Slashdot Mirror

(I can't believe I'm feeding a troll, but I couldn't let this just slip by.)

All evidence shows that Linux is less secure than other operating systems, in particular Windows.

Wrong.

For one thing, this can be explained by the open nature of Linux where anybody has access to all of the encryption algorithms, sources and keys. In the computer world, just like in the human world, it is in environments where anything goes that the worst viruses come to existence.

Linux uses standard encryption algorithms, just like Windows. 3DES and DSA are the same everywhere. Private keys are still private (Linus didn't pack his GPG key into the latest kernel source, if that's what you're thinking), and public keys public.

Also, Linux distributions are filled with various backdoors since anyone, including ill-intended foreigners, can add anything to the kernel base and its surroundings. At some point, there was even a hacked version of a compiler that introduced backdoors in every program that it produced!

OSS isn't run on the Wiki model. All submissions to open-source projects are looked over and verified by the project maintainers. At least with OSS I don't have to worry about backdoors added by certain ill-intended Americans.

Finally, and probably most importantly, Linux growth happens through the actions of the low-key movement of techies that try to replace everything they can in their organisations with Linux. Apart from acting unprofessionally, these zealots let their feelings for the beloved OS trump any kind of common sense behavior, such as using the right tool for the job. Instead they carelessly introduce vulnerabilities in environments that were previously locked down.

Wow! Shocking! A valid point! Not exactly a problem with Linux itself, though...

Yes, this can be a problem. Linux is good, but not perfect for everything. There are some things Windows just does better. The proper response is to fire these idiots. They'd do just as much damage administrating a Windows server

In short, organisations who value computer security should stay away from Linux, and refrain from hiring those who mention Linux in their resume.

Really? You should let IBM know about this.

What really astonishes me is that open source has made such great leaps in other areas yet there's no apparent replacement for Outlook & Exchange. For a huge number of folk in business, having an open office suite is useless if they don't have calendar sharing, resource scheduling and email/contact sharing amongst groups. Is this really so difficult to achieve?

I don't know how difficult it is, but it exists. Ever hear of Lotus Notes?

While it may not be open source or Free (as in beer), it runs on x86 Linux (RH & SUSE at least), as does the Domino server (an Exchange alternative). It's not directly compatible with Exchange, but nothing is going to be except Microsoft mail products which are closed and non-Linux, so if that's your problem then nothing but a Microsoft supported platform can ever satisfy you so why did it take you so long to figure that out and give up?

How about Lotus Notes and it has a Linux client?

FWIW, here http://www-03.ibm.com/servers/eserver/opteron/pdf/ IBM_dualcore_whitepaper.pdf is a paper that talks about the merits of multicore processors (specifically Opterons) in the HPC world.

To be a plot spoiler, they conclude that dual core is a good thing.

"It's not about multi-core processors, it's about the Cell architecture, for parts of which IBM holds many patents and makes a lot of money on licensing."

You are correct. But I should point out that IBM is not the only company producing the Cell either. Sony and Toshiba also own rights to the Cell, but I doubt they will start putting them into servers.

They are, however, the only company working on the Roadrunner project, which TFA was actually about and which TFA (or rather IBM's spokesperson in TFA) says will produce the technological products they will soon be able to provide through a rep near you :D. There are also allusions to the idea that the technology has applications beyond supercomputing; that it will be useful in "normal" multicore systems as well. I don't know how accurate that is, but the hypnotic salespeak does make one feel warm and fuzzy inside.

It is an interesting approach to say the least. I am sure this is a hard requirement in the supercomputing world (though I must confess some ignorance in that area) but frameworks like this do have useful applications, and if implemented properly can make life easier potentially. I think the problem is that any technological abstraction we have tried thus far, be it a soft requirement like apis or libraries or a harder requirement like we have with JVMs tend to be judged suboptimal by experienced developers for one reason or another. When that happens either you get developers coding around the thing (basically poking holes in the abstraction model), which has the effect of negating some of its benefits, or in places where that is not an option (like JVM) they choose not to use it which is probably at least as bad.

I know this is gross oversimplification but so is the article. In any case every time we hear about tech x that will save the world, especially in this particular way, we run into the same set of problems, chief of which are universal adoption as a requirement (which doesn't happen) and whether the tech performs like the brochure said it would in real life (I think it's safe to include a ditto here). It sounds like great tech and I hope it works out well. I can't pretend to understand it at a level that matters for working with it, but it does sound useful in general computing terms. If it can be used on non-supercomputers I don't doubt IBM will use it themselves, maybe integrating it with AIX and even their Linux solutions. Even if it isn't it's probably really great for supercomputing applications, and it's pretty cool that they are using Opterons in the hardware side of the project.

Whether it will be useful for anyone else, or used by anyone else, who knows. The fact remains that the criticism is correct. This article is an announcement of a potential future product given by a spokesperson for the company producing it in which they outline a computing problem, declare that there have hertofore been no real solutions, then propose theirs. That does not make it inaccurate, wrong, or bad, but it is in fact true.

Antiglobalization has been around for years. Also, I think the Trusted Computing Platform is a good thing overall. It lets me prevent an OS hole from opening up my files. It doesn't help DRM that much. It lets me avoid keyloggers. It works on all platforms. The issue is that micrsoft is trying to confuse it with NGSCB/Palladium, which is evil. Look at this site from IBM for more information.

idemix which is the software in question appears to be covered by a number of patent applications submitted by the inventor, Jan Camenisch. What's the point in open-sourcing it if IBM has half a dozen or more patents covering the technology being used? Or will this process grant use of any IBM-owned patents necessary to run the code? And if so, what happens as people start modifying the code; how far can they go and still be indemnified against IBM patent infringement?

Patents and open source don't mix well. I don't see how this is going to work.

can be found here.

Yeah, 'cause clearly, nothing is more secure than a closed source solution. Security by obscurity is the ONLY ANSWER! And advice on computer security by random slashdot posters is far more trustworthy than anything from a company that's been making secure systems for longer than most of us have been alive.

http://www-306.ibm.com/software/info/features/feb1 52005/

This was originally released back in the middle of November 2006!

http://www-128.ibm.com/developerworks/forums/dw_th read.jsp?forum=367&thread=142364&cat=10

Here you go: Notes on Linux. Enjoy :)

I worked on multiple NASA projects in the 1990s. During the mid-90s we used Rational http://www-306.ibm.com/software/rational/ for a short period of time (6 months) then dropped it. IME, the people who want these tools are architect that couldn't program their way out of a paper bag. Since Ive become an architect now, I prefer Visio http://www.microsoft.com/office/visio/ to most of these other tools - that's only if a pencil drawing doesn't cover everything good enough AND I need to make a presentation to someone with money.

IBM has many nice tools and the best bang for your buck hardware, but Rational ought to be buried into a deep, dark hole with a RADIOACTIVE sign outside. http://www.nmsu.edu/~safety/images/signs/sign_caut ion-rad-mat.jpg

In about 1968, IBM had an optical memory where about 2 Km of optical path was folded into something the size of a filing cabinet using mirrors, and 1 bit was circulating endlessly. Optical fibres transparent enought to do this did not happen for years. This geta a brief mention in... http://www-03.ibm.com/ibm/history/history/year_196 8.html

The idea of even firing someone into space was foreign to us only a century ago. Frankly, I would be stunned if, within the next several thousand years, humanity didn't figure out a way to fold space. There are tons of physicists that work on that type of math already (and higher dimensional math to boot). The geeks at IBM, amongst several other labs worldwide, have already figured out quantum teleportation. http://www.research.ibm.com/quantuminfo/teleportat ion/ Frankly, why would anyone ever even assume that someone would travel in a linear fashion, trundling along from star to star? Of course it's a waste of time and would take billions of years - and to assume that all foreign lifeforms would be restricted to a form of travel that we personally, within only half a century of space flight, could conceive of, is arrogant and shortsighted.

Ironically, relational fans (such as I) tend to think that polymorphism and inheritance are usually poor modeling techniques and way over used. Many OO fans have agreed even, recommending composition instead of inheritance. IS-A is out out out of style in OO circles.

You can keep using the phrase 'many OO fans', but that does not stop these features being of real use. Inheritance and polymorphism are fundamentally different ideas. Inheritance is overused, polymorphism underused.

The best GUI system would be mostly declarative because that would allow *multiple* languages to use the same GUI engine.

Multiple languages can already use the same GUI engine on the JVM.

OOP has not figured out a way to do that because each OO language is too different from each other.

No, see below.

If you can prove otherwise, you will be an OO hero instead of wrong. Declarative is more cross-sharable. Declarative == Sharable.

Then I am an OO hero, as I can easily prove you wrong. Here are some examples of multiple languages using the same GUI engine (Swing on the JVM):

Java (well, no examples needed there).
Groovy: http://www.oreillynet.com/onjava/blog/2004/10/gdgr oovy_basic_swingbuilder.html
JRuby:
http://www-128.ibm.com/developerworks/java/library /j-alj09084/
Jython:
http://www.uselesspython.com/Jython_Swing_Basics.h tml
Scala:
http://scala.sygneca.com/code/scalagui

I can provide many more if you wish.

Research? Where?

All the work of Alan Kay, Dan Ingalls, Adele Goldberg and so on.

Again, that is a query language syntax issue, not an OOP issue. (Some dialects support "natural joins" that do just that.) I would like to rework SQL myself, but that is another issue. If you are using a special magic query language that is better than SQL, fine, but that is not an OO victory even if true.

Yes, it is, because it is a query language that allows for inheritance, polymorphism and re-use. I can query for, say, a Contact, and I will automatically get subclasses. This allows for transparent extensibility.

Also, as I said, but you keep ignoring, what you retrieve are instances of classes that have been proxied and advised with additional code to allow transparent persistence. That is an OOP matter.

That is outright bad biz modeling.

Sorry, but I can't take you seriously. You are doing nothing more that claiming that anyone who takes a different approach from you is wrong!

You don't subclass Contact, you reference it. A customer, client, etc, may have multiple contacts. For example, a given company may have a billing contact, a sales contact, and a general (front desk) contact. Even Amazon has multiple contact options per customer. This is what happens when you think in trees instead of sets. I rest my case.

You haven't made any case. You can certainly model by referencing, but it depends on the complexity. In a situation where you don't have multiple contacts, inheritance is far simpler.

Dr. Codd and Charles Bachman faught navigational-versus-relational battles in the 70's and relational is generally considered to have won. That is until OO fans tried to resurrect navigational concepts. Navigational is modern-day GOTO's.

You had better tell CERN that - they use navigational models for the vast majority of their data collection, as relational systems are neither fast enough or flexible enough.

Again, a good many OO fans have lots of complaints and grumbles about OO/relational mappers.

That is true, but I am talking about modern mappers, not the way they were years

It's a little bit more than you state. WebSphere Application Server is the IBM workhorse for their web enabled applications. It's built into hundreds of IBM apps. There are several flavors Express, Base, Network Deployment, Process Server, but looking at the network deployment version, It's a full J2EE compliant application server containing:
- Apache web server
- Caching proxy (including content based routing and dynamic caching of servlets)
- Servlet container
- EJB container (+ JNDI + IIOP)
- Web Service Support
- Service Integration Bus (internal EJB)
- Support for JMS
- Clustering (session failover)
- Central administration of many servers and applications in a "cell" topology
- Strong Security (LDAP and others)
- JMX
- Java Connector Architecture connectivity to backends (databases, enterprise applications like SAP and Siebel, Mainframe connectivity to CICS and IMS, and many others).
- strong integration with other products like Tivoli Access Manager for front end security integration, and other Tivoli management products for monitoring (end to end transaction tracking through the application server - actually cool)

It serves as the base for WebSphere Portal and WebSphere Process Server (BPEL engine for running business processes)
Development can be done with Rational Application Developer (based on Eclipse) with a full WAS as unit test environment.

It's quite a potent thingie; check out:
http://www.redbooks.ibm.com/cgi-bin/searchsite.cgi ?query=websphere+AND+application+AND+server&Search Order=4

Indeed, of the 5 years I worked as a tech sales for the WebSphere brand (I quit IBM two years ago) in the IBM Software Group, Don Ferguson was the most charismatic software architect I've encountered. He has a gift of putting all aspects of software engineering in a perspectif. His style is extremely entertaining even if it is as if he dislikes too much attention.

A link to his blog might be interesting (see http://www-03.ibm.com/developerworks/blogs/page/do nferguson). Not too much information there, though. In particular it is surprising to see that his efforts of the last years are heavily related to PHP (hey, who out there realizes that the "father of WebSphere" is very much into PHP?).

Given that WebSphere and WebLogic are the two largest players in commercial app server market, mainly focusing on formal J2EE (in contrast to Spring, Hibernate and other non-formal frameworks), I interpret the departure of Ferguson towards MS as a continuation of the further decline of importance of formal J2EE development.

It seems they have some sort of pricing voodoo going on. Example:

With this announcement IBM is introducing Value Unit based pricing for the products referenced. Value Unit based pricing will help to align the prices of these products to the principle of the PSLC pricing curve which provides for a lower price per MSU for larger capacities. There will also be a price benefit when customers grow their capacity. Additional capacity will be based on the number of Value Units (MSUs) the customer has already installed. Additional capacity will not be priced starting at the base with a higher price per unit but on the capacity that is already installed.

Proof of entitlements (PoEs) will be based on new Value Units. Value Units of a given product cannot be exchanged/interchanged/aggregated with Value Units of another product.

Anyway it's a webserver and some applets. Here's a direct link to the list of stuff that's been stuffed into the Websphere brand envelope: SW By category

If they're running their website on it I feel sorry for their customers trying to do ecommerce -- getting a price is impossible, you can't proceed from the product page to the purchase, it keeps asking where I'm from, etc. etc.

But my heart really goes out to the poor soul that's got to translate that gibberish into meaningful chinese. I love IBM, but American Geek is my mother tongue and I can't make out what they're saying here.

I think it's more about us as human beings reaching our "biological" limit on how much we can understand the nature of universe.

No, that's not it. The problem is a lack of experimental data. We don't have the capability to conduct experiments at the scale at which superstrings are hypothesized to exist. But maybe somebody will find a way to do that. After all, there was a time when it was considered hopeless to ever take a picture of an atom.

Yes, this is a known issue with XML/SGML. In fact external DTDs can do more than just log the usage, they can also be cause of various security issues. See:

http://www-128.ibm.com/developerworks/xml/library/ x-tipcfsx.html

If you don't predict a branch, the SPE grinds to a halt for some ludicrous number of cycles while it fetches new instructions. The SPE has a huge decode buffer, and you need to warn it well in advance of where new instructions will be coming from if it's going to decode them in time. I don't know why you're saying they ren't pipelined, either; they are.

See also:
http://www-128.ibm.com/developerworks/power/librar y/pa-cellspu/

That means you have to do a fair amount of setup and partitioning specifically around that 256KB limit, which wouldn't apply on a multi-core PPC.

Yeah, right. There's a special spu-gcc compiler and linker to link spu code to PPC programs, and a threading model for loading and unloading code/data segments onto an SPU. Here's a short explanation, along with Linux install instructions:

http://www-128.ibm.com/developerworks/power/librar y/pa-linuxps3-1/?ca=dgr-btw01Linux-and-PlayStation 3

So, that's how Sony expects you to manipulate an SPU... you don't really get direct access to the hardware, you have to follow their threading model. But, since it's C, you can embed inline assembly if you want to go hard core.

The threading model looks pretty broken to me. You've got all these cycles lost in thread context switching, when the best approach is just to let something small run in chip without intervention. This is why I think monte carlos are perfect for Cell. Your real concern then becomes: Can I break my entire calculation down to six or so roughly equal steps, and cram enough of my dataset for each operation down to ... say 32KB or 64KB blocks?

Where I work a lot of folks use GEANT as their main sim engine. It just seems like a no-brainer to try a port to the Cell.

The main difference between the Blade and the PS3 configuration is memory limitations and the number of SPUs available. The Blades have 2 CBEs with 8 SPUs and 1 GB RAM each (so 16 SPUs, 2 GB total). The PS3 has 7 SPUs and 256 MB (under linux). There is also a networking limitation since inifiband can be added to the Blade system.

The memory difference is the biggest hurdle since it will limit your data sets. However it will allow developers to prototype. The code itself will be completely portable between both systems since they both run Linux and have the same interface and SDK to the SPUs.

There are developers using the PS3 for CBE app development. You could ask them directly http://www-128.ibm.com/developerworks/forums/dw_fo rum.jsp?forum=739&cat=46

developerWorks has a good article up about Linux on the PlayStation 3. It includes overview, installation, and first Cell BE processor programming steps.

Developers: grab the Linux Cell SDK from here:

http://www.alphaworks.ibm.com/tech/cellsw?open&S_T ACT=105AGX16&S_CMP=DWPA

Now that there is a readily available and fairly cheap consumer-level computer with a kick-ass Cell processor in it, it's time we start modifying some code to take advantage of it. Some libraries/applications that may benefit from the Cell include libdv, libxvidcore, libjpeg, libGL, libogg, libmp3lame, mplayer, xine, and mythtv, just for starters.

See this FAQ for how hard decimal arithmetic really is. Few programmers understand this.

a) good coding practices
b) formal peer reviews for pre-design, design, code, test specifications, and test results
c) Purify!!! http://www-306.ibm.com/software/awdtools/purify/ A license for every developer AND tester!

I haven't written any code since 1999, but that was how I setup the development team for that company. The reviews also are a form of cross training and team building. Nobody is perfect and showing our individual errors helps everyone fit in. OTOH, there was 1 guy who clearly didn't understand header files and was labled "Mr. Header" for almost a year. After the first 1 week of taunts by his peers, he quickly learned when and how to avoid putting too many header files into his code.

Other code issues were discovered and learned by the entire team. We didn't hide errors, we published them within the team and never told management anything about who caused what to happen.

RMONK, N.Y., October 4, 2006 IBM today revealed a cross-company effort to make the IBM System z mainframe the worlds most sophisticated business computer - easier to use for a greater number of computer professionals by 2011. The goal of this five-year effort, which will include an investment of approximately $100 million, is to enable technology administrators and computer programmers to more easily program, manage and administer a mainframe system - as well as to increasingly automate the development and deployment of applications for the mainframe environment.

I'm sure if someone made it easy to call web services from COBOL

I believe WebSphere does this. Supposedly you can pump COBOL data into EJB applications.

see:
http://www-306.ibm.com/software/data/ims/imsjava/j avcobol.html

Fair enough.

But do these chips come with 32Mb of L3 Cache, have the fastest Fiber Channel Bus Interconnect in the market, and allow for extremely flexible, multi-platform OS true hardware virtualization?

Performance comparisons between x86 and RISC chips in my opinion are really not valid. What you really want to look at is system workload. Scalability is where the POWER chips really perform and these chips are designed for the high-end server market.

see for yourself

You do realize that the CURRENT generation of POWER5+ CPU's are already quad-core, right? Honestly, guys, you all need to read up on what makes POWER different from PowerPC. One is a server or workstation class chip, the other is a desktop class one.

Move back? They were never on them. POWER6 != powerpc (though they are similar in more ways than not).

I think Apple is perfectly happy with the Intel move at this point. One of the reasons for the migration (if you can get past Jobs' reality distortion field of blah blah per watt or whatever) was that IBM wasn't able to keep up with demand, either with getting the speeds up, or with delivering the slow crappy ones they already had.

Additional details regarding minimum JVM versions required to address the timezone changes:

- Solaris/Windows/Linux: 1.4.2_11
- HP-UX: 1.4.2.11
- AIX: 1.4.2 SR 5 20060420

- Solaris/Windows/Linux: 1.5.0_06
- HP-UX: 5.0.3
- AIX: 5.0 SR 1 20060310

References:
HP: http://www.hp.com/products1/unix/java/DST-US.html
Solaris/Windows/Linux: http://java.sun.com/developer/technicalArticles/In tl/USDST/
AIX : http://www-1.ibm.com/support/docview.wss?uid=swg21 232128

The irony is, I was reading recently that Notes was released for Mac OS X :-) Not that I care -- I don't run Notes or Vista -- but it is funny.

Is this the modern update to the old myth that "DOS isn't done until Lotus won't run?"

• IBM
• System76

Debian is vastly popular in the embedded market, and it scales up to the monster s390 nicely

HP isn't the only one that publishes service information for their machines. For example, here's the service manual I've been using recently for Dell Latitude D620s at work:
http://support.dell.com/support/edocs/systems/latd 620/en/SM/index.htm Of course, similar manuals for most of their other models are on the website as well.

IBM has their manuals up for download too. Here's one I found real quick for a new(ish) IBM/Lenovo Thinkpad:
http://www-307.ibm.com/pc/support/site.wss/documen t.do?sitestyle=lenovo&lndocid=MIGR-62866

Now Apple, on the other hand, is usually pretty tight-lipped about how to dismantle their machines, especially the notebooks and the Mini. Not that you still can't find the service manual PDF if you know where to look...

"Push on Leonovo to offer something comparable to...." like the IBM/Lenovo ThinkPad T60p ?

That's not really "supported" for Linux. Read IBM/Leonovo's "Linux Certification - What does it mean? " page:

"Known problem areas and problems experienced in the past have been:

• Wireless support - drivers have not been provided for many wireless cards
• High-end video (3D acceleration for example) - basic video function typically works
• Certain implementations of Ethernet
• Power management - suspend/hibernation has been a problem area in the past
• Modems - driver support
• Mice - optical mice may not be supported by current mice suppliers
• Audio - problems in the past have been related to the core chipsets (some chipsets support Linux and some do not)
• Access IBM button - can define a program to launch, but no Access IBM
• Rapid Restore PC (Rescue and Recovery)
• Access Connections"

And this is a manufacturer writing about their own product., telling customers that, out of the box, it won't work right. That's not support.

Looks like X20 with additional 64 MB RAM module.

Laptop model is from link in his article.

http://www-03.ibm.com/systems/p/?ca=stg_pSeries&me t=search&me=W&P_Campaign=pseries

Sorry, your troll-fu is weak.

One of the big boys is going to come a 'callin: MS, Apple, or IBM. Good-bye "Reiser FS", hello "insert MarketingSpeak here" (I'm thinking someting inoffensive and corporate, like "YouFS" or "MSRULZ4VR").

I think you're confusing bottlenecks, but it's easy to do. At the very least, I may not have been entirely clear about which bottleneck I consider most important.

First, a short primer on how the SPE works. You can find a more in-depth explanation in Al Eichenberger's paper on IBM's site.

The SPEs have flat memory and software managed paging to help hide the latency of starting a new task on an SPE. A separate DMA controller brings code to the SPE's local memory, ideally well ahead of when it is needed. I think you're confusing the SPE's prefetch instruction with a traditional cache prefetch. The SPE uses a single high speed memory port to fetch instructions and data, and I'm pretty sure each can only access its local memory store. The SPE's fetch pipeline can hold 2.5 "fetch packets" of instructions, each packet containing 32 instructions. That prefetch amounts to 80 instructions, or 40 to 80 cycles of execution capability. (The SPE vector architecture can issue 1 or 2 vector instructions per cycle, and that's it.) Also, IIRC, branches can re-hit in this buffer, allowing tight loops to execute entirely from the prefetch buffer structure. This is entirely reasonable.

Yes, the ratio of compute power to memory bandwidth has increased enormously, but in the meantime, the amount of work the CPU does in each byte of memory has also increased noticeably. Furthermore, most interesting workloads have either good locality, or good access predictablilty. If that weren't true, then we wouldn't see noticeable gains on many workloads as CPUs got faster. Instead, we'd build ever wider memory interfaces to try to keep up. Indeed, memory interfaces have grown from 8 bits and 16 bits to now 128 bit and 256 bit. (A dual Opteron system with RAM populated on each memory port has a 256-bit wide memory interface, effectively.)

For graphics workloads, the access pattern ranges from moderately to very highly predictable. Hence the prevalence of specialized DMA engines and/or data prefetch instructions in many programmable graphics engines, including the Cell Broadband Engine. The PowerPC Altivec instruction set defines a set of streaming prefetch instructions for the same purpose. So, both PS3 and Xbox360 have well defined, well understood and effective ways to hide memory latency and to make the most of the bandwidth they have.

The RAM bottleneck I was referring to does not concern bandwidth or latency (though both are certainly an issue). It has more to do with working set. As scenes get more complex, it takes larger numbers of textures, vertices and everything else. (I hesitate to say "triangles," because they're not the only primitive you might deign to render.) Keeping all that render state in addition to world state and program code now becomes the challenge. Now the PS3 has a leg up here: The Xbox360 may not have a hard-drive, whereas the PS3 always has at least some HD. Paging textures and world data from optical media is tremendously painful. At least the PS3 can use its HD to page some of its state. Sure, hard drives are much slower than main memory, but optical media is much, much, MUCH slower than that. Think 10s of milliseconds vs. 100s to 1000s of milliseconds, depending on how much seeking you end up doing.

The more you can keep in RAM, the richer the world you can build, and the less you need to hit the spinning media. That's the bottleneck I was referring to.

Actually IBM/Lenovo have really improved their update rollout methods.

To find support documentation and downloads, you can go to Lenovo Support & downloads and enter in the model number of the computer in question. Or you may find drivers from the Driver Matrices page. Better yet, you can download the ThinkVantage System Update program which will scan your computer, check for available updates from the Lenovo servers, download, and install them for you.

The links provided are not secrets. They're all available from the "SUPPORT & DOWNLOADS" link off of Lenovo's homepage.

Actually IBM/Lenovo have really improved their update rollout methods.

To find support documentation and downloads, you can go to Lenovo Support & downloads and enter in the model number of the computer in question. Or you may find drivers from the Driver Matrices page. Better yet, you can download the ThinkVantage System Update program which will scan your computer, check for available updates from the Lenovo servers, download, and install them for you.

The links provided are not secrets. They're all available from the "SUPPORT & DOWNLOADS" link off of Lenovo's homepage.

Actually IBM/Lenovo have really improved their update rollout methods.

To find support documentation and downloads, you can go to Lenovo Support & downloads and enter in the model number of the computer in question. Or you may find drivers from the Driver Matrices page. Better yet, you can download the ThinkVantage System Update program which will scan your computer, check for available updates from the Lenovo servers, download, and install them for you.

The links provided are not secrets. They're all available from the "SUPPORT & DOWNLOADS" link off of Lenovo's homepage.

I thought the PS3 had an all-new processor called the Cell, nothing to do with PPC but designed from scratch to be massively parellisable and distributed - a cluster in a box in fact. Was I dreaming?

While the [Synergistic Processing Unit, or SPU] ISA is a novel architecture, the operations selected for the SPU are closely aligned with the functionality of the Power(TM) VMX unit.

My goodness people. Here we have a fairly inovative new designed "cell" processor designed by IBM toshiba and Sony. Of course its complexity more difficult to get run at full efficency, but the geek in me thinks the arcitecture is interesting. The compiler optimizations will come with time, like the PS2 whos games did end up looking significantly better over time.

http://www.research.ibm.com/cell/

Since Apple went x86 and Itanium sank, there are very few companies even trying anything new in the cpu world. My hats off to them and I hope it works out well.