Sadly, the real lesson we need to learn from Battlestar Galactica is not this. They weren't saved by old software, they were saved by not having critical systems on the network unnecessarily.
Our water and power utilities need to re-watch the pilot.
Why are our water and power utilities connected to the internet? Is it so that the employees controlling them can work from home? If so they are accepting too much risk relative to the benefit.
think about a grid responding to demand. The old model (like the titanic) had a remote manager phoning the site lead who phoned the control room who phoned the engineering room.
Remote management has clear benefits, don’t be obtuse. I totally agree on the security risks, but the benefits are obvious.
No, and it's not strictly impossible to correctly build out a SOC / SIEM and ingest all the logs you want and pay for the right engineers to make sure it all works correctly. But damned if anyone manages to do a great job in this area. It's too complex and too expensive, so almost everyone settles for "best effort."
A lot of problems are easy conceptually, but we can't manage to tackle them.
To be fair, it was both. Those ships that got the recent upgrades also got trojans that Caprica Six put in the defense mainframe with Baltar's access codes.
From a efficiency perspective centralized anything seems better at first glance, however decentralized anything just seems more resilient in the long run like nature and/or our universe.
Makes it harder to catch you in a general sweep, but it probably makes it much easier if anyone decides to target you specifically. How good a tradeoff that is varies from person to person. And the difficulty of targeting a specific person is changing in an AI world. I wouldn't try to characterize it any farther than that, just to observe that it is changing. "Hey AI go individually research and target county clerks for getting direct access to their {specific system}" is much more practical today than it was 10 years ago, and will be even more so 10 years from now.
It reminds me of an interview with police detectives. They say that some criminals think that using weapons like swords and bows will make the police job harder because police is so used to deal with guns that they may have a blind spot.
The thing is: police actually have data on swords and bows, and the fact it is so unusual to commit crimes with these limit the search space and tend to make the investigation easier.
It's incredible how far back the surveillance state goes - GSM mobile phones have an IMEI number that's tied to the handset - and the SIM is tied to the subscriber, and your phone broadcasts imei to neighboring towers constantly.
I haven't really gotten really into this, but from what I can tell, anything that has to do with mobile phones is strictly worse in terms of anonimity than Wifi.
At least anything that ties a Wifi connection to you you can change in an OS setting, but if you get into faking IMEI/SIM stuff, that can very quickly get you charged with an actual crime.
One of the things about mobile phone is that there's 2 OSes on the phone: one that handles the phone stuff and the other that you as a user actually interact with. What's really scary about it is that the OS we do not think about is often licensed to the manufacturer (Apple, Samsung, etc.), closed-source, and dated. There's a lot of security firms that sell fake cell towers to law enforcement and governments and that gives them access to all of the data you send over the network. In older formats like 2G there was no encryption on anything sent over the wire.
If you dig into the fake cell tower rabbit hole you'll find what you're talking about to be an even worse problem.
I think the argument is more the IMEI side of things, one might naively expect that they could could simply change SIM and that would change ones identity and that anything like an IMEI could be easily configurable like how a MAC is on a NIC is.
This seems largely to be about security by obscurity.
I was hoping it was gonna be about how our modern practices are making things less secure.
For instance, we claim we need to be able to rapidly update clients so that we can patch security vulnerabilities as they are discovered (often without involving the user at all). And there are a lot of companies that have an incentive to push this narrative because they have products which facilitate this whack-a-mole approach to security. But there's no reason to believe that new software is more secure than old software. Old software is just more likely to be known to be insecure. So anything written before it became trendy to update without your user's consent is more secure in at least one way because it is not configured to automatically update to whatever comes down the pipe from "the vendor".
> Interest in magnetic tape is only increasing further today because the cost of random access memory (Ram), a form of computer memory that doesn't rely on tape, is skyrocketing.
I am one of those emacs users who use it for everything, it always worries me the way that it could be attacked given (particularly email some how evaling some elisp or something), Despite how easy emacs would be to exploit in many aspects I do not know of any attack done on emacs either through supply chain attack (think about how many third party libs an emacs mode might use) or one of the many ways that it's no doubt possible to get emacs to eval something the user would not want it to do. If anyone knows why or if their has been an attack on emacs I would love to know.
I understand BBC may not have the technical background to critically assess this approach, but whoever using Eudora (I loved it in 2001 btw) for security should get their credentials removed via King's order. Security through "obsolescence" is no different from security through obscurity - therefore, it doesn't work. Somebody not bothering to look for holes in your software doesn't mean they don't exist - in the age of Claude - I am pretty sure I can destroy your legacy software in minutes.
> I understand BBC may not have the technical background...
This is BBC Future - the BBC's tech clickbait publisher - not BBC News.
BBC consists of the non-profit news bureau as well as at least a dozen for-profit clickbait and listicle publishers. BBC's ad-free mandate is only for the UK.
> in the age of Claude - I am pretty sure I can destroy your legacy software in minutes
Yep. One of our PortCos has unrestricted access to Anthropic and GPT models. With whitebox testing, it's trivial to identify vulns in legacy environments. With blackbox testing, it takes some effort but it doable with the right steering.
>But the team picked an older software version for their honeypot that had been upgraded multiple times since its release and, consequently, no hackers bothered to target it. "When we upgraded it to the new one, we had all the attacks we wanted," recalls Bishop. "I thought it was so amusing."
It feels like fantasy imo
While I could understand using outdated or not popular protocols, then software with known vulns seems crazy when we are living in world of automated, ai assisted scanners
Let's say your security budget is pretty limited, and you can say with relative certainly that there's only a 0.1% chance of any adversary bothering to check the known flaws in your FossileWare(tm) Firewall in a given month.
Vs.how many dire flaws have been found in big-name security products in the past year, that even a low-budget adversary is likely to be regularly checking for?
There is no human making a decision on what attacks to run. They check for literally everything. 20 years ago, I ran a web server and would run 'tail -f' on the access logs. I watched the bots sending requests for files that only exist on linux, then send requests for files that only exist on windows. They don't know what kind of system they are hacking. They just know that there is a computer at that IP address.
There are databases that list every known flaw in existence, and the bots just go down the list checking each and every one. And if it doesn't work, in 5 minutes, they'll run those same tests again as if something changed in those 5 minutes.
If you're getting scanned, keep logs, and have a decent way to tag log entries with the associated vuln, then you can start building statistical models of how exploit attempt probabilities fade as the target software gets older and lower market share. No, the 0.1% won't be exact. Yes, such data and models can be shared.
Paying $$$ for a LatestGreatest(tm) Firewall won't make you 100% secure either. And the $$$ might better be spent on an extra layer of swiss cheese, or better recovery capability.
Reminds me of a sci-fi book I read where the protagonist goes to hack a computer but can't because it's running a 50 year old alien OS that none of his tools will run on.
That said, as a strategy it will only protect you from bots going after low-hanging fruit, not someone targeting you specifically.
Money defines the targets, and usually people running old software are thought not to have money. It is a zero sum game that shifts with the technology and ends when technology no longer is the theater where people can make or steal money.
Eh not really true at all with my experience. 100 million dollar industries commonly have old crap running for years that becomes nearly untouchable as staff rotates and everyone gets afraid to touch it.
Another reason is that a lot of tech from 20 years ago uses the QNX microkernel operating system, which is highly secure even when the software is never patched.
The problem is that most of these older devices are about to be replaced and many manufacturers have switched to Linux for their operating system, which is cheaper and requires no licensing, but is also far more insecure. It needs almost constant patching to stay secure because the attack surface is so large.
I know someone who works at a data center for the IRS, and a lot of their records are all stored in PDFs. He thought it was disgusting when he first learned about it, but eventually realized that it's essentially resistant to a large scale data leak. Retrieval just isn't fast enough.
Presumably the pdfs are so large that moving any significant chunk of them over available bandwidth takes so much time, and requires so much storage, that doing so just isn't useful for the attacker.
While I don't know any of the relevant numbers involved, imagine a scenario where exfiltrating even 10% of the total data would take years. What would be the point?
A malicious actor can target individual records, but leaking a lot of them becomes impractical.
virus source code published in books and magazines. We don't have such threats anymore.
--
No, we do not need this anymore - today you have CAPTCHAs, convincing people to copy & paste PowerShell code on their machine and execute it with Admin privileges! :-D
Sadly, the real lesson we need to learn from Battlestar Galactica is not this. They weren't saved by old software, they were saved by not having critical systems on the network unnecessarily.
Our water and power utilities need to re-watch the pilot.
Why are our water and power utilities connected to the internet? Is it so that the employees controlling them can work from home? If so they are accepting too much risk relative to the benefit.
think about a grid responding to demand. The old model (like the titanic) had a remote manager phoning the site lead who phoned the control room who phoned the engineering room.
Remote management has clear benefits, don’t be obtuse. I totally agree on the security risks, but the benefits are obvious.
It makes sense to have them connected for a lot of reasons.
which reasons?
Remote monitoring is one of them presumably
But is it completely impossible to separate the monitoring and equipment/management?
Stupid example like the equipment itself is not networked, but you could watch via a webcam and/or get metrics via image recognition.
Maybe it's safer just to have somebody work the night shift.
No, and it's not strictly impossible to correctly build out a SOC / SIEM and ingest all the logs you want and pay for the right engineers to make sure it all works correctly. But damned if anyone manages to do a great job in this area. It's too complex and too expensive, so almost everyone settles for "best effort."
A lot of problems are easy conceptually, but we can't manage to tackle them.
To be fair, it was both. Those ships that got the recent upgrades also got trojans that Caprica Six put in the defense mainframe with Baltar's access codes.
The point is NOT older, rather decentralized!
From a efficiency perspective centralized anything seems better at first glance, however decentralized anything just seems more resilient in the long run like nature and/or our universe.
Makes it harder to catch you in a general sweep, but it probably makes it much easier if anyone decides to target you specifically. How good a tradeoff that is varies from person to person. And the difficulty of targeting a specific person is changing in an AI world. I wouldn't try to characterize it any farther than that, just to observe that it is changing. "Hey AI go individually research and target county clerks for getting direct access to their {specific system}" is much more practical today than it was 10 years ago, and will be even more so 10 years from now.
It reminds me of an interview with police detectives. They say that some criminals think that using weapons like swords and bows will make the police job harder because police is so used to deal with guns that they may have a blind spot.
The thing is: police actually have data on swords and bows, and the fact it is so unusual to commit crimes with these limit the search space and tend to make the investigation easier.
It's incredible how far back the surveillance state goes - GSM mobile phones have an IMEI number that's tied to the handset - and the SIM is tied to the subscriber, and your phone broadcasts imei to neighboring towers constantly.
I haven't really gotten really into this, but from what I can tell, anything that has to do with mobile phones is strictly worse in terms of anonimity than Wifi.
At least anything that ties a Wifi connection to you you can change in an OS setting, but if you get into faking IMEI/SIM stuff, that can very quickly get you charged with an actual crime.
One of the things about mobile phone is that there's 2 OSes on the phone: one that handles the phone stuff and the other that you as a user actually interact with. What's really scary about it is that the OS we do not think about is often licensed to the manufacturer (Apple, Samsung, etc.), closed-source, and dated. There's a lot of security firms that sell fake cell towers to law enforcement and governments and that gives them access to all of the data you send over the network. In older formats like 2G there was no encryption on anything sent over the wire.
If you dig into the fake cell tower rabbit hole you'll find what you're talking about to be an even worse problem.
I think this how modern CPUs work also.
Anything billed by the second has good (-enough) cost attribution built in.
> surveillance state
How do you suppose mobile phones are meant to work without subscriber info?
> I haven't really gotten really into this
Clearly.
I think the argument is more the IMEI side of things, one might naively expect that they could could simply change SIM and that would change ones identity and that anything like an IMEI could be easily configurable like how a MAC is on a NIC is.
This seems largely to be about security by obscurity.
I was hoping it was gonna be about how our modern practices are making things less secure.
For instance, we claim we need to be able to rapidly update clients so that we can patch security vulnerabilities as they are discovered (often without involving the user at all). And there are a lot of companies that have an incentive to push this narrative because they have products which facilitate this whack-a-mole approach to security. But there's no reason to believe that new software is more secure than old software. Old software is just more likely to be known to be insecure. So anything written before it became trendy to update without your user's consent is more secure in at least one way because it is not configured to automatically update to whatever comes down the pipe from "the vendor".
> Interest in magnetic tape is only increasing further today because the cost of random access memory (Ram), a form of computer memory that doesn't rely on tape, is skyrocketing.
hmm
Sign of quality journalism? LLM's probably would not have made such a misstake ..
I am one of those emacs users who use it for everything, it always worries me the way that it could be attacked given (particularly email some how evaling some elisp or something), Despite how easy emacs would be to exploit in many aspects I do not know of any attack done on emacs either through supply chain attack (think about how many third party libs an emacs mode might use) or one of the many ways that it's no doubt possible to get emacs to eval something the user would not want it to do. If anyone knows why or if their has been an attack on emacs I would love to know.
I understand BBC may not have the technical background to critically assess this approach, but whoever using Eudora (I loved it in 2001 btw) for security should get their credentials removed via King's order. Security through "obsolescence" is no different from security through obscurity - therefore, it doesn't work. Somebody not bothering to look for holes in your software doesn't mean they don't exist - in the age of Claude - I am pretty sure I can destroy your legacy software in minutes.
> I understand BBC may not have the technical background...
This is BBC Future - the BBC's tech clickbait publisher - not BBC News.
BBC consists of the non-profit news bureau as well as at least a dozen for-profit clickbait and listicle publishers. BBC's ad-free mandate is only for the UK.
> in the age of Claude - I am pretty sure I can destroy your legacy software in minutes
Yep. One of our PortCos has unrestricted access to Anthropic and GPT models. With whitebox testing, it's trivial to identify vulns in legacy environments. With blackbox testing, it takes some effort but it doable with the right steering.
wait until you find out that a large percentage of operating system developers use mutt, alpine or mail.
>But the team picked an older software version for their honeypot that had been upgraded multiple times since its release and, consequently, no hackers bothered to target it. "When we upgraded it to the new one, we had all the attacks we wanted," recalls Bishop. "I thought it was so amusing."
It feels like fantasy imo
While I could understand using outdated or not popular protocols, then software with known vulns seems crazy when we are living in world of automated, ai assisted scanners
Let's say your security budget is pretty limited, and you can say with relative certainly that there's only a 0.1% chance of any adversary bothering to check the known flaws in your FossileWare(tm) Firewall in a given month.
Vs.how many dire flaws have been found in big-name security products in the past year, that even a low-budget adversary is likely to be regularly checking for?
There is no human making a decision on what attacks to run. They check for literally everything. 20 years ago, I ran a web server and would run 'tail -f' on the access logs. I watched the bots sending requests for files that only exist on linux, then send requests for files that only exist on windows. They don't know what kind of system they are hacking. They just know that there is a computer at that IP address.
There are databases that list every known flaw in existence, and the bots just go down the list checking each and every one. And if it doesn't work, in 5 minutes, they'll run those same tests again as if something changed in those 5 minutes.
Yep, this kind of stuff is easily automated these days and bots don't get bored.
> a 0.1% chance of any adversary bothering to check the known flaws in your FossileWare(tm) Firewall in a given month.
How would you arrive at this percentage in your risk analysis?
If you're getting scanned, keep logs, and have a decent way to tag log entries with the associated vuln, then you can start building statistical models of how exploit attempt probabilities fade as the target software gets older and lower market share. No, the 0.1% won't be exact. Yes, such data and models can be shared.
Paying $$$ for a LatestGreatest(tm) Firewall won't make you 100% secure either. And the $$$ might better be spent on an extra layer of swiss cheese, or better recovery capability.
Reminds me of a sci-fi book I read where the protagonist goes to hack a computer but can't because it's running a 50 year old alien OS that none of his tools will run on.
That said, as a strategy it will only protect you from bots going after low-hanging fruit, not someone targeting you specifically.
Here is a good example where this is not so. Unfortunately, my datacenter runs this vulnerable VMS release on an emulated VAX.
https://www.theregister.com/on-prem/2018/02/06/ghost-in-the-...
Things that are plenty and cheap to replace are easy to hack because if you broke them, you could always try again...
Money defines the targets, and usually people running old software are thought not to have money. It is a zero sum game that shifts with the technology and ends when technology no longer is the theater where people can make or steal money.
Eh not really true at all with my experience. 100 million dollar industries commonly have old crap running for years that becomes nearly untouchable as staff rotates and everyone gets afraid to touch it.
Another reason is that a lot of tech from 20 years ago uses the QNX microkernel operating system, which is highly secure even when the software is never patched.
The problem is that most of these older devices are about to be replaced and many manufacturers have switched to Linux for their operating system, which is cheaper and requires no licensing, but is also far more insecure. It needs almost constant patching to stay secure because the attack surface is so large.
The Irish example seems good. Keeping the old tech in parallel rather than replacing
This is terrible advice and I feel like this article will be used by managers of outdated IT systems to defend not upgrading their tech stack.
There are hospitals out there still running on windows XP. Someone's going to skim this article and say "look, we're actually being prudent"!
I know someone who works at a data center for the IRS, and a lot of their records are all stored in PDFs. He thought it was disgusting when he first learned about it, but eventually realized that it's essentially resistant to a large scale data leak. Retrieval just isn't fast enough.
How is it resistant? Can't all the pdfs just be leaked?
Right, the entire PDF.
You can't just create a tiny csv of the records you want. You need to upload GBs of data which is easily noticed by any of the routers involved.
Although I mean just because somebody noticed data is being leaked they might not have the authority to stop that doge employee or whomever.
Presumably the pdfs are so large that moving any significant chunk of them over available bandwidth takes so much time, and requires so much storage, that doing so just isn't useful for the attacker.
While I don't know any of the relevant numbers involved, imagine a scenario where exfiltrating even 10% of the total data would take years. What would be the point?
A malicious actor can target individual records, but leaking a lot of them becomes impractical.
Daily reminder that connecting everything to the cloud was a terrible idea and should never have happened like it did.
MS-DOS: Over 45 years, and STILL NO remote holes in the default install!
Well, there is no protection against compromised floppy disk sent by snail mail.
I have also seen virus source code published in books and magazines. We don't have such threats anymore.
--
virus source code published in books and magazines. We don't have such threats anymore.
--
No, we do not need this anymore - today you have CAPTCHAs, convincing people to copy & paste PowerShell code on their machine and execute it with Admin privileges! :-D