Trust and privacy reasons, marketing backlash, and recent court decisions mainly. You have to trust that Facebook employees will leave your private data alone, their digital/robotic minions will categorize you in a non-offensive way, and that government officials will never take an interest in you. Some people just don't trust a job like that to another person with unknown motives. Also, if the device does most of the heavy lifting (collection, maintenance, and security) for you, then it becomes much less onerous a task.
Why couldn't it do some kind of integration with existing services, like Facebook and possibly Google? The device/service exposes as much information about you as you allow it to to Google or synchronizes that with Facebook/Myspace/Twitter/SocialNetworkDuJour and then people use that existing service to find you. We would still have the integration problems (Most of my family is non-technical and they love Facebook because it's so brain-dead simple) but it would certainly fix the "how to find me" problem you mentioned.
Really? I would think that's the easiest part to understand. Your device controls your data. Period. Otherwise it's not really your data. It being your server, you can actually decide which external accounts (other friends) get access to particular data. You can sort of ape that setup on existing social networking sites, but since they don't run on your hardware, the admins of those sites have access to everything you have posted on there, regardless of (lack of) access controls.
Think about all the stories and comments on Slashdot about employers claiming ownership of projects developed or housed on company hardware. Have you not read the comments about "It's their computer, so it's their data"? Every cloud computing story seems to have at least a dozen related comments in it.
Not every Nokia phone is going to have a 600MHz ARM and half a gig of RAM behind the screen. Not every phone user is going to shell out the prices higher-end components are going to require. Nokia has the market share they have because of the diversity of their products. They need to cover the high-end as well as the low-end. I could easily see Nokia moving from S40 on the low end and S60 on the high end to S40 for low-end and pre-paid phones (If they don't drop it altogether), S60 in the low-to-medium phones (I think their E63 is a step in that direction), and Maemo 5 on their high-end phones that compete with Androids and iPhones.
I have a Garmin eTrex Legend and I love it. It is WAAS-enabled so I get resolutions down to ~7 feet with clear skies. I like its B/W LCD screen (Easier to read in direct sunlight). It is water proof down to about 3 feet (Though the manual says to take the batteries out after a dunking and let the unit dry) which means it is rain proof, which has been a real boon for me.
I have a bicycle attachment for it which stays on my handle bars and the unit clips into/out of in less than a second. It requires the battery cover of the unit be swapped out, but that's easy enough to not bother me much.
I have a phone (Blackberry Curve) and a PDA (Nokia N810) with GPS but neither are very visible in direct sunlight. They also don't have very good battery life compared to my Legend (15-18 hours on a pair of AA batteries).
I paid $120 for the unit, but I've seen them down to about $80 USD. Many units also allow you to upload maps, which come in quite handy at times.
I use my n810 as an eBook reader all the time. The screen works quite well, and FBreader works like a champ (Even has its own Maemo repository). It doesn't have a book repository, but I haven't need one. The n800 should work about the same, and has a slightly larger screen. I just needed the keyboard.
But then again, I've been using eReaders since my original Palm Pro.
The Borg are not xenophobic. They do not wish merely to crush their enemies or protect themselves. They wish to incorporate other races' biological and technological distinctiveness into the Collective. Destroying Earth before it gives rise to the Federation and thus is a threat to the Borg is a response to fear, one the Borg do not feel.
The story is a bit convoluted, but I think a lot of that is necessary for the reboot they wanted to accomplish. There is a lot of story telling in there, but it's sprinkled around and not played up in most parts. If you're looking carefully, you see it in several spots.
Maybe I was just not distracted because I haven't seen a lot of the other shows that these actors played in, thus I was not experiencing the "Agent Smith" phenomenon.
I thought Spock was well done, very much in line with what I remember of Spock from TOS, Vulcan with enough Human in him to drive him in ways other Vulcans could never grasp. Kirk was very much a young bulldog just coming into his prime, full of testosterone and bravado, but with enough brains to see what no one else seemed to be able to.
This doesn't tear apart the foundations of Star Trek so much as it sweeps aside most everything built on those foundations: Honor, fear in the face of death, duty in the face of insurmountable odds, there is no such thing as a "no win" solution--those are still there. Even when beings die by the planet-load (Hopefully that's not too much of a spoiler), all is not lost and with perseverance, the good guys can still win and the universe can still be saved.
Possibly, especially since its allegorical value has plummeted along with gas prices. Or they may have a character spout a one-liner to cover it. I'm just a little bit curious.
I have to agree. The coolest part of Beast with a Billion Backs was how it tied directly to the end of Bender's Big Score. Other than that, it's my least favorite of the three.
I'm just not sure if or how they'll recover from the Dark Matter situation (Trying to not spoil it for anyone).
I don't know what kinda busted-ass network you run, but I have used Gobby over a local network and over the Internet to work on everything from documentation to source code to HTML and CSS files. It works like a champ. Several other admins and developers have started using it at my company for collaboration, both "extreme programming" type and "can you help me figure out the problem with X?" things.
The ONLY thing I want from Gobby at this point is an easy way to see who is where within the document.
Linux and Windows users (And I think there's an OS X port too) can use Gobby, which is like SubEthaEdit, but free, written in GTK+, includes a free server for collaboration over the net, and zeroconf support for finding users on the local network. Since it's based on GTK+, it has things like syntax highlighting, spellcheck, etc. already available. It should also be in most popular distros' repos already.
You may need a better music player. Mine plays for 16-20 hours continuously and charges in three from a USB port. I took it to Moab last weekend and after probably six to eight hours continuous use, I was still at 3/4 charge.
I do have a solar charger, but mine actually charges an internal LiOn battery, so I don't have to keep connected and keep it in the sun to charge my music player. I just strapped it to the top of my trunk rack (I was headed primarily north) and by the end of the day I still had plenty of power for small devices.
If you get a solar cell, I highly recommend one with an internal battery for power storage (Preferably one that's replaceable, as it WILL wear out eventually) and one that has changeable charging tips so you don't have to carry too much charging gear.
I also hike and have strapped the solar charger to the top of my pack for extra/emergency power. It's really a great tool.
DRDB is nice for a two-system active/passive failover solution, but does nothing for larger needs or active/active situations. For that (From everything I've read) you need a SAN and GFS2, which is all fine and good except we'd also like to be able to replicate between sites for a site failover solution.
I apologize for my outdated knowledge. I very much remember them being read only in the LVM1 days. I imagine I have never made an attempt to mount a snapshot read/write since.
They very much were read-only in LVM1. It was one of the design goals for LVM2 to make snapshots writable.
I would agree that it certainly does a lot, and I would be very upset if someone tried to take it away from me. It is just that we are to a point now where we've squeezed most of the possible functionality out of DM+LVM that we can.
This is where we differ. I don't think DM is at its end. Did you know LVM2 can do its own mirroring (i.e. RAID-1)? You probably do know about the DM-crypt, which makes whole filesystem encryption almost trivial.
I can imagine a system (Maybe LVM3) that allows the filesystem sitting on top of it to request changes to the underlying LV, maybe allocating additional PEs for snapshots or space for additional redundancy for specific objects like RAID-Z, or automatic object replication between PVs over iSCSI.
I checked that out last night, and I'm sort of confused about it. Apparently Tux2 was scuttled by patents, and maybe that's got the information I need, but when I tried to see what features Tux3 will have, I wasn't able to find anything that made much sense.
Apparently the main benefit of tux3 is versioning. Great, wonderful. That'll work well on the big company MSA but what about anything else?
Not to persuade you away from JFS (To each their own), but ext4 fixes the slow deletes (Extents free up a lot of metadata over ext3's method, especially for large files) and the long fsck's (fsck only checks the parts of the FS that are in use and can be run while the fs is online), from what I've read.
While I don't have much to add, I have noticed at least two comments from you in this article stating that LVM snapshots are read-only. This is incorrect.
While LVM+ext[3|4] does have shortcomings, namely replication and versioning, along with a few minor things like tail-packing, LVM2 (Which debuted in Linux 2.4 or 2.6 I forget) does have writable snapshots. I have used them many times for backups as well as cloning VMs.
While I dearly love the CRX (I have owned two, second one ('91 HF) is still in my garage while I look for body panels and a new hood) and will keep this car forever (It is now my geek project car), and while I liked the original Insight (With its diminutive CRX-ish looks), I am excited about the possibilities for the CR-Z, which, according to scuttlebutt will be released in Europe soon and (crossed fingers) will make it to the States soon after that.
I wasn't sure why Intel would choose Fedora over Ubuntu either until I remembered the maintainer tools that Fedora has been working on.
It's not just RPM that Intel is after. Fedora has made a concerted effort over the last three or four releases to provide all the tools a group would need to make their own customized Fedora-derivative distro. I can't remember the software names off the top of my head, but groups like Fedora Unity use them to create more updated "spins" of Fedora releases.
So all Intel has to do now is build their own repository manager server and they can have automated testing, building, and packaging of any packages they want, up to and including the entire distro.
Unless for some reason apt thinks the file is in another directory than it is in your path. I can't tell yout the number of times I've had apt/dpkg barf while trying to find the package that owns a particular file.
What package manager silently downgrades packages?
I can see a package mirror (maliciously) refusing to stock updates, but yum at least picks a mirror at random by default. Apt didn't last I saw, but if you picked your own mirror, you already trust them.
Slashdot Mirror
Trust and privacy reasons, marketing backlash, and recent court decisions mainly. You have to trust that Facebook employees will leave your private data alone, their digital/robotic minions will categorize you in a non-offensive way, and that government officials will never take an interest in you. Some people just don't trust a job like that to another person with unknown motives. Also, if the device does most of the heavy lifting (collection, maintenance, and security) for you, then it becomes much less onerous a task.
Why couldn't it do some kind of integration with existing services, like Facebook and possibly Google? The device/service exposes as much information about you as you allow it to to Google or synchronizes that with Facebook/Myspace/Twitter/SocialNetworkDuJour and then people use that existing service to find you. We would still have the integration problems (Most of my family is non-technical and they love Facebook because it's so brain-dead simple) but it would certainly fix the "how to find me" problem you mentioned.
Really? I would think that's the easiest part to understand. Your device controls your data. Period. Otherwise it's not really your data. It being your server, you can actually decide which external accounts (other friends) get access to particular data. You can sort of ape that setup on existing social networking sites, but since they don't run on your hardware, the admins of those sites have access to everything you have posted on there, regardless of (lack of) access controls.
Think about all the stories and comments on Slashdot about employers claiming ownership of projects developed or housed on company hardware. Have you not read the comments about "It's their computer, so it's their data"? Every cloud computing story seems to have at least a dozen related comments in it.
Not every Nokia phone is going to have a 600MHz ARM and half a gig of RAM behind the screen. Not every phone user is going to shell out the prices higher-end components are going to require. Nokia has the market share they have because of the diversity of their products. They need to cover the high-end as well as the low-end. I could easily see Nokia moving from S40 on the low end and S60 on the high end to S40 for low-end and pre-paid phones (If they don't drop it altogether), S60 in the low-to-medium phones (I think their E63 is a step in that direction), and Maemo 5 on their high-end phones that compete with Androids and iPhones.
I have a Garmin eTrex Legend and I love it. It is WAAS-enabled so I get resolutions down to ~7 feet with clear skies. I like its B/W LCD screen (Easier to read in direct sunlight). It is water proof down to about 3 feet (Though the manual says to take the batteries out after a dunking and let the unit dry) which means it is rain proof, which has been a real boon for me.
I have a bicycle attachment for it which stays on my handle bars and the unit clips into/out of in less than a second. It requires the battery cover of the unit be swapped out, but that's easy enough to not bother me much.
I have a phone (Blackberry Curve) and a PDA (Nokia N810) with GPS but neither are very visible in direct sunlight. They also don't have very good battery life compared to my Legend (15-18 hours on a pair of AA batteries).
I paid $120 for the unit, but I've seen them down to about $80 USD. Many units also allow you to upload maps, which come in quite handy at times.
I use my n810 as an eBook reader all the time. The screen works quite well, and FBreader works like a champ (Even has its own Maemo repository). It doesn't have a book repository, but I haven't need one. The n800 should work about the same, and has a slightly larger screen. I just needed the keyboard.
But then again, I've been using eReaders since my original Palm Pro.
The Borg are not xenophobic. They do not wish merely to crush their enemies or protect themselves. They wish to incorporate other races' biological and technological distinctiveness into the Collective. Destroying Earth before it gives rise to the Federation and thus is a threat to the Borg is a response to fear, one the Borg do not feel.
Instrumental plot device, but otherwise not really explained. It just is.
The story is a bit convoluted, but I think a lot of that is necessary for the reboot they wanted to accomplish. There is a lot of story telling in there, but it's sprinkled around and not played up in most parts. If you're looking carefully, you see it in several spots.
Maybe I was just not distracted because I haven't seen a lot of the other shows that these actors played in, thus I was not experiencing the "Agent Smith" phenomenon.
I thought Spock was well done, very much in line with what I remember of Spock from TOS, Vulcan with enough Human in him to drive him in ways other Vulcans could never grasp. Kirk was very much a young bulldog just coming into his prime, full of testosterone and bravado, but with enough brains to see what no one else seemed to be able to.
This doesn't tear apart the foundations of Star Trek so much as it sweeps aside most everything built on those foundations: Honor, fear in the face of death, duty in the face of insurmountable odds, there is no such thing as a "no win" solution--those are still there. Even when beings die by the planet-load (Hopefully that's not too much of a spoiler), all is not lost and with perseverance, the good guys can still win and the universe can still be saved.
Oh the irony...
Possibly, especially since its allegorical value has plummeted along with gas prices. Or they may have a character spout a one-liner to cover it. I'm just a little bit curious.
I have to agree. The coolest part of Beast with a Billion Backs was how it tied directly to the end of Bender's Big Score. Other than that, it's my least favorite of the three.
I'm just not sure if or how they'll recover from the Dark Matter situation (Trying to not spoil it for anyone).
Quite simple: Because there was a penis involved.
I don't know what kinda busted-ass network you run, but I have used Gobby over a local network and over the Internet to work on everything from documentation to source code to HTML and CSS files. It works like a champ. Several other admins and developers have started using it at my company for collaboration, both "extreme programming" type and "can you help me figure out the problem with X?" things.
The ONLY thing I want from Gobby at this point is an easy way to see who is where within the document.
Linux and Windows users (And I think there's an OS X port too) can use Gobby, which is like SubEthaEdit, but free, written in GTK+, includes a free server for collaboration over the net, and zeroconf support for finding users on the local network. Since it's based on GTK+, it has things like syntax highlighting, spellcheck, etc. already available. It should also be in most popular distros' repos already.
You may need a better music player. Mine plays for 16-20 hours continuously and charges in three from a USB port. I took it to Moab last weekend and after probably six to eight hours continuous use, I was still at 3/4 charge.
I do have a solar charger, but mine actually charges an internal LiOn battery, so I don't have to keep connected and keep it in the sun to charge my music player. I just strapped it to the top of my trunk rack (I was headed primarily north) and by the end of the day I still had plenty of power for small devices.
If you get a solar cell, I highly recommend one with an internal battery for power storage (Preferably one that's replaceable, as it WILL wear out eventually) and one that has changeable charging tips so you don't have to carry too much charging gear.
I also hike and have strapped the solar charger to the top of my pack for extra/emergency power. It's really a great tool.
DRDB is nice for a two-system active/passive failover solution, but does nothing for larger needs or active/active situations. For that (From everything I've read) you need a SAN and GFS2, which is all fine and good except we'd also like to be able to replicate between sites for a site failover solution.
I apologize for my outdated knowledge. I very much remember them being read only in the LVM1 days. I imagine I have never made an attempt to mount a snapshot read/write since.
They very much were read-only in LVM1. It was one of the design goals for LVM2 to make snapshots writable.
I would agree that it certainly does a lot, and I would be very upset if someone tried to take it away from me. It is just that we are to a point now where we've squeezed most of the possible functionality out of DM+LVM that we can.
This is where we differ. I don't think DM is at its end. Did you know LVM2 can do its own mirroring (i.e. RAID-1)? You probably do know about the DM-crypt, which makes whole filesystem encryption almost trivial.
I can imagine a system (Maybe LVM3) that allows the filesystem sitting on top of it to request changes to the underlying LV, maybe allocating additional PEs for snapshots or space for additional redundancy for specific objects like RAID-Z, or automatic object replication between PVs over iSCSI.
I checked that out last night, and I'm sort of confused about it. Apparently Tux2 was scuttled by patents, and maybe that's got the information I need, but when I tried to see what features Tux3 will have, I wasn't able to find anything that made much sense.
Apparently the main benefit of tux3 is versioning. Great, wonderful. That'll work well on the big company MSA but what about anything else?
Not to persuade you away from JFS (To each their own), but ext4 fixes the slow deletes (Extents free up a lot of metadata over ext3's method, especially for large files) and the long fsck's (fsck only checks the parts of the FS that are in use and can be run while the fs is online), from what I've read.
While I don't have much to add, I have noticed at least two comments from you in this article stating that LVM snapshots are read-only. This is incorrect.
While LVM+ext[3|4] does have shortcomings, namely replication and versioning, along with a few minor things like tail-packing, LVM2 (Which debuted in Linux 2.4 or 2.6 I forget) does have writable snapshots. I have used them many times for backups as well as cloning VMs.
LVM2 is capable of a lot.
While I dearly love the CRX (I have owned two, second one ('91 HF) is still in my garage while I look for body panels and a new hood) and will keep this car forever (It is now my geek project car), and while I liked the original Insight (With its diminutive CRX-ish looks), I am excited about the possibilities for the CR-Z, which, according to scuttlebutt will be released in Europe soon and (crossed fingers) will make it to the States soon after that.
I wasn't sure why Intel would choose Fedora over Ubuntu either until I remembered the maintainer tools that Fedora has been working on.
It's not just RPM that Intel is after. Fedora has made a concerted effort over the last three or four releases to provide all the tools a group would need to make their own customized Fedora-derivative distro. I can't remember the software names off the top of my head, but groups like Fedora Unity use them to create more updated "spins" of Fedora releases.
So all Intel has to do now is build their own repository manager server and they can have automated testing, building, and packaging of any packages they want, up to and including the entire distro.
Unless for some reason apt thinks the file is in another directory than it is in your path. I can't tell yout the number of times I've had apt/dpkg barf while trying to find the package that owns a particular file.
What package manager silently downgrades packages?
I can see a package mirror (maliciously) refusing to stock updates, but yum at least picks a mirror at random by default. Apt didn't last I saw, but if you picked your own mirror, you already trust them.