Slashdot Mirror

Add to that openlazlo a bunch of other webframeworks which are moving into the cross platform domain, add to that echo2, and http://blogs.sun.com/chrisoliver/category/F3 ...

Id rather see something truly open being supported instead of yet cloning another windows forever mac for a limited time technology into Linux until Microsoft starts suing it into oblivion once they have enough marketshare!

I posted a comment to Jonathan Schwartz's blog asking him to open source it. This was attached to one of his blog entries about open source and java. That comment didn't even make it past moderation.

That's Foo.getAdapter() you moron.

Thanks, but I'll take the consistent syntax of C# properties any day. (And yes, I understand that it's just the legacy of the well-developed language, and that properties are just syntactic sugar. But why not enjoy the sugar? It's tasty...)

It's too bad there's not some way to make Swing look like native widgets. Oh, wait:

http://java.sun.com/docs/books/tutorial/uiswing/lo okandfeel/plaf.html

Of course, there was also the shrine to Apollo at Clarus.

Don't forget that the Oracle at Delphi was a shrine to Apollo, who killed the Python.

Also, Apollo helped the Trojans during their war with Greece. He aided the killing of Achilles, who was a student of Phoenix.

Appollo was born in part due to a necklace of amber.

Apollo was known as the god of the sun which is a star, and he also was known as the one who brought the mice to the people.

Much about Apollo is known from the record of him in the Odyssey and the Illiad. In the Odyssey, the sun god was known as Helios.

here are many advantages to AIX: cheaper hardware, powerful POWER5 architecture to run on (IBM hardware scales quite nicely), decent support, and it is maintained by one of the oldest technology companies in America.
Adding to that list, you can add that they don't try to pull support stunts with hardware and dig their heels in when they transition to a more open form.


IBM still maintains AIX.

They dont play the games in shortening the lifespan like one of their competitors does. Now IBM'd move away from some insecure defaults in authentication...

can you customise the clippyness out of it? http://blogs.sun.com/marigan/entry/how_the_vi_edit or_would

Well, Schwartz must be doing something right. By god, they're actually doing video blogs with lightsabers now!

http://java.sun.com/developer/media/

Sun Ray server runs on Solaris 10 sparc and X86, which BTW includes StarOffice and Mozilla,so editing, email and webbrowsing should be no problem. If you want you can even use some Linux distros. More info at http://www.sun.com/desktop/index.jsp?tab=1

Except the shared bus the Xeons sit on is a seriously limiting factor, no-one in HPC is using Xeons because of it.

A better bet would be a Sun Fire X4600 type of machine, 8 dual-core Opterons and 128GB of memory in a 4U server chassis.

This is well known, and having played with one, it's a very nice machine. Unlike its 24TB cousin ...

Except the shared bus the Xeons sit on is a seriously limiting factor, no-one in HPC is using Xeons because of it.

A better bet would be a Sun Fire X4600 type of machine, 8 dual-core Opterons and 128GB of memory in a 4U server chassis.

This is well known, and having played with one, it's a very nice machine. Unlike its 24TB cousin ...

You must have a 6 core one. You can get 4 and 8 core ones too.

You must have a 6 core one. You can get 4 and 8 core ones too.

24 threads and only one FPU does not make for fast ops at all tasks....

Just a couple of points: Niagara 1 (UltraSPARC T1) has 8 cores, 4 threads per core making 32 threads. Yes, it only has 1 floating point unit (attached to one core), though.

Niagara 2 has 8 cores, but a floating point unit for each core (i.e. 8) and 8 threads per core, making 64 threads.

For floating-point intensive workloads and others, ROCK looks like an interesting and highly-innovative development.

Intel's Core2 architecture is suboptimal and doesn't scale above two cores. Due to the bottleneck inefficiency you lose almost one core when running quad-core. With eight cores it gets even worse.

http://blogs.sun.com/bmseer/entry/woodcrest_scalin g_problems%3F_(fluent_part1)

With appropriate plugins, Office will interoperate with ODF documents --

Like the one mentioned here from Sun.

Lets get the facts right, here. It's not a bug-report, it's an RFE, a Request for Enhancement.

AND IT'S BLOODY FOUR YEARS OLD!!!

See for your self: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id =4802695

To add insult to injury, when the status of this 'RFE' only very recently changed (January 16th 2007), IT WAS TARGETED FOR THE NEXT RELEASE. In other words it'll be another 18 months.

And the first person who tells me I should use a 32-bit browser anyway, I'm going to strangle with Java-code.

Excuse me for losing my cool there, I've been waiting for it to magically apear in every point release of V5 and then in Java 6. Even if this is 'non-trivial', Sun should get their collective heads out of their assess and just do it. There's no telling how many people and projects are being held up by this.

If you have a few minutes, please get an account at SDN and vote for this particular RFE. I'd like the vote to go up to about a thousand at least. Perhaps that'll get them out of snooze-mode. :/

I picked the Common Development and Distribution License for my recently open source projects. I wrote up a rather length justification for this decision (and why I decided against others, like the GPL, LGPL, and BSD license) on my blog.

If I'm not mistaken Solaris is by Sun => of course they have the jdk/re for that. It has long been ranted on the bugs (re: bugtracker ) that 64-bit support is not supported for the 64-bit os's unless you use a binary. It also says at the bottom of that bug that the jdk 1.7 and jre 1.7 will have 64-bit support.

Point re: the java applets was that they are still ubiquitous, thus is very likely someone will have a JVM spawned, besides that, it clearly is not that onerous given the example I gave.
I'm trying to be polite, but you really are a strain on that.

I can't check the 1MiB runtime, so I'll just take your word for it. Implementing multimedia under Linux is not that hard anymore. ALSA has stabilised significantly and making that a requirement would not have been such a big deal.

And not crucial for a web app, we were discussing media streaming. But just typing this makes me feel tired. Feels like you are looking for a fight, so... Goodbye.
http://java.sun.com/products/java-media/jmf/index. jsp

http://java.sun.com/j2se/1.5/pdf/generics-tutorial .pdf one of the best tutorials out there.

I'm sorry, but I don't know what you are smoking. Vector is entirely retrofitted to use generics, and it does it fine. And a comment like "ArrayList is a much better choice nowadays" is complete BS. They are virtually identical, with one important difference, spelled out in the documentation: "This class is roughly equivalent to Vector, except that it is unsynchronized". Meaning that if you ever plan to do anything with threads, stick with Vector. In fact, always stick with Vector. It's a better class, and there is no good reason not to do it. Vector is in no way deprecated, and it never will be.

Second, it's not hard at all to get a Foo[] array from a Vector. Just pass a Foo[] array to the toArray() function and it will fill it. It's dead easy. If you're really lazy you don't even have to size it correctly, the class will do that for you. Just tell it what type it should be in and it will do all the work. This is literally one line of code.

Which explains why Philip Wadler, one of the people responsible for Haskell, was part of a team that wrote GJ (Generic Java), one of the experimental Java mutations (others included PolyJ and Pizza) that, back in the day (late 90s) helped explore how parametric polymorphism could be added to Java, and which formed the basis for the generics introduced in Java 5.

So if you want to understand generics, Wadler is your man.

I think that this a very twisted and silly conclusion. Just because someone wrote or developed something doesn't automaticly imply that he or she is the best person when it comes to explaining everything about it. Quite frankly I think its much more of the opposite. Developers are by definition not the best people to turn to when you wish to really learn whatever they developed. While a developer would try to approach his ideas from a technical point of view most students would be more helped when you begin slow and abstract and eventually work your way up.

Naturally I'm a bit cynical here but heck. To me this has once again nothing to do with an intersting article but simply yet another post to try and push some "interesting" book forward thus hoping that someone is going to make more money out of it.

My opinion in all this is simple. If you're interested in Generics then the best place to start learning is the Sun Java tutorial. In this case the trail about Generics would sound like a good idea to start reading.

And you know whats so good about this tutorial? Its free, its from the company which developed Java (a company which doesn't only have very good developers when it comes to Java but also has people available who are quite skilled on documenting) and you can even download it so that you can read and study even if you don't have a permanent connection to the Internet.

http://java.sun.com/j2se/1.5/pdf/generics-tutorial .pdf

http://java.sun.com/docs/books/tutorial/extra/gene rics/index.html

Also highly worthwhile is http://java.sun.com/docs/books/tutorial/collection s/index.html which describes the generics-enabled standard collection classes, which is 90% of what you'll want generics for anyway.

Share and enjoy.

http://java.sun.com/j2se/1.5/pdf/generics-tutorial .pdf

http://java.sun.com/docs/books/tutorial/extra/gene rics/index.html

Also highly worthwhile is http://java.sun.com/docs/books/tutorial/collection s/index.html which describes the generics-enabled standard collection classes, which is 90% of what you'll want generics for anyway.

Share and enjoy.

http://java.sun.com/j2se/1.5/pdf/generics-tutorial .pdf

http://java.sun.com/docs/books/tutorial/extra/gene rics/index.html

Also highly worthwhile is http://java.sun.com/docs/books/tutorial/collection s/index.html which describes the generics-enabled standard collection classes, which is 90% of what you'll want generics for anyway.

Share and enjoy.

Naftalin and Wadler are also holding a Java One session this year, it is on Wednesday, session id is TS-2890. If you have a Sun Developer Connection account (free) you can watch it online after the conference is over.

I agree with reviewer, the book is very good. It is true that Java generics is a compile time check, and that the generics information is removed (erasure). Nevertheless, that was a deliberate tradeoff for backwards compatibility, and it still makes coding complex Java a lot safer and easier. Look for instance at the 1.5 and 1.6 improvements to the concurrency libraries with Future, Callable and Executors.

It's been years since I've written Java, but
SpecificClass[] sca = scv.toArray(new SpecificClass[0]);
should do it, right?

I've had trouble with generics myself. I don't claim to be a programming superstar though, so I might be overlooking something simple. My problem is that I cannot get a (useful) array out of a vector. No matter what I do, Vector.toArray() will only return an array of Objects. But I don't want Object[], I want something specific. And I thought that is what generics were supposed to do: declare a vector, use generics to tell Java what kind of objects are in it, and get rid of all those nasty casts. Unfortunately, this doesn't seem to be the case. It seems that the information is lost at runtime, do to what Google tells me is "type erasure." There is an open bug on the subject, which I think is related. Regardless, I find it rather annoying and cumbersome to work around.

Even though Intel is probably the industry's biggest proponent of multi-core computing and threaded programming [...]

Even though Intel is probably the industry's biggest proponent of multi-core computing and threaded programming [...]

Why wait for Google. According to the Lightening 0.3.1 Release Notes there's a version you can download with integrated Sun Java System Calendar Server (WCAP) support. And in October you might have another option, given that Apple's Leopard iCal Server supports Sunbird, it should also support Lightening too. Or you could wait for it to be ported to Linux after they've released the source code, which they've promised to do.

I don't know why you think NFS doesn't support failover; check out Red Hat Cluster (PDF) or Sun Cluster. You will need a RAID array that has two host ports, such as VTrak E310s, IBM DS3200, HP StorageWorks 500, or Xserve RAID.

I would not suggest cluster file systems such as Lustre for a small installation; they're generally designed to scale up to hundreds or thousands of servers, but not to scale down to a handful.

If you have the right hardware, you don't need to be on-site. Serious servers come with something called lights out management. This utilizes a self-contained ROM-based system that's always running, even when the main system is shut down (or displaying a BSOD). As long as the system is getting power and there's an Ethernet cable connected to its management port, a remote user can do anything that an onsite user can do, provided it doesn't require opening the cover of the system. You can even re-install the operating system, used remote ISO and floppy images.

I'm the documentation lead for a server with a LOM that's very fancy indeed. There's a graphic terminal service that supports things like interacting with the BIOS, or logging into the server's GUI. There's a LOM command line you can access using a serial connection or over SSH. The LOM also supports IPMI, which is kind of a basic necessity when you have a lot of servers, even if they're all down the hall.

This server is certified for Windows 2003 (and I understand a lot of our customers buy it for that fell purpose), so it would be ideal for Microsoft's container. However, we have a our own competing container product.

And yes, the company I work for is Sun, and yes, we're selling Windows-based systems now. Shocking, isn't it?

If you have the right hardware, you don't need to be on-site. Serious servers come with something called lights out management. This utilizes a self-contained ROM-based system that's always running, even when the main system is shut down (or displaying a BSOD). As long as the system is getting power and there's an Ethernet cable connected to its management port, a remote user can do anything that an onsite user can do, provided it doesn't require opening the cover of the system. You can even re-install the operating system, used remote ISO and floppy images.

I'm the documentation lead for a server with a LOM that's very fancy indeed. There's a graphic terminal service that supports things like interacting with the BIOS, or logging into the server's GUI. There's a LOM command line you can access using a serial connection or over SSH. The LOM also supports IPMI, which is kind of a basic necessity when you have a lot of servers, even if they're all down the hall.

This server is certified for Windows 2003 (and I understand a lot of our customers buy it for that fell purpose), so it would be ideal for Microsoft's container. However, we have a our own competing container product.

And yes, the company I work for is Sun, and yes, we're selling Windows-based systems now. Shocking, isn't it?

If you have the right hardware, you don't need to be on-site. Serious servers come with something called lights out management. This utilizes a self-contained ROM-based system that's always running, even when the main system is shut down (or displaying a BSOD). As long as the system is getting power and there's an Ethernet cable connected to its management port, a remote user can do anything that an onsite user can do, provided it doesn't require opening the cover of the system. You can even re-install the operating system, used remote ISO and floppy images.

I'm the documentation lead for a server with a LOM that's very fancy indeed. There's a graphic terminal service that supports things like interacting with the BIOS, or logging into the server's GUI. There's a LOM command line you can access using a serial connection or over SSH. The LOM also supports IPMI, which is kind of a basic necessity when you have a lot of servers, even if they're all down the hall.

This server is certified for Windows 2003 (and I understand a lot of our customers buy it for that fell purpose), so it would be ideal for Microsoft's container. However, we have a our own competing container product.

And yes, the company I work for is Sun, and yes, we're selling Windows-based systems now. Shocking, isn't it?

Wow, what an original an innovative idea! It's too bad that Google didn't think of this. Or I'd really expect that Sun would have come up with something like this by now. But it took those geniuses from Redmond to deliver true innovation!

I took the tour today, got a neat tee shirt and free lunch too!

You're forgetting: Sun offers CoolThreads technology. Do the math: The power savings and heat reduction on AMD's Opteron servers is considerable, esp when running Solaris. Check the benchmarks and claims - Sun is the solution to use.

Not bad for x86 and x64 hardware that's rated by Sun on over a dozen OSes. Sun is covering their bases well. (Even if you want to run Windows within that portable datacenter.)

With Project Blackbox, it's obvious that Sun is paying attention to their customers. Need to expand the datacenter, but don't have the space? Use their portable container setup. It's sheer genius, esp. for emergency contigencies/disaster situations. If I were a CIO/CTO, I would be taking a SERIOUS look at Sun's product as part of my data/computing landscape.

(And no jokes about hijacking the container with a forklift or breaking into it... That's why you hire 24/7 security if the data is important to you.)

Microsoft seems hellbent on adding their marketing spin to the product arena. This is one instance where they need to SIMPLIFY their verbage. I'm sorry, M$ - I'm far more comfortable putting my IT folks on a laptop, managing a remote UNIX (Solaris) or Linux solution than a Windows-based setup. Not unless I want to keep sending my user to the container's locale every few days for one issue or another.

Microsoft needs to rethink their strategy here. I think Sun ($un?) got it right.

A Novel Datacenter Concept

Project Blackbox packages compute, storage, and network infrastructure capabilities into scalable, modular units outfitted with state-of-the-art cooling, monitoring, and power distribution systems. Customers will be able to order a variety of standard and custom configurations of systems, storage, networking, and software. Housed in a standard 20-foot shipping container for maximum flexibility, Project Blackbox will be easily transported using common shipping methods. Simple hookups for water, AC power, and networking will enable customers to quickly deploy Project Blackbox upon delivery.

and on one physical CPU. And it's cheaper than an Apple.

http://www.sun.com/servers/coolthreads/t1000/

Why is this even news?

1) I find it highly offensive and irresponsible that the discussion on /. for this case (both this story and the one before) has automatically presumed that the disabled community is being used as a witless proxy for larger battles.

2) Having read and digested the entire nuanced thread, particularly posts like this and that, I have come to the reasoned conclusion that disability access is being used as a proxy for larger battles. I also duly note the similarly with the Massachusetts fight over ODF, but disavow that this is a pattern. Even if Peter Korn of Sun thinks differently. <*sighs deeply* />

I would like to quote a well connected individual who this all into perspective for me a week ago:

It is curious that the original inquiry sources a New York Times company product. The NYT Co. and the Sulzberger family that controls it have been highly skeptical in the advancement of independence for people with disabilities. The latest criticism and skepticality from them regards the cost and reliability of accessible voting machines. The NYT co. product below fails to explain that this is a simple contract dispute and does not involve access issues. Diebold believes it met the requirements of the RFP more than ES&S and therefore should be given the contract rather than ES&S.

The difference of course is that the Diebold system for access is completely electronic and would need to blend the results with the paper optical scan ballots while the AutoMark simply prints a completed paper ballot and no blending of results is necessary. The Diebold DRE is a lot less expensive than the Automark, which is their biggest selling point.

This is likely the last big contract outstanding and could add to the sale price of the election division when Diebold decides to sell it. The new CEO already has said in Fortune magazine that the election division is not a long term strategic fit for the company.

And I initially thought he was just being cynical!

Not that anyone at /. cares, but here is a link to Voluntary Voting System Guidelines which both the ES&S and Diebold products fail to completely satisfy.

Hopefully nobody is actually using telnet on an unsecured network...
but if they are they'd better patch Solaris 10.
http://sunsolve.sun.com/search/document.do?assetke y=1-26-102802-1

Let me check the year, oh yeah, 2007!

All of this to show your IT manager how easy it can be. Then show your homework, and you could likely get one justified for every UNIX administrator in your company. Want to make the argument, compare that to what it costs your company for 25 Windows stations for the same people. I think you'll find the argument compelling. Typical lifespan of a thin client is 7-10 years, you won't need to replace it in three years as you would with typical desktops.

Typical Sun Ray cost for 25 users (assuming they would use existing monitors/keyboards, estimated numbers):
Sun Microsystems T1000 w/ 8-core CPU & 8GB RAM w/ 3-yr Support: $8500
Sun Ray 2 Display Clients + RTU + Support (3yr): $500 ea ($12,500 for 25)
Your labour & time to install/manage: $55/hr * 2 hours/week * 3 years = $17,160
Total 3-year Cost: $38,160.00

This should justify itself.

All of this to show your IT manager how easy it can be. Then show your homework, and you could likely get one justified for every UNIX administrator in your company. Want to make the argument, compare that to what it costs your company for 25 Windows stations for the same people. I think you'll find the argument compelling. Typical lifespan of a thin client is 7-10 years, you won't need to replace it in three years as you would with typical desktops.

Typical Sun Ray cost for 25 users (assuming they would use existing monitors/keyboards, estimated numbers):
Sun Microsystems T1000 w/ 8-core CPU & 8GB RAM w/ 3-yr Support: $8500
Sun Ray 2 Display Clients + RTU + Support (3yr): $500 ea ($12,500 for 25)
Your labour & time to install/manage: $55/hr * 2 hours/week * 3 years = $17,160
Total 3-year Cost: $38,160.00

This should justify itself.