I use Cedega on Ubuntu and Debian, so far only to play World of Warcraft. I get 20 FPS with a fairly anemic NVidia card (GeForce 2 MX 400) at 1024x768. Installation was dead simple (they provide.debs). There's usually a minor bug or two when a new release of WoW comes out (Blizzard, understandably, doesn't test on Cedega before releasing patches), but they have consistently been fixed within 24 hours. I have maybe 200 hours in the game, and am completely satisfied with Cedega. I haven't tried Point2Play, but I hear it makes it easier if you're not comfortable with dpkg -i from the command line.
Is RPM working yet, or does it still just tell you the.so or.o that it needs, without telling you the last known name and version number of the RPM in which the library can be found?
I'm being a little facetious, but I have the dubious task of being an unofficial maintainer (ie: it's my responsibility to make them go, but I have no authority to change the OS) of a handful of SuSE ES 9 and RH EL 3 machines. Yast and rc are nice, but it seems a little backward to have to connect to a remote server to discover the meaningful (ie: the actual package and version I should be looking for) dependencies for a package I already have on the local machine. It's a little frustrating to see.dll hell in the most popular version of the OS I keep touting to people. Friends tell me, "Yeah, I tried it. It's impossible to install software." It kills me when I know it can be better. Not trying to evoke a holy war, but.debs include the last known name and version of the.deb(s) on which they depend.
Or is there something I'm missing in RPM that does tell me these answers (entirely possible -.rpm is not my primary package type).
Everyone don your tinfoil hats for this one. Suppose Microsoft, salty about Red Flag Linux and the new gov't edict to shift towards Chinese software, decided to start a smear campaign against China. What would be the best way to whip America into an ideological furor? Report that the Chinese gov't hates Liberty and Democracy.
I'm not saying that is what did happen; clearly China is pretty big on censorship, and it is entirely possible that China did lean on MSN China. It's just amusing to think that the tinfoil hat explanation is really quite possible.
Razor-sharp until you put it side-by-side with a modern LCD. Then it will look dull, and probably dim too. That said, if price is the major consideration, you can't beat a CRT.
Actually, I run my machine dual head. The 997DF CRT is side by side with a modern LCD. A rather nice 17", with which I am very pleased. But I like it for different reasons - instant on and minimal desktop footprint being the two big ones. But it is not a better image - the resolution isn't high enough. And while it is sharper, it also has larger gaps between the pixels - which is why most projectors look a little fuzzy - they intentionally blur the image to hide the black grid. Brightness is pretty much equal. I like the color better on the CRT, but I haven't color calibrated either, so I don't know which is objectively "right".
That's because the majority of Americans are brainwashed from a young age.
I'm not trolling here and perhaps I'm not totally informed, but don't children chant their allegiance to the flag in primary school every day? Isn't it drummed into to everyone to love the constitution? Even though parts of it are hideously outdated and you could argue that every day it is being corrupted further.
Same thing with what was once Adam Smith's Free Market and is now Ayn Rand's Objectivism. Business school in the US is a place to have one message relentlessly repeated until you believe; that profit proves societal advancement - anything that profits is good. This is a twisting of the original, brilliant, concept from Smith that profit is a practical motivator for societal advancement given that greed is a fact. This twisted stepchild, Objectivism, is where all the IP laws come from: They must be good - the progenitors of the laws are making money.
The same is, I'm sure, true of law school and the seminary. Brainwashing in the US is epidemic.
I'm including common applications such as design or photographic work - LCD's work as well or better than CRT's and good ones don't cost much more, if any.
I just bought a new CRT (Samsung 997DF) for $179 that runs razor sharp at 1920x1440. The cheapest LCD I'm familiar with that gets close costs $1499 from Apple (for the 23 inch model). I consider 8X to be much more. Froogle lists your Dell LCD starting at $500 for 1680x1050; nearly three times the price for two thirds the pixels.
But if the world goes to 100% open source, innovation goes to zero. The open source guys hate it when I say this, but it's true.
I didn't RTFA, and I won't (not going to give him the hit). But zero? Zero is a very small number. I can't speak for every OSS developer, or every OSS application. And I haven't done a whole lot of Open Source development, but I did write a GUI tool for Log4J. It competes with Chainsaw, which is also quite good, and which is also Open Source.
Did Lumbermill or Chainsaw get written because of private benevolence? No. Did either copy a commercial equivalent? Well I don't think Chainsaw did, and Lumbermill was written from whole cloth.
Those may not be big innovations, but they are innovations - and what is an operating environment if not a collection of small innovations?
And what is the extent of commercial involvement? The company I work for uses Lumbermill; early in development they would make recommendations and let me work on it on the clock. That's not benevolence, that's rational self interest, the backbone of the free market.
Hmmm, browbeating tone, steadfast refusal to consider whether there may in fact be a problem here, posting as AC... do you work at Sun perchance? If so, I fear your goal is to win the argument, not to understand the problem (a goal which is well served by your hostile tone), but I'll make one more attempt anyway.
Since you seem focused on attacking the specific form of Dr. Kabutz's example, rather than considering the general problem, I've coded a quick example that is a bit more real-world like.
Suppose I wrote a posterize function as shown in the above, and placed it as a static method in a color utility class. I then use that method in a section of code that processes 1600x1200 images (yes, I'm sure there are optimizations that could be done to the posterize method above, but let's assume it's close enough, for the sake of argument) (and yes, I know I'm not processing the alpha channel, but work with me here, OK?). All is well, the image processor is working as fast as I need for my case - I'm processing 5 photos per second on a machine just like the one I'm writing from now.
Now suppose some young developer, the very sort of developer Sun is reaching out to with the programmer-friendliness additions in Tiger, comes along and says, "hey, k3wl, we can use autoboxing here and change two of the three parameters in this method signature to Objects without altering any of the invoking code!" And so he changes the signature of the posterize method to the signature you see in my example as posterizeObj. Performance of the method is cut in half, despite zero code changes except the autoboxing.
So, from your ivory tower at Sun (or whatever ivory tower you live in where every developer has a deep understanding of compiler de-optimization), can you tell me where I went wrong? Should I not give apprentices full access to the source code repository? What if he's not an apprentice, just a newly converted SmallTalker who thinks everything should be an object and doesn't understand the performance problem?
Perhaps there should be a comment on the posterize method saying, "Don't change this to autoboxing, it's a high performance method!" But doesn't that beg the question; if, as of JDK 1.5, I have to add more comments to keep other developers from breaking my code without the compiler telling me, is it really more programmer-friendly?
Auto-boxing changes nothing in terms of performance. If you pass a value type as a parameter to a method that expects an object it will generate code that is precisely the same as if you were wrapping it yourself by hand.
It doesn't even really alter 'strictness' since it basically just promotes value types to their object wrappers, and only where that actually makes sense.
While your statement is true when talking about cases where you intend object conversion, the problem is when you don't want conversion to occur. IBM notes the problem in this article, for the same reason I do. If the compiler uses objects without telling you, it's a problem. Having done a bit of real-time image processing with Java, it is clear that Java is an adequate high performance platform, and that primitives are significantly faster than objects - as long as you can count on them remaining primitive.
The Java community at large doesn't care much about an open source Java.
The vast majority of users of any zero cost technology don't care if it's open source.
While you may be correct about the Java community at large, I think you may be off the mark with respect to senior Java developers. I think a fair number of senior Java developers have, at least once or twice, had language recommendations coldly rebuffed by Sun (and, to Sun's credit, probably have had a few accepted through the JCP). Sun is renowned for sticking their fingers in their ears and chanting, "lalala I cannot hear you lalala" about core language issues. An open source JVM is an opportunity to have our voices heard, or at least to create legally sharable patches for the features we want.
Consider the pre-Tiger dustup over autoboxing; it's a good idea if you want to increase user-friendliness at the cost of strictness and performance. A lot of people didn't want to be forced onto one side of this tradeoff, but Sun's apparent number one priority for 5 was to exceed the perceived user-friendliness of.Net at any cost. Even a command line switch to issue a warning about autoboxing was out of the question.
This is one sort of problem that OSS solves. The vast majority of users of any given technology may not care if it's open source or not, but to the power users it gives a degree of freedom that really matters.
"It's nice to see Microsoft recognizing the power of blogs, but the move is likely going to draw accusations that Redmond is trying to buy off bloggers to hype Longhorn."
Blogging was nice while it lasted. Corporations are quickly going to flood the channel with paid content. If you think the PR machine is powerful in major media, which has lots of people looking for bias, has some regulation, and which does not see $10,000 as any more than pocket change, think what's going to happen to blogs over the next five years.
Suppose Coca-Cola offered to pay Joe Blogpack $2,500 to do a column talking about a dead rat found in a storage container at a Pepsi bottling facility, how quickly do you think he would jump? Do you think he would care if the story is true? And if he did, would he have access to the resources to find out if it's true? Suppose news.google.com is running 200 links to other bloggers who didn't take the time to fact check - our honorable Joe Blogpack checks his facts against the tainted stories and even thinks he's doing the right thing.
That's the core problem with many comments, but it can be avoided. Comments are good when they state the intent or business case for a block of code, acting as a guide to the meaning of a block for subsequent developers. They are bad when they profess to know the actual outcome or implementation of a block; only the code itself can accurately reflect the state of the code.
The above would be far more useful like this:
/** Tells whether this instance meets the magilla criteria. */ public boolean isMagilla() {
return false;// currently, no MyClass meets the magilla criteria. }
Now the intent of the method is clear, and anyone coming along who wonders why it's hard coded will know under what circumstances they should change it to a formula (namely, if MyClass becomes capable of meeting the Magilla criteria).
Comments can be good, but they should always be a guidepost to the intent of a block of code, and not attempt to explain how the code achieves that goal.
I've used Python for several large projects and have never found a situation where Python got me in trouble that would have been easily avoided in some other language.
First, let me be clear, so there's no misunderstanding: I use both statically typed and dynamically typed languages both for work and for play. I like them both.
Why would anyone use static typing?
You may be smarter than the average bear, maybe compiler warnings are nothing but a nuisance to you. However, for many of us, they allow us to focus on the logic and let the machine catch conversion errors and invalid method calls. And even those of us who are above such things may write libraries that are used by other developers. Those other developers may be just starting in the language or problem space we are coding in, or may just be getting into professional computer science altogether. For those people, strictness is helpful. Knowing that you get a loss of precision when you call int = int + float is something some people don't automatically know.
Heck, I've been doing this for a long time, and I still appreciate the warning when I'm hammering through a block of code and use the wrong type accidentally.
Likewise, knowing at compile time that you can't invoke ".foo( Argle bargle )" on an object of type Baz saves you from having to find the error in runtime testing. When you've got a million lines of code to test (and are working with a team, not all of whom have rigorous unit testing habits), there are invariably blocks of code that don't get exercised until the product reaches the customer. Adding strict type checking saves some of those errors. Beyond that, a good set of interfaces with method calls that clearly state their purpose can even make it possible to implement extensions without having to RTFM.
Is it necessary with all teams? No. Is the benefit always greater than the cost? Of course not. But there are situations in which it is beneficial. Saying that strict typing is never beneficial is silly. A lot of people who have a damned site more experience than you and I put together have chosen type safety for some languages, and they did so for good reasons.
Clearly everyone missed my point, so I'll clarify.
I was using an example from the real world to point out why 3 seconds matters. In any significant application there will be some processes that are sufficiently complicated that the choice of algorithm or choice of language will lead to a 3 second delta one way or another. There will also be places where adding a UI shortcut will save 3 seconds.
The real world example talks about UI shortcuts that can save those 3 seconds, and Python makes it easier (according to the common wisdom) to add features. Other languages are more performance centric, and make it easy to save those 3 seconds in operation intensive sections of the code.
I wasn't arguing that Python is bad because it's not as performant. I was saying that both CPU performance and UI friendliness are important, so choosing between Python, Ruby, C#, Java, C++, C or any other language on the spectrum is a question of weighing values.
Ferfucksake people, stop trying to be argumentative and start trying to understand what people are saying. We all claim to be so smart, is our only ability with our intelligence to pick nits? Or can we use our intelligence to seek mutual understanding?
I mean, I can see why the media is turning into a bunch of ranting extremists - they're just a mirror reflecting our own horrible image.
Well, with any other language, if I get a piece of unfamiliar code and have problems reading it due to weird indentation, I just run it through Emacs' indent-region. Can anyone explain to me why this is not just as viable as mandating the indentation policy by embedding it in the language's syntax?
I think maybe you're just looking at it from a different angle than the proponents. The upside of whitespace-based scoping lies in the assumption that we now all use advanced editors like Emacs or IDEs, and so we all now have automatic indent controls. Given that, curly braces are now redundant - the whitespace already contains the block scope information. The curly braces are repetitive - they just restate what the indent already does.
2) Misperceptions about typing Many people think agile languages like Python and Ruby are not strongly typed and therefore present scalability problems and can't be used reliably by large teams. But Python and Ruby are strongly typed (unlike Perl)- you don't get type conversions you don't ask for. The real distinction is that the agile languages are dynamically typed rather than statically typed like Java/C++. To truly grasp the notions of "duck-typing" and lazy evaluation of types is as much a stretch as it was to "get" objects for those of us who were around 15 years ago- it's a basic change in how you think. You'll know when you're there, because you'll see in a flash that Java's static type declarations are not only redundant and painful, but they are also in themselves a key source of brittleness in large programs over time.
The "unwanted type conversions" thing is a nice straw man, but it won't wash. The value of strictly typed languages is in compile time type checking. It's good to have languages that are not type checked, and it's good to have languages that are. The former is better with smaller groups, smaller programs, or more proficient developers. The latter is better with larger groups working on larger programs with more apprentices on the team. The former allows more flexibility and speed. The latter offers more imposed speed limits (and less car crashes).
I don't think Python will ever dent Java's dominance in the enterprise.
I completely agree. Just as we could all tell (all the way back to 1999) that Java would never put a dent in Perl.cgi on the server side. It's just not designed as a server side scripting language the way Perl is.;)
And most enterprise apps are not CPU limited anyway.
Say what? You must be living in a very different world than I. If it's an enterprise app, then it has a few thousand internal users. Multiply a few seconds by a few thousand people by a few times an hour by a few dollars per hour. Performance matters. Middle of last year my company pulled two people off their primary project to add a feature that saved our primary users two mouse clicks. Those 4,000 users now save 3 seconds, 10 times an hour, 8 hours a day. It's like getting an extra 33 employees for free.
Now, I hasten to add that performance isn't the only thing - if it takes three extra months to add a new feature then you've got to multiply 3 months by all those people not doing what they could be doing with the new feature. But claiming it's a non-issue is shortsighted.
http://netpub.cstudies.ubc.ca/oleary/images/python _java_comparison.gif may give you an idea. Quite simply it's easier to program in, and you're more productive using it.
I write code professionally in both Java and Python, and I like them both. First, that comparison graphic was not meant to show two nearly identical programs; the Java version is significantly expanded. Following is a version more suited to direct comparison:
1. public class Hello { 2. public String myName = "Monty"; 3. 4. public void printHello() { 5. System.out.println( "Hello, " + myName ); 6. } 7. 8. public static void main( String[] args ) { 9. Hello h = new Hello(); 10. h.printHello(); 11. } 12.}
So, extra lines for the curly braces, plus the instance parameter declaration, and Java allows the implicit constructor. Not much difference.
But what of that declaration difference? Well, Java is strongly typed. Each class is fully defined before it is used. Python is more flexible.
Which is better? It's like asking if a dump truck or a motorcycle is a better vehicle. Java is better if you want to make libraries that take longer to incorporate, but also make it harder to use them incorrectly. Python is better if you want to make flexible code that lends itself to poor maintainability. (and yes, I know you can write maintainable code in Python, it's just that Java does more to impose maintainability)
Much like the Perl v. Java debates (I do Perl too): Are you writing a 25 line one-off? Use Perl. Are you writing a 250,000 line CRM system? Use Java. Are you writing a 25,000 line log analysis app which will have a very string intensive section? Tough call.
In short, Python is more flexible, and therefore gives you more rope with which to hang yourself. Java is more strict, and so slows you down a bit to reduce your ability to write unsafe code. It's a toolbox. You put the tools in and you do the job.
Added Steven Crocker, a respected Internet pioneer: "It shouldn't have taken that long."
Who the hell is Steven Crocker, and who did he blow to get this ridiculously out-of-place one line yawner crammed into the middle of this article? "It shouldn't have taken that long." Well now that is some poignant and inspiring commentary there, respected Internet pioneer Crocker. How about something on the price of oil? "It's pretty high." Thanks.
Could it be that marketing is always overselling the product? Seriously. I cannot count how many times I have heard (in the past now I am in science), "oh, yeah....well, you need to include feature X because we told customer Y we already had that feature".
It could be worse. At my present company, many of the customer liaison's have gotten in the habit of saying, "It doesn't do X? Well obviously it has to do X or it's completely useless. What were you thinking?" Suggesting that the problem is the idiot developer failing to think like the customer, not the lack of either: a) direct contact between customer and engineer, or at least b) complete specifications. Now I know that complete specifications are hard, and perhaps impossible, but failing that either we need to talk to the customer or the product isn't going to do what they want except through blind luck.
(But don't overdoit and break traction or you'll go down.)
Offtopic (for the initial message, not the thread):
Do you have a tried and true method for knowing how much is too much (or more specifically, how much is just short of too much)? I've been riding for 4 years and have about 10,000 miles behind me, but still feel like I'm very far short of the maximum force I can safely apply to the bars to swerve. I'd like to get to know the limit better so I can use it if I need it.
We are a big section of the opinion makers in computer hardware. We have the ability to affect the public opinion on XDR. To a large extent we were the ones most adversely affected by the last round, and we are the ones who can shift public opinion now.
This should be like a usenet death penalty. The free market is there to reward those companies that serve their customers and punish those that do not. It is a good system, but it tends to have a short attention span. Tell your friends. Tell your purchasing deparment. Keep Rambus from coming back from the dead and send a message to other companies who think about abusing submarine patents. It's the same thing as harsh criminal sentencing, except that the free market has a far better track record of responding to example punishment (that is to say; if you support harsh criminal sentincing, you should support this on the same ideological grounds, and if you don't support harsh criminal sentencing because it doesn't work, you should still support this because it does).
Java's motto shouldn't be "Write once run anywhere" - it should be "Write once, test everywhere". An admirable goal, true, but don't kid yourself about what it really means.
1 MegaLOC. 36 megs of Java source. Swing, JSP, Servlets, SOAP, SOA, Kerberos, LDAP, JDBC, and EJBs (to name just a few). Our newest clients are C#. 4 years I've been working on it. I'm one of 4 developers using Linux. The other 20 or so use Windows. We deploy to Windows 2000, XP, Solaris, SuSE, Debian, and RedHat on the server side, and all those but SuSE and Solaris on the client side. The most I've done to support it is replace backslashes with forward slashes (forward slashes in Java work on any platform).
Slashdot Mirror
I had a lot of respect for him. I used to see him as Peter Jackson, Cinematic Artist. Now I see him as Peter Jackson, Greedcock.
Would you be happier if NewLine got to keep the money?!? The money belongs to someone. I say give it to the guy who created it.
I use Cedega on Ubuntu and Debian, so far only to play World of Warcraft. I get 20 FPS with a fairly anemic NVidia card (GeForce 2 MX 400) at 1024x768. Installation was dead simple (they provide .debs). There's usually a minor bug or two when a new release of WoW comes out (Blizzard, understandably, doesn't test on Cedega before releasing patches), but they have consistently been fixed within 24 hours. I have maybe 200 hours in the game, and am completely satisfied with Cedega. I haven't tried Point2Play, but I hear it makes it easier if you're not comfortable with dpkg -i from the command line.
Is RPM working yet, or does it still just tell you the .so or .o that it needs, without telling you the last known name and version number of the RPM in which the library can be found?
.dll hell in the most popular version of the OS I keep touting to people. Friends tell me, "Yeah, I tried it. It's impossible to install software." It kills me when I know it can be better. Not trying to evoke a holy war, but .debs include the last known name and version of the .deb(s) on which they depend.
.rpm is not my primary package type).
I'm being a little facetious, but I have the dubious task of being an unofficial maintainer (ie: it's my responsibility to make them go, but I have no authority to change the OS) of a handful of SuSE ES 9 and RH EL 3 machines. Yast and rc are nice, but it seems a little backward to have to connect to a remote server to discover the meaningful (ie: the actual package and version I should be looking for) dependencies for a package I already have on the local machine. It's a little frustrating to see
Or is there something I'm missing in RPM that does tell me these answers (entirely possible -
Everyone don your tinfoil hats for this one. Suppose Microsoft, salty about Red Flag Linux and the new gov't edict to shift towards Chinese software, decided to start a smear campaign against China. What would be the best way to whip America into an ideological furor? Report that the Chinese gov't hates Liberty and Democracy.
I'm not saying that is what did happen; clearly China is pretty big on censorship, and it is entirely possible that China did lean on MSN China. It's just amusing to think that the tinfoil hat explanation is really quite possible.
Razor-sharp until you put it side-by-side with a modern LCD. Then it will look dull, and probably dim too. That said, if price is the major consideration, you can't beat a CRT.
Actually, I run my machine dual head. The 997DF CRT is side by side with a modern LCD. A rather nice 17", with which I am very pleased. But I like it for different reasons - instant on and minimal desktop footprint being the two big ones. But it is not a better image - the resolution isn't high enough. And while it is sharper, it also has larger gaps between the pixels - which is why most projectors look a little fuzzy - they intentionally blur the image to hide the black grid. Brightness is pretty much equal. I like the color better on the CRT, but I haven't color calibrated either, so I don't know which is objectively "right".
That's because the majority of Americans are brainwashed from a young age.
I'm not trolling here and perhaps I'm not totally informed, but don't children chant their allegiance to the flag in primary school every day? Isn't it drummed into to everyone to love the constitution? Even though parts of it are hideously outdated and you could argue that every day it is being corrupted further.
Same thing with what was once Adam Smith's Free Market and is now Ayn Rand's Objectivism. Business school in the US is a place to have one message relentlessly repeated until you believe; that profit proves societal advancement - anything that profits is good. This is a twisting of the original, brilliant, concept from Smith that profit is a practical motivator for societal advancement given that greed is a fact. This twisted stepchild, Objectivism, is where all the IP laws come from: They must be good - the progenitors of the laws are making money.
The same is, I'm sure, true of law school and the seminary. Brainwashing in the US is epidemic.
I'm including common applications such as design or photographic work - LCD's work as well or better than CRT's and good ones don't cost much more, if any.
I just bought a new CRT (Samsung 997DF) for $179 that runs razor sharp at 1920x1440. The cheapest LCD I'm familiar with that gets close costs $1499 from Apple (for the 23 inch model). I consider 8X to be much more. Froogle lists your Dell LCD starting at $500 for 1680x1050; nearly three times the price for two thirds the pixels.
What was that you were saying?
But if the world goes to 100% open source, innovation goes to zero. The open source guys hate it when I say this, but it's true.
I didn't RTFA, and I won't (not going to give him the hit). But zero? Zero is a very small number. I can't speak for every OSS developer, or every OSS application. And I haven't done a whole lot of Open Source development, but I did write a GUI tool for Log4J. It competes with Chainsaw, which is also quite good, and which is also Open Source.
Did Lumbermill or Chainsaw get written because of private benevolence? No. Did either copy a commercial equivalent? Well I don't think Chainsaw did, and Lumbermill was written from whole cloth.
Those may not be big innovations, but they are innovations - and what is an operating environment if not a collection of small innovations?
And what is the extent of commercial involvement? The company I work for uses Lumbermill; early in development they would make recommendations and let me work on it on the clock. That's not benevolence, that's rational self interest, the backbone of the free market.
Zero? I think not.
Hmmm, browbeating tone, steadfast refusal to consider whether there may in fact be a problem here, posting as AC... do you work at Sun perchance? If so, I fear your goal is to win the argument, not to understand the problem (a goal which is well served by your hostile tone), but I'll make one more attempt anyway.
Since you seem focused on attacking the specific form of Dr. Kabutz's example, rather than considering the general problem, I've coded a quick example that is a bit more real-world like.
Autobox.java
Suppose I wrote a posterize function as shown in the above, and placed it as a static method in a color utility class. I then use that method in a section of code that processes 1600x1200 images (yes, I'm sure there are optimizations that could be done to the posterize method above, but let's assume it's close enough, for the sake of argument) (and yes, I know I'm not processing the alpha channel, but work with me here, OK?). All is well, the image processor is working as fast as I need for my case - I'm processing 5 photos per second on a machine just like the one I'm writing from now.
Now suppose some young developer, the very sort of developer Sun is reaching out to with the programmer-friendliness additions in Tiger, comes along and says, "hey, k3wl, we can use autoboxing here and change two of the three parameters in this method signature to Objects without altering any of the invoking code!" And so he changes the signature of the posterize method to the signature you see in my example as posterizeObj. Performance of the method is cut in half, despite zero code changes except the autoboxing.
So, from your ivory tower at Sun (or whatever ivory tower you live in where every developer has a deep understanding of compiler de-optimization), can you tell me where I went wrong? Should I not give apprentices full access to the source code repository? What if he's not an apprentice, just a newly converted SmallTalker who thinks everything should be an object and doesn't understand the performance problem?
Perhaps there should be a comment on the posterize method saying, "Don't change this to autoboxing, it's a high performance method!" But doesn't that beg the question; if, as of JDK 1.5, I have to add more comments to keep other developers from breaking my code without the compiler telling me, is it really more programmer-friendly?
Auto-boxing changes nothing in terms of performance. If you pass a value type as a parameter to a method that expects an object it will generate code that is precisely the same as if you were wrapping it yourself by hand.
It doesn't even really alter 'strictness' since it basically just promotes value types to their object wrappers, and only where that actually makes sense.
While your statement is true when talking about cases where you intend object conversion, the problem is when you don't want conversion to occur. IBM notes the problem in this article, for the same reason I do. If the compiler uses objects without telling you, it's a problem. Having done a bit of real-time image processing with Java, it is clear that Java is an adequate high performance platform, and that primitives are significantly faster than objects - as long as you can count on them remaining primitive.
The Java community at large doesn't care much about an open source Java.
.Net at any cost. Even a command line switch to issue a warning about autoboxing was out of the question.
The vast majority of users of any zero cost technology don't care if it's open source.
While you may be correct about the Java community at large, I think you may be off the mark with respect to senior Java developers. I think a fair number of senior Java developers have, at least once or twice, had language recommendations coldly rebuffed by Sun (and, to Sun's credit, probably have had a few accepted through the JCP). Sun is renowned for sticking their fingers in their ears and chanting, "lalala I cannot hear you lalala" about core language issues. An open source JVM is an opportunity to have our voices heard, or at least to create legally sharable patches for the features we want.
Consider the pre-Tiger dustup over autoboxing; it's a good idea if you want to increase user-friendliness at the cost of strictness and performance. A lot of people didn't want to be forced onto one side of this tradeoff, but Sun's apparent number one priority for 5 was to exceed the perceived user-friendliness of
This is one sort of problem that OSS solves. The vast majority of users of any given technology may not care if it's open source or not, but to the power users it gives a degree of freedom that really matters.
"It's nice to see Microsoft recognizing the power of blogs, but the move is likely going to draw accusations that Redmond is trying to buy off bloggers to hype Longhorn."
Blogging was nice while it lasted. Corporations are quickly going to flood the channel with paid content. If you think the PR machine is powerful in major media, which has lots of people looking for bias, has some regulation, and which does not see $10,000 as any more than pocket change, think what's going to happen to blogs over the next five years.
Suppose Coca-Cola offered to pay Joe Blogpack $2,500 to do a column talking about a dead rat found in a storage container at a Pepsi bottling facility, how quickly do you think he would jump? Do you think he would care if the story is true? And if he did, would he have access to the resources to find out if it's true? Suppose news.google.com is running 200 links to other bloggers who didn't take the time to fact check - our honorable Joe Blogpack checks his facts against the tainted stories and even thinks he's doing the right thing.
The above would be far more useful like this:Now the intent of the method is clear, and anyone coming along who wonders why it's hard coded will know under what circumstances they should change it to a formula (namely, if MyClass becomes capable of meeting the Magilla criteria).
Comments can be good, but they should always be a guidepost to the intent of a block of code, and not attempt to explain how the code achieves that goal.
I've used Python for several large projects and have never found a situation where Python got me in trouble that would have been easily avoided in some other language.
First, let me be clear, so there's no misunderstanding: I use both statically typed and dynamically typed languages both for work and for play. I like them both.
Why would anyone use static typing?
You may be smarter than the average bear, maybe compiler warnings are nothing but a nuisance to you. However, for many of us, they allow us to focus on the logic and let the machine catch conversion errors and invalid method calls. And even those of us who are above such things may write libraries that are used by other developers. Those other developers may be just starting in the language or problem space we are coding in, or may just be getting into professional computer science altogether. For those people, strictness is helpful. Knowing that you get a loss of precision when you call int = int + float is something some people don't automatically know.
Heck, I've been doing this for a long time, and I still appreciate the warning when I'm hammering through a block of code and use the wrong type accidentally.
Likewise, knowing at compile time that you can't invoke ".foo( Argle bargle )" on an object of type Baz saves you from having to find the error in runtime testing. When you've got a million lines of code to test (and are working with a team, not all of whom have rigorous unit testing habits), there are invariably blocks of code that don't get exercised until the product reaches the customer. Adding strict type checking saves some of those errors. Beyond that, a good set of interfaces with method calls that clearly state their purpose can even make it possible to implement extensions without having to RTFM.
Is it necessary with all teams? No. Is the benefit always greater than the cost? Of course not. But there are situations in which it is beneficial. Saying that strict typing is never beneficial is silly. A lot of people who have a damned site more experience than you and I put together have chosen type safety for some languages, and they did so for good reasons.
Clearly everyone missed my point, so I'll clarify.
I was using an example from the real world to point out why 3 seconds matters. In any significant application there will be some processes that are sufficiently complicated that the choice of algorithm or choice of language will lead to a 3 second delta one way or another. There will also be places where adding a UI shortcut will save 3 seconds.
The real world example talks about UI shortcuts that can save those 3 seconds, and Python makes it easier (according to the common wisdom) to add features. Other languages are more performance centric, and make it easy to save those 3 seconds in operation intensive sections of the code.
I wasn't arguing that Python is bad because it's not as performant. I was saying that both CPU performance and UI friendliness are important, so choosing between Python, Ruby, C#, Java, C++, C or any other language on the spectrum is a question of weighing values.
Ferfucksake people, stop trying to be argumentative and start trying to understand what people are saying. We all claim to be so smart, is our only ability with our intelligence to pick nits? Or can we use our intelligence to seek mutual understanding?
I mean, I can see why the media is turning into a bunch of ranting extremists - they're just a mirror reflecting our own horrible image.
Feh.
Well, with any other language, if I get a piece of unfamiliar code and have problems reading it due to weird indentation, I just run it through Emacs' indent-region. Can anyone explain to me why this is not just as viable as mandating the indentation policy by embedding it in the language's syntax?
I think maybe you're just looking at it from a different angle than the proponents. The upside of whitespace-based scoping lies in the assumption that we now all use advanced editors like Emacs or IDEs, and so we all now have automatic indent controls. Given that, curly braces are now redundant - the whitespace already contains the block scope information. The curly braces are repetitive - they just restate what the indent already does.
2) Misperceptions about typing
Many people think agile languages like Python and Ruby are not strongly typed and therefore present scalability problems and can't be used reliably by large teams. But Python and Ruby are strongly typed (unlike Perl)- you don't get type conversions you don't ask for. The real distinction is that the agile languages are dynamically typed rather than statically typed like Java/C++. To truly grasp the notions of "duck-typing" and lazy evaluation of types is as much a stretch as it was to "get" objects for those of us who were around 15 years ago- it's a basic change in how you think. You'll know when you're there, because you'll see in a flash that Java's static type declarations are not only redundant and painful, but they are also in themselves a key source of brittleness in large programs over time.
The "unwanted type conversions" thing is a nice straw man, but it won't wash. The value of strictly typed languages is in compile time type checking. It's good to have languages that are not type checked, and it's good to have languages that are. The former is better with smaller groups, smaller programs, or more proficient developers. The latter is better with larger groups working on larger programs with more apprentices on the team. The former allows more flexibility and speed. The latter offers more imposed speed limits (and less car crashes).
I don't think Python will ever dent Java's dominance in the enterprise.
.cgi on the server side. It's just not designed as a server side scripting language the way Perl is. ;)
I completely agree. Just as we could all tell (all the way back to 1999) that Java would never put a dent in Perl
And most enterprise apps are not CPU limited anyway.
Say what? You must be living in a very different world than I. If it's an enterprise app, then it has a few thousand internal users. Multiply a few seconds by a few thousand people by a few times an hour by a few dollars per hour. Performance matters. Middle of last year my company pulled two people off their primary project to add a feature that saved our primary users two mouse clicks. Those 4,000 users now save 3 seconds, 10 times an hour, 8 hours a day. It's like getting an extra 33 employees for free.
Now, I hasten to add that performance isn't the only thing - if it takes three extra months to add a new feature then you've got to multiply 3 months by all those people not doing what they could be doing with the new feature. But claiming it's a non-issue is shortsighted.
I write code professionally in both Java and Python, and I like them both. First, that comparison graphic was not meant to show two nearly identical programs; the Java version is significantly expanded. Following is a version more suited to direct comparison:So, extra lines for the curly braces, plus the instance parameter declaration, and Java allows the implicit constructor. Not much difference.
But what of that declaration difference? Well, Java is strongly typed. Each class is fully defined before it is used. Python is more flexible.
Which is better? It's like asking if a dump truck or a motorcycle is a better vehicle. Java is better if you want to make libraries that take longer to incorporate, but also make it harder to use them incorrectly. Python is better if you want to make flexible code that lends itself to poor maintainability. (and yes, I know you can write maintainable code in Python, it's just that Java does more to impose maintainability)
Much like the Perl v. Java debates (I do Perl too): Are you writing a 25 line one-off? Use Perl. Are you writing a 250,000 line CRM system? Use Java. Are you writing a 25,000 line log analysis app which will have a very string intensive section? Tough call.
In short, Python is more flexible, and therefore gives you more rope with which to hang yourself. Java is more strict, and so slows you down a bit to reduce your ability to write unsafe code. It's a toolbox. You put the tools in and you do the job.
Added Steven Crocker, a respected Internet pioneer: "It shouldn't have taken that long."
Who the hell is Steven Crocker, and who did he blow to get this ridiculously out-of-place one line yawner crammed into the middle of this article? "It shouldn't have taken that long." Well now that is some poignant and inspiring commentary there, respected Internet pioneer Crocker. How about something on the price of oil? "It's pretty high." Thanks.
Could it be that marketing is always overselling the product? Seriously. I cannot count how many times I have heard (in the past now I am in science), "oh, yeah....well, you need to include feature X because we told customer Y we already had that feature".
It could be worse. At my present company, many of the customer liaison's have gotten in the habit of saying, "It doesn't do X? Well obviously it has to do X or it's completely useless. What were you thinking?" Suggesting that the problem is the idiot developer failing to think like the customer, not the lack of either: a) direct contact between customer and engineer, or at least b) complete specifications. Now I know that complete specifications are hard, and perhaps impossible, but failing that either we need to talk to the customer or the product isn't going to do what they want except through blind luck.
(But don't overdoit and break traction or you'll go down.)
Offtopic (for the initial message, not the thread):
Do you have a tried and true method for knowing how much is too much (or more specifically, how much is just short of too much)? I've been riding for 4 years and have about 10,000 miles behind me, but still feel like I'm very far short of the maximum force I can safely apply to the bars to swerve. I'd like to get to know the limit better so I can use it if I need it.
If it matters, I ride an FJR1300.
We are a big section of the opinion makers in computer hardware. We have the ability to affect the public opinion on XDR. To a large extent we were the ones most adversely affected by the last round, and we are the ones who can shift public opinion now.
This should be like a usenet death penalty. The free market is there to reward those companies that serve their customers and punish those that do not. It is a good system, but it tends to have a short attention span. Tell your friends. Tell your purchasing deparment. Keep Rambus from coming back from the dead and send a message to other companies who think about abusing submarine patents. It's the same thing as harsh criminal sentencing, except that the free market has a far better track record of responding to example punishment (that is to say; if you support harsh criminal sentincing, you should support this on the same ideological grounds, and if you don't support harsh criminal sentencing because it doesn't work, you should still support this because it does).
Java's motto shouldn't be "Write once run anywhere" - it should be "Write once, test everywhere". An admirable goal, true, but don't kid yourself about what it really means.
1 MegaLOC. 36 megs of Java source. Swing, JSP, Servlets, SOAP, SOA, Kerberos, LDAP, JDBC, and EJBs (to name just a few). Our newest clients are C#. 4 years I've been working on it. I'm one of 4 developers using Linux. The other 20 or so use Windows. We deploy to Windows 2000, XP, Solaris, SuSE, Debian, and RedHat on the server side, and all those but SuSE and Solaris on the client side. The most I've done to support it is replace backslashes with forward slashes (forward slashes in Java work on any platform).
Stop propagating your ignorance.