Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
Compare cell phone plans using Wirefly's innovative plan comparison tool ×
Government The Military United States Hardware

Join the Hunt For the Government's Oldest Computer (muckrock.com) 147

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.
This discussion has been archived. No new comments can be posted.

Join the Hunt For the Government's Oldest Computer

Comments Filter:
  • by JoshuaZ ( 1134087 ) on Saturday March 05, 2016 @04:04PM (#51644785) Homepage
    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.

    • by bws111 ( 1216812 ) on Saturday March 05, 2016 @04:10PM (#51644831)

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

      • by klubar ( 591384 ) on Saturday March 05, 2016 @05:45PM (#51645247) Homepage

        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.

        • 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 ?

          • 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.

            • 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.

        • by Stripe7 ( 571267 )
          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.
          • by NicBenjamin ( 2124018 ) on Saturday March 05, 2016 @08:56PM (#51646061)

            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.

          • by plopez ( 54068 )

            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

        • by Anonymous Coward

          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 200

      • 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.

        • 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 [fuse-network.com], 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.)

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

    • 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

    • by Crashmarik ( 635988 ) on Saturday March 05, 2016 @06:19PM (#51645413)

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

  • by Scoth ( 879800 ) on Saturday March 05, 2016 @04:23PM (#51644889)

    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.

    • by hey! ( 33014 )

      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 som

    • by DerekLyons ( 302214 ) <[moc.liamg] [ta] [retawriaf]> on Saturday March 05, 2016 @06:33PM (#51645483) Homepage

      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.)

      • by Scoth ( 879800 )

        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.

      • by dbIII ( 701233 )
        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.
    • 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 whic

  • by Anonymous Coward

    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.

    • Re: (Score:3, Insightful)

      by mschuyler ( 197441 )

      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 o

      • 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.

    • 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.

  • Intredasting (Score:5, Interesting)

    by Sax Russell 5449D29A ( 4449961 ) <sax.russell@outlook.com> on Saturday March 05, 2016 @04:35PM (#51644941)

    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.

    • Re:Intredasting (Score:5, Informative)

      by sphealey ( 2855 ) on Saturday March 05, 2016 @05:16PM (#51645103)

      = = = 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

      • 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.

      • 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).
    • by mrbill ( 4993 )
      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.
      • 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 syste

        • by sconeu ( 64226 )

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

    • 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.
  • by creimer ( 824291 ) on Saturday March 05, 2016 @04:38PM (#51644949) Homepage
    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.
  • by klindsay ( 1315185 ) on Saturday March 05, 2016 @04:42PM (#51644969)
    "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.
  • I've got boxes of vacuum tubes that could be available for the right price.

    Just sayin'.

    • 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.

  • by Anonymous Coward

    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.

  • by imac.usr ( 58845 ) on Saturday March 05, 2016 @05:00PM (#51645043) Homepage

    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.

    • by creimer ( 824291 ) on Saturday March 05, 2016 @05:19PM (#51645121) Homepage

      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.

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

      • 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.
    • by Dutch Gun ( 899105 ) on Saturday March 05, 2016 @05:28PM (#51645149)

      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."

    • 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.
    • 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

    • 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

      • 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 suppli

    • 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. :-)
      • 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 q

  • by T.E.D. ( 34228 ) on Saturday March 05, 2016 @05:11PM (#51645081)

    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.

    • by jdeitch ( 12598 )

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

    • 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

    • The DR11-W was not a printer interface, but a computer-to-computer interface for the PDP-11. Details are here [trailing-edge.com].

    • by Lorens ( 597774 )

      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

    • 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 t
  • by mrbill ( 4993 ) <mrbill@mrbill.net> on Saturday March 05, 2016 @05:18PM (#51645109) Homepage
    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.
    • by KGIII ( 973947 )

      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

      • 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.

        • by KGIII ( 973947 )

          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

  • 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

    • > 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 li

      • by eyenot ( 102141 )

        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 insi

  • 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

    • 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 . . .
  • 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)
  • 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.

  • 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 ima

  • 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 belie
  • 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.

  • 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 pr

    • 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

  • 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.]

He's dead, Jim.

Working...