Now, using a WAP or long Ethernet cable to extend your connectivity next door is a different matter.
Just to clarify a point... as long as it's your connectivity, it is perfectly reasonable for you to use it, even if you happen to be next door, down the block, or in another state. Where it should become illegal is when your next door neighbor uses it from next door, down the block, or in another state....
Based on my experience, Earthlink support doesn't even know the structure of their -own- network, much less know how to debug yours. My favorite was that, living in the south bay (San Jose area), every time Oakland went out, so did my network connection (an hour away or so). Or maybe that it took one of their support people an hour just to figure out that I needed to include "@mindspring.com" at the end of my bloody PPPoE login name....
Did their system status message say the San Jose area? Is Sunnyvale remotely the Oakland area? No, and no. I called them about the first two multi-hour outages. By the third one, I gave up. Why sit on hold for an hour waiting to tell them what they're too clueless to figure out on their own?
EarthLink tech support---another one of the top ten reasons I left Earthlink. Wow, the way this thread is going, I will have listed them all by the time I get to the bottom....:-)
I definitely believe it. Most ISPs with half a brain among them support the use of NAT. It just makes sense. I mean, after all, why would they want to waste five IP numbers on a single customer who is likely to only have one machine turned on and downloading files anyway when they could give him/her one IP number and keep the other four for more customers?
In an era where most ISPs are seriously having trouble getting enough IP numbers to fill the demand, we're seeing more and more ISPs suddenly being very enthusiastic about NAT....
The cable companies, however, have a different mentality, often because they have different limitations. If you don't allow servers, there's nothing preventing you from giving everyone a ten-net number (10.x.x.x) and NATting it at the ISP level. Some ISPs do this.
Since they then have virtually an unlimited number of IP numbers that they can assign, they have no financial incentive to allow customers to use NAT.
On the flip side of this, though, the cable company's use of NAT would then be a violation of the law, as it would hide the source of your traffic from -their- upstream provider. If this ever becomes law and goes to court, I suspect people will point that out.:-)
But as many slashdot readers will point out, the purpose of laws is not to perpetuate a bad business model.
Guess what? All those home users use their machines at... pretty much the same time... before work and after work. This whole "we rely on people's bandwidth usage being distributed throughout the day" crap is just that.
Simply put, they either have enough bandwidth to give you the bandwidth stated in your contract or they don't. If they don't have the bandwidth to provide it 24x7, then they also don't have enough bandwidth to provide it during peak usage periods. Period.
Besides, the typical throughput of most people with three computers behind a NAT box tends to not be any higher than that of one person anyway, since one person usually means a teenager or college student downloading lots of {porn, warez, mp3s,.. } while three people tends to be a family viewing web sites. Or if there's only one person involved, then odds are he/she is only using one of those machines at a time and is using NAT to save downtime when switching.
Actually, it is somewhat more complicated than that. You don't own your image. Not in any way. I can take photographs in a public place and use them in a TV show, and you have no right to sue if you happen to be in them, because you have no expectation of a right to privacy in a public place. If I took a photograph of you through your bedroom window, you would have the right to sue for a privacy violation (and possibly for other things like defamation, but that's a separate issue).
There are many reasons why you might have to sign a waiver like the one you mention. The most obvious reason is that you paid the photographer to do a service on your behalf with no expectation that the photographer would claim any rights to said photo. Because it is a work for hire, you have reasonable expectations to own the rights to that photo unless you sign those right away.
You are technically correct that anyting you create (apart from works for hire) is not only yours, but is implicitly copyrighted. This gives you the right to sue for actual damages. To sue for punitive damages, however, you must explicitly file the paperwork.
For example, if someone steals your book and reprints it, you can sue for the profit lost from the sales that would otherwise have gone through your publisher. If you register the copyright, you could also sue for any arbitrary amount above and beyond that to punish the person for said theft and reduce the chances of it happening again.
As for whether the DMCA would protect you from your ISP, the answer is almost certainly "no". It would make it illegal for the ISP to attempt to decrypt the traffic without your knowledge, but that's already probably an invasion of privacy issue, and would be illegal anyway. The ISP, however, would still have the right to refuse to allow you to send any encrypted traffic down the wire, as the packets themselves are not encrypted (with the exception of IPSec, which can still be blocked globally without actually examining the contents of the traffic), only the data within the packets.
More significantly, these laws are criminal, not civil, AFAIK. The ISP couldn't sue under these laws. They could ask the district attorney to prosecute under them, however.
Communications law is extraordinarily complex, and my understanding of it is just from a single class on the topic, though, so the usual "IANAL" caveats apply.
Dial-ups, that's fine, but DSL, such service restrictions suck. That port 25 block is one of the ten or so reasons I left Earthlink. It would be different if they blocked it by default to prevent idiots who don't know what they're doing from leaving open relays, but refusing to remove said blocks is sleazy.
Making it particularly bad is the fact that they didn't start blocking it until well after my first month, at which point I no longer could drop them without penalty until the end of the entire 1 year service term. Make no mistake, Earthlink is the worst ISP I've ever encountered. They make my experience with Pac Bell look positively refreshing by comparison.
Just to pick nits, I've never seen anyone (other than maybe a few open source projects during early development) include two copies of an application. Most apps only benefit from altivec enhancements in a handful of frequently used but easily isolated core processing routines. Thus, they simply take one conditional path on hardware with Altivec and a different path on hardware without it.
Also you don't compile an application for Altivec. You write a separate version of those core routines that is completely rewritten to use vector operations instead of scalar.
But you are right that, to my knowledge, no one has ever bothered to implement altivec emulation traps on the G3. It just doesn't seem that useful.:-)
That's still exhorbitantly expensive. My Epson SC740 costs about $3 for a generic black and about $5 for a generic 3-color tank. And even at that price, it still costs about 100x as much per page as the $200 Brother laser printer that's now sitting next to it... also bought from Dell's web site.
I always thought that news story explained a lot about why he kept babbling incoherently about their award winning customer service when Applehasalmostalwaysratedbetter.
Also, most consumer balk at the cost of toner compared to ink cartidges.
Yeah. Most consumers are so caught up with the initial cost of the cartridge that they forget that they get only about 10% usage out of that ink cartridge unless they use it frequently, but get 100% usage out of the toner cartridge even if the printer sits there for a month without printing.
My cost per page (not including paper stock) for inkjet (based on usage once a week or so): about $0.35/page. My cost per page for a black-and-white laser copy: $0.005/page.
I've stopped even turning on my inkjet printers unless I need color or oversized printing. You'd be amazed how quickly my laser printer will pay for itself.
Low-end laser printers start at under $200. Simply put, the day of the inkjet has passed.
Bad analogy. Cars have laws to prevent just the sort of policies you describe.
If a car company tries to invalidate my warranty because I bought an air filter from Wal-Mart, I have the right to sue the pants off them. They simply don't have the legal right to limit the warranty in this way. (Note that these laws are specific to cars, AFAIK.)
This is in contrast, however, to non-consumables, i.e. repair parts. If I put a part in my engine that isn't an OEM part, sure they won't cover it under warranty. On the other hand, if it is under warranty, the dealer buys those OEM parts as part of the warranty coverage, thus the end-user still doesn't end up buying them.
In either case, your statement is wrong. Or, to coin a phrase based on a line by the late, great, Douglas Adams, computers are almost entirely but not completely totally unlike cars.
Antitrust laws do nothing of the sort. A car manufacturer could buy an oil company if they want to. It's a vertical merger. Antitrust laws would, however, prevent GM and Ford and Nissan and eight other companies from merging into a single huge car company.
A car company, however, can't require you to use only their parts. This has nothing to do with antitrust laws, however, and the laws that prevent such activity only apply to automobiles. Sorry, thanks for playing.
Simply put, what they're doing is legal. Unethical, yes, immoral, probably, consumer-unfriendly, sure, but legal nonetheless. They have a right to do this, just like you have the right to tell them where to shove their products.
That having been said, it wouldn't hurt to do a nationwide advertising campaign that explains to people why they should look for third-party ink refills before choosing a printer---educate the masses. When they see their business drying up, they will reconsider this stupidity....
With one caveat. Express permission can be sold/purchased in the form of a merger, buyout, or other absorbtion of one company by another. If I say that I want info about when the next version of Foo from Foo Systems, Inc. comes out, and they get bought by Real-eBar, and form Real-eFooBar, Inc I still want said information about the Real-eFooBar product.
It's important to be extremely precise when defining these thing to avoid unnecessarily preventing legitimate commerce. If the law is too broad, it will get struck down, and you'll be back to square one---or worse.
Overseas spam is irrelevant unless it is also being generated on behalf of overseas companies. Any well-written spam legislation should make it possible to sue not only the company doing the actual distribution, but also the company on whose behalf they are doing it.
For example, some U.S. company decides they want to spam the world, and contacts an overseas spam outfit. You get the spam. You may not be able to sue the spam outfit, but you can bet your backside that you hvae the right to sue the American company that hired them.
But their problem is still that many people trade mp3s and have no intention of buying the music.
There's a flaw here. Most of the people who trade mp3s who have no intention of buying the music are equally capable of getting that music in other ways --- by taping from the radio, by copying from brick-and-mortar friends, etc.
That means that Napster et al have only served to provide those sorts of people with a broader range of music, as they likely would not have purchased said music even without the existence of such services.
However, the people who download for the purpose of "try-before-you-buy" are also given access to a much broader range of music, and thus are encouraged to buy more music.
File sharing has only hurt music that was not worth buying, in much the same way that listening stations at Blockbuster does, only on a much larger scale. It has changed what music people listen to, not reduced it, and I challenge anyone at the RIAA to prove otherwise.
This year is the 30th anniversary of what we now think of as hard drives, i.e. a sealed box containing the heads and platters, as opposed to separate removable platter stacks.
While many people have said for years that the Winchester drive design would run out of steam "any year now", it has continued to achieve greater and greater areal density with reasonable reliability and steadily decreasing price.
You'd be shocked at the quality of the signal. Those same phone wires can carry DSL signals that shove a megabit plus across the wire. The limitation of phone quality has nothing to do with the line quality, and very little to do with the infrastructure. It has everything to do with the shitty little speakers and microphones they use in consumer telephones.
I used to work in FM radio. We had a relatively low-tech solution for remote broadcasting. It involved a telephone interface box that hooked a decent microphone up to the phone line through hgh quality amp circuitry and an appropriate isolation transformer (or maybe an optical coupler, not sure). The other end was similarly equipped.
The quality was surprisingly good, sufficiently so that we broadcast a (mono only) feed from a live concert once as an experiment. Were it not monaural, I might actually have considered doing it regularly. There was a bit of roll-off in the highs, but it blew the heck out of AM radio quality.
Granted, this was just across campus, and the signal quality of various parts of the telephone network probably varies widely, but most of the quality limitations in telephone are simply poor quality hardware at the endpoints, not inherent limitations in the medium itself.
Yeah, and knowing IBM, I'd imagine their lawyers have already assembled a three foot tall stack of patents that SCO violates.... There's not a company in the world---MS included---that would be more suicidal to sue than IBM.
Good point. When I was in grad school, I ran ftp.mklinux.org out of my dorm room. The IT dept. grumbled a lot about it, but since it was academic research, they let it go. We were noted as serving 20 gigs in a single day on occasion, with typical being between 1 and 3.
If they had set up a similar policy, even if they only charged 1/1000th of a cent per megabyte, it would have cost me on the order of $150/month. At that rate, it's roughly half the cost of leasing your own T1, complete with paying an ISP to service it.
On the other end, if they charged a half cent per meg., my continuous 384 kbps would cost about as much as a full OC3.
Put another way, this is how -not- to keep serious CS students at your university, guys....
Actually, the parent poster is correct. Sony happens to use a different formulation for their DVCAM tapes than they do for their DV tapes. That doesn't prevent other companies from using that formulation for standard DV tapes.
What makes DVCAM superior to DV (mini or otherwise) isn't about the tape itself at all. It's about the tape speed. DVCAM gets 40 minutes on a mini-DV[CAM] tape, while DV gets 60 minutes on the same tape. When you run the transport half again faster, you're gonna get less dropouts. It's that simple.
As to whether "professionals" use mini-DV, the answer is "you bet your ass". For several years, the most popular small field camera for reporters to carry (including at least one national network) was Sony's TR-101, a mini-DV camcorder. It was small, lightweight, offered manual settings on almost everything, and had pretty darn good optical image stabilization. More importantly, it was cheap, so if something happened to it, it wasn't a big loss.
When a reporter's life is on the line, it certainly helps if the reporter isn't afraid to ditch the camera in a pinch.:-)
The point was that the drives "scrub" the disk periodically when they're idle (or so I'm told). Spinning it up and doing a read of a few blocks just doesn't have the same effect, although it would at least minimize the stiction risk.
Now, a periodic "dd if=/dev/hda? of=/dev/null" is another story, and I'd expect that to be sufficient, but I'm not certain....
Even expensive DV tapes used to be pretty unreliable. I'm not sure if they still are. I used Panasonic mini-DV tapes (at the time, $15 apiece) to shoot a movie. At least two of the tapes had missing frames upon replay a few minutes later. That's two out of... oh, ten. Not at the ends of the tape, either. Randomly distributed. Thankfully, I caught the most critical one on the day of the shoot and was able to go back and retake the scene. (The others were covered by using different camera angles for those lines.)
I have two Panasonic DV tapes recorded as recently as December of 2000 with dozens of missing frames caused by severe dropouts. It was one of only two on-tape copies of the master for said movie, and the only copy that's here in California.
Thankfully, when I made the copy of the original, I kept the files on a hard drive. The hard drive data, unlike the DV backup, is intact; the DV original, shot to hell.
Hopefully tapes by other vendors (TDK, Fuji, etc.) are more reliable, but even still, the thought of trusting any valuable data to be adequately backed up by mini-DV tapes makes me shudder.
Slashdot Mirror
Just to clarify a point... as long as it's your connectivity, it is perfectly reasonable for you to use it, even if you happen to be next door, down the block, or in another state. Where it should become illegal is when your next door neighbor uses it from next door, down the block, or in another state....
Did their system status message say the San Jose area? Is Sunnyvale remotely the Oakland area? No, and no. I called them about the first two multi-hour outages. By the third one, I gave up. Why sit on hold for an hour waiting to tell them what they're too clueless to figure out on their own?
EarthLink tech support---another one of the top ten reasons I left Earthlink. Wow, the way this thread is going, I will have listed them all by the time I get to the bottom.... :-)
In an era where most ISPs are seriously having trouble getting enough IP numbers to fill the demand, we're seeing more and more ISPs suddenly being very enthusiastic about NAT....
The cable companies, however, have a different mentality, often because they have different limitations. If you don't allow servers, there's nothing preventing you from giving everyone a ten-net number (10.x.x.x) and NATting it at the ISP level. Some ISPs do this.
Since they then have virtually an unlimited number of IP numbers that they can assign, they have no financial incentive to allow customers to use NAT.
On the flip side of this, though, the cable company's use of NAT would then be a violation of the law, as it would hide the source of your traffic from -their- upstream provider. If this ever becomes law and goes to court, I suspect people will point that out. :-)
Guess what? All those home users use their machines at... pretty much the same time... before work and after work. This whole "we rely on people's bandwidth usage being distributed throughout the day" crap is just that.
Simply put, they either have enough bandwidth to give you the bandwidth stated in your contract or they don't. If they don't have the bandwidth to provide it 24x7, then they also don't have enough bandwidth to provide it during peak usage periods. Period.
Besides, the typical throughput of most people with three computers behind a NAT box tends to not be any higher than that of one person anyway, since one person usually means a teenager or college student downloading lots of {porn, warez, mp3s, .. } while three people tends to be a family viewing web sites. Or if there's only one person involved, then odds are he/she is only using one of those machines at a time and is using NAT to save downtime when switching.
There are many reasons why you might have to sign a waiver like the one you mention. The most obvious reason is that you paid the photographer to do a service on your behalf with no expectation that the photographer would claim any rights to said photo. Because it is a work for hire, you have reasonable expectations to own the rights to that photo unless you sign those right away.
You are technically correct that anyting you create (apart from works for hire) is not only yours, but is implicitly copyrighted. This gives you the right to sue for actual damages. To sue for punitive damages, however, you must explicitly file the paperwork.
For example, if someone steals your book and reprints it, you can sue for the profit lost from the sales that would otherwise have gone through your publisher. If you register the copyright, you could also sue for any arbitrary amount above and beyond that to punish the person for said theft and reduce the chances of it happening again.
As for whether the DMCA would protect you from your ISP, the answer is almost certainly "no". It would make it illegal for the ISP to attempt to decrypt the traffic without your knowledge, but that's already probably an invasion of privacy issue, and would be illegal anyway. The ISP, however, would still have the right to refuse to allow you to send any encrypted traffic down the wire, as the packets themselves are not encrypted (with the exception of IPSec, which can still be blocked globally without actually examining the contents of the traffic), only the data within the packets.
More significantly, these laws are criminal, not civil, AFAIK. The ISP couldn't sue under these laws. They could ask the district attorney to prosecute under them, however.
Communications law is extraordinarily complex, and my understanding of it is just from a single class on the topic, though, so the usual "IANAL" caveats apply.
Making it particularly bad is the fact that they didn't start blocking it until well after my first month, at which point I no longer could drop them without penalty until the end of the entire 1 year service term. Make no mistake, Earthlink is the worst ISP I've ever encountered. They make my experience with Pac Bell look positively refreshing by comparison.
Also you don't compile an application for Altivec. You write a separate version of those core routines that is completely rewritten to use vector operations instead of scalar.
But you are right that, to my knowledge, no one has ever bothered to implement altivec emulation traps on the G3. It just doesn't seem that useful. :-)
That's still exhorbitantly expensive. My Epson SC740 costs about $3 for a generic black and about $5 for a generic 3-color tank. And even at that price, it still costs about 100x as much per page as the $200 Brother laser printer that's now sitting next to it... also bought from Dell's web site.
There's a big difference between a $5 razor and a $150 printer. But then, I might not think so if I still shaved.... :-p
Yeah. Most consumers are so caught up with the initial cost of the cartridge that they forget that they get only about 10% usage out of that ink cartridge unless they use it frequently, but get 100% usage out of the toner cartridge even if the printer sits there for a month without printing.
My cost per page (not including paper stock) for inkjet (based on usage once a week or so): about $0.35/page. My cost per page for a black-and-white laser copy: $0.005/page.
I've stopped even turning on my inkjet printers unless I need color or oversized printing. You'd be amazed how quickly my laser printer will pay for itself.
Low-end laser printers start at under $200. Simply put, the day of the inkjet has passed.
If a car company tries to invalidate my warranty because I bought an air filter from Wal-Mart, I have the right to sue the pants off them. They simply don't have the legal right to limit the warranty in this way. (Note that these laws are specific to cars, AFAIK.)
This is in contrast, however, to non-consumables, i.e. repair parts. If I put a part in my engine that isn't an OEM part, sure they won't cover it under warranty. On the other hand, if it is under warranty, the dealer buys those OEM parts as part of the warranty coverage, thus the end-user still doesn't end up buying them.
In either case, your statement is wrong. Or, to coin a phrase based on a line by the late, great, Douglas Adams, computers are almost entirely but not completely totally unlike cars.
A car company, however, can't require you to use only their parts. This has nothing to do with antitrust laws, however, and the laws that prevent such activity only apply to automobiles. Sorry, thanks for playing.
Simply put, what they're doing is legal. Unethical, yes, immoral, probably, consumer-unfriendly, sure, but legal nonetheless. They have a right to do this, just like you have the right to tell them where to shove their products.
That having been said, it wouldn't hurt to do a nationwide advertising campaign that explains to people why they should look for third-party ink refills before choosing a printer---educate the masses. When they see their business drying up, they will reconsider this stupidity....
Funny, neither of my Epson printers has an integrated print head. Yes, the really, really cheap ones do, but don't over-generalize.
With one caveat. Express permission can be sold/purchased in the form of a merger, buyout, or other absorbtion of one company by another. If I say that I want info about when the next version of Foo from Foo Systems, Inc. comes out, and they get bought by Real-eBar, and form Real-eFooBar, Inc I still want said information about the Real-eFooBar product.
It's important to be extremely precise when defining these thing to avoid unnecessarily preventing legitimate commerce. If the law is too broad, it will get struck down, and you'll be back to square one---or worse.
For example, some U.S. company decides they want to spam the world, and contacts an overseas spam outfit. You get the spam. You may not be able to sue the spam outfit, but you can bet your backside that you hvae the right to sue the American company that hired them.
Mac OS X 11. Oh wait....
http://www.apple.com/macosx/x11/
Umm... not quite what I meant.
There's a flaw here. Most of the people who trade mp3s who have no intention of buying the music are equally capable of getting that music in other ways --- by taping from the radio, by copying from brick-and-mortar friends, etc.
That means that Napster et al have only served to provide those sorts of people with a broader range of music, as they likely would not have purchased said music even without the existence of such services.
However, the people who download for the purpose of "try-before-you-buy" are also given access to a much broader range of music, and thus are encouraged to buy more music.
File sharing has only hurt music that was not worth buying, in much the same way that listening stations at Blockbuster does, only on a much larger scale. It has changed what music people listen to, not reduced it, and I challenge anyone at the RIAA to prove otherwise.
This year is the 30th anniversary of what we now think of as hard drives, i.e. a sealed box containing the heads and platters, as opposed to separate removable platter stacks.
While many people have said for years that the Winchester drive design would run out of steam "any year now", it has continued to achieve greater and greater areal density with reasonable reliability and steadily decreasing price.
You'd be shocked at the quality of the signal. Those same phone wires can carry DSL signals that shove a megabit plus across the wire. The limitation of phone quality has nothing to do with the line quality, and very little to do with the infrastructure. It has everything to do with the shitty little speakers and microphones they use in consumer telephones.
I used to work in FM radio. We had a relatively low-tech solution for remote broadcasting. It involved a telephone interface box that hooked a decent microphone up to the phone line through hgh quality amp circuitry and an appropriate isolation transformer (or maybe an optical coupler, not sure). The other end was similarly equipped.
The quality was surprisingly good, sufficiently so that we broadcast a (mono only) feed from a live concert once as an experiment. Were it not monaural, I might actually have considered doing it regularly. There was a bit of roll-off in the highs, but it blew the heck out of AM radio quality.
Granted, this was just across campus, and the signal quality of various parts of the telephone network probably varies widely, but most of the quality limitations in telephone are simply poor quality hardware at the endpoints, not inherent limitations in the medium itself.
Yeah, and knowing IBM, I'd imagine their lawyers have already assembled a three foot tall stack of patents that SCO violates.... There's not a company in the world---MS included---that would be more suicidal to sue than IBM.
they have to be careful not to stifle innovation
Good point. When I was in grad school, I ran ftp.mklinux.org out of my dorm room. The IT dept. grumbled a lot about it, but since it was academic research, they let it go. We were noted as serving 20 gigs in a single day on occasion, with typical being between 1 and 3.
If they had set up a similar policy, even if they only charged 1/1000th of a cent per megabyte, it would have cost me on the order of $150/month. At that rate, it's roughly half the cost of leasing your own T1, complete with paying an ISP to service it.
On the other end, if they charged a half cent per meg., my continuous 384 kbps would cost about as much as a full OC3.
Put another way, this is how -not- to keep serious CS students at your university, guys....
Actually, the parent poster is correct. Sony happens to use a different formulation for their DVCAM tapes than they do for their DV tapes. That doesn't prevent other companies from using that formulation for standard DV tapes.
:-)
What makes DVCAM superior to DV (mini or otherwise) isn't about the tape itself at all. It's about the tape speed. DVCAM gets 40 minutes on a mini-DV[CAM] tape, while DV gets 60 minutes on the same tape. When you run the transport half again faster, you're gonna get less dropouts. It's that simple.
As to whether "professionals" use mini-DV, the answer is "you bet your ass". For several years, the most popular small field camera for reporters to carry (including at least one national network) was Sony's TR-101, a mini-DV camcorder. It was small, lightweight, offered manual settings on almost everything, and had pretty darn good optical image stabilization. More importantly, it was cheap, so if something happened to it, it wasn't a big loss.
When a reporter's life is on the line, it certainly helps if the reporter isn't afraid to ditch the camera in a pinch.
The point was that the drives "scrub" the disk periodically when they're idle (or so I'm told). Spinning it up and doing a read of a few blocks just doesn't have the same effect, although it would at least minimize the stiction risk.
Now, a periodic "dd if=/dev/hda? of=/dev/null" is another story, and I'd expect that to be sufficient, but I'm not certain....
Even expensive DV tapes used to be pretty unreliable. I'm not sure if they still are. I used Panasonic mini-DV tapes (at the time, $15 apiece) to shoot a movie. At least two of the tapes had missing frames upon replay a few minutes later. That's two out of... oh, ten. Not at the ends of the tape, either. Randomly distributed. Thankfully, I caught the most critical one on the day of the shoot and was able to go back and retake the scene. (The others were covered by using different camera angles for those lines.)
I have two Panasonic DV tapes recorded as recently as December of 2000 with dozens of missing frames caused by severe dropouts. It was one of only two on-tape copies of the master for said movie, and the only copy that's here in California.
Thankfully, when I made the copy of the original, I kept the files on a hard drive. The hard drive data, unlike the DV backup, is intact; the DV original, shot to hell.
Hopefully tapes by other vendors (TDK, Fuji, etc.) are more reliable, but even still, the thought of trusting any valuable data to be adequately backed up by mini-DV tapes makes me shudder.