Because SSH/OpenSSH depend on zlib, if you replace your current libz.so file with another version whoose API has a bit changed, your SSH server won't work any more.
Existing SSH connections, however, will continue working. It's just the opening of new connections by sshd that will break.
So just keep an SSH session open until your sure the upgrade went OK. No need to hand out your passwords over telnet.
I have a hard time believing that this wasn't done expressely to make it harder for alternate OSes to get to work properly.
Microsoft is keeping it hush-hush, but it's well known among CTOs that Redmond is simply running out of crappiness. Customers were already reeling from the general desktop and server usefulness of Win2K, and the spiffiness of WinXP is starting a major exodus to OS/2 among gamers (who are a traditional bellwether for the datacenter).
Enter ACPI. A weighty specification that you can beat a mugger to death with. Big, juicy, complex data structures. States and modes out the wazoo.
All implemented by heroin-addled BIOS writers working in perpetual darkness, in a basement in Taiwan. Mmmmmm....bugs....
ACPI is Ballmer's last hope to return Windows users to the level of crappiness they love and expect.
sk your dialup isp for a static IP account, and they might tell you about another company who will give it to you for $80/month.
That's to squeeze money out of businesses who absolutely need static IPs. At colocation facilities, who are already charging $100+/month, extra static IPs are only a couple of bucks a month each.
...it's illegal to record phone conversations without mutual consent.
Actually it depends on the jurisdiction. For intrastate calls in the U.S. it varies quite a bit. Interstate calls may be recorded (last I heard -- it may have changed). Other countries vary. Also remember that phone calls can be routed through multiple states and nations, even if the participants are in the same place, which makes them interjurisdiction calls.
Unless you really know the law and have perfect control over the telephone system that is used, the only safe assumption is that the call can be legally recorded.
Copyright is yours whether or not you file with the government.
Wrong, dangerously wrong! Without filing, you can only get injunctive relief to prevent them from further using the copyrighted material. If you file the material with the gov't, you can get compensatory damages, and IIRC if the infringement was willful you can get treble punitive damages. (Note: this is based on U.S. law.)
The cardinal rule of copyright is: register before publication. If you don't, you might as well sign away the copyrights for free. Injunctive relief isn't worth a pitcher of warm spit.
Get a good contract with them that says they have to either pay a reasonable fee, or stop using the product after 45 days. (Don't try to get some sort of ludicrous penalties as another poster suggested. Ludicrous fees are routinely thrown out by judges, fair fees are difficult to argue against.) If you need to keep some things trade secrets, make them indemnify you against any losses suffered as a result of their disclosure of the secrets. (That should put a damper on them wanting unrestricted access.)
Why do they want the sources anyway? Are they going to compile the machine-readable sources, or just do a code review for quality, commenting, coherence, good style, etc. The latter could be satisfied by giving their engineers supervised access to print outs of the sources.
And don't trust me, I'm just some random pseudonymous geek on a web forum. Definitely get the assistance of a good lawyer.
The question to the group was comparing WCS to Beowulf from a performance/functionality perspective, not from a licencing perspective.
Licensing is cost, and cost IS performance.
Every man-hour spent reading, auditing, and managing licenses is a man-hour that is not applied to real work (he says, posting to/. from his desk at work;-). Every hour the compute nodes sit idle while licensing is sorted out is a 4.17% performance hit for that day.
All those licenses cost money, which means fewer CPUs. If a compute node costs $400, and licensing is $100/node, you can afford 25% fewer nodes. This is indistinguishable from a free OS that has a 25% performance flaw.
Then there's risk. The software mafia aren't going to audit a Linux cluster, sapping administrative time, and perhaps cease-and-desisting it offline. Linux cluster admins are never going to go to jail because they threw another machine online for the hell of it. Linus Torvalds will never sue a Linux cluster operator into oblivion to make an example of them. These are all possibilities with a proprietary product, and all-too-likely with a notorious lawyer-pit like Microsoft.
The obvious solution is an old laptop. 486 and Pentium laptops can be had cheap, and often have built-in sound. Many models have car adapters available to run off 12V.
Various single-board computers would work too, but there's no way they can beat the price of a used laptop.
B-52s don't carry daisy cutters. A daisy cutter is dropped by a C-130 using the regular parachute cargo delivery mechanism.
I was sure I read that they could on some.mil web page, but I can't seem to find it now. The do have the weight capacity, but I guess they lack the appropriate mechanisms in the bomb bay to carry one. In any event, C-130s are only useful against very, very low tech enemies. They're sitting ducks for people with good missiles.
You don't carpet bomb a radar installation. A single HARM missile will take out a radar site.
True, for a classical radar site that has a single antenna/transceiver/signal processor.
But then you have to ask "Why do classical radar systems have single vulnerable sites?" The answer is that RF electronics used to be extremely expensive and rare. That has changed. The price has fallen through the floor, while the capabilities have flown through the ceiling. 2GHz frequency-agile radios can be had at any department store, and 50GHz stuff will be cheap in 10 years.
That means the enemy of the future won't have centralized, expensive radar installations. He'll have radars or decoys on a 300 meter grid across major cities, with plenty of hot spares sitting around in warehouses. So you either have to carpet bomb the city, or you have to precision zap each antenna as it goes active.
Another thing to keep in mind is that classical radars were one of the only things using their frequencies. When every building in the city is full of cheap, crappy GHz radios (which is a certainty), your HARMs have to deal with much more background noise, potentially at the same frequencies as the radars. Hitting targets and avoiding needless casualties might be much more difficult in 10 years.
Carpet bombing is done to take out large numbers of enemy soldiers dug into wide areas of the surrounding terrain. Carpet bombing is done to supplement artillery or used when artillery is not available.
Or against pervasive high-capability enemies, when you don't have the right kind of weapons to be precise.
So how does the mirror at the back end of the laser's resonance chamber hold up?
Exotic materials. You only need a few square centimeters, so it doesn't really matter how much it costs or weighs. E.g., monoisotopic platinum would not be out of the question. You can't afford either the weight or the cost to cover an entire missile like that, though.
Exotic cooling. A gas dynamic laser like the Coil already needs huge flow rates of (probably cold) oxygen to operate. Pipe it through the optics on its way to the combustor and you've got massive cooling. An ICBM cannot afford to carry 50000L of LOX just in case it gets lasered.
Focusing. The laser can use an expanded beam with lower power density, and focus it to a much smaller area on the target.
Disposability. Depending on the costs, it's OK if the laser destroys itself during its one and only shot. It's no different than guided missiles, which cost $100k minimum and only fire once.
So.... what they're saying is "We didn't see them coming, so we need bigger guns".
Did you even read the article? Did you even know there was an article? The laser does almost nothing compared to, say, a 20mm Vulcan cannon that fires 2500 rounds per minute, or a 105mm Howitzer. To say nothing of a B-52 group loaded with daisy cutters.
What the laser does is hit extremely specific targets. In asymmetric warfare--say, a guerrilla radar installation in the middle of a city you'd rather not carpet bomb--the laser lets you win with greatly reduced carnage. So instead of blasting a couple of city blocks to flinders, there's a loud bang that puts a hole through the radar antenna and breaks a few windows. Instead of carpet bombing a suspected missile installation, just plink them as they launch.
Remember that improvements in force projection almost never come by increasing the total amount of force applied, they come by concentrating the force into the smallest ever-smaller areas. It was true of the first iron sights on rifles, it was true of the first radar fuses on WWII missiles, and it is true of modern battle lasers.
It's no problem at all if a thin layer of paint burns away to reveal the reflective surface only when necessary.... There are effective defenses against this stuff, they just haven't been developed yet because the offense is still a baby.
Reflection can never defend against lasers. No matter how reflective you make the surface, a modest pulse can vaporize a little material from the surface, forming a vapor. The vapor is a strong absorber of light, which means it is efficiently heated by the laser. The vapor then heats the "reflective" surface by direct contact, vaporizing more material and keeping the process going. There's no room for improvement either: the power densities achievable with a modern pulsed laser can vaporize even materials like tungsten and diamond.
For example, signal integrity issues. I can say with complete assurance that Rambus is loaded with signal integrity issues. These issues get -very bad- as the clock frequency goes up.
Bad as in you have to be aware of dielectric losses in the PCB material. I remember seeing reflectometer graph of a Rambus system where the plateaus were noticeably sloping from dielectric loss.
Also Rambus is -not-, strictly speaking, a serial bus.
Serial != 1 bit. Serial == takes more than one clock cycle to transfer a word.
"Security through obscurity" refers to obscurity of the algorithm,...
The original derisive sense referred to the belief that if a security flaw was not publicized, it was harmless and could be ignored. Unfortunately that idea has mutated into a meme that if the details of a technology aren't publicized, it is full of lurking flaws and is therefore worthless.
My point was that all security comes from obscurity, and not just of digital cryptographic keys. Every enhancement to obscurity in any part of the system makes the system more secure.
The whole point is that a strong security system is still secure when everything except the key is known,...
1) True, but keeping the whole thing secret makes it even more secure. Various logistical issues make this expensive to do for practical ciphers, but it's true. 2) Not all security systems are based on symmetric numerical ciphers, and their obscurity equations can be significantly different. In particular, chemical security systems behave as a public key system where even looking at the "public key" (chemical structure) is extremely expensive, nevermind finding a "private key" (synthesis process) that goes with the "public key". Knowledge gained about numerical ciphers does not directly apply to other security methods.
Moreover, different security systems have different goals. Classical numerical ciphers are intended to provide the highest possible secrecy for numerical data. Anti-counterfeiting systems, on the other hand, are intended to raise the cost of counterfeiting high enough that many potential counterfeiters are stopped.
There's nothing keeping anyone from using a more powerful NMR spectrometer to isolate the material and reproduce it.
Nothing but time, skill, money, and luck. Security isn't about building in impregnable fortress, it's about building walls that stop a few people who would otherwise have just strolled right in. As long as a security measure produces more than it costs, it's a winner.
Security through obscurity reigns...
People, stop blindly applying this mantra to everything you come across! Much practical security comes directly from obscurity: passwords only known by a few people and protected against unintended disclosure, metal keys of unknown shapes, PINs that must be used in addition to account numbers, manufacturing processes that would-be attackers don't know how to duplicate, etc.
Evidently the government holds them responsible for them even if they get thrown out. Ie. if someone chokes on a piece of a toy they picked out the garbage, they are can still be sued.
No, there is no warranty for garbage. The real issue is that the toy had probably already been sold in stores. How do you distinguish a toy with a warranty from a toy picked out of the garbage? It's very hard, and the attempt might cause bad publicity. Destroying the toys protects against confusion, which can easily cost more than simple liability.
Absolutely right, and thanks for the correction. I got too focused on the oxidation + hydrogen liberation aspect and my mind's eye wasn't seeing the hydroxyl for some reason. Argh. I *know* that tossing sodium in water makes sodium hydroxide.
Here's another reaction: write equation in haste --> look like fool in public.
We would need higher frequency/lower wavelength light, not just short pulses of it.
You're right, for single photons. Sub-single-wavelength pulses are formed from many photons, of a variety of phases and colors. On average the peaks of the various photons add up to enormous values during the pulse.
If you've taken a class that discussed the Fourier transform, it's analogous to the impulse function, which is composed of all frequencies of sine waves. The sharper you want your impulse to be, the wider the range of frequencies you need to have in your pulse. These zeptosecond guys are using frequencies of light up to x-rays (!), which is how they get such short pulses.
Food is a minor point: you'll freeze before you have to worry about the plants not growing. So you *have* to have a reliable electricity supply.
Phase I: Fission reactors. Two or three fission reactors (for redundancy) can supply heat and electricity for a small town. U.S. Navy submarine reactors would be a likely choice.
Phase II: Polar solar ring. Put a series of photovoltaic arrays around one of the poles, connecting them with a network of AC power lines. Putting the city at the pole minimizes power line length. As a bonus, the poles are likeliest to have water, esp. the crater centered on the south pole.
Phase III: Fission reactors, fusion reactors, or orbiting solar collectors -- whichever is cheapest at the time -- to support heavy industry and larger populations.
And don't discount supply shipments from Earth for the first years of operation. Most supplies can withstand huge accelerations, and an electromagnetic launcher would have a very low marginal cost of operation.
If these things came back, that would be worth it. The learning of how to design spacecraft would be greatly advanced if something came back into Earth orbit, was retrieved by a shuttle, and brought back to a lab on Earth to be tested.
Return trips are very, very, very expensive, and you don't get much more information than you get from good laboratory work.
Maybe then, NASA could learn from their mistakes, and design something that actually works, all the time, as designed.
Better is the enemy of good, and what we have now is good enough to do anything we want to do. What is lacking is the will to do it.
How about pushing the focus of living on Mars, to that of living on a clean Earth?
Beaming energy down from solar collectors would go a long way towards having a clean Earth, and it would greatly cheapen access to space. You're also ignoring the fact that a little pollution (or even a lot) isn't the most serious risk to the Earth: asteroid impacts are.
Personally I favor the Moon over Mars. It has enough gravity that industrial operations aren't inconvenient, yet not so much that landing and launching are overly expensive. Lunar space elevators are also vastly easier than for even Mars; and Lunar rotavators are doable with known materials. The lack of an atmosphere means you don't have to put up with year-long dust storms. It's close enough to Earth that radiation exposure on the trip there isn't a serious problem, and the trip itself is doable by ordinary people. The major downside (might) be the lack of water.
Fusion powerplants of the future will almost certainly use a lithium-water heat exchanger as I stated (and you did not deny) and it's almost a certainty that THESE THINGS LEAK at times and such a leak will probably cause LOTS of hydrogen gas to form.
If it turns out to be a major safety problem they can use a three-stage heat exchanger, with a less-reactive metal or a noble gas as the middle stage.
Your obsession with hydrogen is misguided anyway. The reaction is
alkali metal + H2O --> alkali oxide + H2 + heat
The problem is that the hydrogen is liberated rapidly, making the system explode from overpressure, which causes an overpressure wave (explosion) and blows chunks of hot nasty stuff everywhere. Afterwards the hydrogen might burn a little in the air, or even cause a gas explosion, but that's minor compared to the explosion of the piping and pressure vessels. If the reaction liberated nitrogen it would be just as bad.
My point remains, and it's not paranoia, that there are unintended consequences that come with the use of major technologies like this.
There are very few, if any, unintended consequences in any industrial operation. The bad things you are talking about are predictable and manageable. The only unknown is the will to proprely address them.
Hydrogen gas explosions were serious steps in both Three Mile Island and Chernobyl accidents and have the potential to be one in a lithium cooled fusion reactor accident as well.
Yeah, but those were fission reactors, which have to be as strong as possible because fission ash is so dangerous. The high strength of their pressure vessels and pipes means that more pressure can build up, producing a more spectacular explosion. A fusion reactor is vastly less radioactive, and its systems can be made deliberately weak so that they explode at a lower pressure, causing less damage.
Such a public relations disaster doesn't occur at vegetable shortening of fertilizer plants even when they have explosions there (and they do) because those are "low-tech" kinds of plants in the public eye and they are more conditioned to think of them as "risky".
I suspect you strongly underestimate the political fallout from uncontrolled energy releases at chemical plants, and overestimate panic from 'nookyular radiation'.
But the very fact a fusion reactor group would pick lithium instead of sodium for the coolant shows that somewhere along the line, neutron absorption and tritium production is important...the "hidden agenda" that is neither paranoia on my part or even hinted at in the ABC article.
1) There is no "hidden agenda". The fusion reactors will probably burn a deuterium-tritium mixture. If they didn't breed tritium, you'd need accelerators or fission reactors to get the tritium. If they make their own tritium, you can feed them with deuterium from tap water and lithium, both of which are fairly plentiful and innocuous.
2) Tritium for boosting plutonium bombs? Big deal. Any nation that can afford plutonium and ICBMs can afford tritium no matter how it's made.
With talk of returning to the moon or landing on Mars, reusable rockets would give NASA a little breath of relief in the funding department...
In other words, the lucrative one-shot launch vehicles would be defunded. (Even the NASA shuttle requires so much refurbishment that it is practically one shot.) It is therefore a threat to the gov't gravy train for NASA project managers and prime contractors. Remember that these people habitually block X projects, and they're the ones who destroyed the DC-X.
...The Fast and the Furious? I mean they had everything else: NOS stickers, neon lights on the undercarriage, ad naseum. Why not a laser in the engine?
Re:The best way to convert people from Microsoft..
on
Borking Outlook Express
·
· Score: 5, Insightful
When did we become such elitists? When users are arbitrarily excluded and abused in the name of "free software,"...
RTF email. This isn't about free software, it's about punishing people for using a notoriously-broken email client that causes nothing but headaches and pollution of the infosphere.
I see nothing productive in this article or the attitude of its creator.
And I see nothing productive in the numerous flaws in Outlook's processing of attachments, flaws that Microsoft has known about for several versions and has declined to fix. And I'm specifically not just talking about free software interoperability: different versions of Outlook cannot properly parse each other's attachments.
Outlook internally is one of the most poorly-engineered systems ever created. Its security model is a complete crock of shit that has several times nearly brought down the Internet. Microsoft's "programmers" wrote the attachment parsing code several times, each time being different and broken. (Proof: certain attachments aren't shown to the user because that broken code doesn't properly parse them, but if the user does "File->SaveAttachment" they *can* be saved because the saving code *does* properly parse them.)
The point is not to force them to use it, or to punish those who don't. Where's the freedom in that?
You're free to bathe in a sewer if you want, but that doesn't mean people have to let you into their clubs.
Slashdot Mirror
So just keep an SSH session open until your sure the upgrade went OK. No need to hand out your passwords over telnet.
/me goes off to find a new needle for his irony meter.
Enter ACPI. A weighty specification that you can beat a mugger to death with. Big, juicy, complex data structures. States and modes out the wazoo.
All implemented by heroin-addled BIOS writers working in perpetual darkness, in a basement in Taiwan. Mmmmmm....bugs....
ACPI is Ballmer's last hope to return Windows users to the level of crappiness they love and expect.
Unless you really know the law and have perfect control over the telephone system that is used, the only safe assumption is that the call can be legally recorded.
The cardinal rule of copyright is: register before publication. If you don't, you might as well sign away the copyrights for free. Injunctive relief isn't worth a pitcher of warm spit.
Get a good contract with them that says they have to either pay a reasonable fee, or stop using the product after 45 days. (Don't try to get some sort of ludicrous penalties as another poster suggested. Ludicrous fees are routinely thrown out by judges, fair fees are difficult to argue against.) If you need to keep some things trade secrets, make them indemnify you against any losses suffered as a result of their disclosure of the secrets. (That should put a damper on them wanting unrestricted access.)
Why do they want the sources anyway? Are they going to compile the machine-readable sources, or just do a code review for quality, commenting, coherence, good style, etc. The latter could be satisfied by giving their engineers supervised access to print outs of the sources.
And don't trust me, I'm just some random pseudonymous geek on a web forum. Definitely get the assistance of a good lawyer.
Every man-hour spent reading, auditing, and managing licenses is a man-hour that is not applied to real work (he says, posting to /. from his desk at work ;-). Every hour the compute nodes sit idle while licensing is sorted out is a 4.17% performance hit for that day.
All those licenses cost money, which means fewer CPUs. If a compute node costs $400, and licensing is $100/node, you can afford 25% fewer nodes. This is indistinguishable from a free OS that has a 25% performance flaw.
Then there's risk. The software mafia aren't going to audit a Linux cluster, sapping administrative time, and perhaps cease-and-desisting it offline. Linux cluster admins are never going to go to jail because they threw another machine online for the hell of it. Linus Torvalds will never sue a Linux cluster operator into oblivion to make an example of them. These are all possibilities with a proprietary product, and all-too-likely with a notorious lawyer-pit like Microsoft.
Various single-board computers would work too, but there's no way they can beat the price of a used laptop.
But then you have to ask "Why do classical radar systems have single vulnerable sites?" The answer is that RF electronics used to be extremely expensive and rare. That has changed. The price has fallen through the floor, while the capabilities have flown through the ceiling. 2GHz frequency-agile radios can be had at any department store, and 50GHz stuff will be cheap in 10 years.
That means the enemy of the future won't have centralized, expensive radar installations. He'll have radars or decoys on a 300 meter grid across major cities, with plenty of hot spares sitting around in warehouses. So you either have to carpet bomb the city, or you have to precision zap each antenna as it goes active.
Another thing to keep in mind is that classical radars were one of the only things using their frequencies. When every building in the city is full of cheap, crappy GHz radios (which is a certainty), your HARMs have to deal with much more background noise, potentially at the same frequencies as the radars. Hitting targets and avoiding needless casualties might be much more difficult in 10 years.
Or against pervasive high-capability enemies, when you don't have the right kind of weapons to be precise.What the laser does is hit extremely specific targets. In asymmetric warfare--say, a guerrilla radar installation in the middle of a city you'd rather not carpet bomb--the laser lets you win with greatly reduced carnage. So instead of blasting a couple of city blocks to flinders, there's a loud bang that puts a hole through the radar antenna and breaks a few windows. Instead of carpet bombing a suspected missile installation, just plink them as they launch.
Remember that improvements in force projection almost never come by increasing the total amount of force applied, they come by concentrating the force into the smallest ever-smaller areas. It was true of the first iron sights on rifles, it was true of the first radar fuses on WWII missiles, and it is true of modern battle lasers.
My point was that all security comes from obscurity, and not just of digital cryptographic keys. Every enhancement to obscurity in any part of the system makes the system more secure.
1) True, but keeping the whole thing secret makes it even more secure. Various logistical issues make this expensive to do for practical ciphers, but it's true. 2) Not all security systems are based on symmetric numerical ciphers, and their obscurity equations can be significantly different. In particular, chemical security systems behave as a public key system where even looking at the "public key" (chemical structure) is extremely expensive, nevermind finding a "private key" (synthesis process) that goes with the "public key". Knowledge gained about numerical ciphers does not directly apply to other security methods.Moreover, different security systems have different goals. Classical numerical ciphers are intended to provide the highest possible secrecy for numerical data. Anti-counterfeiting systems, on the other hand, are intended to raise the cost of counterfeiting high enough that many potential counterfeiters are stopped.
Here's another reaction: write equation in haste --> look like fool in public.
If you've taken a class that discussed the Fourier transform, it's analogous to the impulse function, which is composed of all frequencies of sine waves. The sharper you want your impulse to be, the wider the range of frequencies you need to have in your pulse. These zeptosecond guys are using frequencies of light up to x-rays (!), which is how they get such short pulses.
Phase I: Fission reactors. Two or three fission reactors (for redundancy) can supply heat and electricity for a small town. U.S. Navy submarine reactors would be a likely choice.
Phase II: Polar solar ring. Put a series of photovoltaic arrays around one of the poles, connecting them with a network of AC power lines. Putting the city at the pole minimizes power line length. As a bonus, the poles are likeliest to have water, esp. the crater centered on the south pole.
Phase III: Fission reactors, fusion reactors, or orbiting solar collectors -- whichever is cheapest at the time -- to support heavy industry and larger populations.
And don't discount supply shipments from Earth for the first years of operation. Most supplies can withstand huge accelerations, and an electromagnetic launcher would have a very low marginal cost of operation.
Personally I favor the Moon over Mars. It has enough gravity that industrial operations aren't inconvenient, yet not so much that landing and launching are overly expensive. Lunar space elevators are also vastly easier than for even Mars; and Lunar rotavators are doable with known materials. The lack of an atmosphere means you don't have to put up with year-long dust storms. It's close enough to Earth that radiation exposure on the trip there isn't a serious problem, and the trip itself is doable by ordinary people. The major downside (might) be the lack of water.
Your obsession with hydrogen is misguided anyway. The reaction is
The problem is that the hydrogen is liberated rapidly, making the system explode from overpressure, which causes an overpressure wave (explosion) and blows chunks of hot nasty stuff everywhere. Afterwards the hydrogen might burn a little in the air, or even cause a gas explosion, but that's minor compared to the explosion of the piping and pressure vessels. If the reaction liberated nitrogen it would be just as bad. There are very few, if any, unintended consequences in any industrial operation. The bad things you are talking about are predictable and manageable. The only unknown is the will to proprely address them. Yeah, but those were fission reactors, which have to be as strong as possible because fission ash is so dangerous. The high strength of their pressure vessels and pipes means that more pressure can build up, producing a more spectacular explosion. A fusion reactor is vastly less radioactive, and its systems can be made deliberately weak so that they explode at a lower pressure, causing less damage. I suspect you strongly underestimate the political fallout from uncontrolled energy releases at chemical plants, and overestimate panic from 'nookyular radiation'. 1) There is no "hidden agenda". The fusion reactors will probably burn a deuterium-tritium mixture. If they didn't breed tritium, you'd need accelerators or fission reactors to get the tritium. If they make their own tritium, you can feed them with deuterium from tap water and lithium, both of which are fairly plentiful and innocuous.2) Tritium for boosting plutonium bombs? Big deal. Any nation that can afford plutonium and ICBMs can afford tritium no matter how it's made.
In other words, the lucrative one-shot launch vehicles would be defunded. (Even the NASA shuttle requires so much refurbishment that it is practically one shot.) It is therefore a threat to the gov't gravy train for NASA project managers and prime contractors. Remember that these people habitually block X projects, and they're the ones who destroyed the DC-X.
...The Fast and the Furious? I mean they had everything else: NOS stickers, neon lights on the undercarriage, ad naseum. Why not a laser in the engine?
Outlook internally is one of the most poorly-engineered systems ever created. Its security model is a complete crock of shit that has several times nearly brought down the Internet. Microsoft's "programmers" wrote the attachment parsing code several times, each time being different and broken. (Proof: certain attachments aren't shown to the user because that broken code doesn't properly parse them, but if the user does "File->SaveAttachment" they *can* be saved because the saving code *does* properly parse them.)
You're free to bathe in a sewer if you want, but that doesn't mean people have to let you into their clubs.