Slashdot Mirror


Join the Hunt For the Government's Oldest Computer (muckrock.com)

v3rgEz writes: As the saying goes, 'If it ain't broke, don't fix it.' If a machine is doing its job, reliably and without error, then common sense dictates that you just shouldn't mess with it. This is doubly true for computers and quadruply true for government computers. This lends itself to an obvious question: what's the government computer most in need of an upgrade? MuckRock has launched a new FOIA project to find out, and has already started receiving some interesting results.

147 comments

  1. The article has some interesting highlights by JoshuaZ · · Score: 2
    Some government agencies were much more responsive than others. Some of the responses are scary. From TFA:

    According to the FAA, just knowing what kinds of computers the FAA is using would endanger the security of national air traffic. That's pretty bad, both for this project and my confidence in our air traffic system.After all, despite my vague wording, the FAA found 11 pages of documents responsive to the same request. But! They refused to release any of their records to me, citing the blanket Exemption 3 because they deemed, "disclosure would be detrimental to the safety of persons traveling in air transportation."

    He does say in the article that some agencies have confirmed mainframes from circa 1970, but doesn't say which specifically. It should be interesting to see how this project goes over the next few months.

    1. Re:The article has some interesting highlights by bws111 · · Score: 3, Insightful

      It says he 'is catching wind' of circa 1970 computers. That is hardly 'confrmed'.

    2. Re: The article has some interesting highlights by bill_mcgonigle · · Score: 1

      It might be true - some of them supposedly use core memory, which is probably vulnerable to all sorts of EM attacks.

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
    3. Re:The article has some interesting highlights by Crashmarik · · Score: 3, Funny

      Well the FAA has the best security of any federal agency. You need to be an archaeologist to hack into their equipment.

    4. Re:The article has some interesting highlights by Anonymous Coward · · Score: 0

      They were trying for and early/first post. That means they didn't actually /read/ the article so much as skim it. It's surprising they did even that. People don't think clearly when they're in a rush for e-penis. If my ego were attached to moderation, I'd probably be much the same. I suppose I'll post this as an AC that way I won't be seen as fishing for moderation.

    5. Re:The article has some interesting highlights by Darinbob · · Score: 1

      I would not doubt it though. I know that in the 90s at a class reunion one of my friends mentioned they had PDP computers in Onizuka Air Force base (the blue cube). That's torn down now though, but it's not hard to imagine it happening elsewhere. Why upgrade when it's working just fine and it'll cost hundreds of millions of dollars to get a replacement? It's not like you can just go to Newegg and order some software to run the radar arrays.

    6. Re:The article has some interesting highlights by Anonymous Coward · · Score: 0

      I wanna know why the Voyager Probes aren't winning.

    7. Re:The article has some interesting highlights by Anonymous Coward · · Score: 0

      NASA, NOAA, NRO, and the Air Force are probably the most interesting ones, because they likely have the same ground-systems for satellites that they had when they launched them way back when. Ask for tech details of the Goes-3 ground station at NOAA, it was launched in 1978, and I think it's still active.

    8. Re:The article has some interesting highlights by Anonymous Coward · · Score: 0

      I can confirm the AirForce uses computers from the 70s to test missiles.

      AC for obvious reasons.

    9. Re:The article has some interesting highlights by Guy+Harris · · Score: 1

      I know that in the 90s at a class reunion one of my friends mentioned they had PDP computers in Onizuka Air Force base (the blue cube).

      Mentec were developing new PDP-11-compatible processors in the 1990s, so there may well have been PDP-11s in there. (These days, if you still need to run PDP-11 software, it's probably on an x86 box running a simulator, with, if necessary, specialized hardware plugged in to handle peripheral buses of that era.)

  2. In-use vs owned/missing by Scoth · · Score: 4, Insightful

    I'd be curious if he has a way to differentiate active, in-use machines vs. old stuff that may just still be in the inventory roles. I helped a bit with my previous company's effort to clear out old storage and inventory and there was some pretty old equipment that had been in the closet for close to a decade, plus some stuff in their inventory records that had long since disappeared.

    So I wonder if that Gateway Liberty 2000 is actually still sitting on someone's desk where the toil away with WordPerfect and Windows 3.1 every day, or if it got tossed/walked out of the building in the late 90s and no one bothered to update the records.

    1. Re:In-use vs owned/missing by hey! · · Score: 2

      So I wonder if that Gateway Liberty 2000 is actually still sitting on someone's desk where the toil away with WordPerfect and Windows 3.1 every day, or if it got tossed/walked out of the building in the late 90s and no one bothered to update the records.

      j

      It might not be either. One thing I've frequently seen in government offices is really old software they don't have the funding or time to replace or update. So it's not necessarily being used every day, but it may get fired up a couple of times a month to run some old FoxPro database on Windows 3, or maybe even once a year to run a Lotus 1-2-3 spreadsheet that generates a report.

      If so the computer in question would be a good choice, because it's a small laptop. It wouldn't have to take up space on someone's desktop, you could shove it in a filing cabinet.

      --
      Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
    2. Re:In-use vs owned/missing by DerekLyons · · Score: 3, Insightful

      So I wonder if that Gateway Liberty 2000 is actually still sitting on someone's desk where the toil away with WordPerfect and Windows 3.1 every day, or if it got tossed/walked out of the building in the late 90s and no one bothered to update the records.

      Or somebody set it on a shelf in case his 'new fangled' machine died and a backup was needed - and it's simply been handed down from one custodian to the next even since. Nobody cares that it's not needed and probably doesn't work anymore, the paperwork says we have it, and there it is on that self over there... and that's good enough.
       
      When I took over the department test equipment locker at TTF in the late 80's, I had a ton of stuff like that - old and obsolescent equipment squirreled away by previous custodians because "we might need it again some day". The bosses were OK with that because there was no penalty for having excess gear, and it was a massive pain in the ass to get rid of gear. (There's a lot of hoops to jump through to make sure that was in fact excess to requirements and that the person getting rid of it and his chain of command weren't simply trying to take it home themselves or sell it for their own profit or whatever.)

    3. Re:In-use vs owned/missing by Scoth · · Score: 1

      Fair enough. The company I worked for was relatively new (founded 2001 I think) so we didn't have a lot of truly ancient stuff stuck away. Optiplex GX1s, some Sun pizzaboxes (wish I'd been able to get ahold of couple those, but they were technically still in use), some Latitude pre-C-series Dell laptops, etc. A lot of it was mostly just people too time strapped and/or lazy to sort through it all.

    4. Re:In-use vs owned/missing by dbIII · · Score: 1

      True - I've got a couple of Sparcstation 5 machines in storage in case some problems show up in some legacy software in a far newer machine running Solaris10. They may never be needed but one of them was fired up three years ago to check if the legacy software was behaving differently to the way it should.

    5. Re:In-use vs owned/missing by Bing+Tsher+E · · Score: 1

      You should pull and back up the battery backed up NVAM chips in those Sparcstations. There are documented methods to do that on the web. The MAC address of the SparcStation 5 in on the NVAM chips, which have a battery built in. There's a method you can use to 'shave' away plastic to expose contact spots on the NVAM chips and tack solder on an external battery. It's much easier if you do that before the battery is dead and you have to reprogram it (which if I am remembering right can be done in the FORTH-based setup environment of a Sparcstation)

    6. Re:In-use vs owned/missing by serviscope_minor · · Score: 1

      I'd be curious if he has a way to differentiate active, in-use machines vs. old stuff that may just still be in the inventory roles.

      When I was working at Los Alamos, I remember a big fuss in the media about losing millions of dollars worth of computing kit and how much (a) the government sucks and (b) how much scientists suck and how business is obviously better and blah blah blah.

      Turned out that it was essentially all old machines, like tricked out XTs which had purchase prices of $10,000 or something which had got horrendously obsolete and were junked without retiring them from the inventory system properly. But they were still listed as being worth $10,000 in 2005 or whatever, becuase they hadn't always been depreciated correctly.

      Naturally no retractions or corrections were issued.

      --
      SJW n. One who posts facts.
    7. Re:In-use vs owned/missing by dbIII · · Score: 1

      Good point but there's text files on them with their MAC address if all else fails and if I need them again.

  3. Focus on NASA by Anonymous Coward · · Score: 0

    They must have a. 360 running.

    Despite the FAA claiming security concerns,, it took me 10 minutes to find Lora was once bidding on a contract to replace IBM. 9020Es, so they are probably running 360-compatible gear.

    1. Re:Focus on NASA by __aaclcg7560 · · Score: 3, Informative

      What's with. the random periods. in your sentences?

      A double space on the iPhone virtual keyboard becomes a period. Type too fast, random periods appear.

    2. Re:Focus on NASA by sphealey · · Score: 2

      Back in the 1980s my then-employer shared the cost with the FAA of updating IBM's 1401 emulator to the latest architecture (IIRC whatever was the half generation between 370 and 390), so I wouldn't count on that 360 being the oldest!

      sPh

    3. Re: Focus on NASA by Type44Q · · Score: 3, Funny

      random periods appear

      "Scheduled monthlies" are bad enough; I can only imagine how rough "random" must be for those around you...

    4. Re:Focus on NASA by Guy+Harris · · Score: 1

      They must have a 360 running.

      "A 360" in the sense of "an IBM System/360", or "a 360" in the sense of "a machine compatible - except perhaps at the supervisor-mode level - with an IBM System/360"? If the latter, then that's not interesting; my laptop was purchased in 2015, but, as far as I know, its CPU is compatible, even at the ring 0 level, with the CPU from the original IBM Personal Computer (whether I could boot the original version of MS-DOS on it is another matter), and the most recent CPU design capable of running System/360 problem-state code was also announced in 2015.

      Despite the FAA claiming security concerns,, it took me 10 minutes to find Lora

      Presumably you meant "Loral".

      was once bidding on a contract to replace IBM 9020Es,

      Yes, in 1995. IBM replaced the 9020A and 9020D machines in the late 1980s, and then replaced the 9020E's in the 1990s.

      so they are probably running 360-compatible gear.

      What they're running now might well be "360-compatible", but that might be in the same way that the Haswell Core i7 in my laptop is "8088-compatible".

    5. Re:Focus on NASA by Guy+Harris · · Score: 1

      Back in the 1980s my then-employer shared the cost with the FAA of updating IBM's 1401 emulator to the latest architecture (IIRC whatever was the half generation between 370 and 390), so I wouldn't count on that 360 being the oldest!

      Where were 1401's involved? The 9020s were based on 360/50 and 360/65 models, not on 1401s; unless 1401s were used prior to the 9020s (was air traffic control computerized at all before the 9020s), and the 1401 software run in emulator mode or under a simulator (I'm not sure the models 50 and 65 had 1401 emulator microcode - 709x, yes, but not 1401, which may have been used in the smaller models 30 and 40), I don't see how 1401s would have been involved. (As I remember reading, at least some of the 360s had microcode changes to add instructions to the S/360 instruction set to process radar scan lines, suggesting that the code was written in 360 assembler or compiled into 360 machine code, rather than 1401 code.)

    6. Re: Focus on NASA by Anonymous Coward · · Score: 0
      I take it you don't know anyone with PCOS then.

      (Yes, I know you were trying to be funny).

    7. Re:Focus on NASA by Drishmung · · Score: 1

      They must have a 360 running.

      "A 360" in the sense of "an IBM System/360", or "a 360" in the sense of "a machine compatible - except perhaps at the supervisor-mode level - with an IBM System/360"?

      Basically, you just described CICS

      While I may be wrong or simply out of date, my understanding is that CICS is an application running in what is essentially a virtual IBM/360. The virtualisation meant that CICS didn't have to be rewritten for the IBM.370 to take into account newfangled things like memory protection.

      --
      Protoplasm. Quiet Protoplasm. I like quiet protoplasm.
    8. Re: Focus on NASA by Anonymous Coward · · Score: 0

      Sorry you're wrong. CICS is a transaction monitor used to run interactive applications. It can run inside the various flavors of IBM 360/370 DOS, MVS, and z/OS. All of which can be run as VMs.

    9. Re:Focus on NASA by Guy+Harris · · Score: 1

      They must have a 360 running.

      "A 360" in the sense of "an IBM System/360", or "a 360" in the sense of "a machine compatible - except perhaps at the supervisor-mode level - with an IBM System/360"?

      Basically, you just described CICS

      While I may be wrong or simply out of date, my understanding is that CICS is an application running in what is essentially a virtual IBM/360. The virtualisation meant that CICS didn't have to be rewritten for the IBM.370 to take into account newfangled things like memory protection.

      That wasn't newfangled; System/360 had protection even though all models other than the model 67 had no memory mapping, just physical memory access. Oh machines with storage protection (optional in the smaller models, and standard in the larger ones), a 2KB block of memory had a 4-bit "protection key" associated with it, and the processor status word had a "protection key" value giving the key of the current task (process). A block of memory could be stored into only if either the PSW protection key was 0 or was the same as the protection key on the block; with store-and-fetch protection (not available in some models), the block's storage key had an additional "fetch protection" bit which, if set, disallowed fetching from the block, as well as storing into the block, if the PSW protection key is non-zero and doesn't match the block's key.

      And even on early versions of DOS/VS and OS/VS for System/370 models with "dynamic address translation" (i.e., an MMU), all tasks ran in the a single common virtual address space, so it wasn't that different from an S/360 in that regard.

      CICS might have done its own multitasking of requests and responses within a DOS or OS task, and not protected individual "subtasks" from each other, but that wouldn't require virtualization, even on MVS where OS tasks ran in separate address spaces, so if you're referring to "The entire partition, or Multiple Virtual Storage (MVS) region, operated with the same memory protection key including the CICS kernel code." from the Wikipedia article, that doesn't require virtualization to run on later systems.

      What I'm describing is just that the problem-state instruction set of an IBM z13 is, except for anything removed in the original S/360 to S/370 transition (such as ASCII mode), a superset of the original S/360 instruction set.

    10. Re: Focus on NASA by Anonymous Coward · · Score: 0

      NASA has to keep really old stuff online to maintain interface with nonupgradeable hardware, such as Voyager.

    11. Re:Focus on NASA by Anonymous Coward · · Score: 0

      What's with. the random periods. in your sentences?

      A double space on the iPhone virtual keyboard becomes a period. Type too fast, random periods appear.

      Ok, I'll buy that. But where did the double commas come from? It looks like you've got the DTs.

  4. silence speaks volumes by Anonymous Coward · · Score: 0

    since this is the internet and I'm an AC, I'll go ahead and speculate that NASA refused because releasing their computer inventory would reveal their possession and use of all kinds of skynets, cyborgs, and other sci-fi sundries. FAA has some skynet-remote-terminals that it can't disclose either - national security, you know.

  5. Nice way to waste taxpayer money by Anonymous Coward · · Score: 1

    I get highly annoyed when i get an e-mail it takes me 15 minutes to find an answer to, i can imagine how the poor office drone who has to find all this frivolous information feels, especially when it's at the behest of somebody merely curious.

    And that's assuming he can find it on his own, once he has to start bothering several different departments people tend to develop a deep resentment towards you...and trust me i know this from experience.

    1. Re: Nice way to waste taxpayer money by Anonymous Coward · · Score: 0

      Exactly. I don't feel like this is what the FOIA was for. Such a waste of someone's time and energy.

    2. Re:Nice way to waste taxpayer money by mschuyler · · Score: 2, Insightful

      Amen to that. The FOIA has a purpose--to maintain open government. It was not designed to allow the merely curious to indulge their historical fantasies. Replying to this kind of stuff costs money, and guess who pays?

      I've seen cases where, for example, an activist decides the local public library's stance on open access to the Internet violates their stance on "decency" because it allows some people to view pornography, so said activist embarked on a campaign to ask the library for EVERY document they had on budget, personnel, policies, emails, etc. It was so bad that the library--not exactly an over-funded and wasteful public agency, had to hire a full time secretary just to respond to the myriad of requests.

      Basically it all translated to less materials purchased and fewer service hours open to the public while taking an inordinate amount of the library management's time and resources.

      This is just harassment and a misuse of the statute enacted in good faith.

      --
      How about a moderation of -1 pedantic.
    3. Re:Nice way to waste taxpayer money by tlambert · · Score: 1

      Amen to that. The FOIA has a purpose--to maintain open government. It was not designed to allow the merely curious to indulge their historical fantasies. Replying to this kind of stuff costs money, and guess who pays?

      Likely whatever agency repeatedly refuses to answer the request pays and pays and pays and pays.

      Until they get a clue, answer the request, and so can stop paying, because it's "asked and answered".

      So mostly? The agencies who are reluctant to answer the freaking questions they are asked tend to pay the most.

    4. Re:Nice way to waste taxpayer money by 0100010001010011 · · Score: 1

      I think you overestimate how much the average government worker has to do. It's not like the person that did replied to these e-mails was taken off off solving the US debt crisis to answer them.

      If anything they were probably happy to have something to do.

  6. Intredasting by Sax+Russell+5449D29A · · Score: 4, Interesting

    A couple of years ago I heard of a late-70's VAX still being used at a small power plant. To my knowledge it controlled some sort of HVAC systems. Another old system, one I've actually seen, was a mid-80's computer of unknown make/model used to control traffic lights in a small city. It's funny, or actually impressive, to see such old systems still in use. The old-school guys that keep them running tell nice stories about flea market and eBay scavenging.

    --
    -SR
    1. Re:Intredasting by sphealey · · Score: 5, Informative

      = = = A couple of years ago I heard of a late-70's VAX still being used at a small power plant. To my knowledge it controlled some sort of HVAC systems. Another old system, one I've actually seen, was a mid-80's computer of unknown make/model used to control traffic lights in a small city. It's funny, or actually impressive, to see such old systems still in use.= = =

      Those systems were designed to be reliable, maintainable, and understandable by thorough professionals, so it is in no way surprising they continue to work. And in the industrial infrastructure world you don't replace equipment just because there is a shinier new version. 50 year old equipment is not uncommon in the provision-of-electricity industry and I've seen some 80 year old stuff in operation.

      sPh

    2. Re:Intredasting by mrbill · · Score: 1

      Depending on the interface hardware, they might be better off at this point (both power usage and dependability wise) to look at one of the many commercial VAX emulators that run on server-class PC hardware and have available hardware to connect certain types of interface cards.

    3. Re:Intredasting by HotNeedleOfInquiry · · Score: 1

      Back in the late 90's a commercial nuke that I visited was running a Prime mini to monitor core temp. They couldn't simply change it out due to certification issues.

      --
      "Eve of Destruction", it's not just for old hippies anymore...
    4. Re: Intredasting by Bing+Tsher+E · · Score: 1

      I have a rackmount box with a 486 motherboard in it, that has an LSI-11 processor on an ISA bus card. The card has ribbon cables that go out to a card cage with hoary old PDP-11 i/o cards in it. So it's a PDP-11 machine that uses a 'modern' PC as it's supervisor.

      I got it at a university surplus auction with two other systems as a lot. The other two systems were also PDP-11 boxes which I passed on to other collectors. The one I kept is the newest, and probably existed to keep some old data acquisition system going with minimal changes to the hardware/software. It's real PDP-11 hardware running as a hosted coprocessor on a PC. Some day I should power it up.

    5. Re:Intredasting by onepoint · · Score: 1

      While I have not seen stuff from the 40's, I did see recently something from 1953 ( tag date ) at a sub-station, took a picture. reminded me of an old steam engine governor. old stuff, well-maintained works.

      --
      if you see me, smile and say hello.
    6. Re: Intredasting by sconeu · · Score: 1

      Have you ever tried to install the ORIGINAL version of XINU? Comer wrote it for the LSI-11.

      --
      General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
    7. Re:Intredasting by braindrainbahrain · · Score: 1

      I know of a rent-a-storage place where the gate access security is run buy an Apple II! (Not the government, I know, but I was really shocked to find that out).

  7. shouldn't mess with it? by Anonymous Coward · · Score: 0

    On computers, there is no such thing as "ain't broke." That's the old way of thinking.

    In 2016, when developers come to you and say "oh, you're still using last week's version of our app? That one was crap! Here, buy our new version." you dutifully obey, and Upgrade to the Latest Version(TM). It doesn't actually matter what the computer is used for or what the app does. This is the new way.

  8. If you think that's bad... by __aaclcg7560 · · Score: 3, Informative

    When a department gets downsized for one reason or another at my government job, the department manger is supposed to turn in all the unused workstations for redeployment or recycling. They typically don't. A favorite hiding space is the utility closet inside a women restroom. All the field techs are male. Go figure.

  9. waste of tax-payer's money to answer these by klindsay · · Score: 5, Insightful

    "If a machine is doing its job, reliably and without error, then common sense dictates that you just shouldn't mess with it."
    "what's the government computer most in need of an upgrade?"
    You've just given a great reason why some hardware is still in use, it works.
    Why turn around and conclude that it needs an upgrade?

    "knowing which agencies are running hardware older than I am is important"
    Sure, for a very loose definition of important.

    1. Re:waste of tax-payer's money to answer these by Teun · · Score: 2

      Except some, probably a lot, of them are hooked up to a network and the security status is as old as the HW.

      --
      "The likes of Facebook and WhatsApp are free to those whose privacy is of zero value."
    2. Re:waste of tax-payer's money to answer these by tlambert · · Score: 3, Informative

      Meaning you don't have to worry about currently in use attacks, right?

  10. Hey FAA! by PPH · · Score: 1

    I've got boxes of vacuum tubes that could be available for the right price.

    Just sayin'.

    --
    Have gnu, will travel.
    1. Re:Hey FAA! by Guy+Harris · · Score: 1

      I've got boxes of vacuum tubes that could be available for the right price.

      Even if they were still running the old System/360-based 9020s (which they aren't - they're 3 or more generations beyond that), they wouldn't be using vacuum tubes in the CPU. Maybe the 9020 power supplies, or the displays, used them (other than the CRT itself), but even that might be unlikely.

  11. PARCS CLC 1972ish by Anonymous Coward · · Score: 1

    The Central Logic Control Computer (CLC) at AN/FPQ-16 Perimeter Acquisition Radar Attack Characterization System (PARCS) was assembled at Los Alamos in 1968 and installed at Cavalier AFS sometime before 1972. It remains installed and operational - although a replacement computer system is now also installed. The CLC was probably the fastest computer in the world when it was built and may have still been the fastest computer in the world in the lat 1970s. It is compose of 40+ racks of water cooled equipment.

    http://www.srmsc.org/par2010.html
    https://en.wikipedia.org/wiki/AN/FPQ-16_PARCS

  12. "If it ain't broke, don't fix it?" Fuck off. by imac.usr · · Score: 4, Insightful

    I hate that goddamn phrase. When the inevitable time comes when suddenly the old system *does* break, it's no longer under any support, nobody's left at the company who knows how it works, there's no budget for a modern replacement, and it has to be fixed in four hours or the company goes bankrupt. Been there, done that, ate the T-shirt after hours of working with no break for food.

    People who say "if it ain't broke, don't fix it" are the same idiots who brag about uptime.

    Pro tip: *every* system is broken. The trick is being able to repair or work around the broken parts without disruption, not to just seal it behind a wall and rediscover it years later when trying to track down what's still pinging.

    --
    I use Macs for work, Linux for education, and Windows for cardplaying.
    1. Re:"If it ain't broke, don't fix it?" Fuck off. by __aaclcg7560 · · Score: 3, Interesting

      Pro tip: *every* system is broken.

      Every system should be planned for. If a Windows 2003 Server is still on the network at my job, a planned exemption is on file to postpone removal for 30 days or gets pulled off the network. The server owner has six months to migrate to 2008 or 2012, but some people insist on dragging their feet and waiting until they have no choice. The process is somewhat easier these days as all the servers are virtualized across a server farm.

    2. Re:"If it ain't broke, don't fix it?" Fuck off. by Dutch+Gun · · Score: 5, Insightful

      The funny thing about pithy little phrases like that is that there's inevitably a counter-phrase that works to refute it. In this case, perhaps "don't put all your eggs in one basket" would be appropriate?

      There's nothing wrong with "if it ain't broke, don't fix it" as a general rule of thumb (nor of good uptime), but I'd certainly temper it with the notion that a lack of support or replacement parts can also be considered "broken", and thus in need of fixing. There's a lot of damage done by needless upgrades or "enhancements", which is what this phrase is meant to counter. I understand your frustration with idiots who view any one rule as some sort of golden rule handed down from on high, but... well, "don't throw the baby out with the bathwater."

      --
      Irony: Agile development has too much intertia to be abandoned now.
    3. Re:"If it ain't broke, don't fix it?" Fuck off. by YukariHirai · · Score: 1

      When the inevitable time comes when suddenly the old system *does* break, it's no longer under any support, nobody's left at the company who knows how it works, there's no budget for a modern replacement, and it has to be fixed in four hours or the company goes bankrupt.

      That sounds to me more like general incompetence and mismanagement than a fatal flaw in the principle of "if it ain't broke, don't fix it".

      In my workplace, we rely very heavily on a particular piece of software running on a particular server. Said server is a Pentium II running Windows NT 4.0. Amazingly, this software running on this server has been absolutely rock solid.

      Nonetheless, we're of course not relying on it to last forever, and replacing it has been looked at from time to time. Thing is, the same software running on the test new hardware has shown some problems that have proven absolute bastards to figure out, let alone solve. So for the time being, we are better off sticking with what's ancient and working properly, rather than switching over simply because this is new and that is old. When the old does finally fail, the new can be dropped in in short order, and we can sort out the problems that introduces, but in the meantime it's just not worth it.

    4. Re:"If it ain't broke, don't fix it?" Fuck off. by Darinbob · · Score: 1

      Reminds me of the IT guys who told me my macbook was due to be upgraded to Windows 7.

    5. Re:"If it ain't broke, don't fix it?" Fuck off. by Anonymous Coward · · Score: 0

      Thing is, the same software running on the test new hardware has shown some problems that have proven absolute bastards to figure out, let alone solve. So for the time being, we are better off sticking with what's ancient and working properly, rather than switching over simply because this is new and that is old. When the old does finally fail, the new can be dropped in in short order, and we can sort out the problems that introduces, but in the meantime it's just not worth it.

      I'm confused. It almost sounds like you have tested the replacement, and found problems, but are waiting till the original dies to solve them? Perhaps the risk of downtime doesn't justify the near term effort. Either way, I think the point was to have a plan for when the end occurs, which is fair enough. I tried to get Windows 9x running awhile back to run an old program. It was nearly impossible, even in a VM, since I needed direct X to work right and the VM didn't seem to support that.. (Updates were very hard to find, particularly those that met company policy and such.)

    6. Re: "If it ain't broke, don't fix it?" Fuck off. by Bing+Tsher+E · · Score: 1

      Where I work we use about a half dozen old 486 boxes with MS-DOS in the test labs. The machines run test programs written in GW-BASIC to drive stepper motors and read switch outputs. If you buy a truck from a certain one of the big 3 automakers, the dashboard controls were design qualified with this setup. It isn't my "baby" but one of our hoary old test engineers clings to it. I can't even get them to update to Qbasic that came with the update of DOS 5.0, which at least uses a full screen editor.

      Five or so years ago we finally retired some Commodore SX64 systems from testing, and I brought home 4 of them. About two years ago I hauled out about 7 IBM PC-XT systems that were being scrapped and hadn't been used in a few years.

      The 486 boxes still in use weren't even originally lab machines. They were cast off office machines repurposed about 15 years ago. None of the CMOS setup batteries are still working so they are left on all the time and parameters like hard drive type have to be reentered any time they are powered off. Power supply fans failing is the most common issue.

    7. Re:"If it ain't broke, don't fix it?" Fuck off. by YukariHirai · · Score: 1

      I'm confused. It almost sounds like you have tested the replacement, and found problems, but are waiting till the original dies to solve them? Perhaps the risk of downtime doesn't justify the near term effort.

      Kind of. Decisions about this are above my pay grade, so I don't know all the details, but as far as I can tell the thinking is something like this: in testing, these problems have been detected and (I hope) solved, but if the new system is used in production other problems may arise, and we'd like to avoid that disruption, so for now we'll stick with what we know works properly.

      Either way, I think the point was to have a plan for when the end occurs, which is fair enough.

      Exactly. Just up and replacing something important for no better reason than "there's a new model out" is risky and foolish, hence "if it ain't broke, don't fix it". But it's more foolish and risky to not plan around the possibility of it breaking, which is not the same thing as "if it ain't broke, don't fix it", but some people don't know the difference.

    8. Re:"If it ain't broke, don't fix it?" Fuck off. by Anonymous Coward · · Score: 0

      Well, the old equipment is "supposed" to make it's way back to a storage bin in your garage, and then the next time your looking for a job you bring in to your interview this amazing piece of equipment with a deposition or log of your success for saving said company after DAYS of working with no break for food or sleep.

      And you get the job (true story)

    9. Re:"If it ain't broke, don't fix it?" Fuck off. by Anonymous Coward · · Score: 0

      Especially vis-a-vis antique computer systems: The combined power + HVAC bill adds up! 1970s computers achieved roughly 1KFlop/Watt (e.g. Cray-1). Modern computer architectures are targeting 10GFlops/Watt, an improvement by a factor of over ten million.

      Apocryphal: Business has had an IBM-1402 (Or some punch card driven automatic calculator that predated system/360) for many decades. Business FINALLY replaces it (and discovers that the the forearm-sized power cable is 80% rusted through...). Result: power bill savings pay for the new machine in less than one year.

      Not apocryphal: A London subway station recently replaced a mercury arc rectifier that had been in uninterrupted service powering DC trains since before WWII...

    10. Re: "If it ain't broke, don't fix it?" Fuck off. by serviscope_minor · · Score: 1

      Do you modify the software often? I'm not really sure what improvements QBasic would bring otherwise. Also, if the implementation changed, the timings would presumably change, so things might break that way.

      The hardware is super obsolete, but it might be a pain to change. If the timing is not super critical, you might be able to upgrade it to some Vortex86 based industrial PC/104 boards. They'll run DOS just fine, and are built to run in industrial invironments, and will run off much more solid power supplies than those old 486s, so they'll probably be good for another 30 years.

      --
      SJW n. One who posts facts.
    11. Re:"If it ain't broke, don't fix it?" Fuck off. by maestroX · · Score: 1

      It is simply not a fair comparison, the 70s were still more the mechanical age requiring regular maintenance.
      If anything, I think were lucky these things are durable.

    12. Re:"If it ain't broke, don't fix it?" Fuck off. by DutchUncle · · Score: 1

      Do you go around your house changing your light switches every few years as a preventive measure? Or the circuit breakers? I'll bet they're the same ones put in when the place was built. And you replace them when they break, which is rare. A lot of the 1970s vintage systems are simpler and more robust than a modern ultra-small-scale IC SOC. They had transistors and resistors and parts that could survive, rather than a few atoms barely holding each bit. :-)

    13. Re:"If it ain't broke, don't fix it?" Fuck off. by RockDoctor · · Score: 1

      Do you go around your house changing your light switches every few years as a preventive measure?

      Maybe not the best example. If I had to go and find a 20ft step ladder - or worse, a 20ft scaffold tower - to get up to the level needed to change the lights in the living room roof, then I would indeed consider doing a preventative bulb-change, because the cost wouldn't be in the bulbs, but in the hire of the access tools. (I don't have a lobby like that at home, but I've seen people who do have. The house in question being Victorian, when killing a servant wasn't a big problem.)

      But what I'd really do would be to re-design the lighting system to be more maintainable (whether that be lighting suspended to a more convenient altitude, or lighting which can be lowered to floor level ... whatever).

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
    14. Re:"If it ain't broke, don't fix it?" Fuck off. by Anonymous Coward · · Score: 0

      If you need a throw-away phrase to counter "if it ain't broke, don't fix it", try "technical debt".

      I practically sat down and cried when I first heard this. It perfectly captures the danger in "if it ain't broke, don't fix it".

      I also like the following phrase. "Anything that can be put in a nutshell, belongs there."

  13. Footprints of old systems by T.E.D. · · Score: 4, Interesting

    I've found that sometimes old systems leave footprints that last far beyond the computers themselves.

    For example, a couple of years ago we had this working networked system that we wanted to upgrade one computer of. The issue was that the protocol used to talk between the systems was a custom network layer written on top of a serial protocol called DR-11W. The cards were rather hard to find, the hardware very finicky to get talking right, and finding good docs for our custom layer was a real challenge.

    I eventually found out in researching it that DR-11W was in fact the serial printer port on the original PDP-11's back in the 70's. Neither machine was a PDP-11, but since every upgrade ever done was one computer at a time, we've had to maintain this PDP-11 printer port communications interface for the last 40 years. The protocol even required converting all floating-point values to IBM's old format even though neither side used that format! The conversion's not trivial either.

    Our one vendor for these cards has since gone out of business. The story I heard is that they lost their building lease, and didn't feel like it was worth it to move. So it looks like next upgrade, the PDP-11 printer port networking may finally die.

    The moral here is that just because the 40-year-old computer may be physically gone, it might not really be gone.

    1. Re:Footprints of old systems by jdeitch · · Score: 1

      Just minor nitpick, but a PDP-11 is a Digital Equipment Corp system ... not IBM. They were fierce rivals ...

    2. Re:Footprints of old systems by 93+Escort+Wagon · · Score: 2

      Ha! I remember having to do something along those lines back in the 1990s.

      The lab I worked at was running an early 1980s-vintage Micromass mass spectrometer (model 903, IIRC) We'd been paying through the nose for a maintenance contract on an HP 1000 computer, and finally I convinced my boss to let me move the machine control over to a PC (which we could buy for about three months' worth of HP maintenance payments). Well, one of the first things I found out was that Micromass used a serial protocol known as current loop - which, even then, was uncommon. Then, after I finally found a serial card that could handle current loop, I found out that Micromass had implemented it incorrectly. Fortunately, the protocol wasn't complicated, so I was able to physically modify that serial card to work with the mass spec (a much less daunting task back then, back before SoC and high density circuit boards were in everything). But for as long as that mass spec was still in use, that kludged card was necessary to keep the data flowing.

      --
      #DeleteChrome
    3. Re:Footprints of old systems by T.E.D. · · Score: 3

      I didn't say it was an IBM. I said our protocol uses IBM-format floats. I don't think the PDP-11's used that format, so that's probably the footprints of yet another old computer, long since upgraded.

    4. Re:Footprints of old systems by Anonymous Coward · · Score: 0

      That's not a "minor" nitpick. Calling a PDP-11 a piece of IBM equipment... Them there is fighting words.

    5. Re:Footprints of old systems by John_Sauter · · Score: 1

      The DR11-W was not a printer interface, but a computer-to-computer interface for the PDP-11. Details are here.

    6. Re:Footprints of old systems by Anonymous Coward · · Score: 0

      The moral here is that just because the 40-year-old computer may be physically gone, it might not really be gone.

      I'd rather the moral here is that you should abstract communications with the outside world (networking, database, UI, etc.) as much as possible so that it is trivial to replace.

    7. Re:Footprints of old systems by Lorens · · Score: 2

      In my office we have kept a box of punch cards. The program code names are written on the deck edge (which as a bonus served to check that the cards were in order -- I'll spare you the story about the off-site backup having a traffic accident). The people retiring nowadays tell us that they used punch cards when starting out, but it didn't last long, so in a few years no one in the office will have worked with punch cards . . . but programs with the very same code names still run on our mainframe. I haven't checked if they are identical to the punch card versions, but the code to work out things like the amount to pay every month on a thirty-year mortgage probably hasn't changed!

    8. Re:Footprints of old systems by Anonymous Coward · · Score: 0

      That's not a "minor" nitpick. Calling a PDP-11 a piece of IBM equipment... Them there is fighting words.

      At no point did the original poster say that IBM made the PDP-11, he specifically states '..The protocol even required converting all floating-point values to IBM's old format even though neither side used that format'

    9. Re:Footprints of old systems by DutchUncle · · Score: 1

      Remember that the original goal of "internetworking" was "connecting disparate networks (typical from different companies)". At the time this system was set up, someone probably thought it was very clever to standardize on the industry-leading format. The most important part of this story is the reminder that "peephole optimization" can completely miss the point of what the SYSTEM is doing. PS I wonder why nobody ever considered putting a TIP in front of each computer and taking TCP/IP between them, so that as computers were upgraded they could just use TCP . . . oh, right, penny-wise shortsightedness, same problem that almost everyone has.

    10. Re:Footprints of old systems by thejynxed · · Score: 1

      It sounds as if somewhere between the PDP-11 and that other setup, they had an AS/400 sitting in there.

      --
      @Mindless Drivel: 100% of Twitter posts ever Tweeted.
    11. Re:Footprints of old systems by Guy+Harris · · Score: 1

      The DR11-W was not a printer interface, but a computer-to-computer interface for the PDP-11.

      And a company that made DR11-W compatible devices didn't shut down until 2016.

  14. They should check with NASA... by Anonymous Coward · · Score: 0

    I believe they had to refurbish some old 60s era systems to reprocess some lunar mission data that recorded. It may even have been recorded on 7 track tape.

    1. Re:They should check with NASA... by david_bonn · · Score: 1

      Well, there are obviously computer systems on the old Voyager probes, which I believe are still in communications with NASA. The earlier Pioneer probes might still be too.

      Although my bet is there is an old PDP-8 somewhere that is still in use.

    2. Re: They should check with NASA... by Bing+Tsher+E · · Score: 1

      I have some tubes of 6100 processor chips. The 6100 is a Intersil/Harris CMOS 12 bit processor that runs the PDP-8 instruction set. They are static CMOS so can be clocked from about 1 Mhz down to DC if you like.

      People make PDP-8 machines using these processors. They are standard 40 pin DIP package with early 1980s date codes. There is a popular design that uses the 6120 processor which is the next gen version of the 6100, and there are turnkey designs for a hobby single board computer with that processor. People run PDP-8 software, like the FOCAL time sharing interpreter on these boards.

    3. Re: They should check with NASA... by Anonymous Coward · · Score: 0

      I assume that by 'tubes' you are referring to the plastic tubes that ICs were sold in. Looks like some others here have latched onto the word tube and gone off on a tangent.

    4. Re: They should check with NASA... by Bing+Tsher+E · · Score: 1

      Yes. Tubes of, I think, 16 chips per tube.

  15. You would (or wouldn't) be surprised... by mrbill · · Score: 4, Interesting

    I used to run a pair hobbyist/enthusiast sites for fans of DEC's VAX and PDP-11 series of machines.

    Shortly after 9/11, I got a phone call from someone at the Pentagon who was looking for certain parts so they could repair an older VAX that had been damaged in the attack. I was able to get them in touch with a third-party reseller who still had those bits in the back of a dusty warehouse.

    It was surprising that they hadn't upgraded to Alpha (which had been out almost ten years) then; the telco where I worked had one big system that had gone through three company changes (DEC -> Compaq -> HP) and had been upgraded in-chassis from VAX to Alpha.

    I think all large systems sold to the federal government are required to have service/support available for something like 5 to 10 years after final sale availability; can't find concrete details via Google.

    1. Re:You would (or wouldn't) be surprised... by KGIII · · Score: 1

      I owe a lot to Digital. I was just starting out, but still in the Cambridge area, and I actually got a borrowed computer and then a credit line, direct with them. While I did have my own computer at the time, it was nothing like what I needed. We stuck with DEC for quite a while, even after switching to being mostly a Sun shop. They were immensely helpful but sort of stagnated which is why we ended up going to Sun. We still used workstations from DEC for a while before moving to the SPARCstation models.

      I'd worked with DEC with my research project. They literally *loaned* me the computer post-graduation and I didn't pay for it until I got the initial funds for the first contract. The awesome thing was, they didn't even charge me full price for it - it was used, after all. So, they got a bunch of money from us. For what I was doing, it was cost effective to upgrade often.

      I want to say that we had some networking gear from them at one point but I may be misremembering? Did DEC make networking gear? I want to say that, for a while, we even had modems from DEC? Hmm... This was early 1990s through to mid-late 1990s. I think our relationship began in 1988? I'd probably not be far off to say that we had some of their Alpha boxes until 2000 or so.

      Then they ended up with financial issues and were sold/acquired, maybe? Hmm... I want to say Compaq bought 'em but my memory's fuzzy and you probably recall better than I do. I'm not sure if they went through another company before that or not too long after that. Ah well... They were a great company and they made great equipment though I'm surely a little biased. I want to say that I actually still have one of their workstations somewhere, tucked away in storage and held onto for posterity, but I really can't be certain and I'm not exactly in a position to check.

      --
      "So long and thanks for all the fish."
    2. Re:You would (or wouldn't) be surprised... by maxwells_deamon · · Score: 1

      Yes DEC made networking gear. For a long time almost all the ISA network cards I worked with had DEC written on one of the chips. They had people on the committees for eithernet many ideas were taken from decnet.

      Yes DEC got bought by Compaq, Which started the major slide and then when HP and Carly got involved that lead to what is if anything is still left.

    3. Re:You would (or wouldn't) be surprised... by KGIII · · Score: 1

      Awesome, thanks. My memory's fuzzy in my old age. I've put some strange chemicals through that brain of mine. I'm not shocked when I forget, I'm shocked when I remember. I thought I remembered DEC on the chips on MODEMs and NICs. I'm pretty sure that I'd even seen them on stuff that wasn't branded Digital.

      Now that I think about it, I was not a resident at the time, I think they did some of their networking stuff in Augusta, Maine. I didn't live in Maine back then but I've heard people mention it. That makes me curious and I'll have to spend some time with Google tomorrow. I'll see if I can dig up any information about what's left of them.

      I don't know how true it is but I've read that a DEC employee was the first person to send out what we call spam today. Some invite to a conference, trade show, or sales event and was UCE/spam. Something like that. Alas, It's nearly 0100 and Google is so very far away. Meh, it'll give me something else to look up when I'm more alert and more likely to retain it.

      I'll be home in the spring. I'll have to see if I kept one of the old workstations. I'd not be surprised to find out that I had and that it still has its peripherals and an OS installed. As I recall, they were incredibly heavy and sturdy. They were made out of a heavy gauge sheet metal with not a whole lot of plastic. I almost want to say that I've come across one from sometime after the sale as it was just a re-badged workstation. That has to have been somewhere close to 15-20 years ago.

      To Google or not... If I start this trip, I'll never get to sleep. Ah well, I cheated. It looks like it was the AlphaStation that I'm thinking of. Sadly, it doesn't look like it is very well documented:
      https://en.wikipedia.org/wiki/...

      There's a slim chance that I have some paperwork from over the years. If I do then I'll see if I can get it scanned and then send the physical copies off to the Archive.

      That page leads me to here:
      http://web.archive.org/web/200...

      That seems about right. Amusingly it says this:

      Pricing for the XP1000 starts at $7,152 (U.S.) for a system running Windows NT with a 500 MHz Alpha 21264 processor, ELSA GLoria Synergy graphics, 128 MB of RAM and a 4 GB Wide-Ultra SCSI 10,000 rpm hard drive.

      That is $10,170.61 adjusted for inflation. And so much for going to sleep and being bright eyed and bushy tailed in the morning.

      --
      "So long and thanks for all the fish."
  16. Not too surprising... by klubar · · Score: 4, Informative

    There are lots of systems that were designed around embedded PDP-8s and PDP-11s. And given the numbers of Digital VAX sold and specialized software it would not be very surprising if some of these systems are still be used. There were probably over 1.5 million of these machines sold (about 300,000 PDP-8s, 700,000 PDP-11s, 500,000 VAX machines), so there's probably some happily humming away.

    I'm sure the same is true for some earlier IBM 360/370, but they had a better upgrade path and were more expensive to start with. Most of those machines got replaced when they came off lease or the parts availability expired. But probably some of the software from the early 360's is still be used.

    Those were the days when machines were rock solid (and weighed about as much). Unlike today, when electronics are designed to be replaced every two years or so.

    1. Re:Not too surprising... by Crashmarik · · Score: 1

      Those were the days when machines were rock solid (and weighed about as much). Unlike today, when electronics are designed to be replaced every two years or so

      LOL I love the good old days as much as anyone, but the biggest reasons those machines were heavy was the linear power supplies. I'll take a nice modern switching power supply any day.

      Do you have any idea how big your pc would be with a 600 watt linear supply ?

    2. Re:Not too surprising... by Stripe7 · · Score: 1

      Key word here is "maintained", the monthly maintenance costs on those legacy systems probably costs a lot more than leasing a better performing modern computing platform that could do the job better.

    3. Re:Not too surprising... by NicBenjamin · · Score: 3, Insightful

      In theory.

      In practice these tend to be used in arcane, highly technical settings, with fairly old workforces. Setting up a new system that does precisely what the old system does, including bugs that 60-something workers have figured out their way around, but not adding new bugs they won't be able to figure out their way around? Not cheap.

      Particularly since a) you can't pay any developer more then $174k (that's what Congresscritters make), and b) you still have to interface with the department across the hall which isn;t upgrading jack.

    4. Re:Not too surprising... by Anonymous Coward · · Score: 1

      Unlike today, when electronics are designed to be replaced every two years or so.

      Component quality matters. Hugely.

      Last year, I retired an Acer branded desktop purchased in March of 2003, for the sole reason that a single core p4 running at 2GHz is woefully inadequate at running a modern operating system. The only thing I ever had to replace was the PSU (twice) and not since 2006 when I finally put a quality one in. I have no doubt this system would have run until doomsday given a chance.

      Price doesn't equate to quality either. I still have a $90 MSI board running a Phenom II x4 from 2009, makes a decent local file server.

      Also: fun fact about modern electronics that break down after two years...that LED TV that dies on you for no apparent reason probably just needs to have a $2 capacitor replaced and it'll be as good as new. Again...component quality.

    5. Re:Not too surprising... by Anonymous Coward · · Score: 0

      Anybody running a VAX in 2016 needs to be bitch slapped repeatedly. The money that could have been saved in power bills alone over the course of 20 years would be astounding, and don't even get me started on shit like Y2K compliance. I remember way back in the day, our old minicomputer had major issues just trying to deal with the decade changing.

      Of course, anyone who replaced a VAX with anything Windows related will have experienced the increased costs of bullshit licensing and the need to hire ten full time employees to do the work that used to require only one.

    6. Re:Not too surprising... by Anonymous Coward · · Score: 0

      ...and in ten to fifteen years when almost all the "60-something workers" are dead? Will it still look "not cheap" to replace them?

    7. Re: Not too surprising... by Bing+Tsher+E · · Score: 1

      The reason the really old machines were heavy is that they used discrete transistors to make flop flops. So each bit of memory in some cases was an individual circuit board. I am talking about machines like IBM mainframes from before 1970.

      It's kind of dissapointing to see tables filled with Dell laptops on the website. That's dull junk and not at all old.

    8. Re: Not too surprising... by Crashmarik · · Score: 1

      I'm not familiar with the IBM big Iron of the time, but I know the 7400 series logic chips were introduced in 64. Even bit slice wasn't using discrete transistors.

    9. Re: Not too surprising... by Bing+Tsher+E · · Score: 1

      Logic families did exist in the late 60's but I don't think the 7400 TTL logic family is that old. There were DTL families. Minicomputers were commonly made using gate-level logic gates on big wire-wrap panels earlier than the 74xx TTL family, I can't remember the numbering of the chip families but they were pre-TTL. The 7400 chip family brought medium-scale stuff like the 74181 Arithmetic Logic Unit and 8 bit latches and buffer parts like the 74373 parts (which continued to exist long in the late 80s- the 8-bit wide TTL gates are incorporated in the IBM-PC/XT and AT original motherboard designs.) Bit slice is somewhat newer than early TTL, I think it took off in the early 70's. Those AMD 8800 bit slice parts come to mind, I think I still have the datasheet manuals for them somewhere here, and probably a few chips.

      The old discrete transistor stuff probably did end by the late 60's, I remember having cards scrapped from old IBM mainframes that dad brought home where the cards had a handful of transistors configured as flip-flops on them.

    10. Re:Not too surprising... by plopez · · Score: 1

      Replacing the hardware is easy. Making sure you do not throw away what probably amounts to hundreds of years of business rules knowledge is the hard part. Time and again I have seen systems discarded, along with the people running them, and the replacement systems and code monkeys screwing up things like payroll and billing. Do you want to piss people off? Screw up their pay check. What to spend lots of $$$$$? Pay large amounts of fees and penalties because you POS ERP system (*cough* PeopleSoft *cough*SAP) doesn't know how to handle the unique business rules of your industry or your business.

      Then pay for years of customization and bug fixes.

      Then it becomes 01d Sk3wl and legacy and the cycle repeats.

      --
      putting the 'B' in LGBTQ+
    11. Re:Not too surprising... by Anonymous Coward · · Score: 1

      Software is almost always the reason old hardware is not replaced. Seeing how old the hardware is isn't all that interesting. Get info on what software it runs and when that software was written, and when the last bug fix to that software occurred. That will explain why the hardware is still around.

    12. Re:Not too surprising... by Anonymous Coward · · Score: 0

      ...that LED TV that dies on you for no apparent reason probably just needs to have a $2 capacitor replaced and it'll be as good as new. Again...component quality.

      Aye, and it's usually one of the capacitors in the switch mode PSU. Fun fact, the last TV PSU I fixed for a friend, the donor capacitor came from an old scrapped Sun monitor, made sometime in the early 90's. (The Sun monitor itself would have still been operational now had it not been subjected to the attentions of a territorial tomcat..cat urine+electronics=unhappy outcome)

    13. Re: Not too surprising... by Guy+Harris · · Score: 1

      Logic families did exist in the late 60's but I don't think the 7400 TTL logic family is that old.

      Yeah, it wasn't available in 1964; it came out in 1966.

      But, no, the original System/360s didn't use integrated circuits, they used Solid Logic Technology, where a die contained an individual transistor, a few of which were mounted on a wafer.

      The 7400 chip family brought medium-scale stuff like the 74181 Arithmetic Logic Unit and 8 bit latches and buffer parts like the 74373 parts

      But that was after the first 7400 chips, which were SSI.

    14. Re:Not too surprising... by owski · · Score: 1

      There will just be new "60-something workers" that replaced them.

  17. Waste by Anonymous Coward · · Score: 0

    Why are you doing this? What reason do you have? Processing an FOI request costs money, real money; and, it takes up someone's (usually several someones) time. if you're doing this "for fun" then you're wasting time and taxpayer's money.

  18. Posting AC due to NDA by Anonymous Coward · · Score: 0

    I really don't know if NDAs involving high security clearance expire.

    Anyway, once upon a time, when does not matter, I was working at a highly secure government data center. I became kind of buddy buddy with the grey beards in the control room overseeing the big iron. One day, one of them showed me something is a corner of the facility I had never had a reason to venture into. It was a VAX running VMS. Vacuum tubes and all. Funny looking thing, like a small refrigerator kinda sorta. Needless to say, I reached right out and put my hand on it for a good ten-seconds. When I asked what purpose it served, I was told no one was really sure. I thought he was joking. I later found out that in fact no one knew what it was doing if anything at all, so they just let it keep going. Weird.

    1. Re:Posting AC due to NDA by ChrisMaple · · Score: 1

      The VAX was introduced in 1977. It would not have been a tube machine.

      --
      Contribute to civilization: ari.aynrand.org/donate
    2. Re:Posting AC due to NDA by eyenot · · Score: 1

      So either AC is making it up, or AC isn't that familiar with VAX and the computer (if it's real) was something even older?

      That's sort of crazy, to think that somewhere in our government there might be a machine that's older than anybody working around it, that nobody understands even in terms of what function it's processing or even what inputs and outputs it's dealing with, and that if it broke maybe -- who knows -- an entire agency could collapse or a huge part of the infrastructure could fail, or a nuclear war could start or something.

      Pretty weird even if it is speculative fiction.

      --
      "Stratigraphically the origin of agriculture and thermonuclear destruction will appear essentially simultaneous" -- Lee
    3. Re:Posting AC due to NDA by Anonymous Coward · · Score: 0

      "a machine that's older than anybody working around it..."

      B52s are still flying. first flight of the B52 was 1952, and been in active use since 1955.

      And is expected to continue flying until some time after 2040.

    4. Re:Posting AC due to NDA by Darinbob · · Score: 1

      It's possible there were some tubes in the power supply of older VAXen.

    5. Re:Posting AC due to NDA by Bing+Tsher+E · · Score: 1

      Very unlikely. The 'trailing edge' of Tubes in electronics was in RF power applications like large Radio Broadcast equipment. Mundane things like power supply rectifiers and regulators went all solid state by the early 60's for the most part.

      The last tubes in consumer electronics were the high voltage rectifiers in TV sets and the CRT (obviously), and probably Guitar Amplifiers for aesthetic reasons (clipping on the output for classic rock guitar distortion)

    6. Re:Posting AC due to NDA by Trax3001BBS · · Score: 1

      I really don't know if NDAs involving high security clearance expire.

      I was at an government auction with a friend when a SPARCstation came up for bid.. It was him against a contractor who had a customer for it, the final bid cost him $500, the contractor said it was a nice system even with the hard drives they way they were. This was news to the both of us, looking at the two drives installed inside they had taken a bad saw, sawing each hard drive half way through (to the spindle) ; as the system had come out of a secure area.

    7. Re:Posting AC due to NDA by Anonymous Coward · · Score: 0

      Original AC here. The correct assumption is that I do not know enough about VAX technology to have understood it would have been running on integrated circuitry. Although most of the stuff was remarkably old. Perhaps my memory is not serving me, but it was certainly the oldest computer I have ever seen, and truly no one had a straight answer on what it did. The coolest thing about the place was the IBM mainframes. It was always neat to crack one open and see the tiny little OS/2 workstation inside.

  19. The FAA doesn't really have systems from the 1970s by Chibi+Merrow · · Score: 2

    While we had fun telling visitors to the labs that most of systems running Air Traffic Control were "essentially from the 60s and 70s" that's only technically correct.

    The truth is the original HOST mainframes were replaced in the 80s, and then again in the 90s, and now being phased out for ERAM (which is built out of COTS PC parts). In some cases software was brought forward during upgrades, so it may have been possible you'd of been running some assembly code back from the Apollo era but not really. There were still some oddities, though (like the "units" of some fields being in 64ths of a nautical mile), that made interfacing with the ARTCC systems fun.

    There might be some ARTS IIA terminals left? Those are vintage early '80s, but I think they've all been replaced with at least IIE or IIIE ('90s) and are all slated to become some form of STARS (late '90s) by 2020 or so.

    --
    Maxim: People cannot follow directions.
    Increases in truth directly with the length of time spent explaining them
  20. Do Satellites Count? by Anonymous Coward · · Score: 0

    Vanguard 1 has been orbit for 57 years but is a derelict.

    Pioneer 6 was launched 12/16/65 and may still be working, Pioneers 7 and 8 may also be working.

    Voyagers 2 and 1 are still in contact.

    The GOES 3 satellite was launched in 1978 and still functioning.

  21. Which government? by Anonymous Coward · · Score: 0

    I'm relatively new here. I don't know how much latitude Timothy and others have in editing TFS.

    Is Slashdot geared towards an international audience or a mainly-US audience?

    US readers may take this for granted, but as an international reader I'd appreciate summaries like this making it clear that the article is talking about the US government.

    1. Re:Which government? by Anonymous Coward · · Score: 0

      Let's put it this way, when they talk about the world, they mean that one country in the area of America which is called US of A.

    2. Re:Which government? by KGIII · · Score: 1

      Don't take it personal but it's very US-centric. Unless specifically stated otherwise, it's US-related. Most of the time. We don't actually have editors. We just have people who hold that title and get paid for it. I presume they're busy but I'm not actually sure what it is that they do.

      However, this isn't going to be something you see change in the near future. It's the way it is, the way it will be, and the way it has always been. Don't let anyone fool you, Slashdot was never "good." If you don't believe me, read some of the old comments or summaries and compare them with today. No, Slashdot was never good. But, it is what it is.

      Tromping in and thinking that you're going to effect meaningful change isn't going to do you any good. You'll just find it frustrating. You can't change it and it won't change for you. The momentum is strong, the drive of entropy is stronger. It's all and nothing, at the same time. Just when you think you've got it figured out, you'll find out you were wrong.

      Pull up a chair and register an account, if you want. We've new Overlords. As someone said, they bought the site by mistake. I prefer to think they were drunk. So far, so good. They're pretty good people, albeit not the speediest. That's okay, we might not want the speediest. We're a bit risk/change averse. You should see what happens if they move a button. Heh...

      As I said, it is what it is. You either like it or you don't. It's actually possible to hold both views at the same time. It's probably best to hold both views at the same time. Don't worry, there's nothing you can do right. Then again, there's not much you can do wrong. No matter what your motives, someone will find fault, almost invariably. As I said to the new owners, "You'll adjust."

      --
      "So long and thanks for all the fish."
    3. Re:Which government? by RockDoctor · · Score: 1

      It is very US-centric. I just write about thing in my country and make occasional reference to "foreigners like Americans". Not many of them even notice, because they simply don't think of themselves as being foreigners.

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
  22. better reference by John_Sauter · · Score: 1

    The DEC manual for the DR11-W is here.

  23. Oldest computer installations are on classified... by Anonymous Coward · · Score: 0

    sites. Find your nearest land-based ICBM site and ask for a tour.

  24. Oldest I've seen. by Anonymous Coward · · Score: 0

    In my job the oldest I've seen is 12 years old in use. There were some beige boxes in storage that were likely 25 years old as well.

    This was working for MWR under DOD

  25. Re:"If it ain't broke, don't fix it?" by AndyCanfield · · Score: 2

    Reminds me of my first marriage. One night, after twenty years, I discovered that it was indeed broke. A light tap, and the whole thing shattered. The most solid things are often the most fragile. It is wiser to periodically evaluate your universe.

  26. How is this wasteful? by eyenot · · Score: 1

    There are a lot of comments here that this FOIA request is a waste of time and money, all for "fun". But if you think about it, this information shouldn't have to be commanded or browbeaten out of millions of government employees. And if the government is operating the way it *should* be, then the gathering, processing and collating of this data should all be routed through a record so that -- voila -- anytime the government, itself, needs to know what it has on hand then it knows where exactly to look and in one place. This would be very handy if the government, itself, started to wonder if entire groups of machines and protocols needed immediate replacement or not. They could have a specialist go through all of this gathered inventory data and give them a realistic budget and timeline. With the data already on hand, that sort of project would be saved the time and effort it took to compile the data in the first place.

    Now, on top of that, consider the four stated aims of the report:

    * Find which departments are thrifty enough to keep machines in service long past their expiration.

    -- how would this not serve as valuable information to the government if it's assessed and quantified in a way that's easy to understand and makes sense?

    * Learn the technical dependencies that are holding back the effectiveness of public services.

    -- again, how would this not be valuable information?

    * Reveal security problems raised by departments that refuse to upgrade from unsupported, obsolete systems.

    -- again!

    * Critique the allocation and utilization of computer resources by the government

    -- which doesn't offer as much usefulness as the other three points, really just generalizes the entire thing, but still there it is. How is this not valuable information?

    --
    "Stratigraphically the origin of agriculture and thermonuclear destruction will appear essentially simultaneous" -- Lee
  27. I don't know which is worse by RabidReindeer · · Score: 1

    Whether it's the appeal to "common sense" or the "if it ain't broke, don't fix it".

    People seem to think that software is forever. It isn't. It usually rots from the outside in as OS's are updated, peripherals are replaced, and other external factors kick in. To keep these dinosaurs running require more and more desperate measures as time passes (then again, we're talking government, where expensive solutions are considered just another day). Parts that haven't been made in decades wear out, people who know how it worked die off.

    That's not even taking into account stuff like the higher power consumption, lower computing speed, more physical space required for discrete boxes and so forth.

    1. Re:I don't know which is worse by DutchUncle · · Score: 1

      No, it's not that "expensive solutions are considered just another day"; it's penny-wise shortsightedness when things are running, interspersed with crisis emergency spending when the shortsightedness explodes. Besides, what IT project has ever run on time and under budget? So of course the replacement plans get knocked down. Heck, look at the FAA . . .

  28. Sparcstation 10 by mdhoover · · Score: 1

    We still have a sparcstation 10 running something somewhere (and by that I mean noone knows what it does anymore and we can only pinpoint its location down to somewhere in a regional office)

    1. Re:Sparcstation 10 by Handpaper · · Score: 1

      Wide-area version of this? http://bash.org/?5273

    2. Re:Sparcstation 10 by Trax3001BBS · · Score: 1

      Wide-area version of this?
      http://bash.org/?5273

      That's too funny, at least one other person has heaps of laundry laying all over the place.

    3. Re:Sparcstation 10 by Anonymous Coward · · Score: 0

      We still have a sparcstation 10 running something somewhere (and by that I mean noone knows what it does anymore and we can only pinpoint its location down to somewhere in a regional office)

      Sort of reminds me of the time I found a mysterious Sparcstation lurking on the network, I spent hours going round all the offices looking for it, couldn't find it, sooo, thinking someone was up to something, mounted a bit of an attack against it..(this was in my BOFH days).

      How was I to know that 'security' had secretly installed it on the network (spliced into the network by external contractors and hidden somewhere above the dropped ceiling, it was part of an over-elaborate door entry system, it talked to the door controllers via RS-232, and reported back to another sparcstation somewhere, apparently running an Oracle database.)
      Ooops.

  29. rotfl. Opposite of US government by raymorris · · Score: 1

    > And if the government is operating the way it *should* be, then the gathering, processing and collating of this data should all be routed through a record so that -- voila -- anytime the government, itself, needs to know what it has on hand then it knows where exactly to look and in one place.

    Lol. That's quite the opposite of the US federal government.

    In some ways, it's a good thing that the federal government isn't designed for efficiency, that speed and efficiency aren't anywhere on the priority list. As a most obvious example, it would be far quicker and more efficient to have Kim Jong Obama make all of the decisions unilaterally rather than have the whole country debate policy. Heck, it took more than a decade to decide to do Hilarycare, and more than 10 more years to implement it after it was passed. (Including renaming it "Obamacare" along the way.) A dictator would be more efficient, but then you end up with North Korea. So we don't really want efficient government ; we want fair government, we want transparent government, we want big policy decisions to be made carefully, thoughtfully - slowly. Efficiency isn't what we want in government.

    1. Re:rotfl. Opposite of US government by eyenot · · Score: 1

      it would be far quicker and more efficient to have Kim Jong Obama make all of the decisions unilaterally rather than have the whole country debate policy. ...

      we want fair government, we want transparent government, we want big policy decisions

      Um, have you ever heard the term "speak for yourself"? Or the position that using the royal "we" philosophically is not only entirely bullshit and kind of stupid, it's also rude?

      More to the point, nothing I said in parent post was political in any way. The government's bureaucratic process can be more efficient without that having any bearing on the power of the executive branch. Sorry if you got those wires tangled but that's not all of "us" with that problem.

      Sorry but not everything in the world fits inside your soapbox.

      --
      "Stratigraphically the origin of agriculture and thermonuclear destruction will appear essentially simultaneous" -- Lee
  30. Analog by mcswell · · Score: 1

    He needs to qualify his statement: he's clearly looking for digital computers. The Navy (and for all I know other services) has long used analog computers (yes, with gears) for controlling their big guns; we had them on my ship, a DDG retired in 1992. I believe the last of the analog gunfire control computers were retired from the US Navy in the early 2000s, but I could be wrong.

  31. One in the running was melted down for rare metals by Trax3001BBS · · Score: 1

    The government holds auctions on old or now unusable equipment from the nuclear reservation Hanford.

    At one time I was going to bid $300 (US) on a Univac that took up an entire corner; if I'd of gotten it, it would of taken over the house (it's huge, with many pieces).

    I got the bid on another batch of items -when picking them up I asked about the Univac and was told it was pulled as no bid came close to the value of the gold and other heavy metals it could be melted down for.

    That was over 25 years ago so imagine it's part of an engagement ring at this time.

  32. ICBM control computers are over 50 years old by Zibodiz · · Score: 1

    The computers that run the ICBM equipment in Cheyenne, Wyoming are well over 50 years old. If you're interested in seeing antique computers still in use, go to the 'Fort D.A. Russell Days' event, held annually. They give a tour of one of the missile silos, and will tell you all about how old the computers are, yet they keep using them, because they have no need to change.

    Also, in a USDA office somewhere in Western Nebraska (Bridgeport, if I'm not mistaken), I once saw an old IBM server in use. I believe it was a PS/2. I noticed it had a 5 1/4 flippy drive, and asked someone about it, and she told me they use it to store mapping coordinates, and that there was only one person there who knew how to use it (it had a CRT from a PS/1 displaying a CLI), and that they could only use it to print results on an impact printer. That was about 3 years ago, and I'm guessing it's still in use today.

    1. Re:ICBM control computers are over 50 years old by braindrainbahrain · · Score: 1
      Yea, I remember this newscast not long ago about how the ICBM command center used 8" floppy disks:

      http://www.cbsnews.com/videos/...

  33. John Oliver did it by benzh · · Score: 1
  34. Voyager 2 on-board computer by Martti · · Score: 1

    Does Voyager 2 on-board computer count? "Reason for not upgrading: cost, too expensive to send a technician to the edge of the solar system."

  35. Try 60 years... by Alomex · · Score: 1

    A while back we did a site visit to a bank mainframe room. They had one backup holdover of every mainframe model they had ever bought going back to day one, just in case they needed to rerun some old piece of software. You felt like you were traveling back in time as you went down the backup row.

  36. Wait, what? by Etcetera · · Score: 1

    As the saying goes, "If it ain't broke, don't fix it." If a machine is doing its job, reliably and without error, then common sense dictates that you just shouldn't mess with it. This is doubly true for computers and quadruply true for government computers. This lends itself to an obvious question: what's the government computer most in need of an upgrade?

    He goes on to talk about thriftiness, security, "obsolete", and "holding back"... but he doesn't seem to understand the difference. Those indeed are problems, but that doesn't mean the oldest, non-broken, computer is one of them. Windows XP on a public desktop is old; a mainframe data storage retrieval system might be 20 years older. The Windows box is probably "[more] in need of an upgrade."

    This seems like a fun project (finding the oldest government computer system), but I'm not sure the author fully understands what that aphorism truly suggests.

    1. Re:Wait, what? by RockDoctor · · Score: 1
      Agreed. By the definition of the question, the machine in question is still working and doing it's job appropriately, so it does not need replacing.

      Yes, in the event of a hardware failure, it may be a bitch to repair or replace, so appropriate substitution plans should be in place. If the business in question needs a disaster-recovery plan, it should certainly be included (possibly as a "service to be replaced with an equivalent"). That's planning for future events. but until something happens, the machine/ service does not need to be "upgraded" because, by definition, it is working.

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
  37. govt doesn't depreciate by Anonymous Coward · · Score: 0

    The government keeps assets on the books at acquisition cost forever. They use what is essentially cash based accounting (no accrual) so there's no "depreciating". When you excess government property, you say what it's condition is (one of 3 or 4 choices from "functional", to "needs repair", "beyond economic repair", etc.) and then it goes out for public bid, usually as part of a large lot of stuff.

  38. IRS by herbierobinson · · Score: 1

    He should ask the IRS and see if they are off 8 track mag tape, yet. [This is no a a joke, they last I heard, they were trying to get off mag tape.]

    --
    An engineer who ran for Congress. http://herbrobinson.us
  39. Don't need to do it in FORTH by dbIII · · Score: 1

    There is a utility for changing MAC address, hostid etc on those old sparcs for the purpose of such things as running legacy software locked to such details and those things are already running it so they are already pretending to be something other than what is on the chip.

  40. We the People, bitch. Feel free to move to NK by raymorris · · Score: 1

    We the People of the United States, in Order to form a more perfect Union, establish Justice ... do ordain and establish this Constitution for the United States of America.

    If you prefer North Korea-style society, you are welcome to move there.