Oh really? And that applies to software development as well does it? And it means more productivity as well, right - of course many people are happy to sit in a big open office and chat all day, but do they get more work done?
Amen to that. People that are in favor of cubes or the so-called open arrangements have particular models in mind for how people work, both what they must do to do their jobs, and how individuals are best able to get stuff done. The problem is that their models are wrong for many jobs and for many people. If you are in a cube farm in which everyone is working on the same project, then overhearing someone else's conversation (because you can't not hear it) about project stuff may be informative, but if you are working on a different project, it is a distraction. If everyone is working on jobs they can do relatively quietly, then cubes might be alright, but if you are frequently on telecons or in impromptu, loud, contentious discussions, then you may be a disruptions to your neighbors, of whom there are many. If your job requires you to do relatively rote, menial tasks, then external disruptions may not have much impact, but if you need to have some good focused thinking time, the external disruptions will prevent you from ever getting focused, and you will not be able to do your job. If you are an extrovert who enjoys company, then perhaps you want to hear all the conversations around you, but if you are an introvert who needs quiet time to think, those conversations are disruptive. For many people, I would think any savings in going to a cube farm (assuming that the motivation is $$'s) will be more than offset by losses in productivity.
It would have been nice if the article had given these details. Otherwise, you would have expected the software to have been loaded and verified long before launch day. Leaving out the information you provided made this hold seem to be worse than it really was.
This is nice in theory, but in elections with 3 stong candidates, voters always think: I would like to vote for B, but if I do, it is likely that A will win over C, with B in third, and if I had a choice, I'd rather have C than A. In this manner, you can end up with winner that the majority of the people do not want.
This is the reason that some are pushing for Instant Runoff Voting (IRV). In this method, everyone votes for their top choices, in order. Then, say in a 3 candidate election, after the first round, the third place candidates votes would revert to the second choice on those ballots. In this way, you would eventually come down to a two-way race, and you would get a more accurate picture of who the voters want (or don't want) in office. At the same time, people can list their true favorite as their top choice.
I'm sure IRV would be more complicated to implement, but it would eliminate winners with only 34% of the vote (and possible up to 66% against them.)
Isn't The Canopy Group a venture capital firm, and aren't the companies that you list companies that have received investments from Canopy? It may be tempting to lash out at these companies, but these companies probably have nothing to do with the SCO-IBM lawsuit, and in some cases (such as TrollTech) the linux community at large benefits for the contributions of these companies. To try to punish them would be like cutting off your nose to spite your face.
Re:Languages for the Java VM...
on
The Future of Java?
·
· Score: 5, Informative
Here are a few of the more notable Java Processors ones:
www.ajile.com (and www.systronix.com has a few boards built witht the processors).
www.ptsc.com
Also, the Jazelle technology from ARM embeds a java processor section into an ARM chip:
It matters because it provides ammunition to use with people that will insist on only using Windows and parrot the latest MS FUD, while completely resisting even LOOKING at Linux. The next time they say something to the effect that Linux is useless, can't compete with MS products, etc., you just point to a couple of electronic boxes in their home entertainment center and say, "Oh yeah, well your TiVo is using Linux, your MusicServer is using Linux,..."
1) At least one company has solved this problem, with an automatic tool (The name of the tool is J-Integra, from Linar systems, www.linar.com). I believe it uses JNI, but does not require you to write any JNI. 2) With either approach, there has to be a linkup between the DLL and the Java code. MS' approach used an approach that hid everything in the JVM. Could they not have generated JNI code that did this instead? This approach does not necessarily have to make the developer deal directly with the JNI code (i.e. write it). And, it meets the language and JVM specifications, and would work with other Windows-based JVM's.
I suspect that such approaches did not meet MS' business objectives (to kill cross-platform java by promoting a polluted version of java).
3) MS adds extensions for Windows only development, which are optional to developers depending on their target market (HINT: Apple has Cocoa extensions in Java......samething......they are optional)
This is inaccurate and very misleading. The MS JVM added keywords that were not part of Sun's Java, and which Visual J++ would use automatically, meaning that a developer might be writing what he thought was standard Java, when in fact non-standard, MS-only java would be produced. MS was ordered by a court to turn these off, making them an opt in, rather than an opt out. However, these keywords were still not part of the Java language as defined in the language spec. This leads to non-portable Java, and a Java different than the one Sun built and licenses.
Perhaps more important were the J/Direct extensions, which used special comments and the MS compiler to generate annotations the the class files, which the MS runtime used to link in native code, such as windows DLL's. THe problem is not that you could link in native, non-portable code, since standard JNI (Java Native Interface) provides a way for you to do this. THe problem was that the MS approach only worked with MS compilers and MS runtimes. Class files compiled in this manner would not work with Sun or IBM runtimes on Windows environments. Within the Java Virtual Machine specification, it is stated that the class annotations (which MS used) must not change the semantics of the class. In MS' JVM, the annotations would cause a certain behavior, while in the other JVM's, they would be ignored. Clearly, MS' use of the annotations violates the JVM specification.
MS had the option of taking an approach that obeyed the specifications. THere is at least one product that can analyze a DLL and produce the mapping code to make calls on that DLL, using standard Java mechanisms. With their billions in cash, surely they could have easily produced a product that used the same approach. Perhaps a clue came from evidence in the DOJ trial, in which a Microsoft engineer spoke of "blunting the cross-platform java market" by distributing a "polluted" java.
5 years ago Netscape was the king of the browsers -- nothing else came close. When MS released it's browser they out leaped Netscape by offering faster JVM and better compatibility. If Netscape didn't just sit back and watch it's browser being eroded from the market and Sun didn't just put all it's eggs in one basket (Netscape) things could have been different.
This is a revisionist view of history. At least in one instance (Compaq), as revealed in the antitrust trial, MS threatened a computer OEM with withdrawn or more expensive licenses if it did not stop installing Netscape on their computers. In this case, people did not start using IExplorer because of choice (it was not as good as netscape at the time), but because that was what came with their computers. Couple this with (artificially) embedding Iexporer in their OS (even though they continued to offer a standalone version), and dropping the cost to OEM's to zero (losing money on a product and cutting off Netscape's air supply), and you have a clearer picture on why IExplorer started to gain market share.
It seems that these issues constantly need to be brought up. People are either ignorant of these bits of history, or they are deliberately ignoring them. After all, if IExplorer was such a great product, why didn't they just let it gain marketshare purely on its merits?
> > Not to mention Sun childishly refusing to submit Java to standards bodies so they could maintain full control of the language. Here's a hint, Sun: it's either proprietary, or it's a standard. Pick one.
> Not quite so simple; Sun had a huge concern that if they submitted Java to a standards body, Microsoft would gain undue influence over it via MS's participation on the committee. Same result (corrupted Java), just a different path.
Keep in mind the leverage MS has over other members of these standardization committees. Just to use an example, would Hewlett-Packard ignore a request from MS to vote a particular way if MS mentioned that the copies of Windows shipped on HP's computers might have to increase, thus increasing the cost of HP's computers, or lowering the profit? Doubtful.
Hmmm... I also see Java being seven years old and.NET being essentially 0. Not a really fair comparison.
I assume you are responding to the issue of many implementors of JVM's, versus only MS and Ximian as implementors of C#. Do you know of anyone who has announced that they will develop a C# clone? If they were to succeed and make $$$, would they not fear MS coming over and taking over that market, as MS has done in the past? I would think that this is a reason that only the most naive companies would embark on such an exercise. Yes, C# may be technically open, but practically speaking is a MS language.
C# does have one major feature that Java does not, meta data and reflection.
Apparently, you knowledge of Java is as extensive as your knowledge of the relative openness of the two languages. Java does have reflection. Every object has a Class object, and can get it throught the getClass() method. From this, you can query for the class and interface hierarchy, and for the methods.
As for openness, it depends on your definition. C# is standardized, but I'll bet that if a big player, such as IBM, were to make extensions to the language and was able to flood the market with their version, Microsoft would not sit on their hands; they would take to the courts, just as Sun did when MS tried to change the way Java worked with their tools and JVM. I see no difference. I do, however, see dozens of implementations of JVM's on the market that are compatible. I don't seen any company other than Ximian stepping forward to provide C#, and we know why that is: MS would squash them.
Java on the other hand, is C++ with garbage collection. A big deal to some C/C++ers who have been living in a hole for the last 30 years, but nothing that will double your producivity.
I wholeheartedly disagree with this statement, because you are making a gross simplification. I am much more productive with Java than with C/C++ (I know both well), and for many more reasons that just garbage collection. These reasons are:
- a comprehensive, well-thought out library. For instance, doing sockets programming in Java is MUCH more straightforward than in C or C++. The same functionality can be had in C++ with some outside libraries, but this is not a standard library. This is true with many more aspects. - built in threads support that is much easier to use than POSIX threads. - excellent support for distributed programming with RMI and CORBA, and both are standard. - Better exception handling - Easier building. I don't know how many hours I have wasted trying to compile and link libraries (such as ACE/TAO) completely under C/C++. I especially like when I get unresolved link errors, but no clue on where I might find these functions. With Java, if I have a missing class, the fully qualified name (i.e. package name) usually indicated where the class should be found. - Excellent documentation of the libraries. I am much more confortable using the JavaDoc-generated Java documentation than doing man ???? on function calls in C++. If I am using a library feature of Java that I have not used before, the J2SE documentation is usually sufficient for me to figure out what classes to use, the the usage patterns for the classes usually intuitive.
Adding a garbage collector to C++ would not even cover 10% of the improvement Java provides over C++, from a productivity standpoint.
Microsoft is now locked in competition with small start-up companies that originally pioneered the Web services field
Usually the mainstream press makes it seem as though Microsoft invented web services (and the internet, for that matter.) Its refreshing to see a little more accuracy in reporting once in a while.
Judge Thomas Penfield Jackson is smarter than you or I, and he studied this issue for a very long time and heard the best arguements money could buy, and he came to the conclusion that there is something very wrong here, and that it is bad for you and I, and that drastic measures are warranted.
Not only that, but the appeals court, with a very conservative reputation, upheld most of the findings of Judge Jackson. The only important thing they overturned was the remedy, but mostly because of insufficient hearings into possible remedies. This doesn't preclude the present judge from recommending the same remedy, although this appears to be unlikely.
IMHO, Netscape realized that they can't compete in a market where their chief competitor is a free download, and so they folded (complaining even to this day that it was all Microsoft's fault). Microsoft did what it does best: kill a competitor, gain a lot of users for their software, and make a ton of money (isn't this what every company strives to do?).
How many times do I need to repeat this? MS told at least one and probably other OEM's that they would not be able to buy Win95 licenses if they continued to install Netscape on their PC's. This immediately "cut off Netscape's air supply". You may have heard about this anti-trust trial, in which these tactics were judged illegal. A monopoly cannot use its monopoly power to limit its competition. I don't know how I can put it any clearer than this.
In Linux, no one forces you to install any browser, or prevents you from installing a browser. Linus Torvalds does not tell Red Hat to not install Opera, that only Mozilla can be installed. This is simply not possible.
Why is it that everybody objected when IE got pre-installed on Windows, yet nobody objected when Netscape was pre-installed?
OEM's were pre-installing Netscape because that was what their customers wanted. When MS forced IE onto the machine, and Netscape off, it was done against the wishes of the OEM's and their customers, thwarting the wishes of the free market.
Only after version 3.0 did the browsers really got into a fight. So yes, Microsoft was an underdog in its browser market until their IE4 and IE5 effectively buried the competition.
IE4 and IE5 were fair browsers, but IE3 was not. Yet, everyone got IE3 because of MS' strongarm tactics. Without those tactics, IE3 would not have had any market share, since Netscape was synonomous with web browser at that time. Without the inroads made (illegally) by IE3, and the sudden drop in revenue for Netscape (OEM's were paying to install the browser), the landscape may very well have been very different in subsequent years.
You can't just ignore what went on in the browser wars from 1996-1999. It has significan impact to the situation now.
I know I'm gonna get flamed to hell for this, but having tried Mozilla, IE, Konqueror and Opera, IE is tops for me in terms of ease of use and stability.
Good, at least you have come to this decision through a comparison of the features of each product. My point, though, is that MS used its monopoly power to close off a distribution channel for Netscape (installation on new PC's by OEM's), and forced the situation so that only their browser could be installed (through "integration"). This way, the only browser that most people (especially new, unsophisticated users) would ever see is IE. They did not have a head-to-head competition on a level playing field in which the browser that is the most useful to the most people wins. MS didn't win a competition, they changed the rules of the game so that they were substantially in their favor, essentially avoiding competition. If their browser was so good (especially in 1996, when this all started), why did they need to integrate the browser and force the OEM's to not install Netscape?
Still doesn't change the fact that IE kicked the crap out of every other browser.
You're missing my point here. Virtually overnight, the market share went from Netscape to IE, not because the consumer decided IE was better; a majority of people made no such decision. MS used its monopoly power to force only IE to be delivered on new PC's. Unsophisticated new users, who would not be likely to download Netscape or even to know of its existence, would stick with what came on their PC's. Many of them probably think MS invented the internet! My point here is that MS unfairly (and illegally, by forcing OEM's to not install Netscape on outgoing PC's) manipulated the market in their favor. Netscape couldn't have competed by coming out with a better browser, since their primary marketing channel had been closed.
If it wasn't for Microsoft and their unfair power to skew the market, we'll all still be paying for Netscape. Or do you forget that Netscape was actually selling the browser before Microsoft got into the game?
You're making Netscape sound like a victim, but at the time, it was Microsoft who was the underdog.
If they were truly the underdog, they would not have been able to skew the market the way they did.
I'm saying that we cannot say that IE won the browser wars because of its quality and usefulness, since it and Netscape were not on a level playing field in which the consumer makes the decisions. MS avoided a good deal of competition by taking the decision out of the hands of those unsophisticated users that were buying PC's in increasing numbers. If there had been a level playing field, perhaps (or not, who's to say) Netscape would have been able to react to IE with improvements, rather than basically rolling over because they were no longer getting paid by computer OEM's to install their browser.
Microsoft won the browser war because IE4 beat the hell out of any other browser that was available at the time.
And I guess the fact that MS:
1) "Integrated" IE into the OS so that you got it whether you wanted it or not, and 2) Threatened the computer OEM's with withheld Windows licenses if they installed Netscape on computers going out the door, thus forcing them to pull Netscape
had nothing to do with it.
Face it, with actions such as these, in which MS used their power to skew the market by shutting off marketing channels, you do not have a fair fight. If MS had played fair AND achieved the market share they have, THEN they would have something to brag about.
In a similar vein, am I the only one who is sick of RMS whining about the naming of Linux? The accepted name is Linux not GNU/Linux. It's out of your hands, RMS. Live with it.
I'm with you on this. Does RMS' stance mean that everything that was built with GNU tools must have the GNU name attached to it? Does it mean that the name of the OS should be changed to reflect the contributers to the kernel and its distributions? If so the name would be miles long.
I think that most everyone would agree that the GNU tools made it possible to build the Linux kernel, just like a MS or Borland compiler enabled the construction of many Windows apps. Saying that this entitles the FSF to determine the name of the product is really stretching it. This is yet another unreasonable stance by RMS that detracts from some of the good things he has to say, and degrades his credibility (and thus influence) in the computing world in general.
IE sucked at first but they were able to quickly, once they put their minds to it, start adding features to improve it, so much so that Netscape couldn't really effectively keep up).
Well, that and they threatened to withhold Windows licenses to OEM's that pre-installed Netscape on PC's that went out, thus cutting off this sort of revenue for Netscape. Having this monopoly power and (illegally) wielding it changes some economic rules.
Slashdot Mirror
Oh really? And that applies to software development as well does it? And it means more productivity as well, right - of course many people are happy to sit in a big open office and chat all day, but do they get more work done?
Amen to that. People that are in favor of cubes or the so-called open arrangements have particular models in mind for how people work, both what they must do to do their jobs, and how individuals are best able to get stuff done. The problem is that their models are wrong for many jobs and for many people. If you are in a cube farm in which everyone is working on the same project, then overhearing someone else's conversation (because you can't not hear it) about project stuff may be informative, but if you are working on a different project, it is a distraction. If everyone is working on jobs they can do relatively quietly, then cubes might be alright, but if you are frequently on telecons or in impromptu, loud, contentious discussions, then you may be a disruptions to your neighbors, of whom there are many. If your job requires you to do relatively rote, menial tasks, then external disruptions may not have much impact, but if you need to have some good focused thinking time, the external disruptions will prevent you from ever getting focused, and you will not be able to do your job. If you are an extrovert who enjoys company, then perhaps you want to hear all the conversations around you, but if you are an introvert who needs quiet time to think, those conversations are disruptive. For many people, I would think any savings in going to a cube farm (assuming that the motivation is $$'s) will be more than offset by losses in productivity.
It would have been nice if the article had given these details. Otherwise, you would have expected the software to have been loaded and verified long before launch day. Leaving out the information you provided made this hold seem to be worse than it really was.
This is nice in theory, but in elections with 3 stong candidates, voters always think: I would like to vote for B, but if I do, it is likely that A will win over C, with B in third, and if I had a choice, I'd rather have C than A. In this manner, you can end up with winner that the majority of the people do not want.
This is the reason that some are pushing for Instant Runoff Voting (IRV). In this method, everyone votes for their top choices, in order. Then, say in a 3 candidate election, after the first round, the third place candidates votes would revert to the second choice on those ballots. In this way, you would eventually come down to a two-way race, and you would get a more accurate picture of who the voters want (or don't want) in office. At the same time, people can list their true favorite as their top choice.
I'm sure IRV would be more complicated to implement, but it would eliminate winners with only 34% of the vote (and possible up to 66% against them.)
Isn't The Canopy Group a venture capital firm, and aren't the companies that you list companies that have received investments from Canopy? It may be tempting to lash out at these companies, but these companies probably have nothing to do with the SCO-IBM lawsuit, and in some cases (such as TrollTech) the linux community at large benefits for the contributions of these companies. To try to punish them would be like cutting off your nose to spite your face.
Here are a few of the more notable Java Processors ones:
www.ajile.com (and www.systronix.com has a few boards built witht the processors).
www.ptsc.com
Also, the Jazelle technology from ARM embeds a java processor section into an ARM chip:
http://www.arm.com/armtech/jazelle?OpenDocument
It matters because it provides ammunition to use with people that will insist on only using Windows and parrot the latest MS FUD, while completely resisting even LOOKING at Linux. The next time they say something to the effect that Linux is useless, can't compete with MS products, etc., you just point to a couple of electronic boxes in their home entertainment center and say, "Oh yeah, well your TiVo is using Linux, your MusicServer is using Linux, ..."
A definition seen on slashdot: MS partner - a victim that MS hasn't gotten to yet.
1) At least one company has solved this problem, with an automatic tool (The name of the tool is J-Integra, from Linar systems, www.linar.com). I believe it uses JNI, but does not require you to write any JNI.
2) With either approach, there has to be a linkup between the DLL and the Java code. MS' approach used an approach that hid everything in the JVM. Could they not have generated JNI code that did this instead? This approach does not necessarily have to make the developer deal directly with the JNI code (i.e. write it). And, it meets the language and JVM specifications, and would work with other Windows-based JVM's.
I suspect that such approaches did not meet MS' business objectives (to kill cross-platform java by promoting a polluted version of java).
3) MS adds extensions for Windows only development, which are optional to developers depending on their target market (HINT: Apple has Cocoa extensions in Java......samething......they are optional)
This is inaccurate and very misleading. The MS JVM added keywords that were not part of Sun's Java, and which Visual J++ would use automatically, meaning that a developer might be writing what he thought was standard Java, when in fact non-standard, MS-only java would be produced. MS was ordered by a court to turn these off, making them an opt in, rather than an opt out. However, these keywords were still not part of the Java language as defined in the language spec. This leads to non-portable Java, and a Java different than the one Sun built and licenses.
Perhaps more important were the J/Direct extensions, which used special comments and the MS compiler to generate annotations the the class files, which the MS runtime used to link in native code, such as windows DLL's. THe problem is not that you could link in native, non-portable code, since standard JNI (Java Native Interface) provides a way for you to do this. THe problem was that the MS approach only worked with MS compilers and MS runtimes. Class files compiled in this manner would not work with Sun or IBM runtimes on Windows environments. Within the Java Virtual Machine specification, it is stated that the class annotations (which MS used) must not change the semantics of the class. In MS' JVM, the annotations would cause a certain behavior, while in the other JVM's, they would be ignored. Clearly, MS' use of the annotations violates the JVM specification.
MS had the option of taking an approach that obeyed the specifications. THere is at least one product that can analyze a DLL and produce the mapping code to make calls on that DLL, using standard Java mechanisms. With their billions in cash, surely they could have easily produced a product that used the same approach. Perhaps a clue came from evidence in the DOJ trial, in which a Microsoft engineer spoke of "blunting the cross-platform java market" by distributing a "polluted" java.
5 years ago Netscape was the king of the browsers -- nothing else came close. When MS released it's browser they out leaped Netscape by offering faster JVM and better compatibility. If Netscape didn't just sit back and watch it's browser being eroded from the market and Sun didn't just put all it's eggs in one basket (Netscape) things could have been different.
This is a revisionist view of history. At least in one instance (Compaq), as revealed in the antitrust trial, MS threatened a computer OEM with withdrawn or more expensive licenses if it did not stop installing Netscape on their computers. In this case, people did not start using IExplorer because of choice (it was not as good as netscape at the time), but because that was what came with their computers. Couple this with (artificially) embedding Iexporer in their OS (even though they continued to offer a standalone version), and dropping the cost to OEM's to zero (losing money on a product and cutting off Netscape's air supply), and you have a clearer picture on why IExplorer started to gain market share.
It seems that these issues constantly need to be brought up. People are either ignorant of these bits of history, or they are deliberately ignoring them. After all, if IExplorer was such a great product, why didn't they just let it gain marketshare purely on its merits?
> > Not to mention Sun childishly refusing to submit Java to standards bodies so they could maintain full control of the language. Here's a hint, Sun: it's either proprietary, or it's a standard. Pick one.
> Not quite so simple; Sun had a huge concern that if they submitted Java to a standards body, Microsoft would gain undue influence over it via MS's participation on the committee. Same result (corrupted Java), just a different path.
Keep in mind the leverage MS has over other members of these standardization committees. Just to use an example, would Hewlett-Packard ignore a request from MS to vote a particular way if MS mentioned that the copies of Windows shipped on HP's computers might have to increase, thus increasing the cost of HP's computers, or lowering the profit? Doubtful.
Hmmm... I also see Java being seven years old and .NET being essentially 0. Not a really fair comparison.
I assume you are responding to the issue of many implementors of JVM's, versus only MS and Ximian as implementors of C#. Do you know of anyone who has announced that they will develop a C# clone? If they were to succeed and make $$$, would they not fear MS coming over and taking over that market, as MS has done in the past? I would think that this is a reason that only the most naive companies would embark on such an exercise. Yes, C# may be technically open, but practically speaking is a MS language.
C# does have one major feature that Java does not, meta data and reflection.
Apparently, you knowledge of Java is as extensive as your knowledge of the relative openness of the two languages. Java does have reflection. Every object has a Class object, and can get it throught the getClass() method. From this, you can query for the class and interface hierarchy, and for the methods.
As for openness, it depends on your definition. C# is standardized, but I'll bet that if a big player, such as IBM, were to make extensions to the language and was able to flood the market with their version, Microsoft would not sit on their hands; they would take to the courts, just as Sun did when MS tried to change the way Java worked with their tools and JVM. I see no difference. I do, however, see dozens of implementations of JVM's on the market that are compatible. I don't seen any company other than Ximian stepping forward to provide C#, and we know why that is: MS would squash them.
Java on the other hand, is C++ with garbage collection. A big deal to some C/C++ers who have been living in a hole for the last 30 years, but nothing that will double your producivity.
I wholeheartedly disagree with this statement, because you are making a gross simplification. I am much more productive with Java than with C/C++ (I know both well), and for many more reasons that just garbage collection. These reasons are:
- a comprehensive, well-thought out library. For instance, doing sockets programming in Java is MUCH more straightforward than in C or C++. The same functionality can be had in C++ with some outside libraries, but this is not a standard library. This is true with many more aspects.
- built in threads support that is much easier to use than POSIX threads.
- excellent support for distributed programming with RMI and CORBA, and both are standard.
- Better exception handling
- Easier building. I don't know how many hours I have wasted trying to compile and link libraries (such as ACE/TAO) completely under C/C++. I especially like when I get unresolved link errors, but no clue on where I might find these functions. With Java, if I have a missing class, the fully qualified name (i.e. package name) usually indicated where the class should be found.
- Excellent documentation of the libraries. I am much more confortable using the JavaDoc-generated Java documentation than doing man ???? on function calls in C++. If I am using a library feature of Java that I have not used before, the J2SE documentation is usually sufficient for me to figure out what classes to use, the the usage patterns for the classes usually intuitive.
Adding a garbage collector to C++ would not even cover 10% of the improvement Java provides over C++, from a productivity standpoint.
The NY TImes article has the following line:
Microsoft is now locked in competition with small start-up companies that originally pioneered the Web services field
Usually the mainstream press makes it seem as though Microsoft invented web services (and the internet, for that matter.) Its refreshing to see a little more accuracy in reporting once in a while.
Judge Thomas Penfield Jackson is smarter than you or I, and he studied this issue for a very long time and heard the best arguements money could buy, and he came to the conclusion that there is something very wrong here, and that it is bad for you and I, and that drastic measures are warranted.
Not only that, but the appeals court, with a very conservative reputation, upheld most of the findings of Judge Jackson. The only important thing they overturned was the remedy, but mostly because of insufficient hearings into possible remedies. This doesn't preclude the present judge from recommending the same remedy, although this appears to be unlikely.
IMHO, Netscape realized that they can't compete in a market where their chief competitor is a free download, and so they folded (complaining even to this day that it was all Microsoft's fault). Microsoft did what it does best: kill a competitor, gain a lot of users for their software, and make a ton of money (isn't this what every company strives to do?).
How many times do I need to repeat this? MS told at least one and probably other OEM's that they would not be able to buy Win95 licenses if they continued to install Netscape on their PC's. This immediately "cut off Netscape's air supply". You may have heard about this anti-trust trial, in which these tactics were judged illegal. A monopoly cannot use its monopoly power to limit its competition. I don't know how I can put it any clearer than this.
In Linux, no one forces you to install any browser, or prevents you from installing a browser. Linus Torvalds does not tell Red Hat to not install Opera, that only Mozilla can be installed. This is simply not possible.
Why is it that everybody objected when IE got pre-installed on Windows, yet nobody objected when Netscape was pre-installed?
OEM's were pre-installing Netscape because that was what their customers wanted. When MS forced IE onto the machine, and Netscape off, it was done against the wishes of the OEM's and their customers, thwarting the wishes of the free market.
Only after version 3.0 did the browsers really got into a fight. So yes, Microsoft was an underdog in its browser market until their IE4 and IE5 effectively buried the competition.
IE4 and IE5 were fair browsers, but IE3 was not. Yet, everyone got IE3 because of MS' strongarm tactics. Without those tactics, IE3 would not have had any market share, since Netscape was synonomous with web browser at that time. Without the inroads made (illegally) by IE3, and the sudden drop in revenue for Netscape (OEM's were paying to install the browser), the landscape may very well have been very different in subsequent years.
You can't just ignore what went on in the browser wars from 1996-1999. It has significan impact to the situation now.
I know I'm gonna get flamed to hell for this, but having tried Mozilla, IE, Konqueror and Opera, IE is tops for me in terms of ease of use and stability.
Good, at least you have come to this decision through a comparison of the features of each product. My point, though, is that MS used its monopoly power to close off a distribution channel for Netscape (installation on new PC's by OEM's), and forced the situation so that only their browser could be installed (through "integration"). This way, the only browser that most people (especially new, unsophisticated users) would ever see is IE. They did not have a head-to-head competition on a level playing field in which the browser that is the most useful to the most people wins. MS didn't win a competition, they changed the rules of the game so that they were substantially in their favor, essentially avoiding competition. If their browser was so good (especially in 1996, when this all started), why did they need to integrate the browser and force the OEM's to not install Netscape?
Still doesn't change the fact that IE kicked the crap out of every other browser.
You're missing my point here. Virtually overnight, the market share went from Netscape to IE, not because the consumer decided IE was better; a majority of people made no such decision. MS used its monopoly power to force only IE to be delivered on new PC's. Unsophisticated new users, who would not be likely to download Netscape or even to know of its existence, would stick with what came on their PC's. Many of them probably think MS invented the internet! My point here is that MS unfairly (and illegally, by forcing OEM's to not install Netscape on outgoing PC's) manipulated the market in their favor. Netscape couldn't have competed by coming out with a better browser, since their primary marketing channel had been closed.
If it wasn't for Microsoft and their unfair power to skew the market, we'll all still be paying for Netscape. Or do you forget that Netscape was actually selling the browser before Microsoft got into the game?
You're making Netscape sound like a victim, but at the time, it was Microsoft who was the underdog.
If they were truly the underdog, they would not have been able to skew the market the way they did.
I'm saying that we cannot say that IE won the browser wars because of its quality and usefulness, since it and Netscape were not on a level playing field in which the consumer makes the decisions. MS avoided a good deal of competition by taking the decision out of the hands of those unsophisticated users that were buying PC's in increasing numbers. If there had been a level playing field, perhaps (or not, who's to say) Netscape would have been able to react to IE with improvements, rather than basically rolling over because they were no longer getting paid by computer OEM's to install their browser.
Microsoft won the browser war because IE4 beat the hell out of any other browser that was available at the time.
And I guess the fact that MS:
1) "Integrated" IE into the OS so that you got it whether you wanted it or not, and
2) Threatened the computer OEM's with withheld Windows licenses if they installed Netscape on computers going out the door, thus forcing them to pull Netscape
had nothing to do with it.
Face it, with actions such as these, in which MS used their power to skew the market by shutting off marketing channels, you do not have a fair fight. If MS had played fair AND achieved the market share they have, THEN they would have something to brag about.
In a similar vein, am I the only one who is sick of RMS whining about the naming of Linux? The accepted name is Linux not GNU/Linux. It's out of your hands, RMS. Live with it.
I'm with you on this. Does RMS' stance mean that everything that was built with GNU tools must have the GNU name attached to it? Does it mean that the name of the OS should be changed to reflect the contributers to the kernel and its distributions? If so the name would be miles long.
I think that most everyone would agree that the GNU tools made it possible to build the Linux kernel, just like a MS or Borland compiler enabled the construction of many Windows apps. Saying that this entitles the FSF to determine the name of the product is really stretching it. This is yet another unreasonable stance by RMS that detracts from some of the good things he has to say, and degrades his credibility (and thus influence) in the computing world in general.
IE sucked at first but they were able to quickly, once they put their minds to it, start adding features to improve it, so much so that Netscape couldn't really effectively keep up).
Well, that and they threatened to withhold Windows licenses to OEM's that pre-installed Netscape on PC's that went out, thus cutting off this sort of revenue for Netscape. Having this monopoly power and (illegally) wielding it changes some economic rules.
Oh, and I have a problem with dangling prepositions too.
Ending a sentence with a preposition is something up with which we will not put. - Winston Churchill