Why OldTech Keeps Kicking 339
Hugh Pickens writes "In 1991 Stewart Alsop, the editor of InfoWorld, predicted that the last mainframe computer would be unplugged by 1996. Just last month, IBM introduced the latest version of its mainframe, and technologies from the golden age of big-box computing continue to be vital components in modern infrastructure. The New York Times explores why old technology is still around, using radio and the mainframe as perfect examples. 'The mainframe is the classic survivor technology, and it owes its longevity to sound business decisions. I.B.M. overhauled the insides of the mainframe, using low-cost microprocessors as the computing engine. The company invested and updated the mainframe software, so that banks, corporations and government agencies could still rely on the mainframe as the rock-solid reliable and secure computer for vital transactions and data, while allowing it to take on new chores like running Web-based programs.'"
Is it really "old" tech? (Score:5, Insightful)
Why OldTech Keeps Kicking? (Score:4, Insightful)
Basic psychology. People stick with what they're used to, even if it doesn't always make the most sense.
Re:Is it really "old" tech? (Score:4, Insightful)
can be argued for other things too (Score:4, Insightful)
because it works (Score:3, Insightful)
10 years ago, in Byte (Score:5, Insightful)
Ten years gone, and still relevant.
Damn I miss Byte.
Re:Is it really "old" tech? (Score:5, Insightful)
Mainframes... going out of style?! (Score:4, Insightful)
The problem with the vision is that Stewart Alsop didn't take into account the growing complexity of computer programs. We have plenty of (in comparison to the software of 1991) inefficient applications that require ridiculous amounts of computer power to serve and process everything we need done. We have complex server applications like gigantic databases and games and video servers that couldn't exist in the 1991 world.
The mainframe of yesteryear may now fit into the physical space of today's desktop... or smaller, but that doesn't mean there won't be a need for a bigger and faster one to take its place. That's as true now as it was then.
Re:Why OldTech Keeps Kicking? (Score:2, Insightful)
People stick with what they're used to, even if it doesn't always make the most sense.
Legacy mainframes do make sense, though. Even if they're old and the people who know how to program them are retiring/dieing off, they do have 20+ years of debugging behind the code. Many of these systems run highly mission critical banking systems. If some of them fail, worldwide economic collapse is a real possibility. It's worth being very conservative in this case. Even if the going rate for COBOL programmers ends up being five times the amount paid to Ruby/Java/whatever coders (just so that somebody would be willing to work with such an archaic language), it'd still be worth it.
Re:Old Technologies that are still kicking... (Score:3, Insightful)
Because it's easier and less risky than switching (Score:3, Insightful)
Second, there's a lot of software written for the mainframe that works. It does important stuff, and what it does is probably not exceedingly well documented, and porting all that shit to something new is a massive, risky, expensive task.
Why mess with what works, particularly if the vendor seems to be willing to keep the product line going? There's no pressing reason to move, apart from people's prejudices about the mainframe, and the benefits really don't come close to outweighing the costs/risks.
Re:Is it really "old" tech? (Score:3, Insightful)
3. People fear migrations.
Lord knows I do, and I have first hand experience on why.
Re:because it works (Score:3, Insightful)
Many times I've seen historic pieces of IT Architecture in place because the cost to upgrade/train/retain/etc was a lot higher than dusting HAL every few thousand miles.
If the vendor is going o keep supporting it why abandon it?
Re:Is it really "old" tech? (Score:5, Insightful)
The article may as well be asking "Why do personal automobiles keep kicking?". Because they work, and they solve they still solve the problems that they are meant to solve. And when a new problem crops up, (fuel prices/pollution) the solution isn't to get rid of the car, it is to redesign it to address the new concerns; just like IBM and other companies did with mainframes.
Real Old Technology (Score:2, Insightful)
Re:Is it really "old" tech? (Score:4, Insightful)
In the PC world, we're used to revolutions on the desktop every few years. That's the sort of model guys like Apple and Microsoft have relied upon to keep them going. But when you're dealing with infrastructure that in many cases dates back to the 1960s, the idea of incremental change in hardware and software is extremely appealing and quite logical.
Having just done an upgrade to our accounting software this morning, and going through a number of small but still very real headaches, I can appreciate why the guys managing a major bank's information systems is damned glad that IBM does things the way they do.
Re:Why OldTech Keeps Kicking? (Score:4, Insightful)
Ever hear of a "wheel"? (Score:2, Insightful)
How about a bottle or a bucket?
Try an even older and more generic container, a sack.
Old tech hangs around because it does it's job and has not been improved upon in any meaningful fashion by later tech.
Incandescent lights might actually exit the stage soon...
Re:Is it really "old" tech? (Score:3, Insightful)
Radio may some day transform from the traditional AM/FM we've come to know and love (satellite radio, global Wi-Fi streaming, etc ), but the basic idea almost certainly isn't going away anytime soon.
NY Times misses boat again (Score:5, Insightful)
What would be the cost of hiring on top of the existing mainframe admins and developers a team to migrate this stuff to Windows or UNIX? Remember some of this code is written by people who not only have left the company but may have died. Then you have to hire new developers and administrators for the UNIX/Windows systems. Change always creates the potential for problems, so expect a higher percentage of disruptions to the business as you're doling out all this money. If IBM is making it easy for you to keep what you have going, and also allows Linux, web etc. capability, why spend all that money to transition? The answer is that a lot of times companies don't. I worked at a Fortune 100 company that still had plenty of IBM mainframes. They even had a lot of their printing handled by the mainframes, although there were Windows and UNIX gateways into the print queue.
Re:no built in obsolescence (Score:3, Insightful)
Though, most likely they're still in business selling cheap/shoddy products to OEMs.
It's not 'Old Tech' (Score:3, Insightful)
Mainframes constantly evolve.
Mainframes went 64 bit before the PC ever did. Virtualisation is just gaining ground on the PC.
Mainframes have had that for decades with Domains and LPAR's.
Whats old technology, a PC server farm with dedicated server per app, and maybe 10 concurrent users, or a mainframe running many applications with thousands of users, and terrabytes of i/o throughput.
Re:can be argued for other things too (Score:3, Insightful)
drive by wire system using a joystick is 25 lbs.
Such changes all added throughout a car can dramatically improve fuel efficiency.
Re:Why OldTech Keeps Kicking? (Score:3, Insightful)
If software engineers ran businesses mainframes would be gone because they are old and not cool anymore. But software engineers don't run businesses (if they did they'd be business people) and so they're still around, which is a good thing in my book (mainframe models make a lot more sense than individual for many problems).
Note that I'm speaking of the stereotypical software engineer. There are plenty who like old stuff, but the majority of software engineers I've met would rather use a brand new system to do something than an old one, that, or they aren't very good at their job because they hate it. Not that there's anything wrong with either way of course, different ideas != good and bad ideas neccessarily.
Re:Is it really "old" tech? (Score:4, Insightful)
I don't know if it exists yet or not, but it can't be too far off. I can already download podcasts to my iTouch directly over wifi. I would imagine it wouldn't be too hard to make a car radio that did the same thing. You could even make it detect when it's entered a location with a wifi connection, such as the garage, and start downloading new episodes.
Of course, some lame-ass company is probably going to patent this idea, and we'll have to wait until the stupid patent expires before we can actually use it...
Re:because it works! (Score:4, Insightful)
There are Hundreds of Thousands if not millions of dollars of man hours put into that system, and programs. Replaceing them with a new system could lead to a huge mistake. Being that this is a school district. I doubt that anyone is willing to put the job on the line with such a migration. And being a unioned job I doubt that they will hire consultants to do it for them. They are stuck between two political brick walls.
Re:can be argued for other things too (Score:4, Insightful)
Now here's a question for you. Why not drive-by-wire with a steering wheel? There's plenty of examples of it working, I had a steering wheel peripheral for my PS1 not too long ago. If you want to reduce weight without sacrificing utility then duplicate the old interface with new technology, don't re-invent the interface (unless that's what needs to be improved, and steering wheels are a perfectly good interface in my book).
There's very rarely just two options
Re:Is it really "old" tech? (Score:5, Insightful)
Apples and oranges (Score:4, Insightful)
Re:can be argued for other things too (Score:3, Insightful)
What you are doing is removing a significant geometric constraint - you don't need an open straight line between the driver and the steering control mechanism. You may be able to cut a little weight, but more significantly you can decrease the size of the engine compartment.
All of these advantages you get with a joystick you also get with a steering wheel (that isn't physically connected to the tie rods) and the same drive by wire system. The steering wheel is an easier UI because it allows you to reposition your hands on the device at any time.
All of that said, I personally do not want a car that I can't steer when the car is turned off (when I am working on the car), and I would be quite scared to drive a car that I can't steer when the alternator, computer, or power steering unit dies at 80 mph.
Re:Is it really "old" tech? (Score:5, Insightful)
If you deal with money, it doesn't seem like a bad idea at all. I mean, keep your financials database on a Windows system? Are you nuts? Keep your finances in MySQL, running on Linux? I don't think so. Oracle on Linux, maybe, but what about the hardware? Going to buy yourself a nice Dell? (pause for laughter)
Buying a sexy mainframe with real hardware support, the kind where they send out a guy who knows what he's doing, 3 minutes after you call, and he's got the new part installed in an hour and a half or your money back...That stuff is priceless if you really really need your system to be reliable. I can definitely see why they're still around.
Re:can be argued for other things too (Score:3, Insightful)
Any manual steering arrangement can be made lighter than a power assisted system and more efficient (with respect to fuel mileage) than a power-assisted system simply because the steering then doesn't impose a parasitic power draw on the engine.
Removing parasitic loads and saving weight improve fuel efficiency. Replacing a manually (driver) powered system with an engine powered system that requires extra pumps or electric actuators does not do that.
Re:whatever runs old software (Score:3, Insightful)
Heck, I was writing a new COBOL program the week before last. Why? Because it gets the job done, with a minimum of fuss. And the code was based upon a similar COBOL program of about 3 years ago, which works, so I was able to crank out a WORKING program with only an hour or so of work. Convert to Java? Try 3 weeks work, which I just don't have time for.
Why are the mainframes still here? What machines are you going to use to put terabyte sized databases on that require access 24 x 7 for hundreds of users?
If I can line up enough jobs to run a mainframe for 8 hours standalone (yes, take the whole machine and peg it to 100% for 8 hours), just to do maintenance on my databases on Thanksgiving day, when am I ever going to get enough time to do it in a server environment?
Re:can be argued for other things too (Score:3, Insightful)
And yet you seem quite happy with the idea of driving a car that you can't steer when the complex mechanical contraption shatters at 80 mph.
People have this idea in their heads that things with electricity can break while things without electricity can't. You trust the engineer to design a mechanical steering system that won't break, and you don't trust that same engineer to build an electric steering system that won't break. Funny, that.
Re:Is it really "old" tech? (Score:2, Insightful)
Your unusual sig. (Score:3, Insightful)
Total Loss of Knowledge (Score:4, Insightful)
I think old tech survives because of two reasons, one following the other. First, businesses develop inertia along a certain platform. For example, banks write a lot of code that is restricted to run in a mainframe environment (for whatever reason, it can't be moved off). "Inertia," in this case, means that a lot of code and business processes and practice have been developed around that platform. Perhaps even jobs have been created that are primarily concerned with the care and feeding of this platform and all it supports.
Then, time passes. People forget, and people leave. New people take over. At some point, if enough complexity develops and sits over a long enough period of time, the entity that owns the platform and all it supports basically loses control of it. They have no knowledge contained outside the system itself...to make significant changes requires someone to delve into it and tease out the why's and wherefore's of how it works. Either that, or replace it wholesale, abandoning all of the functionality of the code and the stability that comes along with the associated business processes.
If no one quite understands how something works, or even the totality of what it does, then it becomes easier to upgrade an existing platform than replace it. In some cases, the platform can only be upgraded in certain ways that maintains some restrictions of the original platform. And that's why old tech has staying power. No one knows what it does, how it works, or understands the impact of or effort required to replace it.
I think this cycle is inevitable to some extent where complex systems are required to fulfill some needed function. However, I also think there is much that businesses could do to prevent these issues where they are not necessary. I think the fundamental thing that needlessly ties businesses down to old tech is an improper segmentation of responsibilities within the company. Many times, departments and created and responsibilities assigned based not on the actual work that needs to be done, but rather the prejudices of executive management. A work force should be divided up based on areas of related responsibility and the dependencies between those groups, and nothing else. (This is usually how things are done at the low level of organizing groups, but go one or two levels up on the org chart and the concept seems to no longer apply at most places.)
Re:NY Times misses boat again (Score:3, Insightful)
And that's so stunningly inevitable that the whole article is a poor joke. They've started by using misleading labels for the things under consideration - if you fix that, the idiocy of their opinion becomes obvious.
They aren't comparing "old technology" to "new technology". They are comparing "technologies that have survived for 25 years and are still around" to "new technologies". In other words, they are comparing the very best of the older technologies to a random (hence, mediocre) selection of newer technologies. It is no surprise that the better ones are, well, better.
You'd get a different result if you made a fair comparison of "things that were around 25 years ago" to "things that are around today". Nobody ever does that.
It has nothing to do with technology (Score:3, Insightful)
"Mainframe" is servicable, supported, robust, high performance, and reliable. You're buying that when you buy "mainframe," it just so happens that IBM packages that in a larger sized computer. Technology is a fairly small part of that idea. To make "mainframe" go away you have to convince the world that the idea is no good, but it's really really really good.
Re:Is it really "old" tech? (Score:2, Insightful)
Re:Is it really "old" tech? (Score:4, Insightful)
A mainframe is not just a CPU and it's not designed to be a power house of MIPS or FLOPS (or heaven forbid some naive notion of clock speed). Instead a mainframe is an I/O power house. They're designed to handle aggregated data from many different sources and process them efficiently. There are lots of peripheral processors to handle I/O independently of the main processor and each other. The concept of a special purpose computing machine designed for secure, reliable, I/O heavy transaction based processing is still around; and since mainframes do this job cheaper than the alternatives, they're still around.
There was essentially no reason to declare the mainframe "dead" in the first place. Though declaring certain types or models dead makes sense. The original prognostication seemed a bit like noticing that computers were getting faster with more bandwidth while forgetting that mainframes were allowed to improve as well.
Re:Is it really "old" tech? (Score:3, Insightful)
Re:Total Loss of Knowledge (Score:2, Insightful)
I also think older technologies were often designed and implemented to last longer than technologies created today. Just looking at consumer electronics, I think the desire to achieve economies of scale in production often results in products which will fail eventually due to cheap components (e.g., early '90s metal VCRs vs '00s plastic VCRs). In one way, you could say that the desire for a technology to become a commodity makes us design for failure.
Also, sometimes you can't advance through your tech tree as fast as you like because you didn't discover the right technologies in order to move forward... or you didn't build "Leonardo's Workshop".
Re:Is it really "old" tech? (Score:1, Insightful)
I have their Gold support and get somebody on the phone immediately. I've only had two instances in the past 8 years where someone had to come out and they were there way within the 2 hour guarantee, knew what they were doing and fixed the problem. Both of those were tape drive instances by the way. One was our fault, someone put a tape that had been dropped and got bent into the library and it jammed in the drive.
My Dell servers just sit there and work. You ever stop and think that maybe, just maybe the people who write the applications that are used to run a company may be at fault? You know, those guys called software vendors. Or do you just consider the hardware vendor the problem.
Seriously, what's with all the Dell hate, are they the flavor of the month? You want to bash on HP too...I've got a Proliant 1600 still humming along since 1996. Dual Pentium II's baby!
I really wonder how many people who make comments like the above have ever worked in(or for that matter, been allowed into) a real business.
Re:Total Loss of Knowledge (Score:4, Insightful)
In many situations, you can make a solid business case for "if it ain't broke, don't fix it." In many businesses, the mainframes ain't broke, and nobody's in a hurry to fix them. Yes, IBM charges rather phenomenally for support when your machines start to get long in the tooth -- but they have a relatively straightforward upgrade path (to new mainframes) that's cheaper for many people than moving to commodity systems would be.
After all, the people who run mainframes aren't going to buy a bunch of whitebox machines and just cross their fingers and hope they work -- they want support and reliability and equivalent featuresets. By the time you take commodity systems and make them and make them perform like a mainframe, and then make them as reliable as a mainframe, and then you add on the cost of support and maintenance equivalent to what you get with a mainframe
I think most people would be surprised how much stuff that they count on being on-time and correct but don't think about -- things like their bank statements, phone bills, etc. -- are handled on large systems. And not necessarily creaky old 'legacy' ones, either, but bright shiny new ones.
Highly regulated industries (Score:3, Insightful)
Which is why nuclear power plants still rely on mainframe computers, analog control systems and those big bulky institutional green control panels in the control room with lots of blinking lights, dials, knobs and buttons that look like mid 50's science fiction movies. (Nobody wants to stare at that all day- they'll go stir-crazy.)
Contrast that to one coal burning behemoth I visited that had a fiber networked distributed control system running on a modern server system, with a number of large flat screen panels in a modern operations center that looked more like a TV news studio, displaying the status of all the systems; and changes can be initiated with a couple keystrokes or even through a GUI.
The problem with the old systems at nuclear power plants is that many of the people who know them are of retirement age. As one guy who was tasked with maintaining the control systems in one nuke plant's repair shop told me, "Everyone in here is a grandfather". The younger people fresh out of engineering school who are taking their place were schooled on the modern systems like what's at the coal burning plant. There is a crisis going on because a lot of the old-timers are being forced into early retirement (taking their body of knowledge with them) faster than their replacements can learn from them.