Well, the point is that it stores diffs of files for incremental backups, rather than backing up the entire file whenever part of it changes. And it keeps old diffs. rsync is fine if you only care about the latest version of the file--but then you aren't protected against human error like truncation or deletion that goes unnoticed for a while. I almost lost my GPG key that way--thankfully I had an old DVD backup using a 7zip file.
I hope you reported the bug. Ones like that may be rarely encountered, and so may go unnoticed until they hurt someone.
I understand that your needs are different. I think there might be a lot of people interested in a group oriented more toward numbers like 50 GB, 80% uptime, >384kbps upstream. With enough nodes in a system like that, the lower upstream bandwidth might not matter as much--it could be almost like BitTorrent, in that many slow downloads add up to a high effective speed. You might think that bandwidth number is too low, but I don't want to use up 50% of my upstream 100% of the time.
I do have several offsite network backup solutions in use, but I like the idea of a community-based, FOSS solution. Is Tahoe a feasible possibility for numbers like these? How many nodes can it scale to? I'm imagining, basically, two-way BitTorrent for backups.
Well, you might have a point, but even your "disaster" assertion is unfounded. You need to elaborate.
What was it like beforehand? Is what we have today better? Surely it is in some ways--whether it's a net gain is the question. Can people eat better today for less money? (Not "do they".) Is the food supply more stable? Is there more variety throughout the year? Is food safer? (Let's not go down the HFCS/fat-Americans route--that's a question of discretion, not possibility.)
What about things besides groceries? Is living in a city necessarily better than a suburb or rural area? Which has better quality of life? What about the freedom to choose--how does that affect quality of life? Some people would hate living in a cramped city; they'd rather own a little bit of land and a house, maybe in a nice neighborhood.
Well, as a home user, I don't see how I can "commit to maintaining at least 95% uptime year after year." I mean, really, they want a multi-year commitment? I could see some sysadmin with a 150mb Internet connection and a bunch of servers, but I'm just a guy with a few computers and DSL. I don't know what my connection or systems will be like in a few years. I just want to share a few gigs of disk and a few kilobytes of bandwidth to backup some data.
> If the products and services I buy are are delivered by roads and road use taxes go up, so will the prices I pay. If we reduce general taxes accordingly, it is a wash for me.
In theory, that might work. In practice, I don't think it would be so simple.
Hey, it'd be cool if the economy wasn't built around the road system. But what is the alternative? Going back in time a few hundred years and not even having a national economy? Not being free to move about the country independently, depending on planes and trains? Or do we need ubiquitous aircars? Or teleportation?
Besides, your assertion that the economy should not be built around the road system is unfounded. Why is that necessarily a bad thing? What if that's just the way it is? I think that makes about as much sense as saying that our diets should not be built around water-based foods, because then we wouldn't be dependent on water. Hey, roads are the lifeblood of the modern world, just like veins and arteries are necessary in our bodies.
We can disagree on what qualifies as true science--that's fine.
But I do stand by my point that these conclusions drawn from mathematical models based upon extrapolations of physical data that's millions of years old cannot be proven, because the models and extrapolations cannot possibly be verified to be accurate.
If one were studying only recent phenomenon and were using such data to extrapolate from, one could cross-reference it with independent records to see if correlation actually was causation--but since no one was keeping such records millions of years ago, all we have are guesses, inferences that seem right but are by definition unprovable.
Since they are, by definition, unprovable, it is foolish to advocate radical changes based upon them. The short history of science has demonstrated that what seems impossible one day may become commonplace the next, or that what seems to be the only possible answer one day turns out to be completely wrong the next.
We need to consider the bigger picture, and that includes looking outside our own perspective and humbly recognizing our limitations. Science and arrogance do not mix.
You're right--although in this case we're talking about the script unaltered, as written by the author.
All of that is why I think copyright boils down to a farce. Originality is a myth. No idea is born in a vacuum. If royalties were paid to everyone who inspired or influenced or contributed to something, we'd still be paying royalties to descendants of cavemen. The only way to draw the line is arbitrarily, and as the current system demonstrates, that's doomed to failure.
For the greater good, copyright should be eliminated, and freedom advocated. Yeah, big upheaval and all that--the world could adapt. In the end, there would still be jobs making movies and music, etc, and society would continue on. Besides, it might even be better if there was less new media being created--there's way more than can be consumed now, yet the attitude seems to be that there's infinite room for more. Until we add more hours to the day, that's not the case.
Yep, greed is their biggest problem. But it looks like they will gradually get their way. Do you think Hulu will still exist in 3 years, or that it will still have free content that's even slightly recent or worth watching?
I think the only hopes we have are community-maintained blocklists and generic blocking rules or surrogates. I don't think it's practical to have JS controls fine-grained enough to disable animations, because in the end page elements can just be repositioned, and then you're going to have to disable JS, period.
Maybe our only real hope is using sites that don't do stuff like that.
It seems like it will be the most featureful, forward-thinking backup software, ever: deduplication across multiple clients, compression, and strong encryption for untrusted storage. He's very keen on unit tests, too, and it has good verify and restore functions. I'm already using it for some things. GPL, of course, so no proprietary lock-in, ever.
I recommend CrashPlan. Even if you don't use their paid server subscription, the software is free and good for local backups or backing up to friends' systems.
CrashPlan is nice (although they haven't fixed an inotify regression for months now, so I'm stuck doing full rescans several times a day). Java is awful, though, you're right. I've never had it used as much memory as you mentioned, but it's usually 200-400 MB. I have it niced and ioniced down as far as possible. If I need to restart it, a simple $(sudo/etc/init.d/crashplan restart) takes care of it--no need to kill the JVM.
The other killer feature of CrashPlan is, if you subscribe to the + service, you have unlimited backup space. You never need to worry about running out of space or having to delete old backups to make room.
Time Machine has plenty of known bugs relating to failed restores, silent data corruption, etc. Just google up some comments on Slashdot and you'll see what I mean. I sure wouldn't trust it as my only backup solution.
It's too bad that Firefox apparently doesn't fall back to system components for video (e.g. GStreamer, libvlc, mplayer). There are some plugins that do so, but whenever I've tried them, they always had bugs that made them unusable. Maybe with Mozilla's new, pragmatic H.264 approach, something like this will happen so video decoding won't depend on Firefox itself.
Limiting JavaScript resource usage would be nice (but extremely complex). But I think that with NoScript and AdBlock one should be able to remove HTML5 animation, video, and audio, right?
They probably want Hulu to "fail", because then they can say, "We tried but it didn't work," and then push people to their own, private platforms with individual subscription fees--or make deals with ISPs to raise rates and bundle subscriptions that don't count against bandwidth quotas.
I think our best alternative is to just do something else. Make our own TV shows, or don't watch TV altogether. Life goes on.
Slashdot Mirror
Well, the point is that it stores diffs of files for incremental backups, rather than backing up the entire file whenever part of it changes. And it keeps old diffs. rsync is fine if you only care about the latest version of the file--but then you aren't protected against human error like truncation or deletion that goes unnoticed for a while. I almost lost my GPG key that way--thankfully I had an old DVD backup using a 7zip file.
I hope you reported the bug. Ones like that may be rarely encountered, and so may go unnoticed until they hurt someone.
Haha, well, I don't mean using existing franchises--I mean "original" ones. I guess there are plenty of low-budget web-only shows already.
I understand that your needs are different. I think there might be a lot of people interested in a group oriented more toward numbers like 50 GB, 80% uptime, >384kbps upstream. With enough nodes in a system like that, the lower upstream bandwidth might not matter as much--it could be almost like BitTorrent, in that many slow downloads add up to a high effective speed. You might think that bandwidth number is too low, but I don't want to use up 50% of my upstream 100% of the time.
I do have several offsite network backup solutions in use, but I like the idea of a community-based, FOSS solution. Is Tahoe a feasible possibility for numbers like these? How many nodes can it scale to? I'm imagining, basically, two-way BitTorrent for backups.
Well, you might have a point, but even your "disaster" assertion is unfounded. You need to elaborate.
What was it like beforehand? Is what we have today better? Surely it is in some ways--whether it's a net gain is the question. Can people eat better today for less money? (Not "do they".) Is the food supply more stable? Is there more variety throughout the year? Is food safer? (Let's not go down the HFCS/fat-Americans route--that's a question of discretion, not possibility.)
What about things besides groceries? Is living in a city necessarily better than a suburb or rural area? Which has better quality of life? What about the freedom to choose--how does that affect quality of life? Some people would hate living in a cramped city; they'd rather own a little bit of land and a house, maybe in a nice neighborhood.
Bottom line: I think you are oversimplifying.
Well, as a home user, I don't see how I can "commit to maintaining at least 95% uptime year after year." I mean, really, they want a multi-year commitment? I could see some sysadmin with a 150mb Internet connection and a bunch of servers, but I'm just a guy with a few computers and DSL. I don't know what my connection or systems will be like in a few years. I just want to share a few gigs of disk and a few kilobytes of bandwidth to backup some data.
> If the products and services I buy are are delivered by roads and road use taxes go up, so will the prices I pay. If we reduce general taxes accordingly, it is a wash for me.
In theory, that might work. In practice, I don't think it would be so simple.
Hey, it'd be cool if the economy wasn't built around the road system. But what is the alternative? Going back in time a few hundred years and not even having a national economy? Not being free to move about the country independently, depending on planes and trains? Or do we need ubiquitous aircars? Or teleportation?
Besides, your assertion that the economy should not be built around the road system is unfounded. Why is that necessarily a bad thing? What if that's just the way it is? I think that makes about as much sense as saying that our diets should not be built around water-based foods, because then we wouldn't be dependent on water. Hey, roads are the lifeblood of the modern world, just like veins and arteries are necessary in our bodies.
We can disagree on what qualifies as true science--that's fine.
But I do stand by my point that these conclusions drawn from mathematical models based upon extrapolations of physical data that's millions of years old cannot be proven, because the models and extrapolations cannot possibly be verified to be accurate.
If one were studying only recent phenomenon and were using such data to extrapolate from, one could cross-reference it with independent records to see if correlation actually was causation--but since no one was keeping such records millions of years ago, all we have are guesses, inferences that seem right but are by definition unprovable.
Since they are, by definition, unprovable, it is foolish to advocate radical changes based upon them. The short history of science has demonstrated that what seems impossible one day may become commonplace the next, or that what seems to be the only possible answer one day turns out to be completely wrong the next.
We need to consider the bigger picture, and that includes looking outside our own perspective and humbly recognizing our limitations. Science and arrogance do not mix.
You're right--although in this case we're talking about the script unaltered, as written by the author.
All of that is why I think copyright boils down to a farce. Originality is a myth. No idea is born in a vacuum. If royalties were paid to everyone who inspired or influenced or contributed to something, we'd still be paying royalties to descendants of cavemen. The only way to draw the line is arbitrarily, and as the current system demonstrates, that's doomed to failure.
For the greater good, copyright should be eliminated, and freedom advocated. Yeah, big upheaval and all that--the world could adapt. In the end, there would still be jobs making movies and music, etc, and society would continue on. Besides, it might even be better if there was less new media being created--there's way more than can be consumed now, yet the attitude seems to be that there's infinite room for more. Until we add more hours to the day, that's not the case.
Yep, greed is their biggest problem. But it looks like they will gradually get their way. Do you think Hulu will still exist in 3 years, or that it will still have free content that's even slightly recent or worth watching?
I think the only hopes we have are community-maintained blocklists and generic blocking rules or surrogates. I don't think it's practical to have JS controls fine-grained enough to disable animations, because in the end page elements can just be repositioned, and then you're going to have to disable JS, period.
Maybe our only real hope is using sites that don't do stuff like that.
I've heard of some nasty Time Machine restore and silent data corruption bugs.
Well, looks like it's not suitable for home users at all. :(
I think you mean hard links.
I'm very excited (about backup software?) about this new backup program from an old buddy of Linus Torvalds':
http://liw.fi/obnam/
It seems like it will be the most featureful, forward-thinking backup software, ever: deduplication across multiple clients, compression, and strong encryption for untrusted storage. He's very keen on unit tests, too, and it has good verify and restore functions. I'm already using it for some things. GPL, of course, so no proprietary lock-in, ever.
That sounds awesome, thanks for sharing it. I didn't see a quick explanation on the wiki: what are the minimum bandwidth and storage contributions?
I recommend CrashPlan. Even if you don't use their paid server subscription, the software is free and good for local backups or backing up to friends' systems.
CrashPlan is nice (although they haven't fixed an inotify regression for months now, so I'm stuck doing full rescans several times a day). Java is awful, though, you're right. I've never had it used as much memory as you mentioned, but it's usually 200-400 MB. I have it niced and ioniced down as far as possible. If I need to restart it, a simple $(sudo /etc/init.d/crashplan restart) takes care of it--no need to kill the JVM.
The other killer feature of CrashPlan is, if you subscribe to the + service, you have unlimited backup space. You never need to worry about running out of space or having to delete old backups to make room.
Time Machine has plenty of known bugs relating to failed restores, silent data corruption, etc. Just google up some comments on Slashdot and you'll see what I mean. I sure wouldn't trust it as my only backup solution.
But Linux is not "very weak when it comes to backup services." Linux has the most reliable, featureful backup software available, period.
Yeah, but apparently Mozilla isn't interested in implementing Pepper.
Sounds like Lightspark is not:
https://github.com/lightspark/lightspark/wiki/Site-Support
It's too bad that Firefox apparently doesn't fall back to system components for video (e.g. GStreamer, libvlc, mplayer). There are some plugins that do so, but whenever I've tried them, they always had bugs that made them unusable. Maybe with Mozilla's new, pragmatic H.264 approach, something like this will happen so video decoding won't depend on Firefox itself.
Limiting JavaScript resource usage would be nice (but extremely complex). But I think that with NoScript and AdBlock one should be able to remove HTML5 animation, video, and audio, right?
They probably want Hulu to "fail", because then they can say, "We tried but it didn't work," and then push people to their own, private platforms with individual subscription fees--or make deals with ISPs to raise rates and bundle subscriptions that don't count against bandwidth quotas.
I think our best alternative is to just do something else. Make our own TV shows, or don't watch TV altogether. Life goes on.
And as a parting shot at Linux users, Adobe introduces a major regression (hardware accelerated video tints everything blue, e.g. YouTube), claims it can't be reproduced, and closes all bug reports about it, leaving users to implement a nasty hack individually.