I liked this story. I still have a few floppies in storage as mementos of times long since past. Maybe I'll consider sending in to make them useful for someone somewhere else.
I have a 'picture wall' at our house and in between the pictures with good memories I also put some floppy disks (and cassettes). Looks nice in my opinion and reminds me of those times when taking a picture was still somewhat special, and the output was something to be kept, whereas now in the majority of cases it seems pictures are taken to be looked at once and then forgotten for eternity.
I still have a good collection of Atari ST demos on floppies. Even managed to find a working floppy drive (had to try three different drives) for a PC to write some recent demoscene productions (2010-2015) on floppies.
It's interesting to speculate what would have happened if the parallel innovations of CD and internet had not sidelined the floppy, since software was already outgrowing them : the last version of Visual C++ that I installed from floppies came on IIRC several dozen disks, requiring the thing to be carried in a couple of sturdy carrier bags - clearly not a sustainable situation.
They were commonly used in graphic design, at least. (Designers needed to send files to print all the time, and the Zip drive was the standard until CD-R became cheap and fast enough.)
And for the last mile: Moving the downloaded goods from Fast Internet Place (as in the computer class of the school) to Slow Internet Place (home with V.34 only connection). And to send them via snail mail between swapper friends.
There were Zip, there were Jaz, there were LS120, there were various MO formats.
Thinking about it i wonder if what killed it was not just the CD-R, but also that it went from primary storage (early PC etc) to transitory storage (sneakernet etc) as HDDs took over.
I actually bought a floppy drive from Best Buy around 2002 as a replacement. I had some old files I needed from floppy, and Best Buy was the only place open. Cost me $35 at the time. When I got home, I was surprised to find the floppy drive inside had a 40 pin connector instead of the standard 34 pin (MFM) connector. The drive seemed a little different, and I hesitantly connected it to my desktop to find that it was recognized as an LS-120 drive.
By 2002 LS-120 was already past its peak and even Zip drives were pretty much old news. I think I got my first Thumb drive a year later, with about 512 MB of storage.
I still have the LS-120 drive in my primary desktop, and it was a nice drive. It was quieter than other floppy drives, and had a motorized load and eject function.
I used to use Zip disks to sneakernet files to and from computers on an isolated network. This was when USB flash drives were ~8-16MB and we were not guaranteed access to part of the machine were a USB slot was. I also remember using Sysinternals Sync before ejecting the disks. Too many times did I get back to my desk and have corrupt (incompletely written) files otherwise.
As far as I am concerned it wasn't CD and internet but USB storage getting really cheap in 2001 ~ 2002: it fulfilled all the same use cases as the floppy, didn't require special hardware, was more reliable and had more storage capacity.
I remember IIRC AutoCAD 12 on too many floppies... was swapping disks for days in my H.S. cad lab... Though ZIP wasn't so bad... if you removed a few bits, NT4's installer fit on a zip disk.
It's truly frightening to see logic used to justify use of floppy disks in nuke silos: reliability, "tried and true", no need for internet connection! None of these make any technical sense. This software likely is getting no real testing so all these assertions are pointless. The naive assumption that whatever built in 90s was bug-free, most reliable and most secure is completely baseless. It's almost guaranteed that the software possibly has many bugs and issues that wouldn't have been addressed without advances in tools and techniques such as TDD, ability to do massive exhaustive testing, ability to measure code coverage, better safety features in languages, fixes in standard libraries, patches in kernels, provable algorithms for multi-threading and so on. Just think about old flaky FAT disk format VS advanced far more reliable new disk formats. If you can develop new fighter planes, you can sure develop new control software that is at least as reliable as previous generation.
The systems are time-tested. How many accidental launches have happened so far? Zero. In my opinion that's the most important metric of these systems, as long as they are still credible threats otherwise.
Also, what could a better system improve? Slightly faster launches? Slightly more certain launches? I argue that those are not important -- the enemy won't fear the missiles more if they run off USB. But you might increase the chance of a false launch, and certainly incur large testing costs.
From a military point of view, the huge advantage of floppy disks is that they're dumb storage. The modern replacements such as USB storage or SD cards are microprocessor-based systems. How are you going to verify that the SD cards you're buying don't have malicious firmware?
Adding to that, 8 inch floppies have a certain quality to them that no younger medium has ever recreated: try hiding one on your body, try sneaking in an illicit reader to make "personal backups"... Security through unwieldyness goes a long way, when your biggest threats are of the social kind.
You should buy this talisman of mine that wards away tigers. You know it works since I've had it my entire life and I've yet to be attacked by a tiger!
Uhmmmmm, read Eric Schlosser's "Command and Control". The US nuclear missiles had no protection against accidental lunch, you didn't even have to put in a code or a key, and the US government has said, sincerely, that "the US nuclear stockpile is completely safe, based on their observation that there was no accidental lunches yet".
If you have raw electrical cables sticking out of your wall, they are not "safe" just because you haven't electrocuted yourself yet.
"The systems are time-tested. How many accidental launches have happened so far? Zero. "
Hell, if that's your metric just unplug the thing. Exactly zero accidental launches, and you'll only have made the wrong decision once if anything goes wrong...
A floppy disk ejected from the drive and stored in a safe with a Marine with a rifle standing in front of it, is 1000x more secure than anything connected to the Internet. Maybe a million times more.
While true, the phrase "floppy disk" is pretty much immaterial to your sentence, as it's true with almost anything else substituted in there. Floppy disk, hard drive, USB thumb drive, engraved Zippo lighter....
Historically, floppy disks were meant to be removable whereas the majority of hard drives were fixed.
And the problem with USB drives is they are fundamentally not like disks; they are small computers in their own right, and can be made to perform computations you don't want. A floppy is passive.
I remember the various floppies (5.25", 3.5") to be the most unreliable piece of technology I have ever worked with (almost on par with Datassette tapes). Read errors were commonplace.
Having gone through a pile of old floppy disks, I think magnetic media is pretty reliable for long-term storage if treated well. Keep them clean, away from strong magnets, and at moderate temperature and humidity and the data will remain readable for many years.
Where it fails is on re-writing. The disks I had that were most likely to be bad after sitting in storage were the ones I had erased multiple times. If I had only written to a disk once or twice then I usually had no problem reading it again.
rotate it regularly... if you're talking tape drives, re-wind them and always store upright... too much has been lost to poorly storing magnetic and film media.
Fortunately digital conversion is pretty damned good and easy enough to back up... if only "The Wonder Years" hadn't downmixed the music and vocals into their master copies... :-(
Could you make a simulated floppy disk which worked in existing physical drives, but contained flash (and thus lots of more storage, potentially more reliable, etc.)?
It's almost certainly doable, although detecting the head position is going to be a hassle. But you can't increase the capacity much, because you're limited to N sectors and M tracks.
I'm not sure you need to detect the head position, because in the end you need to deliver the appropriate signal at the same location, so it'd be simpler to just have a linear array of 80 "heads" in the "disk" and let the drive head read from whichever one it likes. Then you would only need to track the spindle position.
Head-to-head coupling works in the real world, as anyone who has used a cassette tape adapter in their car stereo knows. I don't know how much "horizontal" tolerance you would need in terms of emitting the signal from the "disk"; if different drives have their heads not quite on center it could present some difficulty (whereas with real disks it would not matter--they could position the head anywhere within the shutter aperture).
Miniaturization would be another challenge--unless you're willing to have a ribbon cable coming out the front of the drive, you'll need to cram some batteries inside what is quite a thin space. Perhaps harvest some energy from the spindle (it would almost certainly be enough if you manage to find or create a thin enough generator). The last piece of the product puzzle might be to add Bluetooth...if for no other reason than being able to get data out as soon as it is written.
Another floppy disk hold out are the many musical keyboards/synths manufactured in the 80s and 90s that used floppy drives to load patches and sequences. Hell, patch, there is another anachronism that has a lot in common with the floppy disk icon :)
I've had to use them up until just recently. To run reverse engineering tests on an SNES console, I've had to use an old copier device. It's a box that plugs into the cartridge port and will load games off of floppy disks. (and of course, it can copy games plugged into the box onto said floppies.)
Usually takes around 5-10 minutes to split an image file into four parts, write them to floppies, and then load them one at a time onto the copier. And it's only that fast because I'd do it in parallel. The disks would usually fail after a few hundred write cycles, so I stocked up on quite a few.
Thankfully, I now have two alternatives. A flash cart that takes an SD card, and soon a boot loader for the expansion port that will let me load my own code through the controller port via serial.
So I'm done with floppies at long last; but now my new fun is sourcing 5V-tolerant components, and card edge connectors for an extremely rare pitch size :D
If you're into firearms and close to a suitable location, seeing a cell phone fly apart from snake-shot (think shotgun-style ammo for a handgun) at close range is pretty gratifying, even if cleanup isn't the most fun... used to do that once a year with expired hardware, more so with the troublesome hardware.
One word that I don't associate with floppies is 'reliable'. I don't know how many times I had bad sectors on my floppies back in the day. There's no protection against magnetic fields, the material on the plastic disk isn't always OK (hence the 'single sided' 3.5" floppies you could buy and which you could transform into double-sided floppies by punching a hole in the corner ;))
So it was surprising to me that floppies were seen as reliable in the article and e.g. usb sticks aren't (they're not mentioned as an alternative at all, which I find a little odd)
I agree that floppies were all but reliable in practice. But I suspect sensitivity to magnetic fields was largely a myth. I did some experiments back in the day with some permanent magnets and 3.5" disks. As I remember it was pretty hard to make the disk unreadable with those. Granted they were probably much weaker than what is available today (neodymium must have been terribly exotic back then). I was much less afraid after that that some carelessly positioned magnet will wipe out my floppy library.
I agree. I tried to erase 3.5" and 5.25" floppies with refrigerator magnets as a kid, and they did nothing.
Magnetism is an inverse-cube law, and the write head is really close to the disk. Even if you hold your magnet to the floppy casing, you have a hard fight against the inverse cube.
I remember having corrupted disks that I had kept too close to the speakers and CRT, but I can't say whether the corruption was actually caused by the magnetic fields or not.
I wanted to learn Perl but lacked an internet connection. The only way was downloading at the library and saving onto floopy disks. It was >12mb windows installer meaning I had to split it across 10+ disks. It took 3 attempts to get a full set of disks without ant errors.
I think this whole reliability talk was about the devices that use the floppy drives not the disk themselves. To get rid of floppy drives some machines would need to be replaced or retrofitted with newer models, and this is the source of unreliability.
> “The floppy disks and associated technology are tried and true,” I was told. “As you can imagine, we want to ensure the utmost in reliability and efficacy when operating such a critical weapon system. Therefore, if a system is ‘old,’ but still reliable, we are inclined to use it.”
That's a quote, not a direct claim by the article, but the article presents it as factual and doesn't follow it up with anything but confirmation.
As long as USB sticks can masquerade as keyboards, having a USB port is not really desirable in a secure environment. And the need for high reliability often goes together with a need for high security.
Many complain here about how bad the disks where, but nobody seem to remember the sudden quality drop in the late 90s. For me it was very different to use a (not very used) 10 year old disk than using a freshly bought one, any brand.
Optical media underwent a similar reliability collapse where long term reliability went from near 100% to maybe 50% when prices dropped.
I haven't used legacy flash media in a long time, I wonder if it followed the same pattern toward the end of its life when prices collapsed. Flash sticks used to be pretty reliable... back when 32 megs was a big flash drive.
Well older flash media likely use larger components on the dies. And larger components means less leakage, iirc.
That said, i wonder how much the quality drop of various media can be traced to outsorcing fever. Meaning that just about every brand just a sticker on the same product made in some Asian sweat shop factory.
The Morbidelli U26 still runs like a tank, this article is bang on. The industrial control units are proprietary as are the drive pinouts. That said, it does support KERMIT, but it is a lot more difficult to train people to use a serial terminal than it is to just keep using floppies. Apparently they are still running 15K Euro:
http://www.kitmondo.com/morbidelli-u-26/ref157473
It's a common source of ridicule amongst German IT workers (esp. commuters) that our high-speed trains ("ICE") still use disks to transfer seat reservations. And it's not that infrequent that you'll hear "seat reservations can't be displayed" being announced at the platform, probably because the train couldn't be anointed by the holy plastic wafer.
>The key word here is reliability — and that’s likely the reason floppy disks are still being used in medical equipment, ATMs, and aviation hardware as Tom mentioned. The cutting edge of technology is fine for your smartphone or a video game console. But when it comes to mission-critical hardware that literally controls a potential nuclear holocaust, “tried and true” carries more weight than “new and improved.”
Great gobs of bullshit. Floppies aren't reliable and they never were. The disks are easily damaged, susceptible to contamination from dust, they wear out quickly; the same goes for the drives. As an example, the Tektronix 3000 series oscilloscopes came equipped with floppy drives, and that was by far the most common reason (like many times more often than any other reason) for repair on the things. Thankfully after they started coming equipped with ethernet, you could forget about the floppy drive. Take note that all the examples mentioned relate to industries which are very conservative WRT implementing technology, and have an onerous approval process to get new tech in use; that's not evidence that floppies are somehow superior, but rather that the process for approving new tech. is sometimes too burdensome. When I was in college, a large fraction of my IEEE chapter's non-dues fundraising was through sales of floppy disks to students; because the damn things weren't reliable.
The "This Morning" segment included is a fluff piece, and completely at odds with recent (well, 2013/2014, not too distant past) reporting of serious problems with the stewardship of nuclear weapons. I am not sure why any sane person would want to gloss over those problems
He is not talking about the reliability of the disks themselves or the drives, but rather the reliability of the massive infrastructure they are connected to.
Seems the problem is they are confusing "reliable" with "tried and true". The onerous approval processes have more to do with sticking with "the devil you know".
Indeed. You will get better mileage out of some reasonably small EEPROM (also, old, tried and true) that has millions of write cycles and a 100 year lifetime. Chip encased in epoxy, no moving parts. No brainer.
The only reason floppies would be used in some medical equipment is that in that kind of niche market, you can get away with holding customers hostage with old crap (and at inflated prices).
> Take note that all the examples mentioned relate to industries which are very conservative WRT implementing technology, and have an onerous approval process to get new tech in use; that's not evidence that floppies are somehow superior, but rather that the process for approving new tech. is sometimes too burdensome.
My own personal experience with this follows: My last job as a sounding rocket telemetry engineer required me to use 3.5" floppies regularly, because the majority of the GSE in use for supporting launches was still running a program called TDPlus. TDPlus, which runs in DOS, is intended for the real-time monitoring of telemetry data. We had lunch-box style computers with the old-style 51 pin parallel port for "high speed" data (up to 10 Megabits/sec) interfacing with experimenter equipment, as well as GDP Space Systems bitsync/decom cards.
The lunchbox computers were always breaking down. Floppies were subject to all the faults you mention and were always failing. I found a hoard of floppies in an old cabinet and spent a day or so seeing which ones I could still use, and then protected them like gold.
All of this was still used because management claimed that "the right color of money doesn't exist for us to upgrade this". They had started to get a few newer units when I left...but depending on such horribly unreliable units for critical launch functions still boggles my mind. There's "too burdensome" and then there's out-right mismanagement.
On the other hand, the floppies I used and the dot-matrix printouts of the data I monitored are neat souvenirs from that time in my life.
The 8" disks I used were actually pretty good for reliability[1]. The 5 1/4" drives were just plain problematic. Doubly so if you used a "Disk Doubler"[2]. The 3 1/2" were pretty nice after the 5 1/4", but I'm still not sure they were ever as good as the 8". I'm glad to be rid of the whole lot of them with USB Flash drives.
I will say it was quite exciting to buy an Indus Floppy Drive for my Atari 400 (upgraded to 48Kbytes). It was amazing going from the 410 cassette player to the floppy. The door popped up in a most satisfying manner.
1) I'm with the other posters that "tried and true" and "know quantity" is probably a better way to look at it. Also, the stuff using the floppy drives probably just works.
2) for the younger crowd, at one time 5 1/4 floppies came in single sided and double sided varieties. So, depending on the floppy drive you bought the correct one. The price wasn't all that different. 5 1/4 floppies have a notch on the side that tells you how to insert it. Put it in wrong and the drive won't close since part of the mechanism goes down through the notch. Now, some "genius" decided that if you had a single sided drive and cut a notch in the other side of the floppy you could use both sides of the disk in much the same way you could play both sides of a record. This 'genius' started selling a device that was basically a hole puncher with a metal attachment that held the floppy properly to punch the hole (really a small square punch). You then could use both sides. Now, inserting the disk upside down would work, but the disk spins the opposite direction. You can guess how this interacted with the fabric that removed dust built into the floppy. http://www.atarimagazines.com/compute/issue10/036_1_FLIPPING... is an old article on it and Wikipedia has pictures https://en.wikipedia.org/wiki/Floppy_disk_variants#Flippy_di...
Serendipity is an amazing thing, as I type this "I'm Much Too Young (To Feel This Damn Old)" just started playing on my music player. I'll leave it to someone else to explain the purpose of the pencil when talking cassette tapes.
I'd imagine that the tolerances are loose enough on an 8" floppy, that barring physical damage, they probably are pretty reliable. 5.25 were somewhat problematic though... I remember in the mid 90's when the quality went through the toilet.. the last 50-pack of 3.5" 2shd disks I bought, I swear about 1 in 4 had errors out of the box... I don't think I used even half of it though.
I've kept a home server and/or nas since the early 00's, and haven't really looked back... I did notice how slow even flash media is though, trying to copy a season of a tv show to my tablet to view while out of town for a couple days recently... took too long, not enough space (got a 128gb microsd for next time), and just plain cumbersome.
It's hard to think about being disconnected these days. I lost over half my workday to a VPN issue.
Vintage electronic music gear, which is still widely popular with certain genres of music, use floppies heavily.
12-bit samplers like the E-mu SP-1200 and Akai MPC60 which are still popular with hip hop producers rely on floppies. A number of other classic samplers and synthesizers use them as well. There's a lot of demand for floppies in that community.
Knowing that group (meaning vintage gear heads rather than hip hop producers) there's doubtless endless online discussions about which particular brand of floppy gives the crunchiest sound. Can't beat the warmth of the MF2-HD.
When I was in high school, some kid's 3.5" floppy didn't read. He was all upset about losing some images he had been working on in a paint program.
I slid open the window and turned the center hub to inspect the surface, which soon revealed a glaring fingerprint; some prankster almost certainly had done that on purpose.
I peeled the black, plastic casing open, and took out the disk, which I then gently washed with cold water and soap in the boys' washroom, drying it with a paper towel.
I slid the disk it back into its casing, and popped it into the drive. It read perfectly.
A lot of CNC machines still use floppies. There are some great USB interfaces (that emulate a floppy drive to the CNC controller) that let you use USB sticks. You still have to format the USB stick to 1.44MB (so... buy lots of very small ones) for it to work, but it's been much better since we've switched to these.
I recently bought a new 3.5" usb floppy drive because a few friends wanted some stuff off their old disks, no telling how much will be recoverable, but willing to try.
It's interesting how little I've thought about the old formats for years. How many floppy/zip/jazz etc disks have useful data on them to this day. I worked doing support for iomega back in the day and the only thing it made me was slightly paranoid about multiple copies in multiple locations for anything really important.
83 comments
[ 0.22 ms ] story [ 139 ms ] threadDefinitely gives more weight to each picture when the turnaround is weeks, and each picture requires hours of physical work.
https://en.wikipedia.org/wiki/Zip_drive
They were commonly used in graphic design, at least. (Designers needed to send files to print all the time, and the Zip drive was the standard until CD-R became cheap and fast enough.)
Thinking about it i wonder if what killed it was not just the CD-R, but also that it went from primary storage (early PC etc) to transitory storage (sneakernet etc) as HDDs took over.
By 2002 LS-120 was already past its peak and even Zip drives were pretty much old news. I think I got my first Thumb drive a year later, with about 512 MB of storage.
I still have the LS-120 drive in my primary desktop, and it was a nice drive. It was quieter than other floppy drives, and had a motorized load and eject function.
"Only one out of every 200 users experiences sudden, catastrophic data loss!" is not the kind of thing you want to hear from your storage vendor.
Also, what could a better system improve? Slightly faster launches? Slightly more certain launches? I argue that those are not important -- the enemy won't fear the missiles more if they run off USB. But you might increase the chance of a false launch, and certainly incur large testing costs.
https://en.wikipedia.org/wiki/Stuxnet
When your storage device itself is smart and compromised, you can't even verify what its contents are. Things get much worse.
If you have raw electrical cables sticking out of your wall, they are not "safe" just because you haven't electrocuted yourself yet.
Hell, if that's your metric just unplug the thing. Exactly zero accidental launches, and you'll only have made the wrong decision once if anything goes wrong...
It's based on the correct observation that after decades of use, we know how good and and how bad it is.
It was the "A:\ is not accessible" message you received when you put it back in.
It's just mice you have to watch out for.
And the problem with USB drives is they are fundamentally not like disks; they are small computers in their own right, and can be made to perform computations you don't want. A floppy is passive.
Where it fails is on re-writing. The disks I had that were most likely to be bad after sitting in storage were the ones I had erased multiple times. If I had only written to a disk once or twice then I usually had no problem reading it again.
Fortunately digital conversion is pretty damned good and easy enough to back up... if only "The Wonder Years" hadn't downmixed the music and vocals into their master copies... :-(
I know there are floppy DRIVE emulators like https://en.wikipedia.org/wiki/Floppy_disk_hardware_emulator but I want to emulate the DISK.
There are also of course software solutions to this in a lot of cases, too, especially if you can virtualize, but sometimes hardware is needed.
Head-to-head coupling works in the real world, as anyone who has used a cassette tape adapter in their car stereo knows. I don't know how much "horizontal" tolerance you would need in terms of emitting the signal from the "disk"; if different drives have their heads not quite on center it could present some difficulty (whereas with real disks it would not matter--they could position the head anywhere within the shutter aperture).
Miniaturization would be another challenge--unless you're willing to have a ribbon cable coming out the front of the drive, you'll need to cram some batteries inside what is quite a thin space. Perhaps harvest some energy from the spindle (it would almost certainly be enough if you manage to find or create a thin enough generator). The last piece of the product puzzle might be to add Bluetooth...if for no other reason than being able to get data out as soon as it is written.
Problem is that FlashPath doesn't work like a normal floppy.
https://en.wikipedia.org/wiki/Synthesizer#Patch
Usually takes around 5-10 minutes to split an image file into four parts, write them to floppies, and then load them one at a time onto the copier. And it's only that fast because I'd do it in parallel. The disks would usually fail after a few hundred write cycles, so I stocked up on quite a few.
Thankfully, I now have two alternatives. A flash cart that takes an SD card, and soon a boot loader for the expansion port that will let me load my own code through the controller port via serial.
So I'm done with floppies at long last; but now my new fun is sourcing 5V-tolerant components, and card edge connectors for an extremely rare pitch size :D
So it was surprising to me that floppies were seen as reliable in the article and e.g. usb sticks aren't (they're not mentioned as an alternative at all, which I find a little odd)
I agree that floppies were all but reliable in practice. But I suspect sensitivity to magnetic fields was largely a myth. I did some experiments back in the day with some permanent magnets and 3.5" disks. As I remember it was pretty hard to make the disk unreadable with those. Granted they were probably much weaker than what is available today (neodymium must have been terribly exotic back then). I was much less afraid after that that some carelessly positioned magnet will wipe out my floppy library.
Magnetism is an inverse-cube law, and the write head is really close to the disk. Even if you hold your magnet to the floppy casing, you have a hard fight against the inverse cube.
> “The floppy disks and associated technology are tried and true,” I was told. “As you can imagine, we want to ensure the utmost in reliability and efficacy when operating such a critical weapon system. Therefore, if a system is ‘old,’ but still reliable, we are inclined to use it.”
That's a quote, not a direct claim by the article, but the article presents it as factual and doesn't follow it up with anything but confirmation.
I can buy the argument that the systems as a whole are trustworthy even if the floppy disks they use kind of suck. But this quote goes beyond that.
I haven't used legacy flash media in a long time, I wonder if it followed the same pattern toward the end of its life when prices collapsed. Flash sticks used to be pretty reliable... back when 32 megs was a big flash drive.
That said, i wonder how much the quality drop of various media can be traced to outsorcing fever. Meaning that just about every brand just a sticker on the same product made in some Asian sweat shop factory.
Great gobs of bullshit. Floppies aren't reliable and they never were. The disks are easily damaged, susceptible to contamination from dust, they wear out quickly; the same goes for the drives. As an example, the Tektronix 3000 series oscilloscopes came equipped with floppy drives, and that was by far the most common reason (like many times more often than any other reason) for repair on the things. Thankfully after they started coming equipped with ethernet, you could forget about the floppy drive. Take note that all the examples mentioned relate to industries which are very conservative WRT implementing technology, and have an onerous approval process to get new tech in use; that's not evidence that floppies are somehow superior, but rather that the process for approving new tech. is sometimes too burdensome. When I was in college, a large fraction of my IEEE chapter's non-dues fundraising was through sales of floppy disks to students; because the damn things weren't reliable.
The "This Morning" segment included is a fluff piece, and completely at odds with recent (well, 2013/2014, not too distant past) reporting of serious problems with the stewardship of nuclear weapons. I am not sure why any sane person would want to gloss over those problems
http://www.npr.org/sections/thetwo-way/2014/01/30/268880352/...
http://www.csmonitor.com/USA/Military/2014/0116/Another-Air-...
The only reason floppies would be used in some medical equipment is that in that kind of niche market, you can get away with holding customers hostage with old crap (and at inflated prices).
My own personal experience with this follows: My last job as a sounding rocket telemetry engineer required me to use 3.5" floppies regularly, because the majority of the GSE in use for supporting launches was still running a program called TDPlus. TDPlus, which runs in DOS, is intended for the real-time monitoring of telemetry data. We had lunch-box style computers with the old-style 51 pin parallel port for "high speed" data (up to 10 Megabits/sec) interfacing with experimenter equipment, as well as GDP Space Systems bitsync/decom cards.
The lunchbox computers were always breaking down. Floppies were subject to all the faults you mention and were always failing. I found a hoard of floppies in an old cabinet and spent a day or so seeing which ones I could still use, and then protected them like gold.
All of this was still used because management claimed that "the right color of money doesn't exist for us to upgrade this". They had started to get a few newer units when I left...but depending on such horribly unreliable units for critical launch functions still boggles my mind. There's "too burdensome" and then there's out-right mismanagement.
On the other hand, the floppies I used and the dot-matrix printouts of the data I monitored are neat souvenirs from that time in my life.
I will say it was quite exciting to buy an Indus Floppy Drive for my Atari 400 (upgraded to 48Kbytes). It was amazing going from the 410 cassette player to the floppy. The door popped up in a most satisfying manner.
1) I'm with the other posters that "tried and true" and "know quantity" is probably a better way to look at it. Also, the stuff using the floppy drives probably just works.
2) for the younger crowd, at one time 5 1/4 floppies came in single sided and double sided varieties. So, depending on the floppy drive you bought the correct one. The price wasn't all that different. 5 1/4 floppies have a notch on the side that tells you how to insert it. Put it in wrong and the drive won't close since part of the mechanism goes down through the notch. Now, some "genius" decided that if you had a single sided drive and cut a notch in the other side of the floppy you could use both sides of the disk in much the same way you could play both sides of a record. This 'genius' started selling a device that was basically a hole puncher with a metal attachment that held the floppy properly to punch the hole (really a small square punch). You then could use both sides. Now, inserting the disk upside down would work, but the disk spins the opposite direction. You can guess how this interacted with the fabric that removed dust built into the floppy. http://www.atarimagazines.com/compute/issue10/036_1_FLIPPING... is an old article on it and Wikipedia has pictures https://en.wikipedia.org/wiki/Floppy_disk_variants#Flippy_di...
Serendipity is an amazing thing, as I type this "I'm Much Too Young (To Feel This Damn Old)" just started playing on my music player. I'll leave it to someone else to explain the purpose of the pencil when talking cassette tapes.
I've kept a home server and/or nas since the early 00's, and haven't really looked back... I did notice how slow even flash media is though, trying to copy a season of a tv show to my tablet to view while out of town for a couple days recently... took too long, not enough space (got a 128gb microsd for next time), and just plain cumbersome.
It's hard to think about being disconnected these days. I lost over half my workday to a VPN issue.
12-bit samplers like the E-mu SP-1200 and Akai MPC60 which are still popular with hip hop producers rely on floppies. A number of other classic samplers and synthesizers use them as well. There's a lot of demand for floppies in that community.
I slid open the window and turned the center hub to inspect the surface, which soon revealed a glaring fingerprint; some prankster almost certainly had done that on purpose.
I peeled the black, plastic casing open, and took out the disk, which I then gently washed with cold water and soap in the boys' washroom, drying it with a paper towel.
I slid the disk it back into its casing, and popped it into the drive. It read perfectly.
It's interesting how little I've thought about the old formats for years. How many floppy/zip/jazz etc disks have useful data on them to this day. I worked doing support for iomega back in the day and the only thing it made me was slightly paranoid about multiple copies in multiple locations for anything really important.