Slashdot Mirror

Yes, but 4 xserves will be far more powerful.

With current systems, ray traced images can take upto 30 minutes for 1 frame. If you could get that frame in 30 seconds, how much could that help? Ya' know?

Also, in the case of a Database. There is no replacing a single box with 72 64bit processors @ 1.2GHz and 600GB of ram or 32 64bit processors @ 1.9GHz with 1TB of memory. Do you really think Sun and IBM would offer those types of configurations if nobody bought it or thought it was important to have everything consolidated vs distributed? You can cluster databases and make smaller nodes with fragmented databases, but not all cases have that option. Sometimes you just need a big bad-ol' box.

When large companies such as Pixar decide to make a large decision like this, they will determine the price/performance ratios of all the available systems. The overall price includes the cost of all the systems including discounts, support/maintenance and the cost of changing over. Performance is measured using benchmark related to the required task. For PIXAR, this is the RenderMark(*). Ultimately, each system will be reduced down to X dollars per RenderMark. The vendor with the most Rendermarks per dollar will be the winner.

At present, Apple has the most powerful systems. That isn't to say SGI, Sun or anyone else won't make an effort to catch up. From: Sun Microsystems

Evaluating Rendering Performance Pixar has developed a benchmark standard to produce a single metric that characterizes a computing system's rendering power. The larger the RenderMark, the greater the system's rendering capacity. The RenderMark is derived from the elapsed time of a set of four jobs that stress important aspects of rendering: Ball. A ball with shading, nubs, and motion blur Pixar. The Pixar logo that includes complex geometry and typesetting designed by Pixar's Typestry software Magic. A RenderMan marketing poster depicting magician's hats and wands, including lots of texture-mapping Bike Shop. A bicycle shop scene from Pixar's Red's Dream, where one of the biggest challenges is the number of spokes to render

From Computer Graphics World A 1000 RenderMark CPU computes the same frame twice as fast as a 500 RenderMark CPU.) The first Toy Story (1995) used 50,000 RenderMarks for rendering; A Bug's Life (1998) needed 700,000 RenderMarks; and Toy Story 2 (1999) took 1.1 million. Monsters, Inc. re quired 2.5 million Render Marks, more than the first three films combined.

This isn't Jurassic Park.

Plus, they only had a 117 Sun workstations in the original Toy Story render farm.

Disney's "Toy Story" Uses More Than 100 Sun Workstations to Render Images for First All-Computer-Based Movi

Say it with me: OpenFirmware

Better yet: Sing it with Mitch Bradley : Firmware, Open Firmware...

Solaris does have a firewall. It is called Sunscreen.

It used to be spearate in a free "lite" version, but is now integrated in Solaris 9 or Trusted Solaris 8.

Documentation is far and away the thing that sets Perl apart from other languages. There are some bad things you can say about Perl's documentation, but most of them are criticisms that you can't even begin to make of other programming languages because they simply aren't in the same ballpark.

I blame POD for this

113476

They are zip files, just extract them and follow the instructions inside. But are you really going to trust some links on slashdot? The way I found these is I went to sunsolve and searched for 108993 and 113476 in PatchFinder.

113476

They are zip files, just extract them and follow the instructions inside. But are you really going to trust some links on slashdot? The way I found these is I went to sunsolve and searched for 108993 and 113476 in PatchFinder.

113476

They are zip files, just extract them and follow the instructions inside. But are you really going to trust some links on slashdot? The way I found these is I went to sunsolve and searched for 108993 and 113476 in PatchFinder.

I wouldn't be at all surprised if this bug was in the PAM library or a module.

Neither would I. From the patch details:

Files included with this patch:

/usr/lib/libpam.so.1
/usr/lib/llib-lpas swdutil
/usr/lib/llib-lpasswdutil.ln
/usr/lib/pa sswdutil.so.1
/usr/lib/security/pam_authtok_check .so.1
/usr/lib/security/pam_authtok_get.so.1
/us r/lib/security/pam_authtok_store.so.1
/usr/lib/se curity/pam_dhkeys.so.1
/usr/lib/security/pam_ldap .so.1
/usr/lib/security/pam_passwd_auth.so.1
/us r/lib/security/pam_unix_account.so.1
/usr/lib/sec urity/pam_unix_auth.so.1
/usr/lib/security/sparcv 9/pam_authtok_check.so.1
/usr/lib/security/sparcv 9/pam_authtok_get.so.1
/usr/lib/security/sparcv9/ pam_authtok_store.so.1
/usr/lib/security/sparcv9/ pam_dhkeys.so.1
/usr/lib/security/sparcv9/pam_lda p.so.1
/usr/lib/security/sparcv9/pam_passwd_auth. so.1
/usr/lib/security/sparcv9/pam_unix_account.s o.1
/usr/lib/security/sparcv9/pam_unix_auth.so.1
/usr/lib/sparcv9/libpam.so.1
/usr/lib/sparcv9/ll ib-lpasswdutil.ln
/usr/lib/sparcv9/passwdutil.so. 1

So there's no workaround ...

How about "chmod ug-s /bin/passwd"? Someone running passwd wouldn't be able to escallate their uid/gid. To change passwords, run su(do) first. On systems wehre users arn't expected to change their passwords (web servers, etc.), this is usually a good preventative step for most setuid programs.

And for the Love of Scott, if you're going to tell the world about a patch, please, oh please, make sure the hyperlinks work.

Here's Sun's announcement, and if I click on the links to get patches,....

Sparc
Solaris 8 with patch 108993-32 or later
Solaris 9 with patch 113476-11 or later

.... the links give me:

Sorry! We couldn't find your document.

The file that you requested could not be found on this server.

G'dammit!

-ez

Karma: Whore (you look at your score after posting)

So there's no workaround ...

How about "chmod ug-s /bin/passwd"? Someone running passwd wouldn't be able to escallate their uid/gid. To change passwords, run su(do) first. On systems wehre users arn't expected to change their passwords (web servers, etc.), this is usually a good preventative step for most setuid programs.

And for the Love of Scott, if you're going to tell the world about a patch, please, oh please, make sure the hyperlinks work.

Here's Sun's announcement, and if I click on the links to get patches,....

Sparc
Solaris 8 with patch 108993-32 or later
Solaris 9 with patch 113476-11 or later

.... the links give me:

Sorry! We couldn't find your document.

The file that you requested could not be found on this server.

G'dammit!

-ez

Karma: Whore (you look at your score after posting)

So there's no workaround ...

How about "chmod ug-s /bin/passwd"? Someone running passwd wouldn't be able to escallate their uid/gid. To change passwords, run su(do) first. On systems wehre users arn't expected to change their passwords (web servers, etc.), this is usually a good preventative step for most setuid programs.

And for the Love of Scott, if you're going to tell the world about a patch, please, oh please, make sure the hyperlinks work.

Here's Sun's announcement, and if I click on the links to get patches,....

Sparc
Solaris 8 with patch 108993-32 or later
Solaris 9 with patch 113476-11 or later

.... the links give me:

Sorry! We couldn't find your document.

The file that you requested could not be found on this server.

G'dammit!

-ez

Karma: Whore (you look at your score after posting)

actually, i also meant to give this link:

http://www.sunsolve.sun.com/pub-cgi/search.pl?mode =results&so=date&coll=fsalert&zone_32=category:sec urity

as i said, there is no such thing as a secure system. so, i really don't understand why news like this make it to the front page. to warn people? as an admin, you better follow these things and if you don't, you deserve to be vulnerable anyways

-- ng

No, but if I wanted to, I could already, thanks.

How about we color it in our editor and do not make the compiler any kind of warning, so that you think it's supported while it really is not?

Interesting. That's not what Sun says, and I'm more inclined to believe them over you.

The general program is Software Express, which is what you described. The specific program which gives access to a preview of Solaris 10 is called Solaris Express. So the article is using the right term.

Actually, the SunBlade 100 did come with an integrated network card, see http://sunsolve.sun.com/handbook_pub/Systems/SunBl ade100/spec.html

When I encountered the problem, I started by looking around at SunSolve, and never really thought of going to straight to comp.unix.solaris from there. I was interrupted in my fiddling with the system by the arrival of my first child, so I reverted Solaris 8 knowing it would work. You raise a good point though, now that I have a bit more time, I should go there and see what the discussions there have to say.

It sounds to me more like a Java Servlet container model than a VM. There's even a "global zone" that can see all the others.

Here's a post about it.

Here's Sun's page on it

It sounds to me more like a Java Servlet container model than a VM. There's even a "global zone" that can see all the others.

Here's a post about it.

Here's Sun's page on it

why does the author constantly mention CORBA? Perhaps I'm not familiar enough with the Java perspective, but I assumed CORBA/COM (etc) were basically dead (or dying) when it came to modern internet programming.

That's a fair question. Looking at the requirements, we needed middleware that is language- and platform-independent. That immediately rules out things such as COM, DCOM, RMI, and .NET. CORBA really was the only candidate. Also, all of us (the people who wrote Ice) had a lot of CORBA experience, and Ice borrows a lot of ideas from CORBA. (No, not everything in CORBA is bad ;-) So, I ended up contrasting things to CORBA because Ice is both similar and different to CORBA.

CORBA isn't exactly dead--people are still building industrial-strength systems with it. And, to be fair, CORBA is the most widely-used distribution technology to date. It's just that CORBA is ageing and that, after more than ten years of experience, we now know how to do better and avoid CORBA's mistakes.

The author and company were only familiar with CORBA.

No, not really. I have experience with Sun RPC, DCE, COM, DCOM, Web Services, and CORBA.

Distributed-objects are an older, more established technology. [ie proven]

There are a lot of existing systems, algorithms, ideas, programs, etc. to build a distributed-object system.

Yes, to are large extent. There is a lot of experience we have gained over the past ten years or so on how to build distributed systems that work.

It masks the network from the API (yes, this should be a CON).

That's an old topic that comes up periodically. The most-quoted paper on that topic is probably A Note on Distributed Computing. Personally, I disagree with many of the conclusions in that paper. True, I can't afford to forget that there is a network underneath my API because the semantics are truly different from the non-distributed case. But that does not strike me as a compelling reason to make a remote invocation substantially different from a local one.

Engineering effort is required to make sure clients can access the model with any langauge, and also to allow designers flexibility in the object model. Most of this should be cheaper with a web-services model.

I don't see where web services would be cheaper than any other language-agnostic platform. In fact, everything I have seen so far about web services is more complex than Ice (or CORBA). And web services are horribly inefficient. If you are interested, look for my postings on the topic in comp.object.corba with Deja.

The reason we have web services (or, more accurately, might have them some day) is a political one rather than a technical one: back in the hey-days of the distributed computing war between CORBA and DCOM, Microsoft were about to lose the battle. They weren't going to simply cede the battlefield to the OMG so, rather than fight a war they couldn't win, they simply created a new battlefield. Sadly, web services is about re-inventing the distributed object wheel all over again and, even more sadly, with even more mistakes and corners on the wheel than CORBA.

You don't get to throw around as many buzzwords.

Unfortunately, in some circles, that seems to be an important criterion in the decision making. Personally, I prefer to use things that work and have substance, instead of chasing after silver bullets :-)

There is duplicated code on both the client and server.

Huh? You haven't been listening to my friend Don Box by any chance? ;-) If that is what you are referring to, it's simply wrong (or certainly, it is wrong as Don presented it in his interview). Neither CORBA nor Ice require duplicated code on client and server--they are pure client-s

JavaSpaces is more or less Sun's Jini take on distributed processing.

From a programmer point of view, you start up a "space", and then you can write objects in, take them out, and read them. And that's all. So there are a very few simple operations, and you structure your app around those.

Anyhow, it seems like a couple of JavaSpaces on a rack of servers might serve as a good way to distribute processing/notification/etc. Of course, you're limited to Java and to moving around Serializable objects....

In case of Red Hat it didn't use to be. Their free (or almost free, for CDs) desktop did drive their high-end solution (agreed, not massive profits... but significant indirect revenue, relative to company size). At least that's my take on it, although RHAT seems to disagree nowadays (I will personally think they are wrong and made a stupid move -- time will tell who's right).

And in case of IBM statement is not wrong at all. IBM professional services, their big bad strategic moneymaker most certainly uses OSS/FS to their advantage, and indeed makes massive profits. Not ONLY using OSS/FS, but as significant part of the mix. And relative size of that portion is increasing, talk to any IBM consultatnt to verify that.

And yet another company that wasn't mentioned but probably should have been is Sun, especially with Star/OpenOffice and their Linux distro, but to lesser degree with free components they include (from Gnome to GCC).

Bah! As computer geeks, we should be willing to just accept the cost of powering our computers as a condition of our lifestyle. After all, its not like we're going to spend that money as readily on fancy cars or anything (which are a lot more expensive).

I did some measurements a few months ago, and my rack of computer equipment has a continuous draw of around 1000W. That's the stuff that runs 24/7, and doesn't include monitors (which I do shut off when I'm not home). And guess what? Soon I'm planning on putting some more power hungry stuff into use, and that number will shoot up quite a bit.

If you really care about power, though, I strongly recommend thin clients. I currently have a bunch of Sun Ray units (yes, they require Solaris/Sparc on the server, but they're just one example of a possible solution). These things use less power than a laptop, make no noise, and basically let you stick a "desktop setup" anywhere you can snake an ethernet connection to. So for the power draw of single decent desktop (I'm using a dual-proc Ultra 60, which has a 350W power supply and is quieter than the PC sitting next to it), you can have as many desktops as you feel able to share CPU power with.

There are many aspects of Java that mean you can't guarantee a response within a certain timeframe. Garbage collection, for example ... there isn't necessarily a 1:1 instruction translation of the JVM's bytecode, and therefor you can't know beyond a shadow of a doubt that a certain operation will get done in a certain amount of time.

you can get real-time behavior out of an interpreted system, if both the interpreter and underlying operating system are designed for it. For Java, GC and instruction translation just have to have that as a design goal. As a matter of fact: here's some information on real time Java:

"With the recently released Real-Time Specification for Java (RTSJ), developed through the Java Community Process by the Real-Time Expert Group, the real-time embedded software developer will be able to use the Java programming language in applications where predictable/hard real-time behavior is a must."

that being said, java probably would run too slowly for most applications on a 20 mhz cpu. But being interpreted or having GC are orthogonal issues to real-timeness.

IBM has it ports only for Linux/PPC and it's just another source of incompatibility problems.

Dunno about you, but I've used the VMs from Sun, IBM, and Blackdown for years. I've never had incompatibility problems after they were out of beta.

Why? Because normal Java doesn't work on such small memory, and J2EE is not ported to J2ME.

Great, you showed one example of when you wanted to run JMS on a router. Why you couldn't run it on another machine, I don't know, but fair enough.

Now, JMS has a public spec. Go write a version that runs on J2ME. Go on, now! Get crackin'!

Looks like your education is based on repeating after someone else, not on your own thinking.

Making assumptions based on no evidence will get you nowhere, sonny. But, let me take a gander at you. From the 1848 comments you've made on slashdot over the past year and a half, (that's over three per day, with a whole lot of Flamebait, Troll, and Offtopic mods in there) I can see that you talk a lot, but don't have much useful to say.

I can see that. So, really, what we need is a solution where an open-source OS (thus alleiviating their fear of M$ practices) could include it on the install disc?

Yes, that would be a considerable improvement. Note though that there are plenty of other benefits of making j2sdk free software, both for the world at large and for Sun Microsystems. I thnk the parent article on LinuxToday makes the case for that quite well.

From the way I understand it, [...] there is an AWT component for the main window.

This matches my (somewhat rusty) understanding of things.

Well, with Swing, you have the option of your application [...] looking like other native applications

I've not tried this, but I suspect that things will not work the way I want them to because, as you pointed out, Swing uses Java code to do the drawing of components. I want the buttons, scroll bars and other widgets to use whatever GTK+ theme I have configured. I also want it to change dynamically at runtime if I alter my GTK+ theme. Using GTK native peers would accomplish this easily, but to do it in Java code seems impractical.

I don't think that [implementing lighweight components in AWT] is technically possible.

I should have been more clear about what I mean. I create subclasses of java.awt.Component and override the paint() and update() methods. According to the API reference this creates a lightweight component with no native peer, much the way Swing does. A JButton is a subclass of java.awt.Component too.

I usually use XML

I do that as well. This has advantages over relational databases in places where absolute performance is not critical. I meant that for user interfaces I use Python CGI scripts that generate a web based user interface. (I keep meaning to try mod_python, but so far in spite of my best efforts I've been unable to make my Python code slow enough for this to be worthwhile. Imagine invoking a new Java VM for every incomming web process!)

I still don't understand. AWT is the one that looks really ugly, Swing has the look-n-feel/theming options.

I didn't mean that I think Swing is ugly (or that AWT isn't ugly -- though really the problem is that Motif is ugly and AWT uses Motif). I meant that the look and feel doesn't match my desktop GTK+ theme as I mentioned above.

I prefer writing all of my BSD software (and am in the process of replacing all my httpd, ftpd, sshd, smtpd, etc) in Java, because then I don't have to be tied to any specific OS.

I suppose time will tell and I wish you luck with the project, but I'm skeptical. Packages such as Apache (and even OpenSSH, which does run on Windows) are quite portable. All one saves with Java is a compile step, and while this is a powerful advantage for applications I think the dynamics of system software are less affected by it. Most people get their operating systems on CDs and through vendor updates, so code mobility is not a concern.

On the other side, I believe performance will suffer, particularly on non-Windows platforms where the Sun VM is grotesquely slow. Non-blocking I/O is indeed efficient (though it is best to combine it with threading for SMP systems), but this feature is at least as easy to use in C.

I am actually working on something like that, because I think that idea is the perfect solution. Make a Java-based OS on CDR(W), and run it on and i386, mac, ps2, dreamcast, etc.

Again, good luck but I'm skeptical. Ultimately you will need something very much like a Unix kernel to talk directly to the hardware and implement the Java VM. I'm also not convinced that maintenence will be much easer. Most C projects separate and minimize platform specific code. A JavaOS would still need testing for each platform, and for the same reasons. Note that Sun and IBM have tried and failed to create a viable JavaOS already. That do

The Java Language Specification
The Java Virtual Machine Specification

The Java Language Specification
The Java Virtual Machine Specification

what is the big deal about downloading java

Great. Hop over to the website and get the download for IBM AIX. Guess what. THERE ISN'T ONE. AIX competes with Sun, so no Java for you.

Java Downloads

If you want Java for AIX, go to the IBM website and get the one that't 3 revs back.

I agree that memory comsuption could be greater than a C program but servers usually are configured with enough memory. Anyway, in the Java 1.5 beta Sun has already implemented class data sharing.

It's important to note that there was a creative and collaborative attitude during the 60's and 70's that does not exist now. You can sense this from folowing the history, for instance that Kay et. al. never tried to copyright or patent any of the design aspects of their work -- the use of the mouse and network, the use of overlapping windows and desktop "objects", etc. If such ideas were developed now they would certainly be treated as precious "Intellectual Property" and patented into oblivion. But the attitude was actually the opposite.

I was there to some extent, and attended some presentations by Kay during the 70's and early 80's. I believe he saw himself as simply continuing the work of other researchers in interactive systems before him, citing, for example Ivan Sutherland and the Sketchpad system with some envy. ( Here, for instance) Sutherland is also credited with inventing the mouse, though I believe it was actually more the product of a movement or a time in history than of a single person.

I really belive the creative and exploratory attitude of the society in general is much more important than the individuals involved. The invention of the 70's was fueled by the extreme exploratory abandon of the 60's in all things -- politics, music, space... We seem now to live in a period of collective creative depression, and we get appropriately depressed results. Now we get security, DRM, and a huge effort to copy and open source a successful but anachronistic human - machine interaction model, MS Windows.

Make sure to pay attention to what Kay himself says while accepting the prize -- it is most significant, and not just an exercise in artificial humility:

"The ARPA/PARC research context and community catalyzed researchers to be incredibly better dreamers and thinkers. This context was itself a great work of art, confirmed by the world-changing results that appeared so swiftly, and almost easily."

Not to mention IBM has its own JDK. Why don't they put their money where their mouth is? Forgive me if its here, and I'm just not seeing it. But Sun's JDK is already open source, but its not free. Where do I get the source to IBM's JDK?

Or maybe IBM is just mimicking SUN?

It's all a PR game. Sun did the same earlier, so why not IBM.

Look at this comment back a few months ago, I think IBM has better a argument anyways.

Well, if it's Turing-complete, all I need is a highly efficient optimizing compiler. ;-) In fact, that's all I'm after.

But then neither can Java.

Why do you say this? Did you ever hear of or see the Grand Canyon demo? Also check out this setup for realtime Java computing with J2SE. Looks pretty good to me. There are also now official OpenGL bindings for Java.

Also, Jython is considerably slower than native Python owing to the overhead of turning Python code into Java forms. The only real advantage of Jython is that it lets you easily call Java routines from an almost Python environment. Sometimes this is enough of an advantage to make it worthwhile, but not if speed is what you are after.

I don't think this is right - the Jython page claims there's a compiler to compile Python to Java bytecode - that should be as fast as native Java, assuming the compiler does a decent job.

Well... Sauron, er I mean Sun, gave out rings at a previous JavaOne conference that contains a microprocessor that runs a JVM, which in turns displays a fractal image at some of the terminals.

More info on the "ring" can be found at:
http://java.sun.com/features/1998/03/rings.html

Well, IMHO, this is exactly what Sun did with Sun Community Source Licensing (SCSL). Check out their principles/a and see how their stand on Open Source. (Scroll down a little bit)

Well, IMHO, this is exactly what Sun did with Sun Community Source Licensing (SCSL). Check out their principles/a and see how their stand on Open Source. (Scroll down a little bit)

>So basically what I want is that the compiled code should be cached and stored on disk, e.g. like a browser loading cached pages when they are still up to date.

Like this?

Sun also grants you a perpetual, non-exclusive, worldwide, fully paid-up, royalty free, limited license (without the right to sublicense) under any applicable copyrights or patent rights it may have in the Specification to create and/or distribute an Independent Implementation of the Specification that: (i) fully implements the Spec(s) including all its required interfaces and functionality; (ii) does not modify, subset, superset or otherwise extend the Licensor Name Space, or include any public or protected packages, classes, Java interfaces, fields or methods within the Licensor Name Space other than those required/authorized by the Specification or Specifications being implemented; and (iii) passes the TCK (including satisfying the requirements of the applicable TCK Users Guide) for such Specification. The foregoing license is expressly conditioned on your not acting outside its scope. No license is granted hereunder for any other purpose.

The conspiracy minded might wonder if Sun sees Java -- not Linux -- as the most important part of the Java Desktop System. As long as people are migrating away from Windows, why not sell them an OS where Java is the preferred system programming language?

One of the facts of life in the enterprise is that it is heterogeneous in terms of platforms, operating systems and (maybe) network technologies.

I agree 100%.

Neither J2EE nor .NET is satisfactory in this environment.

Yea, who would do something stupid like subscription operating system updates?

There also needs to be a way to load bootstrap code remotely. For instance, having a TCP/IP enabled BIOS be able to run TFTP or some other protocol to load a netboot floppy image. Then you could give it a LILO command instructing it where to find a boot image, preferably one on a server in the same hosting center.

Sun hardware has had serial consoles that can boot from the network for years. The syntax for the current OBP (OpenBoot PROM) revisions is here: http://docs.sun.com/db/doc/817-2701/6mibjioqr?a=vi ew

Coupled with with a terminal server/power management module such as this you'll get all those features.

For x86 hardware, some vendors are shipping with serial console capabilities which include network booting, such as Dell's DRAC

Remote floppy boot. DRAC offers remote media access, allowing the server to boot from remote media. DRAC II uses floppy redirection. Administrators can insert a bootable DOS diskette into the diskette drive of the desktop machine and boot a remote server to that floppy. Administrators can then run operations from the floppy, including functions such as flash BIOS to recover servers with BIOS problems.

DRAC III uses Trivial File Transfer Protocol (TFTP) to transfer an image to the card and lets administrators enhance remote floppy performance by downloading floppy images to the memory on the card (see Figure 5 ). Functions on the "diskette" are executed in a DOS environment for 32-bit systems.

Both Solaris 8 and Solaris 9 come with a CD titled "Solaris Companion Software CD". If you can't find that CD, you can download the ISO from the above page, as well as download individual packages.

Thirdly, Java GC is not really "conservative". An object is eligable for collection when (and only when) the reference count is zero.

Minor nitpick, that is not completly correct.
An object can still have references to it, and still be elegible for gc'ing. That is what Weak- and SoftReferences are used for.

From the API docs:

Soft reference objects, which are cleared at the discretion of the garbage collector in response to memory demand. Soft references are most often used to implement memory-sensitive caches.
Suppose that the garbage collector determines at a certain point in time that an object is softly reachable. At that time it may choose to clear atomically all soft references to that object and all soft references to any other softly-reachable objects from which that object is reachable through a chain of strong references. At the same time or at some later time it will enqueue those newly-cleared soft references that are registered with reference queues.

Just because you don't use certain functionality doesn't make it worthless. Have you taken a look at Family Comparison Chart to see the comparisons from 6 through 9? You really didn't like integrated, supported SSH, LDAP, RBAC, KRB, PAM, SRM, SVM (particularly soft partitions), Flash, Live Upgrade, UFS logging, IPMP?

Sun Software Express is the name of their bleeding edge program. "Solaris Express" is what the lazy people call it.

So please, if you want to make intelligent comments about an operating system, be sure to separate it from the hardware upon which it runs, or at least be aware of all platforms on which it runs.