I replied earlier on this topic, but I'll just add that IIRC in related, legitimate research, the laser is used to drive a particle accelerator to provide the necessary slow neutrons to initiate the reaction. So the laser doesn't itself have to cause decay - it's just the gas pedal (to use a truly awful analogy.)
I haven't read TFA yet, but there is a legitimate power-station-scale solid-phase Thorium reactor research project under way in India that uses lasers to assist in initiating a reaction in the reactor. I don't recall the details, but essentially the laser resolves the problem that Thorium does not self-start a chain reaction, as it is not initially a good source of slow neutrons. So a seed of Uranium or something similar must be used to get the thing started. Once it starts running, it produces its own slow neutrons, and so can be sustained without any more Uranium. This characteristic of Thorium reactors is an important safety factor - you can literally turn it off at night if you want, and start it up the next morning. The researchers at AEC in the early 1970s did exactly that. (Google LTFR or 'Liquid Thorium Fluoride Reactor' - there's lots of information online)
So the Indian research appears to show a way to initiate the reaction using lasers. IIRC the laser is used to drive a small particle accelerator, which provides the necessary neutrons.
I don't recall the details, but the mining impact is something like 1/400 as much as uranium, for the same energy production. And processing is both much simpler and much, much less environmentally destructive. And the radioactive waste is very tiny, and does not include significant amounts of bomb-making materials. In fact that's why the AEC stopped looking at Thorium reactors back in the early 1970s - Thorium reactors did not produce 'useful' stuff like Plutonium.
Somewhere on the coast of India there is essentially a long beach made of Thorium-containing sands.
Indeed. Analogy: I mostly drive a stickshift, and when I don't I drive a 4WD F250 set up for trailer hauling. Both are small-percentage niches, and most folks find driving either one either impossible or intimidating. Some folks drive muscle cars, some folks drive Neons. I don't see BMW going out of business because it only has, like, 1% of the auto market. (I don't drive a beemer either - too much $$ for anything I want.) And I used to spend my time doing all the fixing and stuff, but nowadays I let the garage guys do that - it's cheaper for me, considering my time.
Having said that, I think it would be interesting to see if one could build a linux-based laptop that is equivalently shiny, smooth, and fast-and-easy-to-use as an exotic car. That would answer the 'desktop niche' question with, "This is my very niche-y desktop - it runs circles around your MS-Windows Volkswagen." I don't think it can be done (yet).
The first thing that would have to be done would be to unify the GUI (menus and mouse usage) and make inter-communication between apps seamless for every application that is installed - both are a lot better than they used to be, but menu items still often show up in random locations, depending on whether the developer came from Unix (Linux vs BSD), MS Windows or Apple GUI world. Let's see - where are preferences... errr. options are in this app? Do I use ^F, cmd-F, alt-F or ^f to find something in this document/mail/file/??? Ideally the one-true-glue-layer would allow all those things to be left out of the developer's consciousness, and handled entirely by every GUI front end. The developer could just expose the 'Find in document' and 'Find among documents' API calls, and the user could pick whatever GUI layout they prefer. IMHO NeXTStep did this the best of anything I've used - better than OSX IMHO. _Every_ application's menus worked the same.
Diamond is not a good idea in this application. Carbon burns in the presence of Oxygen with an ignition temperature of (IIRC) about 500C (ah, here's a ref: 870-1170K - I guess that's 600C+). If kept away from anything to combine with it, diamond can handle much higher temperatures, but that doesn't apply here.
A case in point, told me by a relative who worked on the Space Shuttle's navigation computer systems for IBM, back in the 1970s. It seems that in the late 1960s some Navy admiral did a favor for a buddy, and authorized a contract for his company to build a bunch of small computers for some project. They didn't work for the project, and weren't used. A large number of them (all that the Navy had ordered) were put on a shelf somewhere. Later, this admiral moved over to NASA, and in order to save his own face he ordered NASA to use these computers to be used for the nav system, thus turning a pig's ear into apparent gold and making the Navy bean counters happy. Of course, these were 1960s technology and hugely underpowered - IIRC they were 4-bit systems with a cycle time of a few kilohertz. These are the much-vaunted triple-redundant navigation computers that were originally used on the Shuttle. At the time of first launch, the calculators carried by the astronauts were more powerful than the nav computers.
Another computer with a different architecture was also designed and built by another company (Raytheon) and programmed in a different language. This was the fourth nav computer, that checked up on the first three.
Yes. IIRC the fuel pumps for the Space Shuttle push a swimming pool of fuel every second (my mind is saying 250,000 gallons per second, but that may be wrong). And the pumps have to be _very_ smooth, as any variation in the flow or pressure is amplified by the burn, causing extreme oscillations in the vehicle. (There's video somewhere on the net of someone using a propane burner as an audio amplifier - just modulate the propane flow.) As I recall, when they were building the Apollo system, they looked all over the world and did a lot of research, for a pump that had the necessary characteristics. They finally found what they needed in the pumps used for firehoses since the late 1800s!:D
True. I have seen articles on astronomy discussing sound waves in the interstellar medium, and the speed of sound in the gas around some astronomical object or another. In fact I won a small debate at work over this.:)
It's hard to say/remember how rigid 'rigid' meant at the time - at least not squishy like a foam mattress, but perhaps not rock-hard rigid. Happy medium and all that.
Or go the other way - toward Church instead of Turing - check out Erlang or Haskell or Ocaml. In fact I recommend learning at least one of the above (I personally like Erlang, but that's just me) just to get a different perspective on computation than any of the 'classic' imperative, memory-location-oriented languages.
A while back (15 years? 25 years?) when the Pinto was still a car, Ford or somebody did an experiment (IIRC it was in Popular Mechanics or some such). First they took a new Pinto and a new Fairlane and crashed them together. The Pinto was, of course, a pancake along with anyone who would have been in it. Then they took two more but filled various body cavities in the Pinto with rigid urethane foam. This time, the Pinto broke even with the Fairlane - nobody in either car would have died.
So just basic methods _can_ have a very good effect.
I was in the Verizon store a couple of days ago, and the Verizon guy told me that, since I have a smartphone with unlimited now, as long as I continue to keep a smartphone my unlimited plan remains. They aren't canceling old plans. But if I go to a dumb phone (even for a month), then I will lose my unlimited plan and won't get it back. I have not switched phones yet, so I don't have any proof, but if you're thinking of switching phones, it's worth asking them.
I've never messed with an iPhone or iPad, so I don't know but I'm astonished about what you say - it's counter to the Macintosh model. That very unification of the user experience was one of the primary hallmarks of the Macintosh computer. It was somewhat difficult for developers but enforced with vigor by Steve Jobs and the rest of the Mac crew. Apple did a lot of research to determine the best ergonomics and UI experience, before the finalized the Mac platform. On the NeXT they went a step further and made it dirt simple to follow the unified model, and difficult not to. As a result, regardless of application, the user could depend on the fact that Apple-H would hide the app, Apple-Q would quit, etc., and commonly used menu items would always be in the same menu locations, Assuming you're correct (which I do), all I can say is "How the mighty have fallen", etc.
Real admins rm -rf / first thing in the morning, just for the exercise - then see if they can rebuild from cold iron in under 15 minutes. Then, done with calisthenics, it's off to work!:D
I'll just add that the coffee pot on a 747 costs (IIRC) $4000 - airplane equipment is just expensive, due to (as the parent noted) low volumes, expensive development and expensive parts. Just about everything electrical on an aircraft has to pass both FAA and FCC, and every time you change a resistor the whole thing has to go through certification again, at cost exceeding $1 million - each. So, amortize that cost over perhaps 500 planes, plus spares, parts, etc. and you are talking about perhaps $1000 per coffeepot just to get federal approval. As the parent noted, for a military application it all gets more expensive. In the one instance I was involved in that, it took the company I worked for over a year, and several engineering person-years, to get through Tempest qualification.
If you think military stuff is expensive, check out medical supplies. In my own experience (a long time ago), a piece of vinyl tubing used in the blood pump for kidney dialysis machines, that has to be replaced for each patient, retailed for $150. This was exactly the same tubing you can buy at the HW store for $1 per foot today. It was sterilized, inspected and packaged by the machine maker. Of course again it had to pass FDA, and no hospital was going to risk a liability suit by buying from any other source, and perhaps 1/4 to 1/3 of the wholesale price of that part was liability insurance carried by the machine maker. So the entire structure of the medical industry has created a cost nightmare. If hospitals could acquire that part on the open market, or (perhaps better) just sterilize their own without risk of liability suits, the cost of that one part might be under $10. (needless to say, this is a summary and doesn't cover all the salient points.)
IANA EE, but... include an additional circuit that switches randomly, imposing a random element on the current flow - if you have some gate space left over from doing the real work.
Back in the late 1700s, the technology behind the textile industry (spinning, looms) was a British state secret. Nobody who had been trained in the technology was allowed to leave Britain. Samuel Slater dressed as a girl, sailed to America, and replicated the British technology. That was a big part of the beginning of the American Industrial Revolution, and the beginning of the end of the British monopoly on cheap textiles.
Some of the mills built in the early 1800s in New England still stand. Of course, the textile industry has moved on - the New England mills started going out of business throughout the 20th century. But in the meantime that industry was a big part of the American economy, politics (directly affecting the reason for and course of the Civil War, for example).
One wonders just what the US would be like had Slater not stolen the British technology.
Tyrannosaurus apparently has the bone structure of a big chicken. So I'm thinking it made VERY BIG clucking and crowing noises!:P "COCK-A-DOODLE-DO, MF!!"
(old joke - Where does an 800 lb. canary sit? Any place it wants too!)
I've always felt there were really three types - not just Nietzche's purported (I've never read it) "Master" and "Slave" but also a third type - "Creative". The creatives just want to create things - software, art, buildings, spaceships - and want to be neither master nor slave. Of the three types, creatives are probably the rarest.
But I'd also like to see these ideas explored from the perspective of hunters vs. farmers (ref.Thom Hartmann's books on ADD). Perhaps the master and slave are just two parts of the farmer phenotype, and this dichotomy does not apply to the hunter phenotype.
Funny, I was just replying (and referring to your comment), and mentioned Senge's "The Fifth Discipline" - and then I noticed you referred to an 'archetypical organizational systems dynamics problem', which is the core topic of the book!:D
I know a couple of friends-of-friends, who have this kind of problem. In all three cases their own beliefs that the system is corrupt, that everyone is out to get them, etc. makes them cynical and uncooperative, and prevents them from ever getting past minimum wage jobs. In all three cases I'd never hire them, because they act out their beliefs. When I visit my friends and these folks are there, they spend the evening ranting about how screwed up the system is, and how nobody appreciates their 'hard work'. Well for them, that's probably true. But if they worked for me I sure wouldn't promote them to anything where I had to depend on them. My friends have them over regularly in hopes of turning them around but so far (several years) nothing has changed. They're toxic, and will remain so.
The other reply (below) is correct. What you describe has NEVER happened to me. And, if you read the most successful and influential business books, the entire thrust of all professional business management is the opposite. I suggest "The Fifth Discipline" by Peter Senge. This book is one of the few books that has been on the best seller list TWICE, 10 or 15 years apart - if your beliefs were true, that book wouldn't have sold 10 copies. But having done it myself, I also recognize that running a business is a lot like being a single mom - there's never enough money to do everything that MUST be done, and not enough hours in the day.
So no, I'm not delusional. In fact, there is an aphorism regarding 'electrical engineers' (and by extension computer engineers hard and soft, since this aphorism long predates computers) that EEs have been working themselves out of a job since 1900, but every time they invent things to make themselves obsolete, the increased productivity of the new systems results in higher usage of electrical/electronic/comuting devices and in a requirement for more engineers. I'm not saying this is true of all jobs or all companies, but it's been true enough for long enough.
I'm not arguing that this all applies in all cases. But I have noticed that we tend to experience what we expect, and what we act out - if you think and act like the 'bosses' are assholes, well people notice that, and they will tend to be assholes - to you.
I've gotten in trouble at work before. My thing hasn't been the troubles you've noted. I'm hard to manage, arrogant, too smart for my own good sometimes, and I don't back down. If I think it should be done this way, I'll say so. This caused a rough patch at my present company but it all got worked out - at their initiative because they wanted me to stay. Now I'm more 'essential' in different ways, none of which are related to knowing where the secret software is hidden!
When I was at Schlumberger a while back, I was fortunate in that my manager was very aware that for our group, 'no news was good news' - he knew that when we were doing our job, nothing would happen. And he kept his manager aware of that for a long time. Later our group became visible due to our success, and got 'adopted' by an idiot VP. I was the last one to leave. I gave them eight months notice, and they were still unable to find a replacement until after I was gone. He couldn't figure anything out. A couple of years later the VP and the group were all gone.
Slashdot Mirror
I replied earlier on this topic, but I'll just add that IIRC in related, legitimate research, the laser is used to drive a particle accelerator to provide the necessary slow neutrons to initiate the reaction. So the laser doesn't itself have to cause decay - it's just the gas pedal (to use a truly awful analogy.)
I haven't read TFA yet, but there is a legitimate power-station-scale solid-phase Thorium reactor research project under way in India that uses lasers to assist in initiating a reaction in the reactor. I don't recall the details, but essentially the laser resolves the problem that Thorium does not self-start a chain reaction, as it is not initially a good source of slow neutrons. So a seed of Uranium or something similar must be used to get the thing started. Once it starts running, it produces its own slow neutrons, and so can be sustained without any more Uranium. This characteristic of Thorium reactors is an important safety factor - you can literally turn it off at night if you want, and start it up the next morning. The researchers at AEC in the early 1970s did exactly that. (Google LTFR or 'Liquid Thorium Fluoride Reactor' - there's lots of information online)
So the Indian research appears to show a way to initiate the reaction using lasers. IIRC the laser is used to drive a small particle accelerator, which provides the necessary neutrons.
I don't recall the details, but the mining impact is something like 1/400 as much as uranium, for the same energy production. And processing is both much simpler and much, much less environmentally destructive. And the radioactive waste is very tiny, and does not include significant amounts of bomb-making materials. In fact that's why the AEC stopped looking at Thorium reactors back in the early 1970s - Thorium reactors did not produce 'useful' stuff like Plutonium.
Somewhere on the coast of India there is essentially a long beach made of Thorium-containing sands.
Indeed. Analogy: I mostly drive a stickshift, and when I don't I drive a 4WD F250 set up for trailer hauling. Both are small-percentage niches, and most folks find driving either one either impossible or intimidating. Some folks drive muscle cars, some folks drive Neons. I don't see BMW going out of business because it only has, like, 1% of the auto market. (I don't drive a beemer either - too much $$ for anything I want.) And I used to spend my time doing all the fixing and stuff, but nowadays I let the garage guys do that - it's cheaper for me, considering my time.
Having said that, I think it would be interesting to see if one could build a linux-based laptop that is equivalently shiny, smooth, and fast-and-easy-to-use as an exotic car. That would answer the 'desktop niche' question with, "This is my very niche-y desktop - it runs circles around your MS-Windows Volkswagen." I don't think it can be done (yet).
The first thing that would have to be done would be to unify the GUI (menus and mouse usage) and make inter-communication between apps seamless for every application that is installed - both are a lot better than they used to be, but menu items still often show up in random locations, depending on whether the developer came from Unix (Linux vs BSD), MS Windows or Apple GUI world. Let's see - where are preferences ... errr. options are in this app? Do I use ^F, cmd-F, alt-F or ^f to find something in this document/mail/file/??? Ideally the one-true-glue-layer would allow all those things to be left out of the developer's consciousness, and handled entirely by every GUI front end. The developer could just expose the 'Find in document' and 'Find among documents' API calls, and the user could pick whatever GUI layout they prefer. IMHO NeXTStep did this the best of anything I've used - better than OSX IMHO. _Every_ application's menus worked the same.
Diamond is not a good idea in this application. Carbon burns in the presence of Oxygen with an ignition temperature of (IIRC) about 500C (ah, here's a ref: 870-1170K - I guess that's 600C+). If kept away from anything to combine with it, diamond can handle much higher temperatures, but that doesn't apply here.
A case in point, told me by a relative who worked on the Space Shuttle's navigation computer systems for IBM, back in the 1970s. It seems that in the late 1960s some Navy admiral did a favor for a buddy, and authorized a contract for his company to build a bunch of small computers for some project. They didn't work for the project, and weren't used. A large number of them (all that the Navy had ordered) were put on a shelf somewhere. Later, this admiral moved over to NASA, and in order to save his own face he ordered NASA to use these computers to be used for the nav system, thus turning a pig's ear into apparent gold and making the Navy bean counters happy. Of course, these were 1960s technology and hugely underpowered - IIRC they were 4-bit systems with a cycle time of a few kilohertz. These are the much-vaunted triple-redundant navigation computers that were originally used on the Shuttle. At the time of first launch, the calculators carried by the astronauts were more powerful than the nav computers.
Another computer with a different architecture was also designed and built by another company (Raytheon) and programmed in a different language. This was the fourth nav computer, that checked up on the first three.
I wonder if any of this was ever replaced?
Yes. IIRC the fuel pumps for the Space Shuttle push a swimming pool of fuel every second (my mind is saying 250,000 gallons per second, but that may be wrong). And the pumps have to be _very_ smooth, as any variation in the flow or pressure is amplified by the burn, causing extreme oscillations in the vehicle. (There's video somewhere on the net of someone using a propane burner as an audio amplifier - just modulate the propane flow.) As I recall, when they were building the Apollo system, they looked all over the world and did a lot of research, for a pump that had the necessary characteristics. They finally found what they needed in the pumps used for firehoses since the late 1800s! :D
True. I have seen articles on astronomy discussing sound waves in the interstellar medium, and the speed of sound in the gas around some astronomical object or another. In fact I won a small debate at work over this. :)
It's hard to say/remember how rigid 'rigid' meant at the time - at least not squishy like a foam mattress, but perhaps not rock-hard rigid. Happy medium and all that.
Or go the other way - toward Church instead of Turing - check out Erlang or Haskell or Ocaml. In fact I recommend learning at least one of the above (I personally like Erlang, but that's just me) just to get a different perspective on computation than any of the 'classic' imperative, memory-location-oriented languages.
Oblig.: A Bunch of Rocks. :D
A while back (15 years? 25 years?) when the Pinto was still a car, Ford or somebody did an experiment (IIRC it was in Popular Mechanics or some such). First they took a new Pinto and a new Fairlane and crashed them together. The Pinto was, of course, a pancake along with anyone who would have been in it. Then they took two more but filled various body cavities in the Pinto with rigid urethane foam. This time, the Pinto broke even with the Fairlane - nobody in either car would have died.
So just basic methods _can_ have a very good effect.
I was in the Verizon store a couple of days ago, and the Verizon guy told me that, since I have a smartphone with unlimited now, as long as I continue to keep a smartphone my unlimited plan remains. They aren't canceling old plans. But if I go to a dumb phone (even for a month), then I will lose my unlimited plan and won't get it back. I have not switched phones yet, so I don't have any proof, but if you're thinking of switching phones, it's worth asking them.
I've never messed with an iPhone or iPad, so I don't know but I'm astonished about what you say - it's counter to the Macintosh model. That very unification of the user experience was one of the primary hallmarks of the Macintosh computer. It was somewhat difficult for developers but enforced with vigor by Steve Jobs and the rest of the Mac crew. Apple did a lot of research to determine the best ergonomics and UI experience, before the finalized the Mac platform. On the NeXT they went a step further and made it dirt simple to follow the unified model, and difficult not to. As a result, regardless of application, the user could depend on the fact that Apple-H would hide the app, Apple-Q would quit, etc., and commonly used menu items would always be in the same menu locations, Assuming you're correct (which I do), all I can say is "How the mighty have fallen", etc.
Perhaps we could update Wikipedia to reference the blog just to confirm it to be the case.
Win!! :D
Ingenuity triumphs! :D
Real admins rm -rf / first thing in the morning, just for the exercise - then see if they can rebuild from cold iron in under 15 minutes. Then, done with calisthenics, it's off to work! :D
I'll just add that the coffee pot on a 747 costs (IIRC) $4000 - airplane equipment is just expensive, due to (as the parent noted) low volumes, expensive development and expensive parts. Just about everything electrical on an aircraft has to pass both FAA and FCC, and every time you change a resistor the whole thing has to go through certification again, at cost exceeding $1 million - each. So, amortize that cost over perhaps 500 planes, plus spares, parts, etc. and you are talking about perhaps $1000 per coffeepot just to get federal approval. As the parent noted, for a military application it all gets more expensive. In the one instance I was involved in that, it took the company I worked for over a year, and several engineering person-years, to get through Tempest qualification.
If you think military stuff is expensive, check out medical supplies. In my own experience (a long time ago), a piece of vinyl tubing used in the blood pump for kidney dialysis machines, that has to be replaced for each patient, retailed for $150. This was exactly the same tubing you can buy at the HW store for $1 per foot today. It was sterilized, inspected and packaged by the machine maker. Of course again it had to pass FDA, and no hospital was going to risk a liability suit by buying from any other source, and perhaps 1/4 to 1/3 of the wholesale price of that part was liability insurance carried by the machine maker. So the entire structure of the medical industry has created a cost nightmare. If hospitals could acquire that part on the open market, or (perhaps better) just sterilize their own without risk of liability suits, the cost of that one part might be under $10. (needless to say, this is a summary and doesn't cover all the salient points.)
IANA EE, but ... include an additional circuit that switches randomly, imposing a random element on the current flow - if you have some gate space left over from doing the real work.
Back in the late 1700s, the technology behind the textile industry (spinning, looms) was a British state secret. Nobody who had been trained in the technology was allowed to leave Britain. Samuel Slater dressed as a girl, sailed to America, and replicated the British technology. That was a big part of the beginning of the American Industrial Revolution, and the beginning of the end of the British monopoly on cheap textiles.
Some of the mills built in the early 1800s in New England still stand. Of course, the textile industry has moved on - the New England mills started going out of business throughout the 20th century. But in the meantime that industry was a big part of the American economy, politics (directly affecting the reason for and course of the Civil War, for example).
One wonders just what the US would be like had Slater not stolen the British technology.
Tyrannosaurus apparently has the bone structure of a big chicken. So I'm thinking it made VERY BIG clucking and crowing noises! :P
"COCK-A-DOODLE-DO, MF!!"
(old joke - Where does an 800 lb. canary sit? Any place it wants too!)
I've always felt there were really three types - not just Nietzche's purported (I've never read it) "Master" and "Slave" but also a third type - "Creative". The creatives just want to create things - software, art, buildings, spaceships - and want to be neither master nor slave. Of the three types, creatives are probably the rarest.
But I'd also like to see these ideas explored from the perspective of hunters vs. farmers (ref.Thom Hartmann's books on ADD). Perhaps the master and slave are just two parts of the farmer phenotype, and this dichotomy does not apply to the hunter phenotype.
Funny, I was just replying (and referring to your comment), and mentioned Senge's "The Fifth Discipline" - and then I noticed you referred to an 'archetypical organizational systems dynamics problem', which is the core topic of the book! :D
I know a couple of friends-of-friends, who have this kind of problem. In all three cases their own beliefs that the system is corrupt, that everyone is out to get them, etc. makes them cynical and uncooperative, and prevents them from ever getting past minimum wage jobs. In all three cases I'd never hire them, because they act out their beliefs. When I visit my friends and these folks are there, they spend the evening ranting about how screwed up the system is, and how nobody appreciates their 'hard work'. Well for them, that's probably true. But if they worked for me I sure wouldn't promote them to anything where I had to depend on them. My friends have them over regularly in hopes of turning them around but so far (several years) nothing has changed. They're toxic, and will remain so.
The other reply (below) is correct. What you describe has NEVER happened to me. And, if you read the most successful and influential business books, the entire thrust of all professional business management is the opposite. I suggest "The Fifth Discipline" by Peter Senge. This book is one of the few books that has been on the best seller list TWICE, 10 or 15 years apart - if your beliefs were true, that book wouldn't have sold 10 copies. But having done it myself, I also recognize that running a business is a lot like being a single mom - there's never enough money to do everything that MUST be done, and not enough hours in the day.
So no, I'm not delusional. In fact, there is an aphorism regarding 'electrical engineers' (and by extension computer engineers hard and soft, since this aphorism long predates computers) that EEs have been working themselves out of a job since 1900, but every time they invent things to make themselves obsolete, the increased productivity of the new systems results in higher usage of electrical/electronic/comuting devices and in a requirement for more engineers. I'm not saying this is true of all jobs or all companies, but it's been true enough for long enough.
I'm not arguing that this all applies in all cases. But I have noticed that we tend to experience what we expect, and what we act out - if you think and act like the 'bosses' are assholes, well people notice that, and they will tend to be assholes - to you.
I've gotten in trouble at work before. My thing hasn't been the troubles you've noted. I'm hard to manage, arrogant, too smart for my own good sometimes, and I don't back down. If I think it should be done this way, I'll say so. This caused a rough patch at my present company but it all got worked out - at their initiative because they wanted me to stay. Now I'm more 'essential' in different ways, none of which are related to knowing where the secret software is hidden!
When I was at Schlumberger a while back, I was fortunate in that my manager was very aware that for our group, 'no news was good news' - he knew that when we were doing our job, nothing would happen. And he kept his manager aware of that for a long time. Later our group became visible due to our success, and got 'adopted' by an idiot VP. I was the last one to leave. I gave them eight months notice, and they were still unable to find a replacement until after I was gone. He couldn't figure anything out. A couple of years later the VP and the group were all gone.