Really? You must have been paying a very low price for Crashplan Home if their Business plan is 5x for you.
I was paying $5 a month, and under the new plan, would be paying $10. That's still a lot better than with Mozy and its high per-gig pricing, and pretty close to Carbonite (which has inferior features, esp. with its lack of Linux support). For my backup needs, Amazon S3 storage would also cost about $10 a month (with bring-your-own backup software), and Backblaze B2 would be about $2.50 a month. For either of those, I would need something like Arq or Duplicati to do the backups, and wouldn't have much in the way of customer service.
I am still weighing my options, but I may well sign up for Crashplan Business, if it looks like the company is doing OK financially.
Nope, Fiber To The Node (DSL or cable head-end), as opposed to Fiber To The Home. FTTN gives almost all the increased bandwidth benefits of FTTH at a fraction of the trenching costs.
Since most homeowners (in the US anyway) still think the max bandwidth of about 20 - 50 Mbps (theoretical) on DSL over twisted-pair copper would be fantastic, ISPs have little incentive to deploy more expensive infrastructure.
You're right, @alphatel, we would never offer Exxon 2 million dollars to reduce domestic oil prices. Clearly that's thinking too small. How about over $3 billion a year in Oil Exploration tax exemptions for Exxon and other oil companies? Nevermind the fact that oil prices have been so high the last decade, the oil companies have been falling over themselves to do oil well exploration and proofing, and would continue to do so, with or without the subsidies.
But yeah, I'm sure *this* $2 million is a waste, even if it serves to reduce the perceived need for that $3 billion annual subsidy to big oil giants.
As long as a quorum of your family/friends are available, you can access the files. Think of a distributed RAID using something like PAR to create enough excess parity to handle the loss of a number of blocks of each file. It is encrypted in such a way that none of the other nodes can reconstruct your files, or you theirs.
> AGW is probably the most extraordinary claim in the history of extraordinary > claims and the proposed solution (seizing most of the world's wealth, > eliminating most of the current industrial base, etc.) is so far beyond > extraordinary
While I feel way too underqualified to judge the science of AGW, I don't understand these oft-repeated hyperbolic claims that carbon reduction strategies will "eliminate most of the current industrial base."
On the contrary, even in the absence of evidence of AGW, most of our strategies are just common economic sense and good health policy -- reduce dependence on oil (which will continue to rise in price as demand outstrips supply); decrease destructive mountaintop coal mining (which imperils the health of countless rural residents); improve public transportation availability and usage; improve our electrical grid and accommodate supply elasticity; etc
Where carbon reduction or carbon taxes would add costs to some manufacturing industries, it will also create economic opportunities in new technology, alternative energy production, new batteries, nuclear energy, etc.
Most of the posts here are focusing on the snapshotting feature of this copy-on-write filesystem. However, if you read the publications from the website, it's pretty clear that the main problems they are trying to solve are data security and secure deletions.
The innovations that they describe in their papers are that they are using authenticated encryption (that is, not starightforward symmetric block cyphers) via a cryptographic device driver, and that they can *securely* delete a file, in one or all snapshots, with very good performance by overwriting just a small portion of the file (the "stub"). This is accomplished, fairly cleverly, by using an encryption scheme dubbed All or Nothing (AON), which will make recover of the plaintext intractible if even a small portion of the ciphertext is missing.
Pretty cool really. Now, I would agree that their current interface of on-demand snapshots, and their semantics for accessing prior versions need a bit of work to achieve a decent level of practical usability. And I definitely would not expect wonderous performance, compared to a unencrypted snapshot system, like LVM.
My main complaint about Google's product releases is not their scattershot approach -- I'm happy to see them try to find ways to improve existing product niches.
But they rarely seem to update their online products:
* Gmail, despite its strong launch and obvious success, has seen little development since. By now, we would expect to see much stronger import/export features, more filtering and junk mail controls...
* Google Video was pretty weak at launch, and amazingly, hasn't improved much since. Details on the videos shown is weak, and 3rd-party review links, imdb links, etc. are nonexistent. Methods for transferring and showing the video on portable devices and Tivo are... completely absent.
* Froogle, News, Maps, and more have stagnated since their beta launch (except that Google's purchase of new imagery for Earth has benefited Maps), despite much improvement from the competition (seen Yahoo Maps lately?).
In fact, pretty much the only products they regularly update are the native apps they purchased from startups, like Earth, Picasa, and Sketchup. These appear to have kept their development teams from pre-acquisition days, and continue to make small but regular improvements.
It's amazing to me that a company with as many employees as Google can make so many online services appear to be the work of one or two developers in their spare time -- strong on concept, but weak on follow-through.
Not if anyone but Sony changes it. Sony owns the CD Trademark AFAIK and thus can simply declare the new format the "CD".
Only if you spell Sony "P-H-I-L-I-P-S", as Philips is the actual owner of the Compact Disc trademark, and is not a record label.
Philips is serious about maintaining CD compatibility, and has forced the purveyors of incompatiple DRM schemes to clearly label that they are not compatible with the standard.
If you RTFA, you would see that Harald Welte is a developer and copyright holder of netfilter, which is used in a number of commercial firewall products. He also has license to prosecute the copyrights of some other developers.
So yes, he has standing to both warn and sue the companies he has given notice to (as well as the companies that have settled with gpl-violations.org).
"Well, then maybe it is up to those who want to use P2P to share legal files to set up a way to police the networks so that illegal materials are NOT shared."
I'm not sure if you're trying to be a troll, or if you RTFA'd, but that is the exact argument these professors are fighting against. The whole point is that if you look for copyright infringement in any file transfer technology, you can find it.
If you force the designers of any such system to preemptively include filtering technology to enforce the copyright for a limited group of people, and to update the technology of those filters to adapt to means taken by users to defeat them, then you will force all applications to suck.
Imagine if the only way to send a file in SCP were to check that it wasn't encrypted or encoded with steganography (encrypted files must be presumed to be hiding copyrighted materials, and be banned), uncompress archives, run a series of energy-sign-fingerprinting technologies on each file, send the resulting hashes to both MPAA and RIAA authorization servers, wait for the OK, and then upload the file. Would you still want to use SCP to archive your system logfiles?
You may think that my argument is a strawman, but it is actually only a logical extension of the plaintiffs' arguments. As soon as they find any large number of people transferring songs via SCP, they would like to force the authors to either pay damages, or incorporate constantly changing filtering technologies into the application.
In the US, you can purchase a Canon Digital Rebel (EOS 300D) for $1000 with the 'kit' lense. Without the lense, I have seen them going for nearly $700 on various web-sites.
I second that. I just got a 300D without lens, added the excellent 50/1.4 prime lense and the 28-135 image-stabilizing macro lens from the folks at BeachCamera. Price after $390 rebate: $1040 shipped.
Now sure, you've still got other accessory costs: a couple 1GB ultra compact flash cards, filters for both sizes of lense, gadget bag to hold it all, tripod...
In all, it's a fair guess that I'll spend more than twice the amount of my $640 camera purchase. But 2x is a good rule of thumb for DSLR, unless you already have all the accessories.
Well, first you would have to known which types of polling equipment was used.
Actually, someone *has* done this. The report considered definitive on defective voting was done by CalTech and MIT:
Using an average of "residual votes" weighted by the prevalence of the voting equipment type, they found that about 2.1 percent of all ballots resulted in an uncounted vote.
I'm going to go out on a limb and say that at least one state (and quite possibly enough electoral votes to swing the election either way) will be "decided" with a margin of victory lower than 2.1%.
Excuse my lack of knowledge on the IP addresses, but cant you determine which country by the IP addy?
Nope. You can at most tell the regional NIC doing the assigning (EU, APAC, NA) and the company to which the block was assigned.
NICs like RIPE and ARIN do not place restrictions upon where within your network you may use the block of addresses. So, if RIPE assigns my ISP a/18 for use within the EU, I could assign addresses from that block to customers in Spain, UK, France, etc.
My peers would only insist (or maybe only ask nicely) that I advertise the block with an EU Autonomous System (AS) also obtained from RIPE.
I agree with the parent. With the exception of multimedia (which I care little for on my desktop -- that's why I have a Tivo), most things are easier to do under Linux.
* pipe and output redirection that work. I just can't function under almost any of the shells available for Win (except by installing cygwin with zsh). * more CLI goodness: history, programmable completion, lots of parsing commands that work well together (cut, grep, awk, sort). * the availablity of software: need a drawing package? "apt-get install inkscape". need to manage your databases? "apt-get install tora". There is just no equivalent under windows. Whenever I need a new tool, I just look on freshmeat and sourceforge to find what I need, and then locate it in an apt4rpm repository. * fabulous built-in handling of postscript. I can generate postscript (and therefore PDF) from any program, view ps with the software that comes with the distro, and import it into many programs. * portability of my environment. As I go through many different computers and distros, I can backup the ASCII files that represent both my global config and personal settings, and install them onto my new computer. I understand WinXP now has a rudimentary ability to move your DocumentsAndSettings between computers, but you're still stuck with the registry and Progra~1.
Wow, do you think we could implement this with headers called Message-ID and In-Reply-To? And allow users to implement filters on the In-Reply-To or References headers?
Perhaps we could even create an RFC and give it the number 2822.
And if someone would only write a document describing how to correctly implement these headers in MUAs, we'd really be in business.
</sarcasm>
Really, it's a wonder that most mail clients make all of this so hard. Even Mozilla gets threading wrong, by refusing to allow them to be sorted by anything but Sent date, and always sorting them in your message list by the date of the *oldest* message in the thread, rather than the newest. It makes threads practically useless.
Despite my caustic comments above, it doesn't help that many popular client (like those by MS) don't properly implement In-Reply-To or References. As a result, most clients simply guess at threads by looking at Subjects.
The question should be, "Why Compulsory Licenses?"
on
Why Only Music?
·
· Score: 2, Insightful
The author never stops to consider whether compulsory licenses are a good thing, instead attempting to find as many forms of content to apply them to as possible.
I don't accept that compulsory licences on music are a good thing, and I *know* they would be bad for software. Compulsory license regimes create a single large collection agency that gets to rake in money for doing nothing, and dole out money to every content creator in the related industry.
What if you write software and don't want to have all payments go through a fee-stealing middleman who allocates payments based upon rigged marketshare numbers? (See Arbitron and Payola)
What if you don't want to pay a tax (on income? on broadband connections? on filesharing software?) in order to continue to subsidize the corrupt middlemen in inefficient industries, and to compensate for the "stealing" (perceived or real) that someone else wants to do? What if the compulsory tax is estimated based upon the RIAA's guess of how many songs were downloaded each day over all filesharing networks? Nevermind whether the downloaded songs had already been paid for once, twice, or a dozen times by the same user.
Compulsory licenses should not be viewed as an ideal way to obtain free access to all of the content that corporations want to lock up indefinitely in closed formats. They are an occasionally necessary evil in relationships between one bloated industry and another (see broadcasters and RIAA).
What gets lost in these arugments of "what's good for the goose is good for the gander" is that the current system of tariffs is not sustainable. Only a several billion dollar-a-year company can afford to comply with a the system of taxation that varies completely from one zip code to the next.
The tariffs were all designed for companies whose service is tied to a physical location, as are legacy telephony services. They flat-out don't work for a small provider that uses VoIP and whose customers could be located just about anywhere. The most inexpensive decent CDR-processing systems for performing called-number rating cost millions. That kind of barrier to entry will guarantee that we never see improvements in our local or long distance phone service.
The only reason to perpetuate the current system of local and state-defined tariffs is to guarantee that every level of government can get a piece of your phone bill.
The reason governments (federal and local) want to tax VoIP like the PSTN is that they want to continue to milk their tax revenues, which they see slipping away as determining the point of service delivery becomes more abstract.
Slashdot Mirror
Rob's humor and contributions to linux advocacy will be missed. His early presence on Slashdot made it my go-to news site for many years.
My condolences to his family. Thanks for sharing him with us for all these years.
Really? You must have been paying a very low price for Crashplan Home if their Business plan is 5x for you.
I was paying $5 a month, and under the new plan, would be paying $10. That's still a lot better than with Mozy and its high per-gig pricing, and pretty close to Carbonite (which has inferior features, esp. with its lack of Linux support). For my backup needs, Amazon S3 storage would also cost about $10 a month (with bring-your-own backup software), and Backblaze B2 would be about $2.50 a month. For either of those, I would need something like Arq or Duplicati to do the backups, and wouldn't have much in the way of customer service.
I am still weighing my options, but I may well sign up for Crashplan Business, if it looks like the company is doing OK financially.
Nope, Fiber To The Node (DSL or cable head-end), as opposed to Fiber To The Home. FTTN gives almost all the increased bandwidth benefits of FTTH at a fraction of the trenching costs.
Since most homeowners (in the US anyway) still think the max bandwidth of about 20 - 50 Mbps (theoretical) on DSL over twisted-pair copper would be fantastic, ISPs have little incentive to deploy more expensive infrastructure.
Nice set of practices.
You're right, @alphatel, we would never offer Exxon 2 million dollars to reduce domestic oil prices. Clearly that's thinking too small. How about over $3 billion a year in Oil Exploration tax exemptions for Exxon and other oil companies? Nevermind the fact that oil prices have been so high the last decade, the oil companies have been falling over themselves to do oil well exploration and proofing, and would continue to do so, with or without the subsidies.
But yeah, I'm sure *this* $2 million is a waste, even if it serves to reduce the perceived need for that $3 billion annual subsidy to big oil giants.
You're describing the key ideas of Tahoe LAFS:
http://en.wikipedia.org/wiki/Tahoe_Least-Authority_Filesystem
As long as a quorum of your family/friends are available, you can access the files. Think of a distributed RAID using something like PAR to create enough excess parity to handle the loss of a number of blocks of each file. It is encrypted in such a way that none of the other nodes can reconstruct your files, or you theirs.
> AGW is probably the most extraordinary claim in the history of extraordinary
> claims and the proposed solution (seizing most of the world's wealth,
> eliminating most of the current industrial base, etc.) is so far beyond
> extraordinary
While I feel way too underqualified to judge the science of AGW, I don't understand these oft-repeated hyperbolic claims that carbon reduction strategies will "eliminate most of the current industrial base."
On the contrary, even in the absence of evidence of AGW, most of our strategies are just common economic sense and good health policy -- reduce dependence on oil (which will continue to rise in price as demand outstrips supply); decrease destructive mountaintop coal mining (which imperils the health of countless rural residents); improve public transportation availability and usage; improve our electrical grid and accommodate supply elasticity; etc
Where carbon reduction or carbon taxes would add costs to some manufacturing industries, it will also create economic opportunities in new technology, alternative energy production, new batteries, nuclear energy, etc.
I found the following logon URL within the Chase site, and bookmarked it. I now use it as my exclusive means of signing on:
s o_logon.jsp
https://chaseonline.chase.com/chaseonline/logon/s
[Warning, Slashcode inserts whitespace within long URLs, though not in the href]
I don't trust their unsecured frontpage worth a damn.
Thanks,
--kirby
Most of the posts here are focusing on the snapshotting feature of this copy-on-write filesystem. However, if you read the publications from the website, it's pretty clear that the main problems they are trying to solve are data security and secure deletions.
The innovations that they describe in their papers are that they are using authenticated encryption (that is, not starightforward symmetric block cyphers) via a cryptographic device driver, and that they can *securely* delete a file, in one or all snapshots, with very good performance by overwriting just a small portion of the file (the "stub"). This is accomplished, fairly cleverly, by using an encryption scheme dubbed All or Nothing (AON), which will make recover of the plaintext intractible if even a small portion of the ciphertext is missing.
Pretty cool really. Now, I would agree that their current interface of on-demand snapshots, and their semantics for accessing prior versions need a bit of work to achieve a decent level of practical usability. And I definitely would not expect wonderous performance, compared to a unencrypted snapshot system, like LVM.
--kirby
My main complaint about Google's product releases is not their scattershot approach -- I'm happy to see them try to find ways to improve existing product niches.
But they rarely seem to update their online products:
* Gmail, despite its strong launch and obvious success, has seen little development since. By now, we would expect to see much stronger import/export features, more filtering and junk mail controls...
* Google Video was pretty weak at launch, and amazingly, hasn't improved much since. Details on the videos shown is weak, and 3rd-party review links, imdb links, etc. are nonexistent. Methods for transferring and showing the video on portable devices and Tivo are... completely absent.
* Froogle, News, Maps, and more have stagnated since their beta launch (except that Google's purchase of new imagery for Earth has benefited Maps), despite much improvement from the competition (seen Yahoo Maps lately?).
In fact, pretty much the only products they regularly update are the native apps they purchased from startups, like Earth, Picasa, and Sketchup. These appear to have kept their development teams from pre-acquisition days, and continue to make small but regular improvements.
It's amazing to me that a company with as many employees as Google can make so many online services appear to be the work of one or two developers in their spare time -- strong on concept, but weak on follow-through.
--kirby
http://seamless.sigtronica.org.nyud.net:8090/
--kirby
Philips is serious about maintaining CD compatibility, and has forced the purveyors of incompatiple DRM schemes to clearly label that they are not compatible with the standard.
See, e.g., http://www.spectacle.org/0702/evan.html
--kirby
If you RTFA, you would see that Harald Welte is a developer and copyright holder of netfilter, which is used in a number of commercial firewall products. He also has license to prosecute the copyrights of some other developers.
So yes, he has standing to both warn and sue the companies he has given notice to (as well as the companies that have settled with gpl-violations.org).
--kirby
"Well, then maybe it is up to those who want to use P2P to share legal files to set up a way to police the networks so that illegal materials are NOT shared."
I'm not sure if you're trying to be a troll, or if you RTFA'd, but that is the exact argument these professors are fighting against. The whole point is that if you look for copyright infringement in any file transfer technology, you can find it.
If you force the designers of any such system to preemptively include filtering technology to enforce the copyright for a limited group of people, and to update the technology of those filters to adapt to means taken by users to defeat them, then you will force all applications to suck.
Imagine if the only way to send a file in SCP were to check that it wasn't encrypted or encoded with steganography (encrypted files must be presumed to be hiding copyrighted materials, and be banned), uncompress archives, run a series of energy-sign-fingerprinting technologies on each file, send the resulting hashes to both MPAA and RIAA authorization servers, wait for the OK, and then upload the file. Would you still want to use SCP to archive your system logfiles?
You may think that my argument is a strawman, but it is actually only a logical extension of the plaintiffs' arguments. As soon as they find any large number of people transferring songs via SCP, they would like to force the authors to either pay damages, or incorporate constantly changing filtering technologies into the application.
--kirby
I second that. I just got a 300D without lens, added the excellent 50/1.4 prime lense and the 28-135 image-stabilizing macro lens from the folks at BeachCamera. Price after $390 rebate: $1040 shipped.
Now sure, you've still got other accessory costs: a couple 1GB ultra compact flash cards, filters for both sizes of lense, gadget bag to hold it all, tripod...
In all, it's a fair guess that I'll spend more than twice the amount of my $640 camera purchase. But 2x is a good rule of thumb for DSLR, unless you already have all the accessories.
--kirby
Reliability of Existing Voting Equipment
Using an average of "residual votes" weighted by the prevalence of the voting equipment type, they found that about 2.1 percent of all ballots resulted in an uncounted vote.
I'm going to go out on a limb and say that at least one state (and quite possibly enough electoral votes to swing the election either way) will be "decided" with a margin of victory lower than 2.1%.
--kirby
Nope. You can at most tell the regional NIC doing the assigning (EU, APAC, NA) and the company to which the block was assigned.
NICs like RIPE and ARIN do not place restrictions upon where within your network you may use the block of addresses. So, if RIPE assigns my ISP a /18 for use within the EU, I could assign addresses from that block to customers in Spain, UK, France, etc.
My peers would only insist (or maybe only ask nicely) that I advertise the block with an EU Autonomous System (AS) also obtained from RIPE.
--kirby
I agree with the parent. With the exception of multimedia (which I care little for on my desktop -- that's why I have a Tivo), most things are easier to do under Linux.
* pipe and output redirection that work. I just can't function under almost any of the shells available for Win (except by installing cygwin with zsh).
* more CLI goodness: history, programmable completion, lots of parsing commands that work well together (cut, grep, awk, sort).
* the availablity of software: need a drawing package? "apt-get install inkscape". need to manage your databases? "apt-get install tora". There is just no equivalent under windows. Whenever I need a new tool, I just look on freshmeat and sourceforge to find what I need, and then locate it in an apt4rpm repository.
* fabulous built-in handling of postscript. I can generate postscript (and therefore PDF) from any program, view ps with the software that comes with the distro, and import it into many programs.
* portability of my environment. As I go through many different computers and distros, I can backup the ASCII files that represent both my global config and personal settings, and install them onto my new computer. I understand WinXP now has a rudimentary ability to move your DocumentsAndSettings between computers, but you're still stuck with the registry and Progra~1.
--kirby
Wow, do you think we could implement this with headers called Message-ID and In-Reply-To? And allow users to implement filters on the In-Reply-To or References headers?
Perhaps we could even create an RFC and give it the number 2822.
And if someone would only write a document describing how to correctly implement these headers in MUAs, we'd really be in business.
</sarcasm>
Really, it's a wonder that most mail clients make all of this so hard. Even Mozilla gets threading wrong, by refusing to allow them to be sorted by anything but Sent date, and always sorting them in your message list by the date of the *oldest* message in the thread, rather than the newest. It makes threads practically useless.
Despite my caustic comments above, it doesn't help that many popular client (like those by MS) don't properly implement In-Reply-To or References. As a result, most clients simply guess at threads by looking at Subjects.
The author never stops to consider whether compulsory licenses are a good thing, instead attempting to find as many forms of content to apply them to as possible.
I don't accept that compulsory licences on music are a good thing, and I *know* they would be bad for software. Compulsory license regimes create a single large collection agency that gets to rake in money for doing nothing, and dole out money to every content creator in the related industry.
What if you write software and don't want to have all payments go through a fee-stealing middleman who allocates payments based upon rigged marketshare numbers? (See Arbitron and Payola)
What if you don't want to pay a tax (on income? on broadband connections? on filesharing software?) in order to continue to subsidize the corrupt middlemen in inefficient industries, and to compensate for the "stealing" (perceived or real) that someone else wants to do? What if the compulsory tax is estimated based upon the RIAA's guess of how many songs were downloaded each day over all filesharing networks? Nevermind whether the downloaded songs had already been paid for once, twice, or a dozen times by the same user.
Compulsory licenses should not be viewed as an ideal way to obtain free access to all of the content that corporations want to lock up indefinitely in closed formats. They are an occasionally necessary evil in relationships between one bloated industry and another (see broadcasters and RIAA).
--kirby
The tariffs were all designed for companies whose service is tied to a physical location, as are legacy telephony services. They flat-out don't work for a small provider that uses VoIP and whose customers could be located just about anywhere. The most inexpensive decent CDR-processing systems for performing called-number rating cost millions. That kind of barrier to entry will guarantee that we never see improvements in our local or long distance phone service.
The only reason to perpetuate the current system of local and state-defined tariffs is to guarantee that every level of government can get a piece of your phone bill.
The reason governments (federal and local) want to tax VoIP like the PSTN is that they want to continue to milk their tax revenues, which they see slipping away as determining the point of service delivery becomes more abstract.
--kirby