The real thing is: why wouldn't you expect this industry to have maintenance? Pretty much every other industry does, and many product companies also sell a 'verified' maintenance service.
For the same reason I can copy a file till my hard drive crashes or the universe ends. In fact, even when the hard drive crashes, I can load a backup onto a similar hard drive and copy the same file, using the same copy program, until *that* hard drive crashes. My point is, computer software shouldn't "wear out" over time. There are no physically moving parts.
Cars, washers, dryers and air conditioners all have moving parts that are susceptable to friction, heat and general wear. Software is more like a book. Once a book is printed, it will stay printed, completely readable until it's purposely destroyed or the papers physically degrades away in thousands of years.
After thinking about it for a second, it's in the Chinese governement's best interest to know who's using the Internet. If you connect to a secure wireless network, you need to have a key, so someone has to know who you are. You can't arrest dissidents if any idiot with a wireless card can connect to the Internet anonymously.
I'm not saying I agree with how they could use it, but it is in their best interest to have secure wireless.
I dislike the idea of backdoors in general, but in this situation, it probably doesn't matter too much. It's just as likely (perhaps even more) that Intel's protocol has backdoors and it's almost guaranteed that, if it's there, the US government, the Chinese government, and any other government who asks will get access to it. If WAPI has a back door, the only people with access will be the Chinese government, and it seems unlikely they'll be in a hurry to give out access to it. At least with the Chinese government you know that they're spying, and you know what they're looking for. It's unfortunate for the Chinese people, but they can take it up with the Chinese government.
Again, I'm not saying I agree with China's policies, but WAPI might not be as bad as all the people screaming about it simply because it's from China.
I'm not sure I agree with most of what you say, but you're right about economics being a very useful thing to study.
I majored in CS and got a minor in econ (and in math), and I use the the econ stuff as much, if not more than the CS stuff. I don't know if minoring in it was entirely worthwhile, but going through intermediate microeconomics and intermediate macroeconomics was possibly the smartest thing I did while in college.
I couldn't read your post, because the massive wall of text hurt my eyes, but I'm guessing you didn't mention all the English classes you're taking?
Holy shit! You went off on that entire little diatribe without using a single <p>. That has to be some kind of record: "Longest post by somebody too stupid to break their post into paragraphs." For fuck's sake, man! If you want people to read what you write, learn to use paragraphs.
And if enough people do it you'll se more and more sites switching to subscription models and ditching ads completely. Of course then you won't have the option of shoplifting content by blocking ads.
It would be nice, but I don't think it will happen. If a website consistently offers useful content, that I can't get anywhere else, I'ld gladly subscribe if the price isn't outrageous. I'm sure many people wouldn't like it, but at the same time, I believe most people don't mind paying for information that's useful to them. If someone wants to make money off of the information they have, they can charge me. Until then, I don't owe them shit, "ad-views" or otherwise.
The only people who really have to fear a subscription based internet are people making money off of worthless content hidden by tons of ads.
And while that may be the case, I don't see the internet going to the subscription model any time soon. It'd have to happen all at once. I don't mind paying for decent info, but I'm not going to pay for it if I can get similar content free on another site.
The appeal to me is precisely that I don't have to watch the ads if I don't want to.
The appeal to me is that these ads will probably get blocked by my existing filters, just like the rest of Google's ads. Let them show blinking pop-ups with sound and Flash for all I care, I'll never see them. Sometimes I almost forget that websites even have ads.
If you are compiling to bytecodes (*.class files), the "compiler" doesn't matter much. If you are comparing the runtimes of JVMs, which produce the binary executable ostensibly after the program starts, then we can talk. As for comparing GCJ's direct compilation to Sun Hotspot engine, GCJ loses almost every time and usually by a noticeable margin. GCJ at this point should be seen for its convenience in packaging, not its raw speed. (I still like GCJ, but let us call a spade a spade.)
You missed my (poorly stated) point. I wasn't arguing for or against one compiler or language, but trying to emphasize that the languages themselves are separate from their implementations. It's quite possible to make a C++ compiler that compiles to JVM bytecode or a Java compiler that compiles to native code.
First of all, no it doesn't load every jar into memory. It scans every jar for the classes and resources available, but it does not actually load their entire contents just for "Hello World."
No contest on this one. I mis-remembered this article. I'm not sure if you're right, but I was definitely wrong.
From a pure math point of view you are correct. In the real world you're quite wrong. Except maybe (MAYBE!) for the top assembly coders in the world, a non-trivial application will be faster written in C than assembly. Not only will the assembly language version take orders of magnitude longer to write than a C version, humans make mistakes. While one or twenty particular sections will be optimized ahead of the equivalent C version -- commonly through the exploitation of side effects -- what about ten or twenty thousand code segments? A C compiler will do a damn fine job and as good a job with the hundred thousandth code segment as it does with the second. That's the gain. The same is becoming (has become?) true of Java when compared to C.
You're right, it would take a huge amount of time to do that stuff in assembly language, or even C. But to state that something simply can't be done in assembly or C is wrong. It may not be done easily, but it can definitely be done. That's the point I was trying to make.
So yes, anything you do in a JVM can be done in assembly. However I am not aware of any non-trivial assembly projects that are completely rewritten to run faster on newer Athlon processors and completely rewritten again to run faster on Intel Xeons.
Of course not. Any gain in execution speed would be lost to the huge development time. Any sane person would write it in a higher level language.
On a side note, any such C re-compiler would want the original C source code for reference. Once you compile down to machine code, you lose almost all context as to what the programmer intended. The source code contains context. So perhaps you would create a meta language that stored the program structure? Maybe you could call them bytecodes? Then perhaps your C re-compiler would read these bytecodes in and optimize them for the platform and environment you're running in?
That's exactly my point. I didn't say it was a good idea, I said it's possible in C. But nobody does it because there are better ways to speed up code.
No, C++ is a multi-paradigm language. It supports OO as well as generic programming as well as procedural as well as template meta programming as well as... Even Bjarne Stroustrop calls C++ a multi-paradigm language, not OO. Where have you been? Hell, many top C++ language architects actively disdain OO usage in C++ like the original author of the STL, the hallmark of C++ generics.
Yes, it's multi-paradigm, and one of those paradigms happens to be object oriented. Whether it's good to use it as such is irrelevant. The OP made the blanket statement that OO
Except for certain math co-processor functions used for 3-d graphics, Java is rarely less than 50% as fast as C and often significantly faster.
Is that Java compiled with gcj? With javac? Is it C compiled with gcc? Visual C++? Intel C compiler? egcs? Digital Mars? One language can't be "faster" than another. Only one compiler produces faster executables than another.
If that seems counter-intuitive, consider that java compiles into machine code (at run time) that is run time-optimized.
For instance, if it notices that it's frequently calling a series of functions with a certain set of values and those functions don't change state, it could calculate the return and skip calling the function altogether.
Does it? What if the state of the object can affect the return value?
Although that may not be an exact representation of what's going on, some of the optimizations are much more tricky, and they are across the board.
Oh, no, I guess it doesn't actually do that. I think you meant to say "I pulled that example out of my ass, but Sun's JVM does some optimization at run-time."
These are things that C and C++ (and even assembly) physically can't compete with.
Bullshit. Assembly language, almost by definition, can do anything that a program on a particular architecture can do. If it can't be done in assembly language, it can't be done.
What do you think Sun's java binary is written in? It's C (or C++, I'm too lazy to check). If Sun can write a program in C that implements those optimizations, then it's possible to write a C program that performs the same optimizations on itself.
Here's another trick, one C++ can't compete with. In an OO language, you often allocate thousands of objects a second--that's extremely typical. In C++, it is necessary to spend execution time to physically recover the space of each object that has been allocated. In Java, the management process has gotten to the point where all it does is reserve the few objects it wants to keep then wipe the rest of the objects at once (like returning from a call, you recover the entire stack at once but for permanent heap memory instead of stack memory)
So...many...things...wrong... First, C++ *IS* object oriented. Second, I can think of several object oriented languages which do NOT lazily allocate memory. Third, do you even know for a fact that Java does this? Fourth, the fact that all of your OO code needs to create thousands of objects a second seems a little unusual. Fifth, "extremely typical" does not make sense. Sixth, there are very few things that are computationally possible, that C++ can't do. I can make a C++ program with lazy allocation in about 5 minutes. Just because you don't know how to do something in C++ doesn't mean it can't be done.
Or are you talking about the physical size of the JRE? It's 13mb, I'm not sure how much it expands to but considering that it's mostly a jar (compressed already), it should be much more efficient that C/C++ libraries. On top of that, since none of it's statically linked, each nontrivial Java app placed on the laptop should be smaller than the equivalent C/C++ app (Especially since byte code was invented to reduce size first--it was used in Excel to reduce code size long before Java was invented)
On my machine, the Java run-time alone, "rt.jar", is a 38 MB file. The GNU C++ library, "libstdc++.a", is a 7.6 MB file. Both of them are compressed, and both are loaded dynamically.
BUT, Sun's Java loads *every* jar file on the classpath entirely into shared memory. The *.jar files in my jre/lib directory are 51 MB by themselves. It's shared memory, so 20 JVMs can use a single copy, but it also means a single "Hello World" needs to load at least 51 MB, just to run. Unless there
I always forget about Project Gutenberg. I think it'd be nice if some popular current authors would place some of their books there for free. I don't know how copyrights work in the publishing world, so I don't know if they're even allowed to, but it would be nice. Maybe not right after a book is published, but maybe a year or two after it goes on sale. That way most people who really want the book would already have bought it, and the price would be down enough so that it would be cheaper to buy it than use ink to print it at home.
That wouldn't exactly work with technical and reference books, because their prices are usually relatively high, but maybe putting them up without figures or something would work.
I guess there are people who would read entire books online, but I think a lot (most?) people would prefer to read things over a few pages on printed paper.
It'd be nice, but I doubt it'll ever happen. Publishing companies are probably too paranoid about lost profit.
Sigh. Your post is so stupid it almost physically hurts me to think about it.
The issue is you refuse to install one of the main web technologies to your browser. The site IS cross-browser compatible; Flash has penetrated over 90% of the userbase of both browsers.
I can think of about 15 web browsers, which two would you be talking about?
I think you're missing the GP's point. The "Remember Segregation" page shows a shit brown page of nothing when Flash is turned off. The problem isn't tasteful use of Flash. The problem is making the whole page in Flash, and not showing up at all if Flash is turned off. Not even a "You need Flash" warning. The web is based on HTML and HTTP. Not plugins. So it would make sense that the winners of an award for websites would at least show something if you don't have the correct plugins installed.
Internet Explorer has a Word Document plugin that lets you view Word documents right inside IE. Would you have been so quick to spout this nonsense if most of the sites needed to be viewed on Windows in IE with the Word plugin turned on? It's a plugin 85% of people have...
You're like a video collector... who wants everything to be in one codec. That just ain't gonna happen. The more codecs you install, the more videos you can play (and no, VLC doesn't play everything...).
Not only is that analogy stupid, it doesn't even apply here. This is more like someone writing a C program in Java. I'm sure those 100% Flash pages are great, but they kind of miss the point.
I don't understand why you're oligated to fix this person's computer.
Are they paying you? Raise the price. If that doesn't work, raise it again. Problem solved.
If they're not paying you, tell them to fuck off. It's not your problem that they're too lazy to figure it out themselves or too cheap to pay someone else to do it.
For starters, mplayer.c is 4000 lines long. Apparently only one man really knows what's going on in there, and he's not taking a look at it. Making sense of it was beyond what my cursory overview of the code could muster. Near as I could tell most of it was written to deal with bugs.
You've got to be kidding? I was skeptical of your post, so I looked at the mplayer source. After 10-15 minutes of looking at mplayer.c, I think I have a fairly good idea of what most of it is doing. There's a lot of stuff for portability, but it's definitly not mostly written to deal with bugs.
The main developers are from eastern europe, I think. They have a pechant for three letter variables, and not a character over. Terse and unreadable code is also preferred. I remember being asked why I dond't compress a three line, readable piece of code into a once liner, line noise version. Comments have long since passed into myth. I sometimes wondered if their compilers supported them.
Yeah, cryptic, three character variable names like "osd_show_percentage", "stream_dump_type", "too_fast_frame_cnt" and "frame_time_remaining". How cryptic! Whatever could those mean?!?
The mplayer system is based on plugins. Written in c code that is hacked to the limit to introduce, insofar as it is possible, object orientation into c. Void pointers abound, and are probably the most common datatype in the respository.
Bullshit. I just checked. mplayer.c has 3 pointers to void, and one pointer to pointer of void. A quick search through some other files found zero void pointers. The code in the loader section does have a few, but it's hardly the most common datatype.
The only part of your post that's even remotely true is "All that said, the program is fantastic." On that we agree. mplayer kicks ass.
Yeah, well it's "uneducated arm-chair-space-exploration critics" like myself who pay NASA's bills. Perhaps if they don't want my uneducated ass to write my congressmen about NASA wasting my money, they should better explain their objectives and why they chose this over other similarly priced options.
Oh, and I like the way you turned my skepticism into my being uneducated. Nice ad-hominem.
I'm not really concerned about destroying the moon. I'm more curious of why they want to do it at all. Isn't the moon covered with craters? Many of which are a lot more than 16 ft deep? Can't they observe the dust and debris around those craters?
Even if they crash it into the deepest crater on the moon, is it really likely to find water by going another 16ft? There's no water in or around craters x/8, x/4, x/2, or x feet deep, so how likely is it at x+16 feet?
After RTFA, it sounds like they're just trying to stir up a lot of material so their probe can fly through it before also crashing. Wouldn't it be more effective to just send a rover down to the area? Look at the Mars rovers... those things have lasted months longer than planned, and have sent back boatloads of data. If this moon probe lasts twice as long as planned, it means the probe takes 30 minutes to crash instead of 15.
Maybe it's just me, but this seems like a strange way to spend $73 million, with little chance of success.
On the other hand, what are the chances the probe won't slam into the moon, and NASA will anounce "HA! Got ya! We only said that in case it failed and did slam into the moon. It's actually a robotic rover!"
We'll for starters, you get developers convincing the biologists that they need Oracle...and it only goes downhill from there.
Is it possible the developers are saying something like "It starts out with the biologists saying they need 30 TB of data available 24/7 with 99.999% uptime and 200-250 concurrent users, and goes downhill from there..."
Simply saying the developers are idiots because they suggest Oracle really doesn't make sense without more context. If more than one group of developers suggest Oracle, they might have a point. Are you sure you're not exagerating the requirements? If you need really high uptime and lots of concurrent users with really, really large amounts of data Oracle is one of the better choices. There's a reason it's so expensive...
Think of it - the kernel code is out there - it boots without any system libraries already being loaded. The printk (kernel print function) doesn't call the os at any point - it has enough knowledge to talk to the hardware directly.
Any real operating system (Windows NT and up, Linux, *BSD,...) prevents you from doing any of the stuff you mentioned. You can't just load a program and start doing low level IO to ports. You can't just bypass the MMU and paging system and write directly to physical memory. You can't just write directly to video memory. You can't just have your program load and start acting like it's the operating system. Any operating system worthy of being called an operating system prevents that. Device drivers would come closest, but they most definitely need system calls.
So, do the same, and add enough functionality to read and write to the various filesystems out there, and you're home - you don't even need to do it in assembler... just #include the appropriate pre-existing header files, and #include any code that you would normally call from a shared library.
Yeah...okay. For the moment, lets pretend it's possible to directly access the disk and filesystem from a user program, without system calls. To be any use at all as a cross platform virus, the program would need access to NTFS, ext2, FAT32, and ReiserFS. Writing filesystem code isn't trivial. I would be very impressed if a single person could implement any one of those filesystems in a reasonable amount of time, and all 4 would be nothing short of impossible. Let's just say that if somebody had the skill to do it, they'd be too busy making buttloads of money to waste their time.
#include the appropriate pre-existing header files, and #include any code that you would normally call from a shared library.
I'd love to hear your explanation on how to do that...
But, the whole thing is pointless. Even if you did manage to write filesystem support for all the required filesystems and were able to #include the code from the shared libraries the operating system would still stop you the instant you tried to read or write directly to the disk.
If this pans out, it's one huge way to give MS the finger...
It's also one way to really piss Microsoft off...
Lucent better have a really water tight case, because Microsoft has tons of patents, tons of cash and tons of lawyers. Until now they haven't done much with the patents, but now they're going to have legions of lawyers combing through every feature of every Lucent app they can get their hands on, looking for anything that even remotely resembles something MS has patented. I don't think half a dozen counter suits over stuff like the "IsNot" operator would be too surprising.
If it's a really blatant case of infringement, MS might settle. Otherwise they're going to pull out all the stops, throw their whole legal department at it, and send a powerful message to everyone else who considers suing them over patents.
I'd love to see Lucent win this, but it's not going to be easy.
The whole article assumes several things... 1. Macro-evolution occurs. A quote: "Our emergence on to the land is one of the more significant rites of passage in our evolutionary history." This quote (and specifically the use of the term "history") shows that he had a pre-conceived belief about what happened, and he was going to interpret whatever evidence he found in that light. As I said in my first post, there's no way that it can be shown that this fossil is a stage of development between two (or more) species. The only thing we know is that this is an animal that we haven't seen before.
That's the whole point of the article, and part of the reason this is so significant. They did calculations, checked with some geologists, made some predictions (or "assumptions" as you want to call them), and looked into them. It's called science.
Do you think the whole scientific community is a bunch of crazy people in lab coats, with weird hair, randomly mixing chemicals, occasionally shouting "Eureka"? He wouldn't have been on Ellesmere Island if he didn't have a good idea of what he might find and how it would show evidence for or against his hypothesis.
As to whether anything actually evolved from this creature isn't relevant because nobody is claiming that. It's more of: "Look at this, a fish with small legs! So at some point fish with small legs existed, as evolution predicted." Not: "Look! This is THE creature we all evolved from."
2. The time ranges that are being used. A quote: "We knew that the rocks on Ellesmere Island offered a glimpse into the right time period." They assumed the time period, and then you say that from the newly discovered animal they could "predict... the timeperiod." How did finding that animal in that area tell anything about the time period?
It's called "Carbon dating". Carbon dating is usually accurate to within a few hundred years. Hell, with a timeframe in the hundreds of millions of years, it'd only have to be accurate to a few tens of thousands of years to be useful.
If you have one fossil from 420,000,000 ± 8000 years ago, another fossil from 350,000,000 ± 8000 years ago, and then find a new fossil from 375,000,000 ± 8000 years ago, you can be pretty sure of the order they occured. That's the main point of the time line.
One more thing...you said in your logic steps: "but still looks a lot like a fish." It seems like quite a logic leap to say that one animal that resembles some other animal must have been a descendent of that second animal.
Is it really? If an old man walks down the street and says "My son is coming", you could make the prediction, or assume, that in a few minutes, a younger person who looks similar to the old man will walk past. If such a person walks past five minutes later, you can be fairly confident it's the son, and that your assumption was probably correct. You can never be 100% sure it's the son, but you can be fairly confident.
That's the point of science: using observations and previous knowledge to verify or disprove assumptions and predictions. Nobody will ever be able to say with 100% certainty that evolution happens, but so far, it meshes with reality a lot better than the alternative theories.
Which has a heavier home page? If it's 2 seconds on a 28.8 KB modem to access a homepage for one of them, it's preferred to 10 mins on the same. I only use 28.8 as an example because it really gets the point across.
Good point. I just checked this out. Ask.com's homepage is a bit heavier than Google's, but not enough to really notice, even on 28.8k. But for a search returning 100 results, Ask.com's results were roughly 30% heavier than Google's.
...enjoy google's constant service upgrades...
What constant service upgrades? Tweaking their algorithms? To most people, Google search has barely changed in years. They've added quite a few new products, but those hardly count as upgrades to their search engine.
And just to nitpick, there IS a download. In fact, it has to be downloaded every single time it's used.
Hardy 's saying is a bit of slight against all female mathematicians too, come to think of it...
Using the word "man" to refer to people in general goes back thousands of years. But, you know, I think you're right. In THIS case, he meant it as a clever, ambiguous attempt at women bashing.
Slashdot Mirror
For the same reason I can copy a file till my hard drive crashes or the universe ends. In fact, even when the hard drive crashes, I can load a backup onto a similar hard drive and copy the same file, using the same copy program, until *that* hard drive crashes. My point is, computer software shouldn't "wear out" over time. There are no physically moving parts.
Cars, washers, dryers and air conditioners all have moving parts that are susceptable to friction, heat and general wear. Software is more like a book. Once a book is printed, it will stay printed, completely readable until it's purposely destroyed or the papers physically degrades away in thousands of years.
Exclusive pictures from one of the fights.
After thinking about it for a second, it's in the Chinese governement's best interest to know who's using the Internet. If you connect to a secure wireless network, you need to have a key, so someone has to know who you are. You can't arrest dissidents if any idiot with a wireless card can connect to the Internet anonymously.
I'm not saying I agree with how they could use it, but it is in their best interest to have secure wireless.
I dislike the idea of backdoors in general, but in this situation, it probably doesn't matter too much. It's just as likely (perhaps even more) that Intel's protocol has backdoors and it's almost guaranteed that, if it's there, the US government, the Chinese government, and any other government who asks will get access to it. If WAPI has a back door, the only people with access will be the Chinese government, and it seems unlikely they'll be in a hurry to give out access to it. At least with the Chinese government you know that they're spying, and you know what they're looking for. It's unfortunate for the Chinese people, but they can take it up with the Chinese government.
Again, I'm not saying I agree with China's policies, but WAPI might not be as bad as all the people screaming about it simply because it's from China.
I'm not sure I agree with most of what you say, but you're right about economics being a very useful thing to study.
I majored in CS and got a minor in econ (and in math), and I use the the econ stuff as much, if not more than the CS stuff. I don't know if minoring in it was entirely worthwhile, but going through intermediate microeconomics and intermediate macroeconomics was possibly the smartest thing I did while in college.
I couldn't read your post, because the massive wall of text hurt my eyes, but I'm guessing you didn't mention all the English classes you're taking?
Holy shit! You went off on that entire little diatribe without using a single <p>. That has to be some kind of record: "Longest post by somebody too stupid to break their post into paragraphs." For fuck's sake, man! If you want people to read what you write, learn to use paragraphs.
It would be nice, but I don't think it will happen. If a website consistently offers useful content, that I can't get anywhere else, I'ld gladly subscribe if the price isn't outrageous. I'm sure many people wouldn't like it, but at the same time, I believe most people don't mind paying for information that's useful to them. If someone wants to make money off of the information they have, they can charge me. Until then, I don't owe them shit, "ad-views" or otherwise.
The only people who really have to fear a subscription based internet are people making money off of worthless content hidden by tons of ads.
And while that may be the case, I don't see the internet going to the subscription model any time soon. It'd have to happen all at once. I don't mind paying for decent info, but I'm not going to pay for it if I can get similar content free on another site.
The appeal to me is that these ads will probably get blocked by my existing filters, just like the rest of Google's ads. Let them show blinking pop-ups with sound and Flash for all I care, I'll never see them. Sometimes I almost forget that websites even have ads.
You missed my (poorly stated) point. I wasn't arguing for or against one compiler or language, but trying to emphasize that the languages themselves are separate from their implementations. It's quite possible to make a C++ compiler that compiles to JVM bytecode or a Java compiler that compiles to native code.
No contest on this one. I mis-remembered this article. I'm not sure if you're right, but I was definitely wrong.
You're right, it would take a huge amount of time to do that stuff in assembly language, or even C. But to state that something simply can't be done in assembly or C is wrong. It may not be done easily, but it can definitely be done. That's the point I was trying to make.
Of course not. Any gain in execution speed would be lost to the huge development time. Any sane person would write it in a higher level language.
That's exactly my point. I didn't say it was a good idea, I said it's possible in C. But nobody does it because there are better ways to speed up code.
Yes, it's multi-paradigm, and one of those paradigms happens to be object oriented. Whether it's good to use it as such is irrelevant. The OP made the blanket statement that OO
Is that Java compiled with gcj? With javac? Is it C compiled with gcc? Visual C++? Intel C compiler? egcs? Digital Mars? One language can't be "faster" than another. Only one compiler produces faster executables than another.
Does it? What if the state of the object can affect the return value?
Oh, no, I guess it doesn't actually do that. I think you meant to say "I pulled that example out of my ass, but Sun's JVM does some optimization at run-time."
Bullshit. Assembly language, almost by definition, can do anything that a program on a particular architecture can do. If it can't be done in assembly language, it can't be done.
What do you think Sun's java binary is written in? It's C (or C++, I'm too lazy to check). If Sun can write a program in C that implements those optimizations, then it's possible to write a C program that performs the same optimizations on itself.
So...many...things...wrong... First, C++ *IS* object oriented. Second, I can think of several object oriented languages which do NOT lazily allocate memory. Third, do you even know for a fact that Java does this? Fourth, the fact that all of your OO code needs to create thousands of objects a second seems a little unusual. Fifth, "extremely typical" does not make sense. Sixth, there are very few things that are computationally possible, that C++ can't do. I can make a C++ program with lazy allocation in about 5 minutes. Just because you don't know how to do something in C++ doesn't mean it can't be done.
On my machine, the Java run-time alone, "rt.jar", is a 38 MB file. The GNU C++ library, "libstdc++.a", is a 7.6 MB file. Both of them are compressed, and both are loaded dynamically.
BUT, Sun's Java loads *every* jar file on the classpath entirely into shared memory. The *.jar files in my jre/lib directory are 51 MB by themselves. It's shared memory, so 20 JVMs can use a single copy, but it also means a single "Hello World" needs to load at least 51 MB, just to run. Unless there
I always forget about Project Gutenberg. I think it'd be nice if some popular current authors would place some of their books there for free. I don't know how copyrights work in the publishing world, so I don't know if they're even allowed to, but it would be nice. Maybe not right after a book is published, but maybe a year or two after it goes on sale. That way most people who really want the book would already have bought it, and the price would be down enough so that it would be cheaper to buy it than use ink to print it at home.
That wouldn't exactly work with technical and reference books, because their prices are usually relatively high, but maybe putting them up without figures or something would work.
I guess there are people who would read entire books online, but I think a lot (most?) people would prefer to read things over a few pages on printed paper.
It'd be nice, but I doubt it'll ever happen. Publishing companies are probably too paranoid about lost profit.
In 1.5? More like since it was called Pheonix around 0.3 beta. In my experience, it's always been pretty bad with memory.
This is great! A group of geeks who can't get dates, using the power of open source, to make sure they never will.
Sigh. Your post is so stupid it almost physically hurts me to think about it.
I can think of about 15 web browsers, which two would you be talking about?
I think you're missing the GP's point. The "Remember Segregation" page shows a shit brown page of nothing when Flash is turned off. The problem isn't tasteful use of Flash. The problem is making the whole page in Flash, and not showing up at all if Flash is turned off. Not even a "You need Flash" warning. The web is based on HTML and HTTP. Not plugins. So it would make sense that the winners of an award for websites would at least show something if you don't have the correct plugins installed.
Internet Explorer has a Word Document plugin that lets you view Word documents right inside IE. Would you have been so quick to spout this nonsense if most of the sites needed to be viewed on Windows in IE with the Word plugin turned on? It's a plugin 85% of people have...
Not only is that analogy stupid, it doesn't even apply here. This is more like someone writing a C program in Java. I'm sure those 100% Flash pages are great, but they kind of miss the point.
I don't understand why you're oligated to fix this person's computer.
Are they paying you? Raise the price. If that doesn't work, raise it again. Problem solved.
If they're not paying you, tell them to fuck off. It's not your problem that they're too lazy to figure it out themselves or too cheap to pay someone else to do it.
Yeah, cryptic, three character variable names like "osd_show_percentage", "stream_dump_type", "too_fast_frame_cnt" and "frame_time_remaining". How cryptic! Whatever could those mean?!?
Bullshit. I just checked. mplayer.c has 3 pointers to void, and one pointer to pointer of void. A quick search through some other files found zero void pointers. The code in the loader section does have a few, but it's hardly the most common datatype.
The only part of your post that's even remotely true is "All that said, the program is fantastic." On that we agree. mplayer kicks ass.
I'm waiting for the "Hippy Hamster", "Kinky Koala" and "Pretty Pony" releases, myself.
Yeah, well it's "uneducated arm-chair-space-exploration critics" like myself who pay NASA's bills. Perhaps if they don't want my uneducated ass to write my congressmen about NASA wasting my money, they should better explain their objectives and why they chose this over other similarly priced options.
Oh, and I like the way you turned my skepticism into my being uneducated. Nice ad-hominem.
I'm not really concerned about destroying the moon. I'm more curious of why they want to do it at all. Isn't the moon covered with craters? Many of which are a lot more than 16 ft deep? Can't they observe the dust and debris around those craters?
Even if they crash it into the deepest crater on the moon, is it really likely to find water by going another 16ft? There's no water in or around craters x/8, x/4, x/2, or x feet deep, so how likely is it at x+16 feet?
After RTFA, it sounds like they're just trying to stir up a lot of material so their probe can fly through it before also crashing. Wouldn't it be more effective to just send a rover down to the area? Look at the Mars rovers... those things have lasted months longer than planned, and have sent back boatloads of data. If this moon probe lasts twice as long as planned, it means the probe takes 30 minutes to crash instead of 15.
Maybe it's just me, but this seems like a strange way to spend $73 million, with little chance of success.
On the other hand, what are the chances the probe won't slam into the moon, and NASA will anounce "HA! Got ya! We only said that in case it failed and did slam into the moon. It's actually a robotic rover!"
Is it possible the developers are saying something like "It starts out with the biologists saying they need 30 TB of data available 24/7 with 99.999% uptime and 200-250 concurrent users, and goes downhill from there..."
Simply saying the developers are idiots because they suggest Oracle really doesn't make sense without more context. If more than one group of developers suggest Oracle, they might have a point. Are you sure you're not exagerating the requirements? If you need really high uptime and lots of concurrent users with really, really large amounts of data Oracle is one of the better choices. There's a reason it's so expensive...
Sorry, I should've clarified. My intention was that you can't do that without system calls. Which is what the GP was implying should be done.
Any real operating system (Windows NT and up, Linux, *BSD, ...) prevents you from doing any of the stuff you mentioned. You can't just load a program and start doing low level IO to ports. You can't just bypass the MMU and paging system and write directly to physical memory. You can't just write directly to video memory. You can't just have your program load and start acting like it's the operating system. Any operating system worthy of being called an operating system prevents that. Device drivers would come closest, but they most definitely need system calls.
Yeah...okay. For the moment, lets pretend it's possible to directly access the disk and filesystem from a user program, without system calls. To be any use at all as a cross platform virus, the program would need access to NTFS, ext2, FAT32, and ReiserFS. Writing filesystem code isn't trivial. I would be very impressed if a single person could implement any one of those filesystems in a reasonable amount of time, and all 4 would be nothing short of impossible. Let's just say that if somebody had the skill to do it, they'd be too busy making buttloads of money to waste their time.
#include the appropriate pre-existing header files, and #include any code that you would normally call from a shared library.I'd love to hear your explanation on how to do that...
But, the whole thing is pointless. Even if you did manage to write filesystem support for all the required filesystems and were able to #include the code from the shared libraries the operating system would still stop you the instant you tried to read or write directly to the disk.
It's also one way to really piss Microsoft off...
Lucent better have a really water tight case, because Microsoft has tons of patents, tons of cash and tons of lawyers. Until now they haven't done much with the patents, but now they're going to have legions of lawyers combing through every feature of every Lucent app they can get their hands on, looking for anything that even remotely resembles something MS has patented. I don't think half a dozen counter suits over stuff like the "IsNot" operator would be too surprising.
If it's a really blatant case of infringement, MS might settle. Otherwise they're going to pull out all the stops, throw their whole legal department at it, and send a powerful message to everyone else who considers suing them over patents.
I'd love to see Lucent win this, but it's not going to be easy.
That's the whole point of the article, and part of the reason this is so significant. They did calculations, checked with some geologists, made some predictions (or "assumptions" as you want to call them), and looked into them. It's called science.
Do you think the whole scientific community is a bunch of crazy people in lab coats, with weird hair, randomly mixing chemicals, occasionally shouting "Eureka"? He wouldn't have been on Ellesmere Island if he didn't have a good idea of what he might find and how it would show evidence for or against his hypothesis.
As to whether anything actually evolved from this creature isn't relevant because nobody is claiming that. It's more of: "Look at this, a fish with small legs! So at some point fish with small legs existed, as evolution predicted." Not: "Look! This is THE creature we all evolved from."
It's called "Carbon dating". Carbon dating is usually accurate to within a few hundred years. Hell, with a timeframe in the hundreds of millions of years, it'd only have to be accurate to a few tens of thousands of years to be useful.
If you have one fossil from 420,000,000 ± 8000 years ago, another fossil from 350,000,000 ± 8000 years ago, and then find a new fossil from 375,000,000 ± 8000 years ago, you can be pretty sure of the order they occured. That's the main point of the time line.
Is it really? If an old man walks down the street and says "My son is coming", you could make the prediction, or assume, that in a few minutes, a younger person who looks similar to the old man will walk past. If such a person walks past five minutes later, you can be fairly confident it's the son, and that your assumption was probably correct. You can never be 100% sure it's the son, but you can be fairly confident.
That's the point of science: using observations and previous knowledge to verify or disprove assumptions and predictions. Nobody will ever be able to say with 100% certainty that evolution happens, but so far, it meshes with reality a lot better than the alternative theories.
Good point. I just checked this out. Ask.com's homepage is a bit heavier than Google's, but not enough to really notice, even on 28.8k. But for a search returning 100 results, Ask.com's results were roughly 30% heavier than Google's.
What constant service upgrades? Tweaking their algorithms? To most people, Google search has barely changed in years. They've added quite a few new products, but those hardly count as upgrades to their search engine.
And just to nitpick, there IS a download. In fact, it has to be downloaded every single time it's used.
Using the word "man" to refer to people in general goes back thousands of years. But, you know, I think you're right. In THIS case, he meant it as a clever, ambiguous attempt at women bashing.