"Cattle can graze on unirrigated badlands where crop lands are impractical."
The problem with your argument is that in reality the vast, vast majority of cows are raised on flat land and are fed corn instead of grass.
No the problem with your counter argument is that it is factually false and way off. Most cattle spend less than a season of their lifespan on grain lots. The rest of the time they graze. Not all cattle are free-range. For example managed intensive grazing restricts cattle's roaming range so that they don't overgraze or damage lands. But those are not feedlots and the grass grows naturally.
Your argument would hold more water IF cows were free to range on land that's not suitable for farming instead of being force feed corn. I'm glad to hear you must now agree with me.
As a rule of thumb, for each step you go up in the food pyramid, only 10% of energy is retained. Putting cows between a person and the vegetable (corn in this case) simply adds inefficiency in the energy transfer. So instead of feeding 10 vegetarians, you're now feeding 1 meat eater. Again your fact are wrong. Consider the cow. It propels it self between bites of grass then delivers it self to a way point. No fuel is used trucking nutrients to the cow, plowing the cows field. No water is pumped to irrigate the fields. Your cow does use energy to transport itself to the food and too the water but that's more efficient that delivering the food and water to the cow. Your vegetables require lots of energy to tend and feed them and it's trucked in and out. And unless you live on raw carrots, Even cooking vegetables takes more water and energy. Most vegetarians need to eat nuts and cheeses and other sources of protein and the sustainable vegetarian diet is likely to be less efficiently trucked than a burger.
But even setting the efficiency argument entirely aside the bigger problem with your argument simply ignores the fact that you can't grow crops most places you can raise a cow. SO the cow is harvesting all the photosynthesis for you that was previously not available.
Vegetarianism requires irrigation that uses up precious water resources. Cuts down forests to make crop land. And drains pesticides and nirates into our rivers. When done poorly, as in many places it leads to salted or barren lands. Huge amounts of fossil fuels are consumes planting, harvesting and manufacturing fertilizer. Even the employees have to drive to work.
Then unless you want to live just on what your area can grow, you have to truck all those crops 2000 miles per bite, lentil breath.
Take the midwest which used to support 200 million buffalo on natural unirrigated grasses. That was all torn up to produce crops.
Cattle can graze on unirrigated badlands where crop lands are impractical. No pesticides. Take away cows and many people would starve. Rather than trucking low energy food thousands of miles, cows can collected centrally then just the high energy tasty bits actually transported.
Now as for methane. We now have 100 million cows in the US and each cow is producing less methane than the buffalo. Moreover grain fed beef produces 1/3 of grass fed. And lets not forget all the other gazers like deer that are less plenitful than the old days. So the amount of methane being produced now from cow farts is a small fraction of the pre-whiteman era.
Moreover all those 100 million cows produce less methane every year than a gigwatt coal firesplant. And their building 1 a week in china right now. By 2040 that may rise to 3 a day if energy demands continue at present rates.
Suppose they just make some shit up and tell you it's Your genome. How will you know? With only two companies, even if you were to have it done by another company you would not know which version was right if either. I suppose you could submit the same thing twice to one company under different fake names but they could also be devious too. Just sample your genome for a signature region (e.g. a restriction digest) and they essentially have a hash code for recognizing you if you try to sneak a second one in. SO they give you back the same thing they gave you the first time.
In the early days, MS gave the impression of tolerating piracy. Whether they did or not it's widely believed it helped them more than it hurt them. Centos is not piracy but it can help Redhad spread itself.
You can. There's a defaults setting for this and a third part gui to make setting it easy. Saw it on mac osx hints a couple years ago. However I would warn you against it. Better to learn the mac way. Homogenous behaviour pays off in the long run if it does not hinder you too much. You can get some of the hover focus funtionality with click to focus by the way. I forget the key combos but you can cut and past text from a background or into it without raising the focus.
That's the problem with these technology breakthroughs. I fell bad for blue ray and HD dvd, cause now everyone will have a drive big enough to rip to. That is if the 40 Gig version of this new memory costs more than a blue ray media.
This seems to fit some other piece of the puzzle. For example, Why to UK iphones cost so much more? Presumably because of a lower subsidy.
The UK market is actually a very good example. A Macbook in the UK costs the equivalent of $1436 in the US it costs $1099. Why does it cost $337 more in the UK? Because that's the optimal price for the UK market. They can charge that much and people will pay it. Because the UK price includes sales tax. The US price does not.
NASA had many cost overruns and did pay to test out the wazzoo. They even used old technology sometimes when they were not sure. (e.g. The First Mars landers used Core memory because they were not sure of the radiation hardness of chip memory)
You also have to realize that when this thing was being rolled out they were in the midst of a recall on sony batteries, and Dells were catching fire. There has to be a risk margin. I was just surpised how large it was. Turns out it was fractionally less than I thought.
People say things like "it's Apple's right" and "good for them". Of course, it's Apple's right to do those deals.
Nevertheless, where do you think this money is coming from? Do you think that AT&T is giving that to Apple because they are such good buddies?
No, you are paying for it one way or another (e.g., by paying a premium for their sluggish EDGE service). But I pay the same price for ATT no matter which phone I use. So tell me again how I am paying more. Seems like I just learned I'm pay $18 per month less that the chumps with non iPhone who get the same service.
I'm failing to see where the number 831 comes from. Exactly, the headline is rubbish. Apple does not make 831 per Iphone. They collect 400+431 over the 2 year contract. They "make" considerably less.
Now another way of saying this is I am paying 431 dollars less than the true sales price of the iphone. Or another way of saying it is, AT&T is giving me an $18 a month discount for using an Iphone on their network. All upside to me. Of course that mean I should be upset about the unlockers who are preventing them from giving me an even larger discount.
This seems to fit some other piece of the puzzle. For example, Why to UK iphones cost so much more? Presumably because of a lower subsidy. And why is apple booking the iphone revenue as deferred subscription income? Because they are probably not making any money on the sales, but on the 18$ per month.
Finally, this also helps axplain the anomolous $200 price drop. My original guess, which this reinforces, was that apple took a huge gamble on the technology. Craploads could have gone wrong. The screens might have scratched to easily, the batteries might have died prematurely, the OS might have blue screened. . So many untested things you can't really adequately Q/A before the roll out. Plus it might not have been popular. There were a few look-alikes in the pipeline, what if one had rolled out earlier?
So they had a huge risk margin built into the price. Once the risk dissipated they could remove that. But at the time this hypothesis seemed a little off. Sure a risk margin is there in any product but how could they overestimate by 50% of the propert phone price? that seems way too high. But now realize the true sales price of the phone was 1031$ and they lowered it by 20% to 831. Now it does not seem quite so absurd.
my bad. APE acts like a kernel mod which is why I think of it as one. It lets you do things you "shouldn't" be able to do without a kernel mod by playing some funky games.
On macs apps are self contained entities (normally) thus archiving means moving all the apps out of the application folder and into another folder. You can then at your leisure drag them back one by one. It also moves all your preference settings and fonts to another folder, and cleans out the startup items list. (fonts and preferences are the only two things that normally leak beyond the app container (there's a few other exceptions for special apps, but as a rule installing an app does not barf all over your computer with DLLs and Registry hooks. On macs, the OS polls the App in Applications folder for its capabilities and files it likes to open, so the app does not need to modify a registry. as a result unistalls are normally as simple as dragging the app out of the application folder.
So after an archive and install you drag back the things you want to keep. The only hassle is sometimes you need to look for the associated preference setting which is where most apps store their activation key.
A few non apple apps behave linux like and shove things into/usr/local/bin or libs or/share/man. So restoring those takes a bit more work. But on the otherhand those are exactly the kind of Cruft you want to blow out between major system updates. So archive and install is the way to go in previous updates
Rumor was that apple worked hard to make the non-archive and install mode (update) smart enough to remove all the bad crap and leave the okay crap. Apparently not!
But if it is APE then well then that's kinda understandable its the singlemost invasive kernel mod of all.
If the OLPC really increases their market share dramatically, and since the OLPC model shifts all marketing a distribution costs onto the country purchasing them, and because all orders are bulk orders in the millions, maybe they could cut the price dramatically. It's all marginal profit to them and a margin of say $20 might be just dandy. It would be a cut feature version to avoid cannibalizing 1sth world nation sales. Maybe they could even open source the low end version and turn the world into their kernel developer.
The other possibility is that they just want a low power system they can embed in the iLoo 2008.
File X which lives in path P goes through 8 revsions over time. x1,x2,x3....x7,x8, which time machine dutifully stores. Note that these revsions are new inodes so the only relationship between these is the filename is the same "X", and the path is the same "P".
At some point x8 is gets moved to path Q so it becomes Q/x8.
Next three hardlinks are made. Q/x8, R/X8, and one back in P/x8. And I also make a revision to R/x8 to make it R/x9, so it does not share an inode with Q/x8 or P/x8
If I tell it to delete all backed up versions of R/x9, what happens to x1..x8 in time machine? If it deletes those then Q/x8 and P/x8 have lost there entire backup history. If it chooses not to delete those then Unless I was aware that Q/x8 and P/x8 existed I'd be very puzzled. If it tries to warn me there are hardlinks then basically it becomes useless if there are any significant number of hardlinks. If you make a rule up, do I get the same behavior if I delete Q/x8 or P/x8?
So I'm curious about how it handles this. It's not an idle question for me since I actually have more hardlinks than actual files. (I use these as my own time machine for snapshoting).
I got the skynet link of course, and it's apt. What we are seeing is the slow transition from single cellular behaviour to a multi cellualr organism. That is instead of being fighting on it's own, it now has a global immune response to an invader (security researcher). With the advent of virtual machine detectors last year these things now commit apoptosis when they detect they have been invaded by the security researcher.
In other words we have changed roles. Instead of us being the host and them being the virus, it now is behaving like a host and us as the invasive organism.
These things certainly have enough global cpu strength to do some serious artifical intelligence. even if it were not efficient, they have millions of cpus to harness. Some already do have code changing algorithms to hide their signature. And the ones that survive, are the fittest in an evolutionary sense. At some point they may actually start changing their own design, and eventually their own requirements.
So skynet may evolve itself naturally, not as an actual construction.
Most applications with a persistent (across sessions) undo behave like this because they can do all their unsaved changes in a temporary file then simply rename it when you do a save. Additionally many applications with a regular undo also behave this way. Some do not. bbedit being one example of one that does not. But most do.
So the point is, that on tiger, when you hard link an document then edit it, the hardlink gets split to two files. That's the behavior. As I said, its a property of a tiger app not the hardlink per se, but this is irrelevant to the greater point.
The greater issue is how does Time Machine know what to delete when the users asks it to delete all version of an file? Well the seemingly obvious logic would be to somehow track which file begat which file. But this logic seems to break down when considers copies, moves and hardlinks. Even if you invent a new rule to handle copies and moves, it would seem to break down at hardlinks. That was what the original post with the subject "hardlinks" was trying to describe.
Sorry to tell you this but you are mistaken. In most application (but not all) that modify files the program unlinks the file before writing the new file. Try it. However there are also many cases where this is not the case, an example is "cat > filename" or pico or bbedit. Those will write to single file and both hardlinks are changed.
Now you might say, well yes that is true but that's just the application doing that not the hardlink. True, but re-read my original question. How does the user know (or the OS know what the user meant) when he asks it to erase all version of a file? In the example I gave the result would change depending if a hardlink existed or not. The alternate answer is that the system would not be able to backtrack versions for any application that did that (which would be basically all major apps).
Sorry but you are utterly and completely mistaken. If you modify a hardlinked file in Tiger, it copies the file and modifies the copy. Thus one does not have two links to the same modified file but two different files.
How do it know? When is a file a version and not a new one?
That's easy. It tracks the changes to the files. If you create a new file by using "save as" that won't be deleted and neither will it's history, but that is obvious because the original file still exists. If you move a file, it is still the same file. If you copy a file, you've made a new file, based upon the old one.
Okay try this one on for size. Make a hard link of a file. Now edit one of the hardlinks and save it (not save-as, just save). Now which one is the copy? From the file systems POV the edited one will be a copy. But from the users point of view it might be the original, especially if they had no way of knowing the hard link had been made.
For example, since I don't have Time Machine yet I currently snapshot my home directory by making a image of it populated by hardlinks. this happens in the background so I don't even know it is happening. Nor do the other users on the computer. You can't really say which is the link since it's a hard link not a softlink or alias. A hard link is an identical file system entry as the original and should be indistinguishable. The save will sever the link.
Watch the Apple leopard video. I believe in there, they talk briefly about how TM has the option to permanently remove all versions of a file. It should also be mentioned on the TM feature page Apple has on the web site... in any case it's possible.
It's such an obvious feature it's no surprise it's included. This is versioning 101 stuff. How do it know? When is a file a version and not a new one? For example if I have a configuration file for some data processing program I use. I edit it in different ways for different runs. Is this a version or a different file. Or how about a generic reference letter I go in and change the names in for another use. version or different file? What if I move or copy a file. Are these versions?
Slashdot Mirror
The problem with your argument is that in reality the vast, vast majority of cows are raised on flat land and are fed corn instead of grass.
No the problem with your counter argument is that it is factually false and way off. Most cattle spend less than a season of their lifespan on grain lots. The rest of the time they graze. Not all cattle are free-range. For example managed intensive grazing restricts cattle's roaming range so that they don't overgraze or damage lands. But those are not feedlots and the grass grows naturally. Your argument would hold more water IF cows were free to range on land that's not suitable for farming instead of being force feed corn. I'm glad to hear you must now agree with me. As a rule of thumb, for each step you go up in the food pyramid, only 10% of energy is retained. Putting cows between a person and the vegetable (corn in this case) simply adds inefficiency in the energy transfer. So instead of feeding 10 vegetarians, you're now feeding 1 meat eater. Again your fact are wrong. Consider the cow. It propels it self between bites of grass then delivers it self to a way point. No fuel is used trucking nutrients to the cow, plowing the cows field. No water is pumped to irrigate the fields. Your cow does use energy to transport itself to the food and too the water but that's more efficient that delivering the food and water to the cow. Your vegetables require lots of energy to tend and feed them and it's trucked in and out. And unless you live on raw carrots, Even cooking vegetables takes more water and energy. Most vegetarians need to eat nuts and cheeses and other sources of protein and the sustainable vegetarian diet is likely to be less efficiently trucked than a burger.But even setting the efficiency argument entirely aside the bigger problem with your argument simply ignores the fact that you can't grow crops most places you can raise a cow. SO the cow is harvesting all the photosynthesis for you that was previously not available.
excellent satire.
Vegetarianism requires irrigation that uses up precious water resources. Cuts down forests to make crop land. And drains pesticides and nirates into our rivers. When done poorly, as in many places it leads to salted or barren lands. Huge amounts of fossil fuels are consumes planting, harvesting and manufacturing fertilizer. Even the employees have to drive to work.
Then unless you want to live just on what your area can grow, you have to truck all those crops 2000 miles per bite, lentil breath.
Take the midwest which used to support 200 million buffalo on natural unirrigated grasses. That was all torn up to produce crops.
Cattle can graze on unirrigated badlands where crop lands are impractical. No pesticides. Take away cows and many people would starve. Rather than trucking low energy food thousands of miles, cows can collected centrally then just the high energy tasty bits actually transported.
Now as for methane. We now have 100 million cows in the US and each cow is producing less methane than the buffalo. Moreover grain fed beef produces 1/3 of grass fed. And lets not forget all the other gazers like deer that are less plenitful than the old days. So the amount of methane being produced now from cow farts is a small fraction of the pre-whiteman era.
Moreover all those 100 million cows produce less methane every year than a gigwatt coal firesplant. And their building 1 a week in china right now. By 2040 that may rise to 3 a day if energy demands continue at present rates.
Any people worry about cows??
Knowing someone's password can be handy. Most folks use the same password on multiple machines or entire networks. Moreover they seldom change them.
Suppose they just make some shit up and tell you it's Your genome. How will you know? With only two companies, even if you were to have it done by another company you would not know which version was right if either. I suppose you could submit the same thing twice to one company under different fake names but they could also be devious too. Just sample your genome for a signature region (e.g. a restriction digest) and they essentially have a hash code for recognizing you if you try to sneak a second one in. SO they give you back the same thing they gave you the first time.
In the early days, MS gave the impression of tolerating piracy. Whether they did or not it's widely believed it helped them more than it hurt them. Centos is not piracy but it can help Redhad spread itself.
is what came to mind. Now it's hard to sit down.
You can. There's a defaults setting for this and a third part gui to make setting it easy. Saw it on mac osx hints a couple years ago. However I would warn you against it. Better to learn the mac way. Homogenous behaviour pays off in the long run if it does not hinder you too much. You can get some of the hover focus funtionality with click to focus by the way. I forget the key combos but you can cut and past text from a background or into it without raising the focus.
That's the problem with these technology breakthroughs. I fell bad for blue ray and HD dvd, cause now everyone will have a drive big enough to rip to. That is if the 40 Gig version of this new memory costs more than a blue ray media.
£1299 - £227 VAT = £1072
NASA had many cost overruns and did pay to test out the wazzoo. They even used old technology sometimes when they were not sure. (e.g. The First Mars landers used Core memory because they were not sure of the radiation hardness of chip memory) You also have to realize that when this thing was being rolled out they were in the midst of a recall on sony batteries, and Dells were catching fire. There has to be a risk margin. I was just surpised how large it was. Turns out it was fractionally less than I thought.
Nevertheless, where do you think this money is coming from? Do you think that AT&T is giving that to Apple because they are such good buddies?
No, you are paying for it one way or another (e.g., by paying a premium for their sluggish EDGE service). But I pay the same price for ATT no matter which phone I use. So tell me again how I am paying more. Seems like I just learned I'm pay $18 per month less that the chumps with non iPhone who get the same service.
Now another way of saying this is I am paying 431 dollars less than the true sales price of the iphone. Or another way of saying it is, AT&T is giving me an $18 a month discount for using an Iphone on their network. All upside to me. Of course that mean I should be upset about the unlockers who are preventing them from giving me an even larger discount.
This seems to fit some other piece of the puzzle. For example, Why to UK iphones cost so much more? Presumably because of a lower subsidy. And why is apple booking the iphone revenue as deferred subscription income? Because they are probably not making any money on the sales, but on the 18$ per month.
Finally, this also helps axplain the anomolous $200 price drop. My original guess, which this reinforces, was that apple took a huge gamble on the technology. Craploads could have gone wrong. The screens might have scratched to easily, the batteries might have died prematurely, the OS might have blue screened. . So many untested things you can't really adequately Q/A before the roll out. Plus it might not have been popular. There were a few look-alikes in the pipeline, what if one had rolled out earlier?
So they had a huge risk margin built into the price. Once the risk dissipated they could remove that. But at the time this hypothesis seemed a little off. Sure a risk margin is there in any product but how could they overestimate by 50% of the propert phone price? that seems way too high. But now realize the true sales price of the phone was 1031$ and they lowered it by 20% to 831. Now it does not seem quite so absurd.
my bad. APE acts like a kernel mod which is why I think of it as one. It lets you do things you "shouldn't" be able to do without a kernel mod by playing some funky games.
On macs apps are self contained entities (normally) thus archiving means moving all the apps out of the application folder and into another folder. You can then at your leisure drag them back one by one.
/usr/local/bin or libs or /share/man. So restoring those takes a bit more work. But on the otherhand those are exactly the kind of Cruft you want to blow out between major system updates. So archive and install is the way to go in previous updates
It also moves all your preference settings and fonts to another folder, and cleans out the startup items list. (fonts and preferences are the only two things that normally leak beyond the app container (there's a few other exceptions for special apps, but as a rule installing an app does not barf all over your computer with DLLs and Registry hooks. On macs, the OS polls the App in Applications folder for its capabilities and files it likes to open, so the app does not need to modify a registry. as a result unistalls are normally as simple as dragging the app out of the application folder.
So after an archive and install you drag back the things you want to keep. The only hassle is sometimes you need to look for the associated preference setting which is where most apps store their activation key.
A few non apple apps behave linux like and shove things into
Rumor was that apple worked hard to make the non-archive and install mode (update) smart enough to remove all the bad crap and leave the okay crap. Apparently not!
But if it is APE then well then that's kinda understandable its the singlemost invasive kernel mod of all.
If the OLPC really increases their market share dramatically, and since the OLPC model shifts all marketing a distribution costs onto the country purchasing them, and because all orders are bulk orders in the millions, maybe they could cut the price dramatically. It's all marginal profit to them and a margin of say $20 might be just dandy. It would be a cut feature version to avoid cannibalizing 1sth world nation sales. Maybe they could even open source the low end version and turn the world into their kernel developer.
The other possibility is that they just want a low power system they can embed in the iLoo 2008.
Let me give you an example.
File X which lives in path P goes through 8 revsions over time. x1,x2,x3....x7,x8, which time machine dutifully stores. Note that these revsions are new inodes so the only relationship between these is the filename is the same "X", and the path is the same "P".
At some point x8 is gets moved to path Q so it becomes Q/x8.
Next three hardlinks are made. Q/x8, R/X8, and one back in P/x8. And I also make a revision to R/x8 to make it R/x9, so it does not share an inode with Q/x8 or P/x8
If I tell it to delete all backed up versions of R/x9, what happens to x1..x8 in time machine? If it deletes those then Q/x8 and P/x8 have lost there entire backup history. If it chooses not to delete those then Unless I was aware that Q/x8 and P/x8 existed I'd be very puzzled. If it tries to warn me there are hardlinks then basically it becomes useless if there are any significant number of hardlinks. If you make a rule up, do I get the same behavior if I delete Q/x8 or P/x8?
So I'm curious about how it handles this. It's not an idle question for me since I actually have more hardlinks than actual files. (I use these as my own time machine for snapshoting).
I got the skynet link of course, and it's apt. What we are seeing is the slow transition from single cellular behaviour to a multi cellualr organism. That is instead of being fighting on it's own, it now has a global immune response to an invader (security researcher). With the advent of virtual machine detectors last year these things now commit apoptosis when they detect they have been invaded by the security researcher.
In other words we have changed roles. Instead of us being the host and them being the virus, it now is behaving like a host and us as the invasive organism.
These things certainly have enough global cpu strength to do some serious artifical intelligence. even if it were not efficient, they have millions of cpus to harness. Some already do have code changing algorithms to hide their signature. And the ones that survive, are the fittest in an evolutionary sense. At some point they may actually start changing their own design, and eventually their own requirements.
So skynet may evolve itself naturally, not as an actual construction.
Most applications with a persistent (across sessions) undo behave like this because they can do all their unsaved changes in a temporary file then simply rename it when you do a save. Additionally many applications with a regular undo also behave this way. Some do not. bbedit being one example of one that does not. But most do.
So the point is, that on tiger, when you hard link an document then edit it, the hardlink gets split to two files. That's the behavior. As I said, its a property of a tiger app not the hardlink per se, but this is irrelevant to the greater point.
The greater issue is how does Time Machine know what to delete when the users asks it to delete all version of an file? Well the seemingly obvious logic would be to somehow track which file begat which file. But this logic seems to break down when considers copies, moves and hardlinks. Even if you invent a new rule to handle copies and moves, it would seem to break down at hardlinks. That was what the original post with the subject "hardlinks" was trying to describe.
What you are saying is mistaken for the reason I give in this other reply
What you propose, inode tracking, would totally fail in the hardlink example for the reason I give in this other reply
Sorry to tell you this but you are mistaken. In most application (but not all) that modify files the program unlinks the file before writing the new file. Try it. However there are also many cases where this is not the case, an example is "cat > filename" or pico or bbedit. Those will write to single file and both hardlinks are changed.
Now you might say, well yes that is true but that's just the application doing that not the hardlink. True, but re-read my original question. How does the user know (or the OS know what the user meant) when he asks it to erase all version of a file? In the example I gave the result would change depending if a hardlink existed or not. The alternate answer is that the system would not be able to backtrack versions for any application that did that (which would be basically all major apps).
Sorry but you are utterly and completely mistaken. If you modify a hardlinked file in Tiger, it copies the file and modifies the copy. Thus one does not have two links to the same modified file but two different files.
That's easy. It tracks the changes to the files. If you create a new file by using "save as" that won't be deleted and neither will it's history, but that is obvious because the original file still exists. If you move a file, it is still the same file. If you copy a file, you've made a new file, based upon the old one.
Okay try this one on for size. Make a hard link of a file. Now edit one of the hardlinks and save it (not save-as, just save). Now which one is the copy? From the file systems POV the edited one will be a copy. But from the users point of view it might be the original, especially if they had no way of knowing the hard link had been made.For example, since I don't have Time Machine yet I currently snapshot my home directory by making a image of it populated by hardlinks. this happens in the background so I don't even know it is happening. Nor do the other users on the computer. You can't really say which is the link since it's a hard link not a softlink or alias. A hard link is an identical file system entry as the original and should be indistinguishable. The save will sever the link.
It's such an obvious feature it's no surprise it's included. This is versioning 101 stuff. How do it know? When is a file a version and not a new one? For example if I have a configuration file for some data processing program I use. I edit it in different ways for different runs. Is this a version or a different file. Or how about a generic reference letter I go in and change the names in for another use. version or different file? What if I move or copy a file. Are these versions?