively inexperienced programmers (and thus draw more people towards the language).
That has backfired as now more experienced people complain that the exam is too easy but Sun can't go and change the existing exam requirements as that would be unfair towards new candidates (or make certified people loose their certification if the new standards were applied retroactively against the existing base of certified people).

any old stereos around the house?
Maybe an old VCR you no longer use?

That microwave looks ready for replacement.

come to think, maybe I shouldn't have returned that broken down microwave to the store when buying a new one a few months ago...
Remove the glass door and put a 15" flatscreen in its place, with a DVD writer mounted next to it...
Everything else vertically in the electronics bay, should be more than enough room.

hmm, whatever :)

Even if OpenGL hardware support in the first generation cards won't be on a par with Avalon support in those cards I think that's only a matter of time.
OpenGL is platform independent, it will still be possible to bypass Avalon completely (just as it is now possible to bypass GDI+ completely, which is exactly what both OpenGL and DirectX are doing after all).
It might not be the officially recommended way of doing business with the graphics subsystem, but then there's nothing new there either :)

The sky isn't falling, software writers just need to adept to a new toolkit like we have so often in the past.
I can still remember the same FUD and ranting when Windows 95 came around and it was not possible to write directly to the hardware anymore (yes, I've been around that long, I wrote a videodriver for a 32 bit DOS extender around that time in Assembler and Pascal).
Of course in those days it was mainly tech auditors in magazines as there was no large web of uniniated people all jumping in with doomsday claims that "Microsoft is trying to destroy the competition" which is what the entire "Microsoft is deliberately making OpenGL a lot slower than DirectX" thing reads like.

Highly detailed technical questions about often obscure parts of the language specification.

Turn your brain into a perfect language parser and you're halfway there.

They're not out to trick you into giving the wrong answer but the ARE out to get you to give the wrong answer if you're not sure what the correct answer is (for example, the correct answer may well be the one that SEEMS the least likely to be correct).

Did you get that information from Microsoft or from some anti-Microsoft rant site?

Your quotes are quite different in style from anything Microsoft themselves write. Like I said, trust the source as only Microsoft know what they're planning for Vista.
Maybe someone has access to an alpha release leaked at some point in which OpenGL support is lacking and that's what sparked the rants, but that's not to say that that's the final version...

People, unless and until you see an announcement from Microsoft themselves don't believe anything you read...
There have been many rumours about Vista/Longhorn about how evil Microsoft is (I think I even saw one stating it would only support Microsoft software and hardware and that everything else would no longer work, a rumour also floating the net when XP was due for release).

The ONLY statement on OpenGL I could find on Microsoft.com has to do with running an Avalon (the new graphics system) application on top of OpenGL (so NOT running only OpenGL):

You can host an Hwnd inside an Avalon application. However, having Avalon render on top of OpenGL-based content which is accelerated through an ICD is likely to cause flicker and it is going to be hard to have things to register properly temporally, given the two independent rendering pipelines

Seems reasonable as you're effectively having every calculation done twice...

Existing applications will continue to run without requiring modification. In order to use new Avalon features, developers or designers will need to write to the Avalon classes.

Remember that many graphics cards (especially the better ones, not talking about the $25 ones you get in supermarket PCs) do OpenGL natively in hardware it shouldn't really matter.
As always, the person creating the application has to know what he's doing.

Avalon also does not replace DirectX or sit between DirectX and the hardware:

No: Avalon is built on top of DirectX …

It's a well known and rather old hoax.
Typically it's accompanied by "send this message to all your friends, MSN will not stay free unless this message is sent to 50 million people in the next 3 months" or something like that.
ALL such things are hoaxes.

Or write a Java program that runs constantly and can communicate in some way (probably through SOAP) with your PHP application and use that as middleware.

But indeed, moving to a JSP/Servlet environment is the best option by far.

Richard, there are indeed a LOT of certifications like that.
MCSE is one of them, in fact the main culprit in killing the good name of certifications overall was the NT4 MCSE exam which was a disgrace.
This got so bad that many companies (including the one I worked for at the time) rejected applications from people with MCSE certifications on general principle unless they could also show several years of realworld experience and came with excellent references.

The SCJP 1.4 exam is indeed too easy to get without realworld experience, I won't disagree with that.
But it's a good deal harder than is MCSE and the 5.0 exam is harder still.

SCJP alone indeed isn't enough to make a good Java programmer, but what it DOES say is that the person knows his basics and knows them well (unless the person is extremely good at cramming for an exam and then extremely lucky at answering obscure questions).

Studying for that exam (and now the SCJD exam which I'm currently working on and is a LOT tougher) taught me a lot about the language I didn't know though, and that despite having 7 years experience of writing Java as a professional fulltime.
That's the real value of this exam.

And yes, I've seen people with certifications that are a disgrace to the profession as well as brilliant people that never saw the inside of a university.
But overall the truth is …

Rice is an important staple food, it should get more attention.

Well said Zeroth. Microsoft and IBM (IBM did so unintentionally, but they did it) together comoditised the computer (Microsoft the software, IBM the hardware) to what it is today.
I too remember the early days, when you had the choice between a ZX81 with less power than a scientific programmable calculator or a mini costing a hundred grand and needing a large airconditioning unit (and of course eating loads of electricity).

Then IBM came with their PC and everything changed. Suddenly the hardware became somewhat affordable.
Then Microsoft changed the price point of office applications from thousands per application to a few hundred for the entire suite and the vision of "a computer on every desk" started to become economically feasible (though it would take people like Dell to really lower the hardware prices from thousands of dollars to hundreds).

Without the vision of IBM of comoditised hardware and Microsoft's idea of comoditised software, lowering profit margins to get income out of bulk instead of sheer profit on a few units, we'd still be connecting to mainframes using dumb terminals over 2400 baud telecouplers.

Smoking (or rather nicotine) IS physically addictive but to a lesser degree than the mental craving indeed.
That's why some people (those with strong minds) who smoke only very occasionally don't get seriously addicted.

Of course smoking isn't the only cause of lung cancer, but it is a main cause.

In your scenario you'd still have to read in the entire stream to create your object before you can start to serialize it.
Unless you can read in the input in small bits and flush your output before reading the next bit (or at least periodically) you're going to run out of memory at some point.

requesting garbage collection periodically can delay that a bit but it won't stop it (as garbage collection is guaranteed to run before the JVM runs out of memory, which is the only guarantee you're getting with the gc, but in your case as it stands now it'll do you little good as you're holding on to all that memory so it can't be reclaimed).

doesn't matter, having a 256MB heap won't help you contain a 600MB file in memory :)
You don't NEED to retain all that data in memory so the real solution is to not retain the data at all.

NPH, DO NOT do peoples' homework for them. It makes sure these people never learn a thing (or rather that all they learn is that there's always a sucker stupid enough to do their work for them).

Are you by any means trying to run the java source file directly?
If you're commanding "java HelloWorld.java" you're actually trying to execute a class called 'java' in a package called 'HelloWorld'.
Compile the class using javac and then execute it using 'java HelloWorld' (while making sure your classpath is correctly set in either case).

Try flushing the buffer after every write.
The only thing I can think of that's eating up memory here is the outputbuffer. It may not be flushing automatically, causing the JVM to attempt to map the entire output into RAM instead of writing it directly to disk (which you seem to be attempting to do).

Think about what you want to do and how you'd do that if you were to do it with buckets full of marbles.
Write that down, then translate that into code.

Your entire algorithm is wrong, and your description of what you want to do is so vague it seems to me you don't even know yourself what you're trying to achieve here.

Try GMax (from the makers of 3DS Max), I think it can import ACAD directly (or through a plugin).
I notice it's now also an Autodesk product...
While not a renderer itself you can use it to create models for several popular games which can act as a renderer for you.

Well, you never change searchValue inside the loop so if it's not the same as your loop counter you're entering an eternal loop.

Maybe a bit late, but here's a new article on java.net that does what you're looking for:
http://weblogs.java.net/blog/kirillcool/archive/2005/08/reflection_and.html

The last part is irrelevant as it deals with some obscure XML generation from Java objects, the first part deals with dynamic compilation and classloading.