Slashdot Mirror

The only mac I've got around is a PowerPC mac, which Apple abandoned while people were still using them to do work. On the down side, Apple abandoned it long ago, because they don't give one tenth of one shit about customers who aren't willing to keep forking out new money whether their old hardware was doing the job or not. On the up side, the only POWER/PowerPC processors vulnerable to MELTDOWN or SPECTRE are Power7 through Power9 processors. Sadly, they are vulnerable to both attacks, suggesting that IBM played the same kind of games with security as Intel.

They've reinvented the IBM SyNAPSE chip.

https://www-03.ibm.com/press/us/en/pressrelease/44529.wss

Seriously, haven't they rediscovered the neural network chip? For about the third or fourth time?

I read nothing in that link that says Power processors are vulnerable to Variant 3, commonly called Meltdown.

That article outright states that Power7 through Power9 processors are vulnerable to MELTDOWN, so no. And it links to IBM's statement on MELTDOWN and SPECTRE, in which they outright state in regards to MELTDOWN that "Complete mitigation of this vulnerability for Power Systems clients involves installing patches to both system firmware and operating systems." So it both makes the statement, and provides the citation. Learn to internet, son.

POWER: https://www.ibm.com/blogs/psir...
SPARC: Still no official public statement from Oracle

IBM Mainframes not impacted.

Itanium handles speculation so differently that it's unlikely to be vulnerable. IBM has released an advisory indicating that Power7+/8/9 are vulnerable to some extent (they don't distinguish between Meltdown and Spectre) and that patches would be rolling out soon.

I've worked at places that used Clearcase and it's a brilliant version control system - with dynamic views/MVFS you could do a binary search on a version tree of a file to see where a bug was introduced.

E.g if you know a bug was introduced between version 0 and version 100, try version 50. If it's still there try version 25 and so on. MVFS, dynamic views and config specs makes this easy. And they even had a version of make called clearmake that knew how to avoid doing a rebuild all

https://www.ibm.com/support/kn...

And the tools had the concept of a trivial merge - if there was only one set of changes, cleardiff was able to automerge them.

And they had a nice graphical version tree browser so you could see all the branches and merges for a file. And all the labels too. I.e. which versions of the file had passed QA tests.

It was all great, better than any version control system I've used since.

Then again it's like vxWorks really - if your client is paying for the infrastructure and licence fees, it's great. If you had to pay yourself you'd slum it with git or svn.Then again if you're the only one working, you probably don't need to do a binary search in version space to find out where someone broke something in a very subtle way.

I know it's /. and people don't read the summary, but

Why are we still clinging on to Helvetica?

is a big hint, which is easily confirmed: Helvetica is what IBM currently uses as its primary typeface. Comparing to anything else would therefore make less sense.

Shake your head and wake up man, you're in the Apple bubble. Swift will never be used much on non-Apple platforms

I'm afraid it is you who seem to be very much asleep.

Wellll.... I very much enjoy my hacking in Swift. But I don't give it a big chance of taking off among the open source crowd. Yeah it runs on Linux, and I love it for that. But I think we'll see adoption like Mono: used here and there, but not in any major way.

That said, Swift strikes a number of very nice balances in my opinion. I really like the expressiveness of the language, especially how easy it is to throw in some .filter {} , .map {}, and what have you.

Shake your head and wake up man, you're in the Apple bubble. Swift will never be used much on non-Apple platforms

I'm afraid it is you who seem to be very much asleep.

just like Objective-C before it

That is why I know I'm right on this, because I fully agree with you about ObjC - I could tell that was not going to be of use outside of Apple's platforms (though I liked working in it anyway).

But having done a wide range of programming in the past, from Java to C and C++ and Javascript and Lisp and various flavors of assembly, sometimes all server, sometimes all client, sometimes dedicated hardware... because of the broad range of past experience I have the ability to tell now when a language will be a good fit for a project. And Swift is a VERY good fit for both client and server work, and after some time, even low level work.

Swift is not just controlled by Apple, at all - it has a thriving community driving development, and not just from Apple devs at all. It's just a really good mix of ides from many different modern languages, with a very well thought out syntax and pragmatic approach to development.

But it's way more than that, it's one of the few languages (maybe the only one?) that really embraces what LLVM can do and takes full advantage of it...

"Still, a practical reversible computer has yet to be built using this or other approaches."

Since quantum computers of any kind have to be reversible due to the very nature of QM, every realization of quantum computation is a reversible computer.

This includes the controversial D-Wave machine as well IBM's QC chip that you can play with online.

Last year they funded the Center for Cognitive Computing Systems Research at the University of Illinois Urbana-Champaign. http://www-03.ibm.com/press/us...

Clearly, you are not following their quantum computing research.

https://www.research.ibm.com/i...

Stauffer Chemical and IBM Think magazine https://www-03.ibm.com/ibm/his...

There you go: available for both z/OS and AIX!

http://www-03.ibm.com/software...

Nope, really. On encountering /* in the first two columns you stop reading and go to the next line.

Continuing to maintain an entire operating system platform to support software that was written before this year's college grads were even born is just plain insane. That's way too much effort for what should be almost zero benefit.

Ever heard of MVS, VM, VMS, or UNIX? All of them (and many others) were developed 40+ years ago and still support software written 4 decades ago today - and are running in many, many commercial environments on current hardware.

Prior to Windows 10 you could disable automatic updates entirely

That's kind of the point regarding the Win10 you must update statement that started these discussions.

Absent WSUS, which we were only without for a year and a half,

Really? (W)SUS didn't come out until 2005. Last I recall, XP was released in 2001.

So, you mean, by the time most businesses would have upgraded to XP, WSUS was out and this whole back-and-forth is largely pointless?

Nice rewording there. I said companies remained on XP. Nothing about XP's own take up could be inferred from that statement.

I stipulated competent IT teams, not just dedicated.

So MS Services isn't competent? I'll be sure to note that next time.

You're living in MS fairytale land. I can assure you that small businesses don't as a rule run WSUS, nor have IT folks that deal with it.

Does that mean they can't? I mean, if all of this is really a concern and there is a solution available, why can't they utilize that solution?

Sure, they can, if they'd prefer to run in the red.

Then they should have an AD to manage logins, at the very least. It costs less to pay someone to click a few buttons to add and remove accounts on a central server than it costs to have them walk across the building to do the same thing. Bonus if they install even a low-end SAN solution and store user profiles and documents on it; then they don't even have to reimage machines when someone leaves the company. These are things that should be considered once a company reaches about 20-25 workstation users and should certainly be in place by 50.

It's a solid argument. Many won't pay for it. I've seen 5 year old Dell laptops with busted keys and cracked screens in daily use. If they won't spend $300 for a new base laptop, you really think you're going to get them to pay $50+/hr for IT support?

Apple, on the other hand, killed off the server version of OS X and never bothered migrating the management tools; those are just gone now. If I recall correctly, Microsoft has actually stepped up to fill that gap on the Mac platform.

It's IBM.

Most companies don't run local mail anymore; it's too much of a hassle to deal with RBL bullshit and spam. What's mail got to do with this, anyway?

Essentially, pointing out that while they use computers, they aren't IT shops in any sense of the word.

You can't, on one hand, say downtime costs tens of thousands of dollars (30% of a Fortune 100's workstation users being unable to work for 2 days), then turn around and say $5000 is too much to pay to fix it. If something is going to cost me $10k to ignore or $5k to fix, the reality is that it's actually going to save me $5k to fix it.

Who said anything like that? Honestly, those kind of remarks are bordering on Trumpian claims. A small shop that uses computers won't be idle for days if their systems are down. It'll be inconvenient, maybe, but not serious. A Fortune 100 has a dedicated IT staff. This portion of the discussion doesn't apply to them.

You've been able to use Python for a while in Postgres, MySQL, SQLite, and even DB2.

I can't quite figure out why anybody would want to use Microsoft SQL Server.

B52s and IBM mainframes are not even closely related.

The only original part of a B-52 bomber today is the airframe, as everything else got replaced, modernized or removed over the last 50+ years.

If you say your mainframes are over 40 years old, you are talking about 360 or 370 machines.

Or an IBM 370-compatible mainframe that came out.... 40 years ago.

https://www-03.ibm.com/ibm/history/exhibits/mainframe/mainframe_FT2.html

...and it is, according to nearly every engineer in the autonomous vehicle business, including the head of Google's autonomous vehicle project, unsolvable. It is at the core of the current regulatory conflict between legislators, who want to keep a human in the loop, and most autonomous vehicle makers, who want humans out of the loop because of the unsolvability of the hand-off problem. Google has already stated they will not produce their autonomous vehicles until the government agrees to remove the human-in-the-loop requirement for operating autonomous vehicles on public roads.

The major exception to this received wisdom is Elon Musk at Tesla Motors, who pretty much believes that no problem is unsolvable. Yay for him; we need incurable optimists. If we are ever going to think our way out of the crap-sack world we are currently headed for, it will be because of technological optimists like Musk.

I realize that betting against Elon Musk is probably not a good strategy, but I do think the hand-off problem is not solvable, for any value of "solved" you care to assert. Human task switching is not fast enough at pedestrian velocities, let alone autobahn velocities. Dragging a human from their porn/spreadsheet/email/phonecall/whatever and expecting him/her to correctly grok a traffic problem in a fraction of a second is not realistic.

Or even several fractions of a second, if IBM's cognitive modelling of human drivers, which is what this expert system is actually doing, successfully expands the window for a driver to react. I am not sure that replacing a human with a robot that is programmed to learn to drive like a human is a win, but we'll have to see what happens. If we don't have a spike in traffic fatalities in BMWs (IBM will be rolling this package out in BMW models first, because they have already announced they are giving Watson to BMW) I will happily revise my estimation of the hand-off problem's solvability. :)

Any idea what some IBM employees found objectional in CEO Ginni Rometty's letter to US president Donald Trump? https://www.ibm.com/blogs/poli...

Was it that he congratulated Trump on his victory? Otherwise it seems very business focused and not political at all.

Dang.

https://www.ibm.com/developerw...

Of course as with many IBM products, the miraculous setups are always in IBM labs.

I'm all for distributed systems, but for many big companies, mainframes still make a lot of economic sense:
http://www-03.ibm.com/systems/...
"While some believe that smaller distributed servers provide the agility needed in today's fast-moving cognitive era, the IBM mainframe is the preferred solution for many of the world's most competitive businesses, including:
92 of the top 100 banks worldwide
70%+ of the world's largest retailers
23 of the world's 25 largest airlines"

And see also, on a smaller scale:
https://en.wikipedia.org/wiki/...
"IBM designed IBM i as a "turnkey" operating system, requiring little or no on-site attention from IT staff during normal operation. For example, IBM i has a built-in DB2 database which does not require separate installation. Disks are multiply redundant, and can be replaced on line without interrupting work. Hardware and software maintenance tasks are integrated. System administration has been wizard-driven for years, even before that term was defined. This automatic self-care policy goes so far as to automatically schedule all common system maintenance, detect many failures and even order spare parts and service automatically. Organizations using i sometimes have sticker shock when confronting the cost of system maintenance on other systems.[1]"

In general:
"Why on Earth Is IBM Still Making Mainframes?"
https://www.wired.com/2015/01/...
"Business is more mobile than ever. Yet however lightweight those mobile devices feel in your pocket, they can still make good use of a big, powerful machine chugging away in a back room, not going anywhere."

Mainframes are also more than just hardware. Mainframes are in a sense a culture of 100% uptime and reliability.

That said, distributed computing continues to improve... And distributed computing culture continues to improve...

As to the original article, IBM is still shooting itself in the foot with this move away from supporting remote work... What IBM needs to be creative is not colocation but "slack" in the Tom DeMarco sense:
https://www.amazon.com/Slack-G...
"Why is it that today's superefficient organizations are ailing? Tom DeMarco, a leading management consultant to both Fortune 500 and up-and-coming companies, reveals a counterintuitive principle that explains why efficiency efforts can slow a company down. That principle is the value of slack, the degree of freedom in a company that allows it to change. Implementing slack could be as simple as adding an assistant to a department and letting high-priced talent spend less time at the photocopier and more time making key decisions, or it could mean designing workloads that allow people room to think, innovate, and reinvent themselves. It means embracing risk, eliminating fear, and knowing when to go slow. Slack allows for change, fosters creativity, promotes quality, and, above all, produces growth."

That was the great thing about IBM Research when I worked there around 2000 -- a bit of slack to be creative and good work/life balance. But, IBMers even then said the rest of IBM was not like Research...

The good news: new harddrives will have the size of a pin. The bad news: it'll require the scanning tunneling microscope attached to work: http://www-03.ibm.com/ibm/hist...

So, now, it'll require more 10 years to work to reduce the scanning tunneling microscope to the size of current harddrives.

You've really missed the point. You are after complexity of the OS so that you can do complicated things with the OS. But most device users are after simple yet powerful solutions for the complex problems they face. Like airline pilots who use IBM's Plan Flight app to calculate fuel loads. I'm pretty sure pilots could cope with a sub-folder, but they really don't *want* to, nor do they *need* to -- they'd rather have big-data-driven decision support for optimising fuel loads presented in an intuitive, fast-to-use format.

http://www.ibm.com/mobilefirst...

You think you're arguing for sophistication and intellect, but really you're arguing about the line between "as simple as possible" and "no simpler". Your use cases differ wildly from most of the billions of the users of iOS devices in where you feel the need for complexity.

Yes, you can RAID memory. See https://www.ibm.com/developerw...

Exactly: The Apple computer open hardware and software architecture was the key of its success. Same thing happened with Commodore 64 and ZX Spectrum. The success was so huge that IBM with its Personal Computer made the same ope hardware stance. For them making a closed system was not a problem, they did actually made some models.
Unfortunately now the typical computer buyer is totally different from the one of the '70, actually the typical computer buyer prefers a smartphone and doesn't try to make specialized things.
A closed system isn't a problem for the typical user that theat its computer like an appliance and is trained to throw away it even if it's working.

Assuming that you are talking about IBM's Common User Access Spec, there's a copy on the internet at http://publib.boulder.ibm.com/...

Given that it dates from the 1980s, it's pretty impressive really. CUA compliant software might be a bit clunky by modern standards, but it would clearly be reasonably usable -- at least on a device with a multiple line text capable display and a typewriter keyboard with function keys. That's more than can be said for a lot of more modern stuff.

The management gets bonuses based on short term goals.

Read up on a company and their corporate governance / executive compensation policies. IBM and Nokia both reward both short and long term goals - and each describe it in extensive detail:
https://www.ibm.com/investor/s...
https://www.nokia.com/sites/de...

I'm a long-term investor in IBM, so I'll focus on them in some detail. IBM has a policy to spend 6% of their annual revenue on research & development. Most long investors (myself included) would rather see IBM increase that number to match their peers in the industry (~10-15%) vs. using that same money to pay out dividends every quarter. Additional concerns include the constant turrnover of talent - both voluntary and involuntary - which most long investors agree has lead to the demise of what was once a world class consulting organization. Some argue that the Indian IT firms (Tata, Wipro, Infosys, etc.) with their inherent lower cost structure was the root cause - but I disagree, IBM dropped the ball on this and is now quietly trying to become a "cloud" company to hide their mistake. A third concern within the past few years is their company reorganization. You cannot compare an annual report from 2012 to the one that will be released in 2017 and attempt to determine how well the company has done. Why? All business segments of the company that existed in 2012 no longer exist due to IBM reorganization. I'm keeping an eye on how well they perform over the next few years post reorganization with the new businesses they are diving into. May well sell my stake if their key financial metrics do not pick up by 2020.

Long post to say - you can criticize a company all you want for the way they do business - but please do not try to simplify it down to "its because of the evil management doing cost cutting to make fat bonuses!". Its a bit more complex than that.

. . . is an interesting, and potentially lucrative idea

Urban farming is already extremely lucrative! Except, the crops grown are only rarely eaten, and more often smoked.

German politicians are even trying it out on their own rooftops in Berlin, as can be seen in this Ice Bucket Challenge video:

https://youtu.be/REOA3xXR8tI

Hmmm . . . now what is that plant next to German politician Cem Özdemir . . . ?

"E-I, E-I, O, jawohl!" : http://www.ibm.com/support/kno...

This is a good opportunity to talk about why security through obscurity is bad:

Your typical credit card number has a theoretical 16 digits that are available. That's a huge number (9,999,999,999,999,999) that makes it look effectively impossible to guess. Let's pare that number down to size.

First, you aren't guessing anywhere near 16 digits. It turns out there's a lot you already know (1st digit is 4 for visa, 5 for mastercard etc.). That reduces the typical address space from 16 to 15 digits. That first number turns out to actually just be part of the bank identification number which is typically 6 digits long. All of the rest of it except for last digit is the actual account number. The last number itself is used for a checksum (Luhn) that is used to verify the number is good.

In other words to get the account number right you've only got an address space of 999,999,999. That's a significant reduction in magnitude to start with. Now let's go back to that Luhn checksum (it isn't a hash). Due to this detail you can easily validate the number to make sure that you haven't mistyped it (Luhn precedes using magnetic tape for credit cards).

The Luhn check uses a Mod 10 algorithm that excludes 90% of the previous address space. You now have 99,999,999 numbers to guess against. Your malicious actor isn't starting work in a quadrillion space number, they're working in the millions. All of that is just from the industry standards themselves. Now remember that each bank is going to have their own formulas for generating credit card numbers and that card thieves have data sets of the tens of millions - old dumps are good for providing data that can show patterns. This is a good example of how data at the aggregate level carries risk that it doesn't at the micro level.

Chances are the account number for the card itself isn't at all random. Chances are really good that the formulas used to generate these numbers for a number of large popular banks have been reverse engineered by any number of parties. You also have policies at many banks such as never reusing a number that also reduce this address space. All the malcious actor has to do is look for patterns. Patterns have a way of reducing the order of magnitude once you learn them.

The expiration dates themselves are typically within 2 years giving a range of only 24 to pick from for the typical transaction. Guess a valid account number, try it at 24 websites and chances are really good one of them will work. That leaves the CVC2 number itself, which of course isn't random either.

The system is broken, it's just a matter of time before industry must recalibrate how it works.

More below for those who are curious:
http://www.creditcards.com/cre...
http://datagenetics.com/blog/j...
http://www.darkcoding.net/cred...
http://blog.opensecurityresear...
http://www.ibm.com/support/kno...

I'm creating one software about this Subject and need good places to search...

Roger
Blog Programa de Reconstrução Capilar
http://dhtequedadecabelo.quetudo.com.br

This is a good opportunity to talk about why security through obscurity is bad:

Your typical credit card number has a theoretical 16 digits that are available. That's a huge number (9,999,999,999,999,999) that makes it look effectively impossible to guess. Let's pare that number down to size.

First, you aren't guessing anywhere near 16 digits. It turns out there's a lot you already know (1st digit is 4 for visa, 5 for mastercard etc.). That reduces the typical address space from 16 to 15 digits. That first number turns out to actually just be part of the bank identification number which is typically 6 digits long. All of the rest of it except for last digit is the actual account number. The last number itself is used for a checksum (Luhn) that is used to verify the number is good.

In other words to get the account number right you've only got an address space of 999,999,999. That's a significant reduction in magnitude to start with. Now let's go back to that Luhn checksum (it isn't a hash). Due to this detail you can easily validate the number to make sure that you haven't mistyped it (Luhn precedes using magnetic tape for credit cards).

The Luhn check uses a Mod 10 algorithm that excludes 90% of the previous address space. You now have 99,999,999 numbers to guess against. Your malicious actor isn't starting work in a quadrillion space number, they're working in the millions. All of that is just from the industry standards themselves. Now remember that each bank is going to have their own formulas for generating credit card numbers and that card thieves have data sets of the tens of millions - old dumps are good for providing data that can show patterns. This is a good example of how data at the aggregate level carries risk that it doesn't at the micro level.

Chances are the account number for the card itself isn't at all random. Chances are really good that the formulas used to generate these numbers for a number of large popular banks have been reverse engineered by any number of parties. You also have policies at many banks such as never reusing a number that also reduce this address space. All the malcious actor has to do is look for patterns. Patterns have a way of reducing the order of magnitude once you learn them.

The expiration dates themselves are typically within 2 years giving a range of only 24 to pick from for the typical transaction. Guess a valid account number, try it at 24 websites and chances are really good one of them will work. That leaves the CVC2 number itself, which of course isn't random either.

The system is broken, it's just a matter of time before industry must recalibrate how it works.

More below for those who are curious:
http://www.creditcards.com/cre...
http://datagenetics.com/blog/j...
http://www.darkcoding.net/cred...
http://blog.opensecurityresear...
http://www.ibm.com/support/kno...

It's an "open" version of IBM's Coherent Accelerator Processor Interface which is basically a specification for how programs can interface with specialized "accelerators" (usually FPGAs) by writing to designated sections of RAM. This method minimizes the number of alternations that need to be made to motherboards, allows the use of standard CPUs and standardizes the unifies the changes that need to be made to the kernel.

TL;DR: supercomputing nerd stuff

Share of users, share of traffic, and share of revenue differ.

Safari has a 4.28% market share on the desktop

Among people willing to buy products and services from Internet businesses, the market share is probably bigger than 4.28%.

and a 14% share on mobile.

Among people willing to buy products and services from Internet businesses, the market share is probably bigger than 14%. Tim Stenovec of Tech Insider summarized a report from IBM Commerce stating that iPhone and iPad users outspent Android users over 3 to 1 on Black Friday 2015. The average iOS user also spends over 9 times as much on paid apps and IAPs than the average Android user.

Assume for a moment that the average iOS user spends 20 times as much money online as the average Android user. If that's the case, then 14% of the users represent 14*20/(14*20+86)*100 = 77% of the money. I don't remember where I read this 20-to-1 figure, but it at least appears consistent with the IBM Commerce report.

If you're into it (and like languages), you might be interested in learning APL. Gnu APL seems to work alright, and there's some good documentation. I've had to make my own key-mapping, but all the needed characters are in unicode, and that's part of the fun.

Admitted, I'm just another guy debating a topic I don't know much about, but won't layering components on top of each other result in massive heating issues? I mean, the heat from each layer has to go somewhere, right?

Yes. That's why IBM, among others, has been fabricating cooling capillaries into chips. They're experimenting with inter-layer liquid cooling through tubes just a few microns wide, imitating physical shapes found in the smallest of blood vessels to keep the fluid moving.

Print up a petition, have your other local neighbors all sign it. Take this in to your state senator's actual office as opposed to just having a curbside conversation. Send a copy to all your local and state newspapers and your city council as a "letter to the editor" and tell the Senator that this is now an "official issue". Curbside chats aren't "official business", petitions delivered as a "constituent concern" in-office usually are. I'd make sure to list in the petition that eminent domain should be a "last resort" if some type of agreement can't be realized. I'd also make the petition as specific as possible, using some SLA-type language in it.

Here are some SLA examples to work off of. Good luck!

You could always give that IBM quantum computing service a shot if you're interested...

A foreign language has been creeping into many of the presentations I hear and the memos I read. It adds nothing to a message but noise, and I want your help in stamping it out. It's called gobbledygook. There's no shortage of examples. Nothing seems to get finished anymore it gets "finalized." Things don't happen at the same time but "coincident with this action." Believe it or not, people will talk about taking a "commitment position" and then because of the "volatility of schedule changes" they will "decommit" so that our "posture vis-à-vis some data base that needs a sizing will be able to enhance competitive positions." That's gobbledygook. (February 19, 1970)

Also on topic: the turbo encabulator.

This is not a new phenomenon, unfortunately.

Now that we're seeing some real progress with making Swift a cross-platform language, I think it's time for Mozilla to drop their Rust project.

For those who don't know, Rust's home page describes it as "a systems programming language that runs blazingly fast, prevents segfaults, and guarantees thread safety." Well that sounds an awful lot like what Swift is!

While Swift has already seen major adoption and use in the real world, we haven't seen that from Rust. There have been a handful of projects using Rust, but that's it. And the ones that have used it, including the Rust implementation itself and Mozilla's Servo browser engine, have not been very impressive so far.

Despite lofty goals and a lot of hype, we haven't seen anything much of substance come out of Rust. Many have complained that it's an awkward language to use, even more than C++ is. Others have pointed out that C++14 and C++17 actually make much of Rust redundant. Yet others point out that Rust's safety guarantees are only as good as its implementation, which suffers from thousands of bugs despite much of it being written in Rust by the creators of Rust!

There was recent discussion about Google adopting Swift for Android. IBM has also taken an interest in Swift. Instead of continued failure with Rust, which is essentially a proprietary language at this point even if it's developed in the open, Mozilla should really consider using Swift, too.

It wouldn't just be better for Mozilla. It would be better for all developers. Swift is quickly becoming a universal language. The large number of developers proficient with Swift could contribute to Mozilla's code bases, rather than just a small handful of niche developers who know Rust.

Swift is the future. It's where software development is heading. Mozilla can do the sensible thing by getting rid of Rust and moving to Swift early on, or they can continue to waste time and effort on Rust. I sure hope they make the sensible decision! A modern web browser written in Swift would be much more useful than a web browser written in Rust.

We really should applaud them. Imagine how hard it will be to figure out how to write code to hack this.

I infer from IBM's use of "micro-operation" in papers about recent z/Architecture processors that it's also true of z/Architecture.

IBM is huge on microcode,

So I guess they abandoned that whole silly "801" "RISC" project when they found it didn't have any microcode. Good thing they did, or they'd have ended up with a line of RISC processors.

I presume that POWER itself when implemented for RS6k has about a normal amount of microcode; I know there is some.

And how do you know this?

Last I looked, AS/400 systems even when they were still being called that were being implemented as a wrapper around POWER or even PowerPC, depending on model.

The predecessor to AS/400, System/38, was an interesting system. They had two layers that they referred to as "microcode" but, as Frank Soltis, one of the chief architects if not the chief architect, noted in one of his books on AS/400, that was a legal fiction for the higher level of "microcode". The underlying processor had a System/3x0-ish instruction set, implemented on a processor that used horizontal microcode to implement it. It then had something called "vertical microcode", made up of instructions in that System/3x0-ish instruction set, running from main memory. However, the uber-CISCy instruction set generated by compilers/a> doesn't get interpretively executed by the "vertical microcode"; instead, the "vertical microcode" translates the MI instructions into the same instruction set as the "vertical microcode" uses, and runs it from main memory as well.

Soltis says this was done because IBM were worried that they might be forced to sell their OS software for use on compatible hardware, and they wanted to make sure that didn't happen with System/38, so they put the development of the low-level OS and binary-to-binary translation code under a hardware manager, and called it "vertical microcode", so they could claim it wasn't system software and didn't have to be unbundled.

That continued with AS/400 - which allowed them to replace the old CISC instruction set with an extended PowerPC instruction set without breaking binary compatibility - the OS would just retranslate the MI instructions to PowerPC code and run that. (The result of the translation are cached on disk along with the "source" MI code; apparently, AS/400 allowed the "source" code to be discarded, presumably to save disk space, and if you did that, you'd have a problem switching to the RISC machines.)

And AFAIK that is basically true of all their hardware now, it's all POWER-based at some level.

Nope, the z/Architecture microprocessors natively run the z/Architecture instruction set; the designers of the z/Architecture and POWER chips might talk to each other and use some hardware in common, but the z/Architecture CPUs are not POWER-based.

Website about the research(note, there is a list of papers too... this "news" is actually from 2015) https://www.research.ibm.com/l... Youtube video: https://youtu.be/q3dIw3uAyE8 And yes... they only have a prototype. There are still lots of possibly, maybe, somedays in this announcment.

Except for IBM mainframes. Support diversity, buy IBM mainframes!

Oh, wait...

They can't. They will be changing the name shortly, once IBM lawyers get involved.

Here is the link for those that have never heard of AIX .. http://www-03.ibm.com/systems/...

I pass the exam, and you? M2150-728,P2065-749 I recite dumps before examination, I think it is very valid and accurate for your reference, P2065-749 Free Demo Download: http://www.itcertkey.com/P2065... NO.1 When importing a file in ANB, which functionality is used to separate e.g. first name and last name into different columns? A. Column Actions B. Conditional Formatting C. Assign Columns D. The file must be prepared externally, in excel Answer: A P2065-749 Real Exams NO.2 The event listener is part of: A. Write server B. None of the above C. Analyst desktop D. Read Server Answer: B P2065-749 test questions NO.3 Within the i2 Intelligence Analysis Platform the primary location for good quality information and intelligence is the group analysis repository. What are the three modes for creating and augmenting information? A. None of the above B. Manually, on demand and Ad-hoc C. XLS. CSV and TXT D. Auto, bulk, and csv Answer: D P2065-749 test NO.4 What are the benefits of deploying IAP on multiple servers? A. Middleware, allows access to intelligence from anywhere at any time B. Response times can be very slow: data is not easily available over a WAN C. All of the above D. With two servers, the internal database locks cannot occur and the system structure is consequently much more robust. Answer: B P2065-749 exam dumps Reference: http://www935.ibm.com/services... _whitepaper.pdf NO.5 In order to utilize multiple repositories either within IAP or external, what application within NDI would a user use? A. Dynamic import tool B. Data must reside within IAP C. None of the above D. Intelligence Portal Answer: D P2065-749 exam simulations Reference: http://www-01.ibm.com/support/... Test information Number of questions: 15 Time allowed in minutes: 30 Required passing score: 66% Languages: English, German, Italian, Japanese, Spanish Castilian IBM Value Package sales mastery test benefit Want to take this test for free? You can if your company has a Value Package. Go here for more details! PartnerWorld: After you pass a test, be sure to add your Candidate ID to your company's PartnerWorld profile! You will receive an e-mail from the test vendor after the test which will provide your Candidate ID.

I pass the exam, and you? M2150-728,P2065-749 I recite dumps before examination, I think it is very valid and accurate for your reference, P2065-749 Free Demo Download: http://www.itcertkey.com/P2065... NO.1 When importing a file in ANB, which functionality is used to separate e.g. first name and last name into different columns? A. Column Actions B. Conditional Formatting C. Assign Columns D. The file must be prepared externally, in excel Answer: A P2065-749 Real Exams NO.2 The event listener is part of: A. Write server B. None of the above C. Analyst desktop D. Read Server Answer: B P2065-749 test questions NO.3 Within the i2 Intelligence Analysis Platform the primary location for good quality information and intelligence is the group analysis repository. What are the three modes for creating and augmenting information? A. None of the above B. Manually, on demand and Ad-hoc C. XLS. CSV and TXT D. Auto, bulk, and csv Answer: D P2065-749 test NO.4 What are the benefits of deploying IAP on multiple servers? A. Middleware, allows access to intelligence from anywhere at any time B. Response times can be very slow: data is not easily available over a WAN C. All of the above D. With two servers, the internal database locks cannot occur and the system structure is consequently much more robust. Answer: B P2065-749 exam dumps Reference: http://www935.ibm.com/services... _whitepaper.pdf NO.5 In order to utilize multiple repositories either within IAP or external, what application within NDI would a user use? A. Dynamic import tool B. Data must reside within IAP C. None of the above D. Intelligence Portal Answer: D P2065-749 exam simulations Reference: http://www-01.ibm.com/support/... Test information Number of questions: 15 Time allowed in minutes: 30 Required passing score: 66% Languages: English, German, Italian, Japanese, Spanish Castilian IBM Value Package sales mastery test benefit Want to take this test for free? You can if your company has a Value Package. Go here for more details! PartnerWorld: After you pass a test, be sure to add your Candidate ID to your company's PartnerWorld profile! You will receive an e-mail from the test vendor after the test which will provide your Candidate ID.

We know now that we don't need any big new breakthroughs to get to true AI.

Grossly exaggerated claim. The following article worth reading on this subject by no one else than two of authorities in the field, one did work on the backgammon game in the 90s and the other one on the IBM Deep Blue program that win over the world chess champion Garry Kasparov in 1997. http://www.ibm.com/blogs/think... In particular:

"However, research in such “clean” game domains didn’t really address most real-life tasks that have a “messy” nature. By “messy,” we mean that, unlike board games, it may be infeasible to write down an exact specification of what happens when actions are taken, or indeed what exactly is the objective of the task. Real-world tasks typically pose additional challenges, such as ambiguous, hidden or missing data, and “non-stationarity,” meaning that the task can change unexpectedly over time. Moreover, they generally require human-level cognitive faculties, such as fluency in natural languages, common-sense reasoning and knowledge understanding, and developing a theory of the motives and thought processes of other humans."

in the US:

http://www.predpol.com/

http://www.ibm.com/software/an...

http://www.motorolasolutions.c...

http://computerstories.net/mic...

PDF from RAND:

https://www.ncjrs.gov/pdffiles...

and so on and so on.

Is there a compelling reason to prefer Linux to the BSDs when it comes to the mainframe?

Well, there's the fact that the mainframe manufacturers (read: IBM) actively support Linux on their systems, and will happily sell you a mainframe with Linux pre-installed. In fact, that's more-or-less the standard configuration these days, as I understand it. http://www-03.ibm.com/systems/...

I know none of the mainframes had Unix running on them [...]

Your knowledge is extremely out of date. RH, SUSE, and even Debian have been actively supporting IBM mainframes for years, with active help and support from IBM. Linux has been running on mainframes in datacenters for over a decade.

does [AIX] have more similarities to Linux than it does to BSD?

While this is a less relevant question than you thought, the answer is still yes. Linux—or, more specifically, GNU—generally steered a middle course between SysV and BSD, and, where it could, would implement compatibility with both. So, overall, Linux—or, at least, the flavor of Linux sometimes referred to as "GNU/Linux"—is closer to both SysV and BSD than either is to the other. (Although the differences between BSD and SysV have also diminished over time as BSD has adapted to become more flexible itself.)