This feels like such an easy fix, right? I mean, if the content on the floppy disk is transferred to a usb stick for example. Alternatively if they still run IDE they could have an IDE to USB or...
This feels like such an easy fix, right? I mean, if the content on the floppy disk is transferred to a usb stick for example. Alternatively if they still run IDE they could have an IDE to USB or SATA controller board and hook up an SSD or older harddrive. Just path the systems to the new location and boombadabing.
That sets you back maybe 100euro's and a few hours of work max.
Insane how they proclaim to it costing them millions and millions.
The issue is that they want to overhaul the entire system. That requires years of planning and research. They're not just building something new, but they also need to meet the needs of the...
The issue is that they want to overhaul the entire system. That requires years of planning and research. They're not just building something new, but they also need to meet the needs of the existing train operations. Which adds a lot of complexity.
They could do your solution, but that's more or less a bandaid. The whole floppy disk thing is only a part of it and just to get clicks on the article.
This is fairly common. There's a lot of businesses and government organizations that have been around for a long time. As they grow, they became reliant on certain tech & systems that were in place before the 2000's. To do a mass upgrade or overhaul would mean being out of service for an extended time and losing money. So they waited until it became a necessity.
Comment box Scope: information Tone: neutral Opinion: not really Sarcasm/humor: none Transportation system overhauls are expensive because they have unique and complex operations. Signaling is a...
Exemplary
Comment box
Scope: information
Tone: neutral
Opinion: not really
Sarcasm/humor: none
Transportation system overhauls are expensive because they have unique and complex operations. Signaling is a lot more dynamic than people think. Trains are constantly moving throughout different environments with different physical conditions (often tunnels) which affect power delivery (catenary, third rail, other), communication effectiveness, etc. When you factor in automatic train control (at all) and street-running lines, any technical solution has to be essentially foolproof.
You can find short-term solutions to individual problems, but if trains are running on old software and hardware, it's unrealistic for them to interface with other systems effectively. This includes both back-end systems that help with real-time positioning and miscellaneous problem detection (track or power issues, unanticipated conflicts), and user-facing systems that control people's access to schedules (station arrival screens, transit apps). Inefficient operational processes dramatically increase operational costs which in turn decreases an agency's ability to provide useful services, such as fast and frequent trains. The more funding spent on COBOL maintenance, the less funding can be spent on operations, accessibility upgrades, etc.
Because we're talking about light and heavy rail, people's lives are also at risk. New signaling systems typically undergo thorough safety testing to ensure that trains aren't going to run into each other, derail, and kill 300 people. Such changes typically require some amount of certification involving computer models and then real-world testing (which is constrained by routine passenger operations). That process takes time (easily months; often years) and money, and transit agencies are incredibly strapped for funding in this country.
Depending on the kind of signaling upgrades needed, the traincars themselves are likely to require non-trivial upgrades. Often, they need to be replaced altogether.
Insane how they proclaim to it costing them millions and millions.
Transportation spending inefficiency in the United States is higher than in peer developed countries, but "a few hundred million" for a complete technical overhaul to a regional system's technology seems relatively benign.
New York City's signaling overhauls and associated modernization work, all things considered, are easily in the billions (and Byford is someone who knows what he's doing). San Francisco's network isn't that large, old, or complex, but it's not nothing.
One of the (many) reasons that this country fails to make efficient use of funding for public transit is because many solutions are engineered around tech debt piece-meal style, and do not attempt to solve the underlying problem. So I welcome efforts to holistically re-architecture transportation systems.
You're right. I was merely focused on the aspect of the drive alone. If they were to completely make their systems up-to-date that will easily cost what they have predicted it to be. Thanks for...
You're right. I was merely focused on the aspect of the drive alone. If they were to completely make their systems up-to-date that will easily cost what they have predicted it to be. Thanks for your insight.
I'm thinking the issue is not reliance on floppies (the storage media), but on floppy drives. What probably needs changing is to make the computers use something other than floppy drives (in some...
I'm thinking the issue is not reliance on floppies (the storage media), but on floppy drives. What probably needs changing is to make the computers use something other than floppy drives (in some way like you suggested); or replace the computers altogether. Maybe there's something special about what's housing/using those drives (but, like you, I doubt it).
But I don't understand why they're worried about physical degradation of the media. Surely they've made backup copies (onto other floppies)?
Comment box Scope: information, considering the issue Tone: neutral Opinion: some Sarcasm/humor: none If a signaling system experiences a catastrophic data failure while trains are in active...
Comment box
Scope: information, considering the issue
Tone: neutral
Opinion: some
Sarcasm/humor: none
If a signaling system experiences a catastrophic data failure while trains are in active operation, it could lead to improper track occupancy such as a head-on collision.
Redundant floppy drives could theoretically address such issues, but any physically equivalent backup would have similar weak points. And if there is some major issue with the hardware reading the floppies (rather than the disks themselves), a redundant set of floppies in the same machine wouldn't necessarily be useful.
I'm not too much of a computer person, but my feeling is that it would be safer and more efficient to use systems that are less prone to failure to begin with. In general, when we're working with systems that require relatively specialized knowledge, the opportunity for an intern to blow something up is kind of high.
Might be the servers themselves are equally as old and getting that USB adapter driver could be just as difficult without getting someone to develop a custom driver for it.
Might be the servers themselves are equally as old and getting that USB adapter driver could be just as difficult without getting someone to develop a custom driver for it.
if that 5.25 driver is hooked up to anything it has to be IDE or SCSI. Safe to say any daughterboard tricking the system into thinking it's just another IDE or SCSI connection with a heavily...
if that 5.25 driver is hooked up to anything it has to be IDE or SCSI. Safe to say any daughterboard tricking the system into thinking it's just another IDE or SCSI connection with a heavily reduced but acceptable data throughput will make any modern hardware work with that.
Actually IDE is a 40 pin header that post-dates and is not compatible with the 32 pin floppy header. Would be great if that weren’t true though, could just pop a compact flash card in the system...
Actually IDE is a 40 pin header that post-dates and is not compatible with the 32 pin floppy header. Would be great if that weren’t true though, could just pop a compact flash card in the system with nothing but a little adapter and you would be good to go.
There really isn’t a good commercially viable way to implement a physical adapter for a floppy drive to something more modern like a flash drive unfortunately. They’re out there, but usually only sold in low quantities to collectors and small businesses.
Actually there are inexpensive floppy disk emulators available on the market today which you can get for about $50 last time I checked. Those usually only have firmware that supports the 3.5 inch...
Actually there are inexpensive floppy disk emulators available on the market today which you can get for about $50 last time I checked. Those usually only have firmware that supports the 3.5 inch disks, though. There are more capable ones on the market which will doubtlessly work; they are much more expensive but not unobtainable by any means.
They tend to be much cheaper iirc, bought a few for my old retro gaming machines and yeah, they work flawlessly. I have a dozen games on a USB stick I can just install quickly without needing to...
They tend to be much cheaper iirc, bought a few for my old retro gaming machines and yeah, they work flawlessly. I have a dozen games on a USB stick I can just install quickly without needing to write floppies and whatnot.
Why don't they use 5.25 inch floppy emulators? Unless said hardware for whatever reason is unable to interface with the floppy emulator... Many people use floppy emulators in the retrocomputing...
Why don't they use 5.25 inch floppy emulators? Unless said hardware for whatever reason is unable to interface with the floppy emulator...
Many people use floppy emulators in the retrocomputing community and they work extremely well.
This feels like such an easy fix, right? I mean, if the content on the floppy disk is transferred to a usb stick for example. Alternatively if they still run IDE they could have an IDE to USB or SATA controller board and hook up an SSD or older harddrive. Just path the systems to the new location and boombadabing.
That sets you back maybe 100euro's and a few hours of work max.
Insane how they proclaim to it costing them millions and millions.
The issue is that they want to overhaul the entire system. That requires years of planning and research. They're not just building something new, but they also need to meet the needs of the existing train operations. Which adds a lot of complexity.
They could do your solution, but that's more or less a bandaid. The whole floppy disk thing is only a part of it and just to get clicks on the article.
This is fairly common. There's a lot of businesses and government organizations that have been around for a long time. As they grow, they became reliant on certain tech & systems that were in place before the 2000's. To do a mass upgrade or overhaul would mean being out of service for an extended time and losing money. So they waited until it became a necessity.
Comment box
Transportation system overhauls are expensive because they have unique and complex operations. Signaling is a lot more dynamic than people think. Trains are constantly moving throughout different environments with different physical conditions (often tunnels) which affect power delivery (catenary, third rail, other), communication effectiveness, etc. When you factor in automatic train control (at all) and street-running lines, any technical solution has to be essentially foolproof.
You can find short-term solutions to individual problems, but if trains are running on old software and hardware, it's unrealistic for them to interface with other systems effectively. This includes both back-end systems that help with real-time positioning and miscellaneous problem detection (track or power issues, unanticipated conflicts), and user-facing systems that control people's access to schedules (station arrival screens, transit apps). Inefficient operational processes dramatically increase operational costs which in turn decreases an agency's ability to provide useful services, such as fast and frequent trains. The more funding spent on COBOL maintenance, the less funding can be spent on operations, accessibility upgrades, etc.
Because we're talking about light and heavy rail, people's lives are also at risk. New signaling systems typically undergo thorough safety testing to ensure that trains aren't going to run into each other, derail, and kill 300 people. Such changes typically require some amount of certification involving computer models and then real-world testing (which is constrained by routine passenger operations). That process takes time (easily months; often years) and money, and transit agencies are incredibly strapped for funding in this country.
Depending on the kind of signaling upgrades needed, the traincars themselves are likely to require non-trivial upgrades. Often, they need to be replaced altogether.
Transportation spending inefficiency in the United States is higher than in peer developed countries, but "a few hundred million" for a complete technical overhaul to a regional system's technology seems relatively benign.
New York City's signaling overhauls and associated modernization work, all things considered, are easily in the billions (and Byford is someone who knows what he's doing). San Francisco's network isn't that large, old, or complex, but it's not nothing.
One of the (many) reasons that this country fails to make efficient use of funding for public transit is because many solutions are engineered around tech debt piece-meal style, and do not attempt to solve the underlying problem. So I welcome efforts to holistically re-architecture transportation systems.
You're right. I was merely focused on the aspect of the drive alone. If they were to completely make their systems up-to-date that will easily cost what they have predicted it to be. Thanks for your insight.
I'm thinking the issue is not reliance on floppies (the storage media), but on floppy drives. What probably needs changing is to make the computers use something other than floppy drives (in some way like you suggested); or replace the computers altogether. Maybe there's something special about what's housing/using those drives (but, like you, I doubt it).
But I don't understand why they're worried about physical degradation of the media. Surely they've made backup copies (onto other floppies)?
Comment box
If a signaling system experiences a catastrophic data failure while trains are in active operation, it could lead to improper track occupancy such as a head-on collision.
Redundant floppy drives could theoretically address such issues, but any physically equivalent backup would have similar weak points. And if there is some major issue with the hardware reading the floppies (rather than the disks themselves), a redundant set of floppies in the same machine wouldn't necessarily be useful.
I'm not too much of a computer person, but my feeling is that it would be safer and more efficient to use systems that are less prone to failure to begin with. In general, when we're working with systems that require relatively specialized knowledge, the opportunity for an intern to blow something up is kind of high.
Might be the servers themselves are equally as old and getting that USB adapter driver could be just as difficult without getting someone to develop a custom driver for it.
if that 5.25 driver is hooked up to anything it has to be IDE or SCSI. Safe to say any daughterboard tricking the system into thinking it's just another IDE or SCSI connection with a heavily reduced but acceptable data throughput will make any modern hardware work with that.
Actually IDE is a 40 pin header that post-dates and is not compatible with the 32 pin floppy header. Would be great if that weren’t true though, could just pop a compact flash card in the system with nothing but a little adapter and you would be good to go.
There really isn’t a good commercially viable way to implement a physical adapter for a floppy drive to something more modern like a flash drive unfortunately. They’re out there, but usually only sold in low quantities to collectors and small businesses.
Actually there are inexpensive floppy disk emulators available on the market today which you can get for about $50 last time I checked. Those usually only have firmware that supports the 3.5 inch disks, though. There are more capable ones on the market which will doubtlessly work; they are much more expensive but not unobtainable by any means.
They tend to be much cheaper iirc, bought a few for my old retro gaming machines and yeah, they work flawlessly. I have a dozen games on a USB stick I can just install quickly without needing to write floppies and whatnot.
When they'll transition to 3.5 inch!
Why don't they use 5.25 inch floppy emulators? Unless said hardware for whatever reason is unable to interface with the floppy emulator...
Many people use floppy emulators in the retrocomputing community and they work extremely well.