I'm interested to see if TransVirtual posts anything on their website about this, since news.com seems to not have the whole picture.
Here's what I know:,/p>
Kaffe certainly isn't a new product. As some other commenter mentions, it is a stardard part of Red Hat's distribution, and I'm sure others as well.
Kaffe is open sourced under the GPL. Which is interesting, since that means (to the best of my understanding) that TransVirtual could not use its own code to make a Windows version of Kaffe unless it releases that Windows version as an open source project. And actually, according to TransVirtual they already have a GPL version of Kaffe available for Windows95.
TransVirtual doesn't mention being funded by Microsoft on its website. That doesn't make it untrue, of course. Maybe Microsoft funded them so that there would be non-Sun Java implementations out there a long time ago. And now have turned the screws a little bit to make them make a press announcement about their Windows version.
TransVirtual's web server is running awfully slow this morning. They appear to have been Slashdotted/News.commed.
The bottom line: it sounds like a publicity play with nothing particularly new or interesting happening.
Yes, there are non-Sun implementations of Java. Yes, some of them work on Windows. No, this isn't anything new. Perhaps, M$ is making Kaffe support its little non-standard extensions. But, it doesn't really matter since no one will want to use them anyway. And they'll be open sourced as a twisted little bit of irony.
I didn't buy this book for a long time, for this very reason. If I needed to jog my memory, or look up an argument, then I'd use the man pages.
But now I've started working from home. And I do a lot of my work on a Windows box. (There's some applications that are only available under Win32 that I just can't get around needing. It usually doesn't bother me too much since UNIX is just a telnet client away anyway.) So I spend a lot of my time telnetted into UNIX boxes over a dial-up connection. Which is a pain since searching through a few man entries, with a half dozen screens of text each, can be slow enough to be annoying.
I used to get around this by just reaching over to the next keyboard, which is my home Linux box. But on a whim, and based off of my success with Java in a Nutshell, I decided to pick up UNIX in a Nutshell. And I've used it more than I thought I would. Sometimes it's just handy to have a dead-tree version of the docs around. Screen real estate is just too valuable sometimes. If I'm in the middle of typing a long command line, sometimes it's easier to flip open a book to verify the parameter instead of opening a new telnet session or window to fire up man. I've even considered getting the UNIX CD Bookshelf in which UNIX in a Nutshell is included in both hard copy and electronic copy. I haven't, since UNIX in a nutshell is probably the most valuable reference book of the bunch, and I've already forked over my $20 to O'Reilly for that one.
The bottom line: There's nothing groundbreaking about this book. All of the information is available elsewhere, and for free. Because of that, it's probably not an O'Reilly "gem". But if spend a decent amount of time in man, and you have 20 bucks (plus shipping) to spare, it's a well compiled reference that might be worthwhile to add to your library.
>I'm not surprised that a Netscape employee won't >admit this, but there are lots of benchmarks from >unbiased sources that show IIS to reign surpreme >serving in both the static and dynamic web page >arenas.
I did unleash a little fury on IIS, didn't I. I didn't mean to turn this thread into IIS bashing. I've just had some really bad experiences with IIS. It's way too integrated into the operating system for my taste. (Anytime that you are using the operating system's user database to authentiate web users, something is just plain wrong.) And the fact that the operating system that it is integrated into is NT just makes it worse. It's just not stable, manageable, or scalable enough.
Your claim that IIS is the fastest webserver is pure flamebait. If you would like to point me to a URL, I'd be happy to look at it. But, I'll guarantee that the fastest webserver isn't any webserver running on NT. Apache, Zeus, and Netscape can all crush IIS, simply by the fact that they can run on high-end UNIX machines with a dozen or more processors.
It doesn't really matter anyway though. I doubt that anyone makes their choice of web servers based on performance tests anyway. The difference in performance is small enough that features, managability, and stability are going to be decision points.
For those of you who brought up PHP's effect on the webserver comparisons, I concede that PHP may be significant development. To be honest, I don't know enough of the details about PHP to comment on its strengths and weaknesses. I'm a servlet kind of guy myself, so I haven't checked it out. But it does seem to be a simpler solution than mod_perl. (Not that Perl isn't cool.) At first look, it just looked like another server-side scripting language to me. Which is cool in it's own right since it's open source. But people tell me that it rocks.
By the way, since I didn't explicitly say it before: my opinions are mine, not Netscape's. I don't work on the development team for Enterprise server, and am not a professional webmaster either, so my opinion probably isn't any good anyway.
Ballmer is not saying that Apache is better, merely conceding the virtual hosting market to Apache. Which Apache wins hands down. His remarks are an interesting insight into the competition between the three major webservers. You have only three serious competitors: IIS, Netscape Enterprise Server, and Apache. And each has its own niche that it dominates in, leaving only a small intersection of real competition.
Apache is an interesting success story. It's one of the most successful open source projects of all time. Apache dominates certain spaces of the web server market, such as the virtual hosting space that Ballmer is conceding, along with the ISP market, and the "Slashdot" space. By the Slashdot space, I mean sites that have highly talented administrators and programmers, lots of traffic, but little money. mod_perl gives Apache an enormous adantages. Shashdot is doing things in mod_perl that other webservers can only offer through expensive application servers. But Apache is reasonably complicated, especially when you starting adding things like mod_perl to the mix. If you know what you are doing, it's great, but there's an intimidating learning curve.
Netscape Enterprise Server Since I work for Netscape I won't say much about Netscape's webserver, since people would think that I'm too biased anyway. But, in short, Netscape seems to cover the exact opposite of Apache's market. People who have the money to spend on a webserver (Netscape is the only webserver you have to pay money for), and are willing to pay that money in exchange for easier manageability and a formal support agreement.
Microsoft's IIS is the most interesting of the three. It's the quicksand solution. There are only three reasons to use IIS:
It was pre-installed on my server. What a lame reason. But lots of people use IIS for a workgroup solution because of this reason.
I'm building a Microsoft-centric solution. Once you start building an ASP-based solution, you're stuck. You have nowhere else to turn. The cost of converting to Perl or JavaScript is just too high. You've walled yourself into a proprietary solution.
It's what I know. Microsoft makes a lot of money on the database server and workgroup server market, because "any idiot can use it".
In short, M$'s webserver is a piece of crap. And everyone knows it, but some people are in a world where they either don't care, or don't have a choice.
So Ballmer can concede the virtual hosting market all that he wants. It doesn't matter to him. Those people need a real solution anyway. His customers are the uneducated and the ones that are boxed in to M$ solutions.
The competition between Apache and Netscape is interesting because they so rarely compete. The best way to sum up this up is to quote a sysadmin friend of mine. He said, "If I was going to run a website myself, I'd pick Apache. It's powerful and fast and stable. But if I had to run a site for a company, I'd pick Netscape. Hiring people qualified with whatever flavor or Perl/Apache/UNIX/mod_perl I choose is too hard and too expensive."
I don't know if I agree with him completely. The actual feature to feature comparison is interesting, but I'm too biased to make that comparison here. But it shows the marketplace's attitude: that the decision between Apache and Netscape is a choice between whether to spend money on a product or on administrators.
I misunderstood the ruling at first, and this parent post helped me really understand what the judge was trying to indicate that M$ was allowed to do. I'm glad to hear that there was no disruption of Sun's patent on Java.
So, that leaves us with the question of what benefit a Java-clone would give Microsoft. (I believe that M$ did announce a long time ago that future versions of MSJ++ might not use the Java name. I wish I could remember more about that.) We all know that M$ has decided to use a "embrace, extend, extinguish" strategy on Java. But M$ has little to gain in continuing this strategy. Especially since Java seems to have found a sweet spot different than many people thought. (More on that later.)
Assume that M$ does develop a Latte language. And invests the large amount of developer effort to develop a clean room clone of JDK 1.1 and a clean-room clone of a Win32 JVM 1.1. M$ then has to decide how quickly to extend. Like remande points out above, this will be very tricky for M$. No IT manager will trust M$ to keep "Latte" compatible, or even to keep investing money in the "Latte" product (since VB is blantantly M$'s flagship development tool). So M$ will have to keep "Latte" relatively close to the Java standard, since few people will be willing to use the extensions in the beginning.
But this is a double-edged sword for M$. If "Latte" is just a straight clone of Java, then what will the compelling reason be to use it. "Latte" will become the white elephant that client-side VBScript has become. No one uses client-side VBScript because it's only supported by a fraction of the population and there's a suitable competitor that's widely supported. (Yes, I'm sure there are a few client-side VBScript exceptions, but they are very rare.)
There is one place that "Latte" could be successful: and that's on the middle-tier of NT-based web applications. Java is becoming enormously popular in both application and database servers for capturing business logic. It's just a great language for doing that kind of work, especially since its so extensible and cross-platform. (Servlets rock!)
"Latte" could be built with extensions relevant to the NT server market, and do reasonably well in the IIS Servlet market. Building a "Latte"-based servlet engine into IIS that the Latte++ development tool integrated with would be a powerfully strong move from M$. But, M$ has already shown its direction for the NT server-based web application: Visual Basic!
Bill would view "Latte"-based servlets as a threat to his world-view of a VB dominated world. Strangely enough, M$ seems to be promoting Visual Basic as its web development language. In his little COM-based dreamland, Visual Basic is easier to develop in than Java. And it's better suited for the web. Until Bill sees a need to ditch that strategy, (probably when Java developers outnumber VB developers) VisualJ++ and "Latte" will just be sidetracks to him.
I don't get this ruling. As everybody here has already pointed out, a "clean room" project is one where the clean room project is designed to create a functionally identical product that does not infringe on the intellectual design property of the original company.
The classic examples (already cited) were the IBM BIOS and the Intel x86 designs. Competitors created "knock-offs" that were functionally compatible.
Sun's already allowing this kind of clean-room effort. I think HP is already working on a version. The catch is that you have to be able to prove to Sun that you are functionally compatible in order to use the Java brand name. And Sun is charging quite a bit of money to perform the functional compatibility tests.
Perhaps the judge is saying that Microsoft can use the brand name of Java without passing the functionality compatibility tests, as long as they are actually compatibile. (In other words, Sun can't charge Microsoft money for the privilege of using the name.)
If, on the other hand, the judge is saying that Microsoft can do whatever it wants with Java, as long as it doesn't use Sun's source code; this stinks. It means that Microsoft will continue its "not quite compatible" track, and 100% Java programs may or may not work on Windows. (Which is exactly the kind of FUD, M$ is trying to create.)
Slashdot Mirror
A direct result of the court ruling I guess.
I'm interested to see if TransVirtual posts anything on their website about this, since news.com seems to not have the whole picture.
Here's what I know:,/p>
The bottom line: it sounds like a publicity play with nothing particularly new or interesting happening.
Yes, there are non-Sun implementations of Java. Yes, some of them work on Windows. No, this isn't anything new. Perhaps, M$ is making Kaffe support its little non-standard extensions. But, it doesn't really matter since no one will want to use them anyway. And they'll be open sourced as a twisted little bit of irony.
I didn't buy this book for a long time, for this very reason. If I needed to jog my memory, or look up an argument, then I'd use the man pages.
But now I've started working from home. And I do a lot of my work on a Windows box. (There's some applications that are only available under Win32 that I just can't get around needing. It usually doesn't bother me too much since UNIX is just a telnet client away anyway.) So I spend a lot of my time telnetted into UNIX boxes over a dial-up connection. Which is a pain since searching through a few man entries, with a half dozen screens of text each, can be slow enough to be annoying.
I used to get around this by just reaching over to the next keyboard, which is my home Linux box. But on a whim, and based off of my success with Java in a Nutshell, I decided to pick up UNIX in a Nutshell. And I've used it more than I thought I would. Sometimes it's just handy to have a dead-tree version of the docs around. Screen real estate is just too valuable sometimes. If I'm in the middle of typing a long command line, sometimes it's easier to flip open a book to verify the parameter instead of opening a new telnet session or window to fire up man. I've even considered getting the UNIX CD Bookshelf in which UNIX in a Nutshell is included in both hard copy and electronic copy. I haven't, since UNIX in a nutshell is probably the most valuable reference book of the bunch, and I've already forked over my $20 to O'Reilly for that one.
The bottom line: There's nothing groundbreaking about this book. All of the information is available elsewhere, and for free. Because of that, it's probably not an O'Reilly "gem". But if spend a decent amount of time in man, and you have 20 bucks (plus shipping) to spare, it's a well compiled reference that might be worthwhile to add to your library.
>admit this, but there are lots of benchmarks from
>unbiased sources that show IIS to reign surpreme
>serving in both the static and dynamic web page >arenas.
I did unleash a little fury on IIS, didn't I. I didn't mean to turn this thread into IIS bashing. I've just had some really bad experiences with IIS. It's way too integrated into the operating system for my taste. (Anytime that you are using the operating system's user database to authentiate web users, something is just plain wrong.) And the fact that the operating system that it is integrated into is NT just makes it worse. It's just not stable, manageable, or scalable enough.
Your claim that IIS is the fastest webserver is pure flamebait. If you would like to point me to a URL, I'd be happy to look at it. But, I'll guarantee that the fastest webserver isn't any webserver running on NT. Apache, Zeus, and Netscape can all crush IIS, simply by the fact that they can run on high-end UNIX machines with a dozen or more processors.
It doesn't really matter anyway though. I doubt that anyone makes their choice of web servers based on performance tests anyway. The difference in performance is small enough that features, managability, and stability are going to be decision points.
For those of you who brought up PHP's effect on the webserver comparisons, I concede that PHP may be significant development. To be honest, I don't know enough of the details about PHP to comment on its strengths and weaknesses. I'm a servlet kind of guy myself, so I haven't checked it out. But it does seem to be a simpler solution than mod_perl. (Not that Perl isn't cool.) At first look, it just looked like another server-side scripting language to me. Which is cool in it's own right since it's open source. But people tell me that it rocks.
By the way, since I didn't explicitly say it before: my opinions are mine, not Netscape's. I don't work on the development team for Enterprise server, and am not a professional webmaster either, so my opinion probably isn't any good anyway.
Ballmer is not saying that Apache is better, merely conceding the virtual hosting market to Apache. Which Apache wins hands down. His remarks are an interesting insight into the competition between the three major webservers. You have only three serious competitors: IIS, Netscape Enterprise Server, and Apache. And each has its own niche that it dominates in, leaving only a small intersection of real competition.
Apache is an interesting success story. It's one of the most successful open source projects of all time. Apache dominates certain spaces of the web server market, such as the virtual hosting space that Ballmer is conceding, along with the ISP market, and the "Slashdot" space. By the Slashdot space, I mean sites that have highly talented administrators and programmers, lots of traffic, but little money. mod_perl gives Apache an enormous adantages. Shashdot is doing things in mod_perl that other webservers can only offer through expensive application servers. But Apache is reasonably complicated, especially when you starting adding things like mod_perl to the mix. If you know what you are doing, it's great, but there's an intimidating learning curve.
Netscape Enterprise Server Since I work for Netscape I won't say much about Netscape's webserver, since people would think that I'm too biased anyway. But, in short, Netscape seems to cover the exact opposite of Apache's market. People who have the money to spend on a webserver (Netscape is the only webserver you have to pay money for), and are willing to pay that money in exchange for easier manageability and a formal support agreement.
Microsoft's IIS is the most interesting of the three. It's the quicksand solution. There are only three reasons to use IIS:
In short, M$'s webserver is a piece of crap. And everyone knows it, but some people are in a world where they either don't care, or don't have a choice.
So Ballmer can concede the virtual hosting market all that he wants. It doesn't matter to him. Those people need a real solution anyway. His customers are the uneducated and the ones that are boxed in to M$ solutions.
The competition between Apache and Netscape is interesting because they so rarely compete. The best way to sum up this up is to quote a sysadmin friend of mine. He said, "If I was going to run a website myself, I'd pick Apache. It's powerful and fast and stable. But if I had to run a site for a company, I'd pick Netscape. Hiring people qualified with whatever flavor or Perl/Apache/UNIX/mod_perl I choose is too hard and too expensive."
I don't know if I agree with him completely. The actual feature to feature comparison is interesting, but I'm too biased to make that comparison here. But it shows the marketplace's attitude: that the decision between Apache and Netscape is a choice between whether to spend money on a product or on administrators.
I misunderstood the ruling at first, and this parent post helped me really understand what the judge was trying to indicate that M$ was allowed to do. I'm glad to hear that there was no disruption of Sun's patent on Java.
So, that leaves us with the question of what benefit a Java-clone would give Microsoft. (I believe that M$ did announce a long time ago that future versions of MSJ++ might not use the Java name. I wish I could remember more about that.) We all know that M$ has decided to use a "embrace, extend, extinguish" strategy on Java. But M$ has little to gain in continuing this strategy. Especially since Java seems to have found a sweet spot different than many people thought. (More on that later.)
Assume that M$ does develop a Latte language. And invests the large amount of developer effort to develop a clean room clone of JDK 1.1 and a clean-room clone of a Win32 JVM 1.1. M$ then has to decide how quickly to extend. Like remande points out above, this will be very tricky for M$. No IT manager will trust M$ to keep "Latte" compatible, or even to keep investing money in the "Latte" product (since VB is blantantly M$'s flagship development tool). So M$ will have to keep "Latte" relatively close to the Java standard, since few people will be willing to use the extensions in the beginning.
But this is a double-edged sword for M$. If "Latte" is just a straight clone of Java, then what will the compelling reason be to use it. "Latte" will become the white elephant that client-side VBScript has become. No one uses client-side VBScript because it's only supported by a fraction of the population and there's a suitable competitor that's widely supported. (Yes, I'm sure there are a few client-side VBScript exceptions, but they are very rare.)
There is one place that "Latte" could be successful: and that's on the middle-tier of NT-based web applications. Java is becoming enormously popular in both application and database servers for capturing business logic. It's just a great language for doing that kind of work, especially since its so extensible and cross-platform. (Servlets rock!)
"Latte" could be built with extensions relevant to the NT server market, and do reasonably well in the IIS Servlet market. Building a "Latte"-based servlet engine into IIS that the Latte++ development tool integrated with would be a powerfully strong move from M$. But, M$ has already shown its direction for the NT server-based web application: Visual Basic!
Bill would view "Latte"-based servlets as a threat to his world-view of a VB dominated world. Strangely enough, M$ seems to be promoting Visual Basic as its web development language. In his little COM-based dreamland, Visual Basic is easier to develop in than Java. And it's better suited for the web. Until Bill sees a need to ditch that strategy, (probably when Java developers outnumber VB developers) VisualJ++ and "Latte" will just be sidetracks to him.
I don't get this ruling. As everybody here has already pointed out, a "clean room" project is one where the clean room project is designed to create a functionally identical product that does not infringe on the intellectual design property of the original company.
The classic examples (already cited) were the IBM BIOS and the Intel x86 designs. Competitors created "knock-offs" that were functionally compatible.
Sun's already allowing this kind of clean-room effort. I think HP is already working on a version. The catch is that you have to be able to prove to Sun that you are functionally compatible in order to use the Java brand name. And Sun is charging quite a bit of money to perform the functional compatibility tests.
Perhaps the judge is saying that Microsoft can use the brand name of Java without passing the functionality compatibility tests, as long as they are actually compatibile. (In other words, Sun can't charge Microsoft money for the privilege of using the name.)
If, on the other hand, the judge is saying that Microsoft can do whatever it wants with Java, as long as it doesn't use Sun's source code; this stinks. It means that Microsoft will continue its "not quite compatible" track, and 100% Java programs may or may not work on Windows. (Which is exactly the kind of FUD, M$ is trying to create.)