Slashdot Mirror
Oracle and Sun Team Up to Provide .NET Alternative
segphault writes "Ars Technica has an article about the new partnership between Sun and Oracle, designed to provide an alternative to .NET." From the article: "According to Ellison and McNealy, their mutual goal is the production of a complete Java-centric enterprise datacenter architecture that leverages Solaris 10 and Oracle's Fusion middleware. Designed specifically as an alternative to Microsoft's .NET technology stack, the new platform is competitively priced and based on robust frameworks."
"According to Ellison, this is all about providing users and developers with technology based on standards. But what standards is he talking about, and are those the standards that consumers care about? The availability of an open source .NET implementation based on ECMA standards certainly makes Java look more proprietary."
.NET has been proposed to the ECMA, which is not even a standard organization. Mono provides only a small subset of .NET.
The whole JDK1.5 API is public and totaly available to be implemented by anyone (www.jcp.org). Also there is already a 98%-complete implementation of it (www.classpath.org). OTOH, only a small part of
(that said, the most used Java Platform (Sun) is still proprietary)
Million Dollar Screenshot
...is the sincerest form of flattery.
Rather than teaming with Larry Elliscum, a better move for Sun would be to open Java up to the ECMA/ISO for standardization.
Tech, life, family, faith: Give me a visit
i for one am sick of dealing with classpaths and 250 jars inside of jar files inside of war files inside of ear files - catch my drift.
.net has, and something like a GAC.
.net after 6 years of dealing with Sun's bullcrap and i have never looked back.
i'm also sick of J2EE containers with class loaders schemes that are more complicated than my senior year algebraic structures course.
build a linker into java just like
than allow versioning of libraries.
then get rid of checked exceptions so i don't have to do try/catch/wrap/rethrows(or do nothing) in 90% of my J2EE code.
then get rid of stateful, local session beans - how redudant is that???
then find a way to get rid of the 14 million defines i need in my server.xml to specify which implementation of each 'open, standard' interface i need
so, java as a language - it's ok
java as a platform - SUCKS!!!!!!!!!!!!!!!!!!!!!!
left java for
Why exactly would that help? Right now the Java standards are open to input from a wide range of voices, from individual developers through open source communities like Apache to corporations like Oracle and IBM. No voice has overall control, no-one can force through self-serving capabilities and everyone gets to use the specifications royalty free. All of them know their contributions can be implemented as open source yet that the market in which they operate can't be monopolised by any single company.
Sun started ECMA standardisation and then realised half-way through the process that it was going to produce the worst of all worlds; a rubber-stamp for the work Sun had done, with no input from any communities and a freezing of the specs by the ECMA dinosaur, combined with a loss of the ability to enforce the Java trademark and an inevitable embrace-and-extend by companies like Microsoft and IBM. Sun should have worked this out before starting with ECMA but fortunately realised in time and pulled out of the process. The result was the creation of the JCP and the most open, competitive software market the computer industry has yet seen.
Microsoft fully understands the PR value of ECMA and is cynically using it to rubber stamp it's Office 12 XML format to undermine the openness of OpenDocument. That action has done us the good service of showing us just how intellectually bankrupt ECMA actually is. What the Java platform needs is not the destruction ECMA would bring, but rather the further evolution of the JCP, which is working better than pretty much any standards body before it and is only hampered by the public perception of Sun control.
Urrm... having trouble reading, are you?From the message you just replied to: ".net's CLI/CLR may have its flaws, but it fixes much of the brain damage of the original JVM design."
.NET/CLR is very similar. It may have a few more features that better suit other languages, but they are few. Although there have been successful implementations of dynamic languages on the CLR (IronPython), many dynamic languages (like Smalltalk) have struggled to make use of the CLR, and have had to make compromises.
The problem with this statement is that it assumes that the JVM design was 'brain damaged'. It wasn't. There were very specific design considerations which it met well: To be easily JITted/translated to native code, to be a compact code and the well suit procedural/OOP language Java.
The
The idea that the JVM is 'brain damaged' is just ranting. The sheer number of languages successfully implemented on the JVM provide the clearest possible evidence against this statement. Things could certainly be better, and it is likely that new opcodes will be added to future versions to make implementation of dynamic languages easier.
But anyway, one developer's 'brain damage' is another developer's sensible choice, so using such terms is meaningless.