They're also basically identical to AT&T's pricing and tiers, so it's not "taking a crap all over an iPhone killer" like another commenter claimed it to be.
Tell me how this is "taking a giant crap" over an iPhone killer?
Extra charge for a provider-provisioned Exchange server? Yup, AT&T has that too. I'm 90% positive that non-provider-provisioned exchange servers aren't blocked by VZW (they aren't with AT&T).
$60/month total for a 5GB tethering plan? Same as AT&T. (Note, last I checked, there was no official support whatsoever for iPhone tethering.)
FYI I am an AT&T customer, and left VZW two years ago a very unhappy customer, but the summary is unjustly misworded to paint VZW extremely negatively when in reality their data pricing is no different than AT&T. (Which is an improvement, VZW's data pricing used to be AWFUL compared to AT&T.)
Most likely there's an error in the summary. It is more likely:
"An extra charge for Verizon to provide an Exchange server for you."
This is how AT&T works. It's something like $10 extra if you want an Exchange server account provisioned for you by AT&T, but I have no problem using Google Sync with my AT&T account.
Also, AT&T's tethering plan is the exact same price as Verizon's - $60/month (total on top of voice plan) for 5GB.
Actually, the article linked that provoked this discussion pretty much says it:
"The company tried to soften the blow by saying it will add employees in some areas, such as wireless, video and broadband, in order to meet customer demand. Many non-management employees who are affected will have a guaranteed job offer as stipulated in union contracts, AT&T said."
Looks like AT&T's layoffs were in divisions OTHER than wireless. Probably the oldskool long-distance telephone market - not a good place to be nowadays.
Yup. I know of a company out West that's laying off 800 employees, meanwhile in certain departments, they're hiring employees. (General thing seems to be MechEs/general support staff are getting laid off, EEs with RF experience are getting hired.)
Sometimes layoffs are less about downsizing than about skillset readjustment. Many of those 12000 may have been in AT&T's broadband and landline divisions with no possible place in AT&T Wireless.
Also, as others have said, bandwidth doesn't necessarily scale with employees, in this case the limiting factors are: 1) Sites where new towers are permitted. People bitch about coverage and network performance but scream even louder when someone wants to put a cell site nearby. Even if it's "pretty-ized", people bitch about that evil RF that might possibly give them cancer, even though 1 minute in sunlight is going to increase your chances of cancer more than a lifetime of RF exposure at any reasonable distance from a cell tower. (Reasonable distance being defined as "you weren't a dumbass that broke the lock on the fence surrounding the tower and climbed the tower".)
2) FCC licenses for new towers take a lot of paperwork and time
3) Equipment costs - this is nowhere near as much of a problem as 1) and 2) are though, especially 1)
There's one thing about the iPhone situation that really annoys me:
AT&T has tiered pricing for data plans for phones based on its capabilities (i.e. estimating just how much of that "unlimited" data you will actually use depending on whether the phone is a PDA phone, or just a dumbphone, etc.) Tethering users, of course, get hit the hardest ($60/mo and they actually have a specifically documented usage cap). PDA users with keyboards get hit second hardest (formerly $40/mo, now down to $30/mo).
iPhone users, despite being probably the heaviest data users on the network, are in the second least expensive data pricing tier ($20/mo I believe, with "dumbphones" with basic WAP browsers the only thing lower on the list at $15 I think.)
As long as there are graduate students in school worried about whether they'll have a job when they graduate, I have a feeling that providing additional funding for research is not anywhere near a "point of diminishing returns" scenario.
Also, you comment "there's only so much equipment today" - More/better equipment leads to scientists becoming more productive, in addition to the fact that I don't think there's a shortage of scientists to take advantage of additional funding, the lead time on equipment is far less than the lead time for new scientists (time to go through school as a scientist vs. choosing another career path as an undergraduate). The gating factor towards equipment availability for scientists is NOT availability of equipment to purchase, but of money to purchase that equipment with.
There's also the fact that the package management systems of most distributions have made architecture variants a non-issue. They will automatically choose the appropriate package for their architecture. For x86 vs x86_64, most distros have solved that problem with multilib approaches and multilib-aware package managers. I know Ubuntu has.
The summary is wrong compared to some PDFs on the approach which have been summarized here. BOTH conditions need to be met for you to be throttled.
More specifically (and keep in mind I'm summarizing another post that is summarizing the PDF) Comcast is creating two new QoS classes, Priority Best Effort and Best Effort. Everyone starts as PBE If you trip the excessive bandwidth trigger, you get moved from the PBE class to the BE class Being in the BE class doesn't hurt you until the CMTS gets congested
Um, if you're running VoIP using uncompressed 44.1 kHz audio you're an idiot TBH.
Any reasonable VoIP approach (even uncompressed 64kbps phone-quality audio) won't even come remotely close to triggering the 70% for 15 minutes criteria.
Similarly, typical streaming video is 2-5 Mbps, so you won't hit the 70% cap as their service is 15 peak.
All cable modem contracts have been written on the assumption that your bandwidth is shared between multiple users. You can burst up to the advertised rate, but you are never guaranteed to get it 100% of the time.
As much as I hate Comcast, this is in my opinion a pretty reasonable approach. You get throttled *only* if the network is congested (compared to Sandvining which was implemented no matter what the network state) and you get throttled only down to 50% of your maximum (which is a hell of a lot better than Cablevision OptimumOffline's stealthcapping, indefinately at 10% of your initial upstream without notification once you tripped the threshold.) It's a pretty fair scheme.
Of course the key is whether the throttling will be done in a normal traffic shaping manner, or Sandvine style with false RST injection. I am assuming false RST injection is out of the question since that got Comcast sued before.
Despite having a full standalone install of TomTom on my phone, I still have a complete standalone vehicle navigation device (running iGO 8), AND I also have a Garmin Oregon for when I leave the road.
I don't know about Android, but for Windows Mobile (and I'm fairly certain iPhone), phones do.
Take a guess who is providing the maps and software for that approach?
Yup, TomTom and Garmin.
Google "Garmin Mobile XT".
In theory Google could offer a solution with a client side full map database, but it would be inconsistent with how they have done business in the past. Android's new mapping solution won't kill TomTom or Garmin for the same reason Google Earth and Google Maps haven't killed products like: DeLorme Topo USA DeLorme Street Atlas USA Microsoft Streets and Trips etc...
GPS is a free service. You can get GPS receivers dirt cheap (hey, while the signals are free, receiver hardware still costs a little money), but all they tell you is your latitude/longitude.
Knowing your lat/long is honestly not that useful unless you have data to reference it against. Useful cartographic data costs money.
USGS maps don't provide routing info. The US Census Bureau TIGER database has streets but not turn restriction info and severe accuracy problems in some areas.
So when you pay for GPS "services" from most providers, what you are paying for is a mapping/routing solution and access to an accurate dataset that allows for mapping/routing.
Google has in theory had the ability to do this on the PC platform for years, but they haven't.
I don't expect to ever see the ability to install a full Google Maps database to an SD card. Until that happens (which I don't think will happen), TomTom and Garmin will always have a place.
This is why even with "Standalone" map software (e.g. software where all maps are stored on the device, thus not requiring any cellular coverage to work) for my AT&T Tilt 2 (which has a relatively large screen as PDA-phones go), I still use my standalone GPS - while the screen is lower resolution, it's larger. Actually TomTom's PDA software gets hard to use on high-res screens.
I would never use anything that required cellular coverage for basic navigation functions, even though the new Google solution supposedly caches your entire route, that doesn't help you if you miss a turn and go offroute and need a re-calc.
Think of it a different way: If a learning curve is steep, you *need* to learn a lot over a little time.
e.g. assume that you start at "not proficient" and end at "proficient" - The steeper the learning curve, the more you have to learn to reach "proficient".
It fits well with the hill analogy - you want to get to the top of the hill from the bottom. The steeper the hill, the more height you have to gain to get to the top.
They didn't use AES because AES didn't yet exist. (Or, to be specific, was very early on in the algorithm competition to determine which one would become the standard.)
Rijndael was chosen as the AES winner by NIST in 2001. WEP was finalized in 1997.
At that point, I believe DES was already known to have issues.
If I recall correctly, WPA/TKIP was an "interim" solution intended to be more secure than WEP but compatible with most WEP hardware. As such it had to leverage some of the low-level components of WEP, of which TKIP was one of them.
So effectively, WPA/TKIP has vulnerabilities because it inherited them from WEP.
I've been perfectly happy with the state of KDE in 9.04, with the exception of the network manager was quite annoying in that release.
It's vastly improved and I'm very happy with it in 9.10 - 9.10 fixed basically every wireless problem I had except for one PEBCAK issue. (I never realized that the "password" they gave me at one of my favorite places to eat was actually a valid 40-bit WEP key and not a passphrase. Windows automatically detects this and tries it as a key instead of a passphrase, Ubuntu does not.)
I can tell you that 9.10 has MASSIVE improvements in 5 GHz support (as in "it works and it didn't at all before") compared to 9.04. That's why my new laptop has been running 9.10 betas for the past 2-3 weeks.
Slashdot Mirror
They're also basically identical to AT&T's pricing and tiers, so it's not "taking a crap all over an iPhone killer" like another commenter claimed it to be.
Tell me how this is "taking a giant crap" over an iPhone killer?
Extra charge for a provider-provisioned Exchange server? Yup, AT&T has that too. I'm 90% positive that non-provider-provisioned exchange servers aren't blocked by VZW (they aren't with AT&T).
$60/month total for a 5GB tethering plan? Same as AT&T. (Note, last I checked, there was no official support whatsoever for iPhone tethering.)
FYI I am an AT&T customer, and left VZW two years ago a very unhappy customer, but the summary is unjustly misworded to paint VZW extremely negatively when in reality their data pricing is no different than AT&T. (Which is an improvement, VZW's data pricing used to be AWFUL compared to AT&T.)
Most likely there's an error in the summary. It is more likely:
"An extra charge for Verizon to provide an Exchange server for you."
This is how AT&T works. It's something like $10 extra if you want an Exchange server account provisioned for you by AT&T, but I have no problem using Google Sync with my AT&T account.
Also, AT&T's tethering plan is the exact same price as Verizon's - $60/month (total on top of voice plan) for 5GB.
Actually, the article linked that provoked this discussion pretty much says it:
"The company tried to soften the blow by saying it will add employees in some areas, such as wireless, video and broadband, in order to meet customer demand. Many non-management employees who are affected will have a guaranteed job offer as stipulated in union contracts, AT&T said."
Looks like AT&T's layoffs were in divisions OTHER than wireless. Probably the oldskool long-distance telephone market - not a good place to be nowadays.
Yup. I know of a company out West that's laying off 800 employees, meanwhile in certain departments, they're hiring employees. (General thing seems to be MechEs/general support staff are getting laid off, EEs with RF experience are getting hired.)
Sometimes layoffs are less about downsizing than about skillset readjustment. Many of those 12000 may have been in AT&T's broadband and landline divisions with no possible place in AT&T Wireless.
Also, as others have said, bandwidth doesn't necessarily scale with employees, in this case the limiting factors are:
1) Sites where new towers are permitted. People bitch about coverage and network performance but scream even louder when someone wants to put a cell site nearby. Even if it's "pretty-ized", people bitch about that evil RF that might possibly give them cancer, even though 1 minute in sunlight is going to increase your chances of cancer more than a lifetime of RF exposure at any reasonable distance from a cell tower. (Reasonable distance being defined as "you weren't a dumbass that broke the lock on the fence surrounding the tower and climbed the tower".)
2) FCC licenses for new towers take a lot of paperwork and time
3) Equipment costs - this is nowhere near as much of a problem as 1) and 2) are though, especially 1)
There's one thing about the iPhone situation that really annoys me:
AT&T has tiered pricing for data plans for phones based on its capabilities (i.e. estimating just how much of that "unlimited" data you will actually use depending on whether the phone is a PDA phone, or just a dumbphone, etc.) Tethering users, of course, get hit the hardest ($60/mo and they actually have a specifically documented usage cap). PDA users with keyboards get hit second hardest (formerly $40/mo, now down to $30/mo).
iPhone users, despite being probably the heaviest data users on the network, are in the second least expensive data pricing tier ($20/mo I believe, with "dumbphones" with basic WAP browsers the only thing lower on the list at $15 I think.)
As long as there are graduate students in school worried about whether they'll have a job when they graduate, I have a feeling that providing additional funding for research is not anywhere near a "point of diminishing returns" scenario.
Also, you comment "there's only so much equipment today" - More/better equipment leads to scientists becoming more productive, in addition to the fact that I don't think there's a shortage of scientists to take advantage of additional funding, the lead time on equipment is far less than the lead time for new scientists (time to go through school as a scientist vs. choosing another career path as an undergraduate). The gating factor towards equipment availability for scientists is NOT availability of equipment to purchase, but of money to purchase that equipment with.
There's also the fact that the package management systems of most distributions have made architecture variants a non-issue. They will automatically choose the appropriate package for their architecture. For x86 vs x86_64, most distros have solved that problem with multilib approaches and multilib-aware package managers. I know Ubuntu has.
The summary is wrong compared to some PDFs on the approach which have been summarized here. BOTH conditions need to be met for you to be throttled.
More specifically (and keep in mind I'm summarizing another post that is summarizing the PDF)
Comcast is creating two new QoS classes, Priority Best Effort and Best Effort.
Everyone starts as PBE
If you trip the excessive bandwidth trigger, you get moved from the PBE class to the BE class
Being in the BE class doesn't hurt you until the CMTS gets congested
Yup. Havoc's crusade to remove all features from GNOME (if it does nothing, it is easy to use the features!) is why I switched to KDE.
Um, if you're running VoIP using uncompressed 44.1 kHz audio you're an idiot TBH.
Any reasonable VoIP approach (even uncompressed 64kbps phone-quality audio) won't even come remotely close to triggering the 70% for 15 minutes criteria.
Similarly, typical streaming video is 2-5 Mbps, so you won't hit the 70% cap as their service is 15 peak.
All cable modem contracts have been written on the assumption that your bandwidth is shared between multiple users. You can burst up to the advertised rate, but you are never guaranteed to get it 100% of the time.
As much as I hate Comcast, this is in my opinion a pretty reasonable approach. You get throttled *only* if the network is congested (compared to Sandvining which was implemented no matter what the network state) and you get throttled only down to 50% of your maximum (which is a hell of a lot better than Cablevision OptimumOffline's stealthcapping, indefinately at 10% of your initial upstream without notification once you tripped the threshold.) It's a pretty fair scheme.
Of course the key is whether the throttling will be done in a normal traffic shaping manner, or Sandvine style with false RST injection. I am assuming false RST injection is out of the question since that got Comcast sued before.
Yup. Best device for the job.
Despite having a full standalone install of TomTom on my phone, I still have a complete standalone vehicle navigation device (running iGO 8), AND I also have a Garmin Oregon for when I leave the road.
I don't know about Android, but for Windows Mobile (and I'm fairly certain iPhone), phones do.
Take a guess who is providing the maps and software for that approach?
Yup, TomTom and Garmin.
Google "Garmin Mobile XT".
In theory Google could offer a solution with a client side full map database, but it would be inconsistent with how they have done business in the past. Android's new mapping solution won't kill TomTom or Garmin for the same reason Google Earth and Google Maps haven't killed products like:
DeLorme Topo USA
DeLorme Street Atlas USA
Microsoft Streets and Trips
etc...
GPS is a free service. You can get GPS receivers dirt cheap (hey, while the signals are free, receiver hardware still costs a little money), but all they tell you is your latitude/longitude.
Knowing your lat/long is honestly not that useful unless you have data to reference it against. Useful cartographic data costs money.
USGS maps don't provide routing info.
The US Census Bureau TIGER database has streets but not turn restriction info and severe accuracy problems in some areas.
So when you pay for GPS "services" from most providers, what you are paying for is a mapping/routing solution and access to an accurate dataset that allows for mapping/routing.
Google has in theory had the ability to do this on the PC platform for years, but they haven't.
I don't expect to ever see the ability to install a full Google Maps database to an SD card. Until that happens (which I don't think will happen), TomTom and Garmin will always have a place.
This is why even with "Standalone" map software (e.g. software where all maps are stored on the device, thus not requiring any cellular coverage to work) for my AT&T Tilt 2 (which has a relatively large screen as PDA-phones go), I still use my standalone GPS - while the screen is lower resolution, it's larger. Actually TomTom's PDA software gets hard to use on high-res screens.
I would never use anything that required cellular coverage for basic navigation functions, even though the new Google solution supposedly caches your entire route, that doesn't help you if you miss a turn and go offroute and need a re-calc.
As much as I would love to see that, they'd run into trademark conflicts with the ACME grocery store chain.
Think of it a different way: If a learning curve is steep, you *need* to learn a lot over a little time.
e.g. assume that you start at "not proficient" and end at "proficient" - The steeper the learning curve, the more you have to learn to reach "proficient".
It fits well with the hill analogy - you want to get to the top of the hill from the bottom. The steeper the hill, the more height you have to gain to get to the top.
They didn't use AES because AES didn't yet exist. (Or, to be specific, was very early on in the algorithm competition to determine which one would become the standard.)
Rijndael was chosen as the AES winner by NIST in 2001. WEP was finalized in 1997.
At that point, I believe DES was already known to have issues.
If I recall correctly, WPA/TKIP was an "interim" solution intended to be more secure than WEP but compatible with most WEP hardware. As such it had to leverage some of the low-level components of WEP, of which TKIP was one of them.
So effectively, WPA/TKIP has vulnerabilities because it inherited them from WEP.
WPA2/AES eliminates all "WEP heritage cruft".
Netbooks in general seem to be pretty good in terms of crapware.
My 9" Aspire One was the first computer I've ever bought that I didn't immediately feel the need to nuke and repave.
My Eee 1000HE was the second.
My Asus G51vx is the third.
I've been perfectly happy with the state of KDE in 9.04, with the exception of the network manager was quite annoying in that release.
It's vastly improved and I'm very happy with it in 9.10 - 9.10 fixed basically every wireless problem I had except for one PEBCAK issue. (I never realized that the "password" they gave me at one of my favorite places to eat was actually a valid 40-bit WEP key and not a passphrase. Windows automatically detects this and tries it as a key instead of a passphrase, Ubuntu does not.)
Which internal wireless card do you have?
I can tell you that 9.10 has MASSIVE improvements in 5 GHz support (as in "it works and it didn't at all before") compared to 9.04. That's why my new laptop has been running 9.10 betas for the past 2-3 weeks.
What graphics chipset did he have?
Keep in mind that Intel graphics chipset support was AWFUL in 8.10 and 9.04, especially 9.04.