Another point is that they overheard another researcher who disagreed with their assumption, he measured abnormal amounts of radon emissions and was concerned enough to evacuate his own family the day prior to the quake. But his data was not taken into consideration, despite their own model being a guess at best.
All models are guesses at best. For any disaster there will ALWAYS be somebody who saw it coming. The problem is that on every day there is somebody who sees a disaster coming that never arrives.
Society loves scapegoats. When somebody tells you anything about the future, you can be certain that they aren't.
Yes it does and it should (in this case), because we don't really know much about the process that causes earthquakes. That is the whole point really, these scientists went on record saying that nothing would happen, yet it did, to the tune of 307 deaths. Had they not abused their position of authority, and instead told the truth: "we don't know", this would have been a non story.
The problem with that is science doesn't "know" anything at all. We don't "know" if the Sun will explode this afternoon. It seems awfully unlikely, but the fact is that we don't have that many data points where our sun exploding is concerned, and even if we did we couldn't be certain. The best we can do is extrapolate from some observations hundreds of light years away or more.
Would things have been any better if the scientists said that they considered an earthquake extremely unlikely, just as I will go on the record and say that it is extremely unlikely that the sun will blow up before dinner?
Why would you add on the refrigerator as a new zone on existing HVAC? I thought the intention was to vent the heat outside, not into the current HVAC system (since if you want to do that, you can just keep it as is, and let the head convect naturally from the refrigerator into the room).
The suggestion was a refrigerator with a compressor that cycled freon through the fridge on the high-pressure side, and then outside through a heat-exchanger on the low-temp side. That's EXACTLY what I proposed as well, except that the compressor would be shared with the HVAC system instead of having a dedicated one in the refrigerator. I'm not suggesting just blowing cold air from the AC over the back of a refrigerator.
My design eliminates the need for a single compressor, which is the most expensive component of a refrigerator and the part that uses most of the electricity. Sure, the HVAC will run more often, but it means that you can invest in one higher-efficiency unit than two less efficient ones.
The refrigerator would basically last forever, since it would contain no moving parts (well, maybe you stick a fan inside, but if you make that accessible it is $5 to replace). It would only require electricity to power the lights and possible fan.
The main downside would be that you couldn't move around the refrigerator without running plumbing/ducts/etc. However, most kitchen designs don't really let you move around the refrigerator anyway. In fact, given that sizes aren't standard and that you probably have a space in the cabinetry of a specific size it would be a lot less hassle if you could avoid ever having to replace it.
While you could use ducts and cold air, it would probably make more sense to just run an insulated pipe of freon into the thing.
They were not jailed for failing to predict it, but for giving (provably) false assurances that no earthquake was eminent. Subtle difference, and one the "science community" seems to conveniently forget when bringing this story up.
If it isn't illegal to say there will be a quake when none happens, why would it be illegal to say there won't be a quake when one does happen?
The problem with this sort of thinking is that it basically leads to useless perpetual states of alarm. That's why you'll never see the doomsday clock set to 3 hours to midnight, or the now-retired homeland security threat level set to green. There is never incentive to say that everything is fine, and always incentive to tell everybody they could die in 30 seconds so that they can't say they weren't warned.
Can we let this silly myth die already? Radiation from a coal plant is heavily diluted. Radiation is a problem of concentration, i.e. it is harmful when it passes a certain threshold. If you dilute it enough, radiation is not harmful, not any more than cosmic rays or a smoke detector.
If people really believed that then we would simply take our radioactive waste, load it onto ships, and have them slowly dump it in weighted containers with vents that open below a certain depth once they're in open ocean. That would be WAY cheaper than concentrating all of it in a place like Yucca Mt.
Sometimes I wonder if a root cause here is something I've seen at work: systems are designed by the best and the brightest to be operated by people who are fairly average. I'm sure for a reactor they all have college degrees, but as the old saying goes, "What do you call the guy who graduated last place in medical school? Doctor."
A better design would be to simply make the refrigerator an additional zone on the HVAC.
I think the costs you estimate are unrealistic - I doubt the incremental cost to add a zone to an HVAC is nearly that high (aside from the typical markup of multi-zone as a luxury item - which this would not be). Sure, retro-fitting an existing house would be expensive, but for those traditional refrigerators would be fine, just as people tend to use window units on older homes that lack an integrated HVAC.
Consider that cost-wise nothing needs to go into the HVAC to support a refrigerator than what would need to otherwise go into the refrigerator itself, aside from a few more feet of ductwork, or lines, or whatever technology is used to transfer heat.
The reason things are done the way they are is tradition.
It isn't any more difficult to crack. Moreover, the absolute only way it would introduce any difficulty at all is if the NSA is scanning images of text.
Look closer. He did actually create one font which is plain letters, but with the shapes not corresponding to the correct ASCII codes. That would actually create a very slight difficulty in scanning text (granted, one solved by a 3 line mapping function with 26 bytes of lookup).
Of course, if you used that in an actual online message the recipient wouldn't be able to read it without using the same font.
Might as well just use ebcdic while you're at it...
If they could do a job at making platform-specific test suites chances are that they could probably just make their platforms follow the specs properly and you wouldn't need that test suite in the first place...
The Chinese can't even effectively fine polluters and now there's talk of capital punishment for polluting? What next? Decimate school children when their class average isn't up to par because the instructor's scolding has no effect?
They did execute some executives who were selling tainted baby milk a few years ago.
Really in China the crime is embarrassing the government. Environmental damage is just the particular means in this case. As with everything else this will be very selectively enforced. They could care less what you dump in the stacks as long as your incident isn't the one that makes CNN for a week. If that happens, Zàijiàn.
However, I think it's also possible to trace TOR packet origins (and destinations if it's a hidden service) via perfect knowledge of the entire internet. I don't know if they have that yet, but I wouldn't be surprised if they did, or will soon.
Very good point. The only way to prevent this is to greatly increase latency, as in Freenet. Why doesn't anybody use Freenet? Well, that would be the latency...
Mixmaster is a great example of a design that is very hard to trace, because it had very high latency and all packets were identical in size. That means very little side-channel leakage.
You are correct, though I think that network interfaces still operate a bit differently than other character/block devices. I believe that network organization was actually abstracted into the filesystem, much as is done with many FUSE filesystems.
The problem with that site is good luck actually buying any of the cameras that are reviewed. They're basically reviewed by chipset, and they're not sold that way. Or, maybe they're sold that way but only contain the advertised chipset 60% of the time.
Dashcams aren't sold under brand names as far as I can tell, and that makes it really hard to buy anything of decent quality. The only sites I've found so far that seem trustworthy are MUCH more expensive.
When companies like Sony/Panasonic/etc start making them then perhaps we'll actually get some assurance of quality. (Quality being defined as reproducibility - I'm sure many of those cameras work great, but you can't reliably and reproducibly buy one.)
Never said they're the same market. However one market is growing and will cannibalise the other segment.
If they're not the same market, how will one grow and cannibalize the other? That's like saying that GM is concerned because Florida Oranges is making a killing.
Somehow I doubt that teenage boys are going to stop buying the latest COD because their girlfriend just introduced them to Bejeweled 13...
Do you have any Idea how old Java 6 is? It's not a question of keeping two version active at once. It's about it's age more than anything. Java 6 was released in 2006. It's not like their EOLing it after 2 year. Support has to end some time, and 7 years is longer than I would have kept it.
And how old is the win32 API? Now ask yourself why there are still companies that target that rather than Java ?
7 years isn't that long. Last time I checked XP was still on extended support (ie anybody can get security updates for free). Its replacement has been around for almost a decade (it won't be until it has been around for a full decade before XP is EOL).
As someone that's just had to write a long missive to my old car insurance company, I have to wonder why such things are even allowed to get that far without someone stepping in and saying "Hold on, that's just going to cause trouble".
Simple - 99% of the time it doesn't cause trouble. If a company sends out 1000 letters, gains a profit of $1000 for 999 of them, and then gets sued and loses to the tune of $40k on one, just what lesson do you think it will learn? Probably not the one the guy who sued them was trying to make.
I've always felt the solution is to split out the natural monopoly, which is the last mile.
Allow service providers to either charge by the circuit or byte depending on the underlying design, and that is just to provide service from the home to the central office. They can then charge for rackspace within the central office. And that's their entire business - they'd be forbidden for offering any services over their lines.
Then ISPs or video/phone/etc providers can set up equipment in the central office and sell service to homes. They will be entirely unregulated - they can bundle or not bundle and set caps or charge by the byte or resell do whatever they want. Of course, it will be easy to compete at this level, so somebody who offers lousy service will end up going out of business. The monopoly that runs the last mile doesn't care who wins, because they get paid risk-free for the circuits and rackspace (at a MUCH lower price than is paid today).
Such a system minimizes the expense of the last mile and promotes robust infrastructure, and lets anybody start up a business selling services to homes for little capital expenditure. That works out great for everybody.
Such an agreement would be trivial to enforce as well. Just set a rule on the incoming router and all traffic destined for non-Verizon IPs coming in over the link gets dropped.
I wonder what the gas mileage is when drafting a truck? Keep in mind that autonomous vehicles will potentially get much better mileage, since they can organize into "trains." On a highway I'd expect to see clusters of closely packed vehicles, rotating through the lead position.
I'm not sure that you'd need an RV necessarily, but I suspect cars will have better sleeping accommodations. That would greatly reduce the need for air travel. Just leave the night before and sleep in the car - unless you're going cross-country it would work fairly well.
For two lines on T-Mobile WITH data (which will never cost you more regardless of how much you use) and unlimited voice you could be around $80/mo these days I think. T-Mobile is significantly cheaper. The last time I switched contracts I found that going to ATT would have doubled my costs to start, and would have put me at risk of overage costs depending on usage. Only catch with that is that if you want a new flagship smartphone you'll be paying an extra $20/month or so for 18 months. I just paid $350 cash for a Nexus 4 (no other phone comes close to that as far as value goes).
Sure, I'd be willing to believe that Verizon has better coverage. However, you really can't compare them to T-Mobile on cost.
The only downside: I'm testing them this month with an old iPhone 3G and I'm seeing EDGE as often as not.:-(
I'm surprised that an iPhone 3G can connect to T-Mobile as anything BUT EDGE. ATT and T-Mobile use different frequencies for 3G (maybe that started to change when T-Mobile got all that spectrum as part of the failed merger). For the most part you'll need T-Mobile hardware to get 3G on T-Mobile. I think HSPA is more compatible between the two, and LTE might be as well once T-Mobile rolls that out. So you might be able to get 2G and 4G on both networks with the same hardware, but generally not 3G.
In any fairly populated area on T-Mobile you should get 4G coverage these days (and 3G as well if that is all your device handles). However, hardware designed for ATT might not fare as well. My stepdaughter has an iPhone 4 and I think she is basically on EDGE for the most part.
Your comment is strange because we're deploying fiber far faster than in Europe. xDSL seems to be receding.
His reasons were good. Europe probably has less corporate malfeasance, and they CERTAINLY have a higher population density. You can get around in Europe just fine with a train pass and a bicycle (or even on foot). There is no way you can do that in the US. The same things that make trains hard in the US make broadband hard.
However, in fairly dense areas like the US Northeast corporate malfeasance is the big barrier.
No. You are missing the point. One would never try to do what you are proposing. It would be a phenomenally stupid thing to do. OTOH, if you wanted to then it will work fine. You can install the old libs for it to use right alongside the latest. Linux is smart enough to use the right lib version for the right executable.
The problem with that approach is that old libs likely contain old security vulnerabilities, bugs, etc. So, the onus is on the app maintainer to backport all that stuff while maintaining the ancient ABIs.
I'm not saying that the lack of API stability on Linux in general isn't something that doesn't have ANY solutions. However, when MS discovers some security flow in DirectX they fix it such that applications written 10 years ago benefit from the fix without a need for re-compilation. On linux that is not so much the case.
Heck, too many libraries in the Linux world break ABI without even doing an SONAME change - that is a MAJOR pain point that causes issues even if you build everything from source.
You don't have to agree. However, the reason that 95% of corporations are running in Windows is because most of them appreciate that they aren't forced to upgrade applications simply because of abandoned APIs, libraries, etc.
You're seriously talking about a graphical change. This would be about fifteen seconds' work on an IBM mainframe, to diddle the format. If it isn't on SAP, then SAP sucks shit and no amount of your apologizing will change it any more than it would change Lotus Notes.
It is 15 seconds of work on any reporting tool, including SAP. The cost comes in all the release management.
Or do you just log into your production system and routinely edit your report files? If so, then your report fixes certainly will be faster, and that's good because I'm sure you'll be doing plenty of them...
Slashdot Mirror
Another point is that they overheard another researcher who disagreed with their assumption, he measured abnormal amounts of radon emissions and was concerned enough to evacuate his own family the day prior to the quake. But his data was not taken into consideration, despite their own model being a guess at best.
All models are guesses at best. For any disaster there will ALWAYS be somebody who saw it coming. The problem is that on every day there is somebody who sees a disaster coming that never arrives.
Society loves scapegoats. When somebody tells you anything about the future, you can be certain that they aren't.
Yes it does and it should (in this case), because we don't really know much about the process that causes earthquakes. That is the whole point really, these scientists went on record saying that nothing would happen, yet it did, to the tune of 307 deaths. Had they not abused their position of authority, and instead told the truth: "we don't know", this would have been a non story.
The problem with that is science doesn't "know" anything at all. We don't "know" if the Sun will explode this afternoon. It seems awfully unlikely, but the fact is that we don't have that many data points where our sun exploding is concerned, and even if we did we couldn't be certain. The best we can do is extrapolate from some observations hundreds of light years away or more.
Would things have been any better if the scientists said that they considered an earthquake extremely unlikely, just as I will go on the record and say that it is extremely unlikely that the sun will blow up before dinner?
Why would you add on the refrigerator as a new zone on existing HVAC? I thought the intention was to vent the heat outside, not into the current HVAC system (since if you want to do that, you can just keep it as is, and let the head convect naturally from the refrigerator into the room).
The suggestion was a refrigerator with a compressor that cycled freon through the fridge on the high-pressure side, and then outside through a heat-exchanger on the low-temp side. That's EXACTLY what I proposed as well, except that the compressor would be shared with the HVAC system instead of having a dedicated one in the refrigerator. I'm not suggesting just blowing cold air from the AC over the back of a refrigerator.
My design eliminates the need for a single compressor, which is the most expensive component of a refrigerator and the part that uses most of the electricity. Sure, the HVAC will run more often, but it means that you can invest in one higher-efficiency unit than two less efficient ones.
The refrigerator would basically last forever, since it would contain no moving parts (well, maybe you stick a fan inside, but if you make that accessible it is $5 to replace). It would only require electricity to power the lights and possible fan.
The main downside would be that you couldn't move around the refrigerator without running plumbing/ducts/etc. However, most kitchen designs don't really let you move around the refrigerator anyway. In fact, given that sizes aren't standard and that you probably have a space in the cabinetry of a specific size it would be a lot less hassle if you could avoid ever having to replace it.
While you could use ducts and cold air, it would probably make more sense to just run an insulated pipe of freon into the thing.
They were not jailed for failing to predict it, but for giving (provably) false assurances that no earthquake was eminent. Subtle difference, and one the "science community" seems to conveniently forget when bringing this story up.
If it isn't illegal to say there will be a quake when none happens, why would it be illegal to say there won't be a quake when one does happen?
The problem with this sort of thinking is that it basically leads to useless perpetual states of alarm. That's why you'll never see the doomsday clock set to 3 hours to midnight, or the now-retired homeland security threat level set to green. There is never incentive to say that everything is fine, and always incentive to tell everybody they could die in 30 seconds so that they can't say they weren't warned.
Can we let this silly myth die already? Radiation from a coal plant is heavily diluted. Radiation is a problem of concentration, i.e. it is harmful when it passes a certain threshold. If you dilute it enough, radiation is not harmful, not any more than cosmic rays or a smoke detector.
If people really believed that then we would simply take our radioactive waste, load it onto ships, and have them slowly dump it in weighted containers with vents that open below a certain depth once they're in open ocean. That would be WAY cheaper than concentrating all of it in a place like Yucca Mt.
Sometimes I wonder if a root cause here is something I've seen at work: systems are designed by the best and the brightest to be operated by people who are fairly average. I'm sure for a reactor they all have college degrees, but as the old saying goes, "What do you call the guy who graduated last place in medical school? Doctor."
A better design would be to simply make the refrigerator an additional zone on the HVAC.
I think the costs you estimate are unrealistic - I doubt the incremental cost to add a zone to an HVAC is nearly that high (aside from the typical markup of multi-zone as a luxury item - which this would not be). Sure, retro-fitting an existing house would be expensive, but for those traditional refrigerators would be fine, just as people tend to use window units on older homes that lack an integrated HVAC.
Consider that cost-wise nothing needs to go into the HVAC to support a refrigerator than what would need to otherwise go into the refrigerator itself, aside from a few more feet of ductwork, or lines, or whatever technology is used to transfer heat.
The reason things are done the way they are is tradition.
It isn't any more difficult to crack. Moreover, the absolute only way it would introduce any difficulty at all is if the NSA is scanning images of text.
Look closer. He did actually create one font which is plain letters, but with the shapes not corresponding to the correct ASCII codes. That would actually create a very slight difficulty in scanning text (granted, one solved by a 3 line mapping function with 26 bytes of lookup).
Of course, if you used that in an actual online message the recipient wouldn't be able to read it without using the same font.
Might as well just use ebcdic while you're at it...
If they could do a job at making platform-specific test suites chances are that they could probably just make their platforms follow the specs properly and you wouldn't need that test suite in the first place...
The Chinese can't even effectively fine polluters and now there's talk of capital punishment for polluting? What next? Decimate school children when their class average isn't up to par because the instructor's scolding has no effect?
They did execute some executives who were selling tainted baby milk a few years ago.
Really in China the crime is embarrassing the government. Environmental damage is just the particular means in this case. As with everything else this will be very selectively enforced. They could care less what you dump in the stacks as long as your incident isn't the one that makes CNN for a week. If that happens, Zàijiàn.
However, I think it's also possible to trace TOR packet origins (and destinations if it's a hidden service) via perfect knowledge of the entire internet. I don't know if they have that yet, but I wouldn't be surprised if they did, or will soon.
Very good point. The only way to prevent this is to greatly increase latency, as in Freenet. Why doesn't anybody use Freenet? Well, that would be the latency...
Mixmaster is a great example of a design that is very hard to trace, because it had very high latency and all packets were identical in size. That means very little side-channel leakage.
You are correct, though I think that network interfaces still operate a bit differently than other character/block devices. I believe that network organization was actually abstracted into the filesystem, much as is done with many FUSE filesystems.
The problem with that site is good luck actually buying any of the cameras that are reviewed. They're basically reviewed by chipset, and they're not sold that way. Or, maybe they're sold that way but only contain the advertised chipset 60% of the time.
Dashcams aren't sold under brand names as far as I can tell, and that makes it really hard to buy anything of decent quality. The only sites I've found so far that seem trustworthy are MUCH more expensive.
When companies like Sony/Panasonic/etc start making them then perhaps we'll actually get some assurance of quality. (Quality being defined as reproducibility - I'm sure many of those cameras work great, but you can't reliably and reproducibly buy one.)
Never said they're the same market. However one market is growing and will cannibalise the other segment.
If they're not the same market, how will one grow and cannibalize the other? That's like saying that GM is concerned because Florida Oranges is making a killing.
Somehow I doubt that teenage boys are going to stop buying the latest COD because their girlfriend just introduced them to Bejeweled 13...
Do you have any Idea how old Java 6 is? It's not a question of keeping two version active at once. It's about it's age more than anything. Java 6 was released in 2006. It's not like their EOLing it after 2 year. Support has to end some time, and 7 years is longer than I would have kept it.
And how old is the win32 API? Now ask yourself why there are still companies that target that rather than Java ?
7 years isn't that long. Last time I checked XP was still on extended support (ie anybody can get security updates for free). Its replacement has been around for almost a decade (it won't be until it has been around for a full decade before XP is EOL).
As someone that's just had to write a long missive to my old car insurance company, I have to wonder why such things are even allowed to get that far without someone stepping in and saying "Hold on, that's just going to cause trouble".
Simple - 99% of the time it doesn't cause trouble. If a company sends out 1000 letters, gains a profit of $1000 for 999 of them, and then gets sued and loses to the tune of $40k on one, just what lesson do you think it will learn? Probably not the one the guy who sued them was trying to make.
Ah, lovely to see tax dollars so well-spent. Or donations, or whatever that council is funded by.
I've always felt the solution is to split out the natural monopoly, which is the last mile.
Allow service providers to either charge by the circuit or byte depending on the underlying design, and that is just to provide service from the home to the central office. They can then charge for rackspace within the central office. And that's their entire business - they'd be forbidden for offering any services over their lines.
Then ISPs or video/phone/etc providers can set up equipment in the central office and sell service to homes. They will be entirely unregulated - they can bundle or not bundle and set caps or charge by the byte or resell do whatever they want. Of course, it will be easy to compete at this level, so somebody who offers lousy service will end up going out of business. The monopoly that runs the last mile doesn't care who wins, because they get paid risk-free for the circuits and rackspace (at a MUCH lower price than is paid today).
Such a system minimizes the expense of the last mile and promotes robust infrastructure, and lets anybody start up a business selling services to homes for little capital expenditure. That works out great for everybody.
Such an agreement would be trivial to enforce as well. Just set a rule on the incoming router and all traffic destined for non-Verizon IPs coming in over the link gets dropped.
I wonder what the gas mileage is when drafting a truck? Keep in mind that autonomous vehicles will potentially get much better mileage, since they can organize into "trains." On a highway I'd expect to see clusters of closely packed vehicles, rotating through the lead position.
I'm not sure that you'd need an RV necessarily, but I suspect cars will have better sleeping accommodations. That would greatly reduce the need for air travel. Just leave the night before and sleep in the car - unless you're going cross-country it would work fairly well.
For two lines on T-Mobile WITH data (which will never cost you more regardless of how much you use) and unlimited voice you could be around $80/mo these days I think. T-Mobile is significantly cheaper. The last time I switched contracts I found that going to ATT would have doubled my costs to start, and would have put me at risk of overage costs depending on usage. Only catch with that is that if you want a new flagship smartphone you'll be paying an extra $20/month or so for 18 months. I just paid $350 cash for a Nexus 4 (no other phone comes close to that as far as value goes).
Sure, I'd be willing to believe that Verizon has better coverage. However, you really can't compare them to T-Mobile on cost.
The only downside: I'm testing them this month with an old iPhone 3G and I'm seeing EDGE as often as not. :-(
I'm surprised that an iPhone 3G can connect to T-Mobile as anything BUT EDGE. ATT and T-Mobile use different frequencies for 3G (maybe that started to change when T-Mobile got all that spectrum as part of the failed merger). For the most part you'll need T-Mobile hardware to get 3G on T-Mobile. I think HSPA is more compatible between the two, and LTE might be as well once T-Mobile rolls that out. So you might be able to get 2G and 4G on both networks with the same hardware, but generally not 3G.
In any fairly populated area on T-Mobile you should get 4G coverage these days (and 3G as well if that is all your device handles). However, hardware designed for ATT might not fare as well. My stepdaughter has an iPhone 4 and I think she is basically on EDGE for the most part.
Rich
Your comment is strange because we're deploying fiber far faster than in Europe. xDSL seems to be receding.
His reasons were good. Europe probably has less corporate malfeasance, and they CERTAINLY have a higher population density. You can get around in Europe just fine with a train pass and a bicycle (or even on foot). There is no way you can do that in the US. The same things that make trains hard in the US make broadband hard.
However, in fairly dense areas like the US Northeast corporate malfeasance is the big barrier.
No. You are missing the point. One would never try to do what you are proposing. It would be a phenomenally stupid thing to do. OTOH, if you wanted to then it will work fine. You can install the old libs for it to use right alongside the latest. Linux is smart enough to use the right lib version for the right executable.
The problem with that approach is that old libs likely contain old security vulnerabilities, bugs, etc. So, the onus is on the app maintainer to backport all that stuff while maintaining the ancient ABIs.
I'm not saying that the lack of API stability on Linux in general isn't something that doesn't have ANY solutions. However, when MS discovers some security flow in DirectX they fix it such that applications written 10 years ago benefit from the fix without a need for re-compilation. On linux that is not so much the case.
Heck, too many libraries in the Linux world break ABI without even doing an SONAME change - that is a MAJOR pain point that causes issues even if you build everything from source.
You don't have to agree. However, the reason that 95% of corporations are running in Windows is because most of them appreciate that they aren't forced to upgrade applications simply because of abandoned APIs, libraries, etc.
You're seriously talking about a graphical change. This would be about fifteen seconds' work on an IBM mainframe, to diddle the format. If it isn't on SAP, then SAP sucks shit and no amount of your apologizing will change it any more than it would change Lotus Notes.
It is 15 seconds of work on any reporting tool, including SAP. The cost comes in all the release management.
Or do you just log into your production system and routinely edit your report files? If so, then your report fixes certainly will be faster, and that's good because I'm sure you'll be doing plenty of them...