There's one problem with allowing exceptions in portraits: it has the potential to hurt AP's credibility.
You're kidding right? AP lost most of its cred years ago my friend. It's the most left slanted print news org on the face of the earth. Ok, maybe they're tied with the NYT, but it's a close race. I still put AP in front because they try their damndest to hide their leftist slant. But they still fail. NYT is in your face with their slant. To me, this makes AP more dangerous. The terms Manchurian Candidate and 'sleeper' come to mind. This, and the fact that AP claims to be a news org, whereas the NYT is much more of an opinion piece paper these days.
So, 72% invested, 6.7% in marketing, and 21% on development? Or, to put it another way, a third as much spent on marketing as on development? I can't think of many companies (big pharmaceuticals excepted) that have this high a ratio of marketing to product development spending.
Footwear and soft drink companies come to mind, i.e. Nike, CocaCola, and Pepsi Co. I don't have hard numbers for today, but in the 80's Nike spent more on marketing than on the entire rest of their business. IIRC they paid Michael Jordan alone something like $300 Million in endorsement contracts over a 10 year period. He was just the tip of the iceberg. They had dozens of top athletes peddling their shoes and clothes in TV commercials and print ads. And the shoes were manufactured by in essence slave labor in the far east, by people making the equivalent of $1 a day.
The faster clock of the 66 may seem like a win, but the 50 MHz version has the faster bus speed.
Only the DX. The DX2 is actually a double clock speed of a 25 MHZ.
I tried them both with Xwing, and the 66MHZ wins hands down.
The straight DX-50 with a VESA local bus Tseng Labs ET4000 card running at 50 MHz walked all over the DX2-66 with the same vid ard card in Doom and Duke Nukem. The main reason being the DX2-66 setup required a 33MHz VLB timing, dramatically lowering the frequency of the graphics chip. The ET4000 ran at a 1:1 ratio of the VLB frequency. Ahh the good old days....
At least it'll be outdoors. Can't be any more toxic than the ammonia along the baseboards of my living room. Yes, my cat is a little devil bastard, like all cats. My recommendation: spray those metal chunks with Simple Solution and you're good to go.
12 processor chips, 72 (6 cores per chip) 700MHz MIPS64 cores 1.4GLOPS each peak, 48GB RAM, 4GB/s bidirectional peak (yes bytes not bits) Kautz topology RDMA cluster interconnect, less than 300 watts power consumption, the Cray draws up to 1600 watts. Runs MIPS Linux and standard MPI apps. Much much less expensive than the Cray, and this one comes fully loaded. Peak GFLOPS performance is much lower than the Cray "deskside", however the SC072 is a desktop unit, interconnect bandwidth is 4GB/s vs. the Cray's 2.5G/s Infiniband interconnect, meaning a higher percentage of peak parallel performance can be obtained.
Too bad none of their machines are true shared memory designs. One can perform OpenMP and forking across the 6 cores of a node chip as if it were a 6-way SMP. This opens interesting opportunities for hybrid SMP/MPI code, and some applications have been proven to greatly benefit from such a scheme, one being Parallel Ocean Program, IIRC: http://climate.lanl.gov/Models/POP/
Damn straight my mileage varies. My mail server is hosted in a data center and has RAID1 over 4 drives. I pay for it by hosting email for a few organizations and charging them nominal fees just to pay for the colo fees.
A little OT, but RAID 1 is disk mirroring and by definition includes only 2 drives. So, do you have two mirrors setup or...some quasi RAID solution that allows you to mirror the contents of one drive to multiple other drives? I've never heard this called "RAID 1" before.
You have your broadband hardware confused. DSL doesn't require the provider to upgrade the copper, that's the cable TV industry that must do that.
Instead of a DSLAM, I believe you meant to say CMTS, Cable Modem Termination System. Or maybe you just weren't clear in what you're proposing/suggesting.
You've omitted hardware required to build out the infrastructure, such as emergency power generation (UPS and diesel generator), routers, directory and authentication servers, mail servers, the customer database system (billing, service, etc). You've also omitted the organization itself--sales, support, infrastructure management, and their associated office space, salaries, and hardware/software costs.
There is much more to running a successful broadband service company/organization than the customer facing communications gear and line rental on the upstream pipes, which btw, you'll need at least two of for redundancy.
Any broadband provider, to be profitable, must run at a deficit ratio of upstream bandwidth to subscribed bandwidth. That aspect of ISP'dom hasn't changed that much since POTS modem days. A provider *must* over subscribe its upstream bandwidth or it will lose money and go bankrupt.
If you look at the backbone maps of various carriers and providers of broadband and business (T1/T3 and ethernet over fiber) lines, the sum total of backbone bandwidth is orders of magnitude less than the total CPE sum.
I recall looking at a Verio networks backbone map a few years ago, and the phatest national pipes they had were OC3 (155Mb/s), and most of the smaller regional connections were only DS3 (45Mb/s). I know for a fact at that time they had a few hundred T1 customers in St. Louis alone, and IIRC, there was no OC3 connection from St. Louis to their backbone. There were two DS3 connections, one to Chicago, the other I can't recall, maybe K.C. They had hundreds of BRI ISDN customers as well. At minimum they were running a 1:3 ratio, but I'm guessing it was likely closer to 1:10 or more.
The ratios are likely a bit lower today given that since 2000 every venture capitalist and his brother in the US, and the big telcos, have been planting fiber all along the highways and railroads of the country. It's less than 1:1 though, and always will be. End users are never all consuming all their available bandwidth simultaneously, thus there is a huge financial disincentive to build a network for that CPE usage scenario.
Comcast will not likely be the last cable provider to implement such a cap policy. Expect in the not too distant future new routers or firmware for existing models with a monthly bandwidth measurement feature and real time usage monitor in the web GUI management interface.
It will allow setting of an email address to which it will send daily, weekly, and/or monthly usage statistics to said address.
What will be interesting is when Comcast TOS's a customer for exceeding the limit, then the customer fights back with his/her own logs, which show they are under the limit. Then what does the court decide, especially if the measuring equipment at both ends is Cisco, since Cisco owns Linksys?
I have no idea why the US hasn't fallen in love with diesel yet.
A couple of reasons:
1. Most Americans, especially politically liberal metropolis citizens and very especially women, don't know the difference between the fuels and grades of fuel, and would constantly be filling the tank with the wrong pump nozzle, blowing up their engines and racking up repair bills. Call your nearest Uhaul store and ask how many times a year a customer fills one of their diesel trucks with gas (prepare for loud laughter and good stories). Most Americans are stupid when it comes to diesel.
2. For the non stupid, who know the benefits of diesel, the main reasons for not wanting diesel personal vehicles are cold weather performance and engine noise. Gelling/freezing of the fuel in the lines, or the fuel treating 'hassle' required to prevent it is a PITA, making diesel far less than attractive for citizens in say Chicago, NYC, Buffalo, Pittsburgh, Minneapolis/St. Paul, etc. Most Americans want quiet cars, and although modern diesels are much quieter then previous generations, they are still far noisier than gasoline cars. We Americans are busy, selfish, lazy people. We thrive on *convenience*. Diesel personal vehicles are not *convenient* for the mass of Americans.
In warm weather states the convenience factor doesn't come into play as much, but you run into the politically charged "pollution" debate. Take a look at the California Air Resources Board regulations on diesel personal vehicles.
It's pretty lucrative from what I hear, especially if you sign up Kraft Foods as a client. And even when arrested, you may likely beat the rap. Ask Jeremy Jaynes about both. He'll probably be back at it soon.
Just be warned, you'll see me on the other side of the scrimmage line.;-)
This is very common for companies to do; companies like IBM or TSMC will manufacture chips designed by other companies but it's not considered a re-branding.
It's called "contract manufacturing" and it's done in every manufacturing industry, from chips to microwave ovens to cell phones.
Offshore hosting looks like a chance for the banana republics to build their online economies. It will happen there as well as here, not instead, so everybody benefits.
It's already happening. Unfortunately, most of it has been child porn hosting, gambling, scams, and spam spewing, all things which are illegal here. "offshoring" has always been the purvey of criminals and corporate greed. The motivations for moving internet operations offshore is the same old motivation.
I don't buy into the premise of the article anyway. Any company, US or not, that has important data traveling across US electronic networks should already be encrypting it. Thus, why move it "offshore" at this point?
If anything foreign companies should be moving data center ops TO the US due to the weak dollar and quality of the infrastructure.
Slashdot Mirror
Mode parent ^
I'm with ya M1rth. You're not alone on /.
Not everyone here is a liberal pussy. At least we know two of us aren't.
If you typed that from memory you deserve an award.
There's one problem with allowing exceptions in portraits: it has the potential to hurt AP's credibility.
You're kidding right? AP lost most of its cred years ago my friend. It's the most left slanted print news org on the face of the earth. Ok, maybe they're tied with the NYT, but it's a close race. I still put AP in front because they try their damndest to hide their leftist slant. But they still fail. NYT is in your face with their slant. To me, this makes AP more dangerous. The terms Manchurian Candidate and 'sleeper' come to mind. This, and the fact that AP claims to be a news org, whereas the NYT is much more of an opinion piece paper these days.
So, 72% invested, 6.7% in marketing, and 21% on development? Or, to put it another way, a third as much spent on marketing as on development? I can't think of many companies (big pharmaceuticals excepted) that have this high a ratio of marketing to product development spending.
Footwear and soft drink companies come to mind, i.e. Nike, CocaCola, and Pepsi Co. I don't have hard numbers for today, but in the 80's Nike spent more on marketing than on the entire rest of their business. IIRC they paid Michael Jordan alone something like $300 Million in endorsement contracts over a 10 year period. He was just the tip of the iceberg. They had dozens of top athletes peddling their shoes and clothes in TV commercials and print ads. And the shoes were manufactured by in essence slave labor in the far east, by people making the equivalent of $1 a day.
The faster clock of the 66 may seem like a win, but the 50 MHz version has the faster bus speed.
Only the DX. The DX2 is actually a double clock speed of a 25 MHZ.
I tried them both with Xwing, and the 66MHZ wins hands down.
The straight DX-50 with a VESA local bus Tseng Labs ET4000 card running at 50 MHz walked all over the DX2-66 with the same vid ard card in Doom and Duke Nukem. The main reason being the DX2-66 setup required a 33MHz VLB timing, dramatically lowering the frequency of the graphics chip. The ET4000 ran at a 1:1 ratio of the VLB frequency. Ahh the good old days....
Which 486 are they using, the 50 or 66 MHz? The faster clock of the 66 may seem like a win, but the 50 MHz version has the faster bus speed.
They're not using a 486, they're using the cache-less Celeron. It's a little bit faster...
http://en.wikipedia.org/wiki/Celeron#Covington
But only if the command is given by the Vilnius schoolmaster.
At least it'll be outdoors. Can't be any more toxic than the ammonia along the baseboards of my living room. Yes, my cat is a little devil bastard, like all cats. My recommendation: spray those metal chunks with Simple Solution and you're good to go.
Paramount already created such a thing for Minority Report. Why don't they just borrow those? ;)
When you mentioned this, I started Googling around but couldn't find any sources for them. Got any favorites?
http://sicortex.com/products/sc072_pds
12 processor chips, 72 (6 cores per chip) 700MHz MIPS64 cores 1.4GLOPS each peak, 48GB RAM, 4GB/s bidirectional peak (yes bytes not bits) Kautz topology RDMA cluster interconnect, less than 300 watts power consumption, the Cray draws up to 1600 watts. Runs MIPS Linux and standard MPI apps. Much much less expensive than the Cray, and this one comes fully loaded. Peak GFLOPS performance is much lower than the Cray "deskside", however the SC072 is a desktop unit, interconnect bandwidth is 4GB/s vs. the Cray's 2.5G/s Infiniband interconnect, meaning a higher percentage of peak parallel performance can be obtained.
Sicortex's really cool machine is this:
http://sicortex.com/products/sc5832
Too bad none of their machines are true shared memory designs. One can perform OpenMP and forking across the 6 cores of a node chip as if it were a 6-way SMP. This opens interesting opportunities for hybrid SMP/MPI code, and some applications have been proven to greatly benefit from such a scheme, one being Parallel Ocean Program, IIRC: http://climate.lanl.gov/Models/POP/
Damn straight my mileage varies. My mail server is hosted in a data center and has RAID1 over 4 drives. I pay for it by hosting email for a few organizations and charging them nominal fees just to pay for the colo fees.
A little OT, but RAID 1 is disk mirroring and by definition includes only 2 drives. So, do you have two mirrors setup or...some quasi RAID solution that allows you to mirror the contents of one drive to multiple other drives? I've never heard this called "RAID 1" before.
http://www.vpshosting.net/network/net-map.php5
They're now up to redundant path OC12s servicing St. Louis. Clicking on map image. Interesting.
You have your broadband hardware confused. DSL doesn't require the provider to upgrade the copper, that's the cable TV industry that must do that.
Instead of a DSLAM, I believe you meant to say CMTS, Cable Modem Termination System. Or maybe you just weren't clear in what you're proposing/suggesting.
You've omitted hardware required to build out the infrastructure, such as emergency power generation (UPS and diesel generator), routers, directory and authentication servers, mail servers, the customer database system (billing, service, etc). You've also omitted the organization itself--sales, support, infrastructure management, and their associated office space, salaries, and hardware/software costs.
There is much more to running a successful broadband service company/organization than the customer facing communications gear and line rental on the upstream pipes, which btw, you'll need at least two of for redundancy.
Any broadband provider, to be profitable, must run at a deficit ratio of upstream bandwidth to subscribed bandwidth. That aspect of ISP'dom hasn't changed that much since POTS modem days. A provider *must* over subscribe its upstream bandwidth or it will lose money and go bankrupt.
If you look at the backbone maps of various carriers and providers of broadband and business (T1/T3 and ethernet over fiber) lines, the sum total of backbone bandwidth is orders of magnitude less than the total CPE sum.
I recall looking at a Verio networks backbone map a few years ago, and the phatest national pipes they had were OC3 (155Mb/s), and most of the smaller regional connections were only DS3 (45Mb/s). I know for a fact at that time they had a few hundred T1 customers in St. Louis alone, and IIRC, there was no OC3 connection from St. Louis to their backbone. There were two DS3 connections, one to Chicago, the other I can't recall, maybe K.C. They had hundreds of BRI ISDN customers as well. At minimum they were running a 1:3 ratio, but I'm guessing it was likely closer to 1:10 or more.
The ratios are likely a bit lower today given that since 2000 every venture capitalist and his brother in the US, and the big telcos, have been planting fiber all along the highways and railroads of the country. It's less than 1:1 though, and always will be. End users are never all consuming all their available bandwidth simultaneously, thus there is a huge financial disincentive to build a network for that CPE usage scenario.
Comcast will not likely be the last cable provider to implement such a cap policy. Expect in the not too distant future new routers or firmware for existing models with a monthly bandwidth measurement feature and real time usage monitor in the web GUI management interface.
It will allow setting of an email address to which it will send daily, weekly, and/or monthly usage statistics to said address.
What will be interesting is when Comcast TOS's a customer for exceeding the limit, then the customer fights back with his/her own logs, which show they are under the limit. Then what does the court decide, especially if the measuring equipment at both ends is Cisco, since Cisco owns Linksys?
I have no idea why the US hasn't fallen in love with diesel yet.
A couple of reasons:
1. Most Americans, especially politically liberal metropolis citizens and very especially women, don't know the difference between the fuels and grades of fuel, and would constantly be filling the tank with the wrong pump nozzle, blowing up their engines and racking up repair bills. Call your nearest Uhaul store and ask how many times a year a customer fills one of their diesel trucks with gas (prepare for loud laughter and good stories). Most Americans are stupid when it comes to diesel.
2. For the non stupid, who know the benefits of diesel, the main reasons for not wanting diesel personal vehicles are cold weather performance and engine noise. Gelling/freezing of the fuel in the lines, or the fuel treating 'hassle' required to prevent it is a PITA, making diesel far less than attractive for citizens in say Chicago, NYC, Buffalo, Pittsburgh, Minneapolis/St. Paul, etc. Most Americans want quiet cars, and although modern diesels are much quieter then previous generations, they are still far noisier than gasoline cars. We Americans are busy, selfish, lazy people. We thrive on *convenience*. Diesel personal vehicles are not *convenient* for the mass of Americans.
In warm weather states the convenience factor doesn't come into play as much, but you run into the politically charged "pollution" debate. Take a look at the California Air Resources Board regulations on diesel personal vehicles.
It's pretty lucrative from what I hear, especially if you sign up Kraft Foods as a client. And even when arrested, you may likely beat the rap. Ask Jeremy Jaynes about both. He'll probably be back at it soon.
Just be warned, you'll see me on the other side of the scrimmage line. ;-)
And how would they go about fixing their usability problems if those very people didn't use the operating system in the first place?
Easy: Have Novell and IBM hire the top UI developers away from Apple and put them to work on Gnome, FireFox, and ThunderBird.
No, actually. Kent State University is *very* picky about antivirus solutions you can use on campus.
They're probably even more picky about National Guard use on campus.
There were actually five. The fifth turned into metal and absorbs electricity.
This is very common for companies to do; companies like IBM or TSMC will manufacture chips designed by other companies but it's not considered a re-branding.
It's called "contract manufacturing" and it's done in every manufacturing industry, from chips to microwave ovens to cell phones.
Lawyers do not get involved unless they are invited, but, there is always a lawyer who will take any stupid case that some idiot brings to their door.
Only stupid cases that appear to have green potential. No green potential, no case.
And worst of all they're cool!
They sure are, just ask any McDonald's clerk.
Switching could be a pain.
Change always is. But change is inevitable.
Prepare thyself.
Offshore hosting looks like a chance for the banana republics to build their online economies. It will happen there as well as here, not instead, so everybody benefits.
It's already happening. Unfortunately, most of it has been child porn hosting, gambling, scams, and spam spewing, all things which are illegal here. "offshoring" has always been the purvey of criminals and corporate greed. The motivations for moving internet operations offshore is the same old motivation.
I don't buy into the premise of the article anyway. Any company, US or not, that has important data traveling across US electronic networks should already be encrypting it. Thus, why move it "offshore" at this point?
If anything foreign companies should be moving data center ops TO the US due to the weak dollar and quality of the infrastructure.