Then why have them on by default without informing the user??? Think zippy, think.
Why doesn't windows *default* to 640x480 256 color anymore?
Technology progresses, and so should the defaults. Multi-core computers and fast internet connections have been prevalent for several years now. Defaulting these new features to *on* certainly uses more resources, but ultimately makes my e-mail client faster (caching) and more productive (full-text search, offline capability). Who cares if my sixth core is spending a few idle cycles indexing my e-mail, and I'm using a few more kbits on a 20mbit connection.
IMHO, software defaults should target the median machine available at the time of its release. If you choose to run the latest software on a dinosaur, the burden is on YOU to optimize it by tweaking the default settings!
Anand shows that the iphone 4 is SUBSTANTIALLY worse than the 3GS in terms of signal attenuation in four different situations (in one case, by as much as 18db !) This is contrary to Apple's claims that the revolutionary antenna design gives the iphone4 superior RF performance to previous generations.
You are correct that the iphone4 is better than the HTC nexus one in a SINGLE test by 0.5db. It is, however, worse in all remaining tests by 2.5 to 9.1db.
As for signal strength indicators, it has been my casual observation that GSM carriers typically over-estimate the idle "bars" they report. A GSM phone with 1-2 bars is rarely usable for an actual conversation, especially if you are in motion. In contrast, pretty much every CDMA phone I have had could place and receive acceptable calls with 0 bars showing while idle.
Additional thoughts : - I think it's comical that apple will now make the first 3 bars "taller" to make them "easier to see" - I believe the fact that they are still selling the 3GS, and specifically re-state that they are willing to take any undamaged phone back within 30 days for a full refund absolves them of all responsibility for this bruhaha... if you keep or purchase an iphone4 at this point, despite the well-publicized reception problems, you have no right to complain, IMHO.
But what about the other way? Hearing someone when I've got a lot of background noise. My problem with "in-the-ear" has been that, because the distance is so short and the driver so small, the rate of volume increase / decrease is sharp. In short, I quickly go from "can't hear you" to "too loud, but still can't understand".
The jawbone does a bit of passive isolation in the ear it's inserted, provided you pick the right earpiece gel. It's plenty for use in a car, but I'm not sure if it will be enough for your particular application.
I can second the Jawbone Prime suggestion. The background (non-wind) noise cancellation is fantastic. I use one in the car all the time. The party on the other end of the call is never able to discern that I am in a car, even at highway speed. The Prime's noise cancellation is substantially better than any other BT headset I have ever tried.
I can clarify the verizon thing for you. They were only testing for speed, and only where everyone had coverage. Given those restrictions, the results aren't surprising. I have had data cards and smartphones from AT&T, Verizon, and T-mobile data through various jobs (oftentimes simultaneously). AT&T is generally the fastest when it works. Verizon is by far the most consistent (500kbps~1 mbit almost anywhere in the country in my experience) with the largest 3G network by a HUGE margin. T-mobile is just a disaster all-around (virtually no coverage, so 99% of your time is spent on EDGE).
Verizon is pretty much the only carrier that consistently has 3G coverage outside of the major metro areas. Even in major cities, I usually find that the verizon card has better signal inside buildings. Outside of major cities, it is my experience that it's easier to find a verizon 3G signal than an AT&T edge signal.
NOW if they modified the test by driving around the country, and averaging in zeros when they had no coverage on a particular carrier (or EDGE results when they had no 3G coverage), then I anticipate verizon would win this contest by a HUGE margin.
Personally, for mobile computing I find 1mbit is plenty for me... so I'll take reliability over speed any day.
I don't have any anecdotal sources of my own, sorry. That's why I ask you.
And the measles vaccine was introduced in 1963, which seems to go against your claim that it helped the outspread? If it helped, then... it wasn't by much, or was it?
On a side note, I hate taking it inductively because it always comes down to "who has the most numbers" which doesn't add anything to my understanding...
Yes... don't let pesky things like "numbers" get in the way of your "understanding"...
I've actually read every one of your posts with great interest. You are either one of the densest human beings on the planet, or one of the most skillful trolls I have ever come across. I really can't decide which it is.
Vaccines could be a big scam, I don't know enough to assert. However, I do remain unconvinced of their effectiveness.
What would it take to convince you? It might involve numbers.
To my layman's understanding, I feel sanitation and nutrition play a much bigger role in the immune system than exposing your body to certain proteins ever could. The vaccines can certainly help in an alarmist way, to tell the body to "get ready" for infection, but if your whole system hasn't structured itself right in the first place, it's pointless.
Your layman's understanding is wrong...
A) Would you consider sanitation and nutrition in the United States to be acceptable? (based on your previous posts, I presume "yes", since you assert that this is the principle reason why certain infectious disease that we erroneously vaccinate against are on a decline)
B) Are you aware that millions of people catch the flu annually despite our "modern" sanitation and nutrition? How can you explain this? What are your "feelings" on flu trends? Are more people getting the flu every year, less, or does it stay about the same... compare this to 100 years ago. Do you feel more people got the flu then? now how about smallpox?
C) Are you aware that not all vaccines are universally "mandatory" or even "recommended" in all parts of the country (e.g. chickenpox, HPV, etc. etc. etc.)? How do you explain the gross disparity in infection rates between those who are "vaccinated" versus those that are not. On average, we all share the same "modern" sanitation and nutritional benefits.
It seems that you assumed that I wouldn't suggest first to use wpa2, etc. Seeing as the article is about cracking advanced encryption, I would hope that this is already in place. Poor advice? I think not. It adds additional roadblocks. I also said that it 'helps'. Not that it's a foolproof plan. It just makes it more of a pain to break in.
For example, using a MAC address filter would mean that they would have to spoof a MAC address that you have whitelisted. This requires additional effort and information gathering.
Using a SSID that is not broadcasted, and also not easily guessable (not a dictionary word, and a certain length, etc), makes it harder for SSID crackers to pick it up as well.
You may be happy with just using strong encryption, but I very much prefer enabling these additional security features to harden it even further, even if it is just a little bit further.
I will second what the other two people replying to you have said :
#1) SSID just requires a single deauth to any client. This literally takes 2 seconds to do. #2) Your clients are broadcasting their MAC addresses in the clear, and it's a fair assumption that any associated client is on your MAC whitelist... Anyone hacking your wireless network is literally staring at these MACS (and probably continuously typing them back into the console).
Anyone with the technical sophistication to go after WPA already knows this (and can bypass both your MAC and SSID measures in LITERALLY 10 seconds)
If enabling these two features makes you FEEL safer then by all means keep them on. But they offer NO additional protection (not even a teeny tiny bit), and are probably a bit of a hassle for you - Have to add each legit client to MAC table - Some clients barf on the hidden SSID
Your ONLY effective consumer-level protection at the moment is to pick a completely random long WPA PSK! (even then, there are a few attacks that allow a hacker to decypt WPA packets without knowing the key)
See If you can find some moca bridges (like the ones used in early verizon FIOS installs). I got a pair of motorola NIMS, and can pull 100mbit over the coax in my house all day long (internal signaling is 250mbits, but the port on the device is only 100mbit) . WAY more reliable than the wireless bridge it replaced, and WAY faster than a powerline or phoneline network!
Yeah, my old AT&T GSM (non 3G) phone would not only drive speakers crazy, but also my CRT monitor when it was close enough. I've never directly observed interference from a CDMA phone.
Pick the laptop that you travel with (perhaps between home and work). Store all shared data on it. Samba mount that directory on the OS you are currently using (all of the OS's you list support samba).
You can work on your data on that laptop while on the move... but when you are using another computer (e.g. desktop) that laptop just effectively serves as an external disk. As long as you use GigE, there is no performance drop over a single locally-attached disk.
As an added benefit, you can backup this data whenever you mount the share (i use rsync). This ensures that I automatically have redundant, geographically diverse, versioned backups of my important files!
Except that part of establishing secure communications is authenticating the other end, or else you are vulnerable to a man-in-the-middle attack; that is why SSL has a certificate system. If the only intention was to have a means of encrypting communication, then there would be no reason for SSL to have such a complicated protocol that includes identification and capabilities management.
I agree with you in principle, but practically speaking this system is hobbled by the lowest common denominator. A decade+ ago, I remember having to pay many dollars, talk to a rep over the phone, fax over government issued ID's, and incorporation documents in order to get a signed SSL cert that was included in the browser root. This process took several days, and you paid for the verification step! Nowadays I just add one to my shopping cart, pay ~$5, specify any domain name I want, and hit checkout. If you want a signed SSL cert for a MITM attack of paypal.com you can go out and buy one within 5 minutes!
IMHO, practically speaking, the signatures means absolutely nothing anymore!!!
Just tried this on my CDMA HTC touch with the following string sent from google voice %n \%n "%n" '%n' %n . As claimed in the article it did indeed crash TouchFlo! The phone needed a reboot!
This is IDIOTIC. How can any reasonable person possibly buy this argument.
Anyone that wants to bring down a cell phone tower or cell network IS NOT GOING TO CARE whether or not it's LEGAL to screw with the cell radio baseband software. They are ALREADY attempting to do something much worse.
Let's be honest here, the "security" aspect of this argument is a smokescreen. It's blatantly all about the profit!
Furthermore, the cellular network should NOT be so fragile that a single rogue cell phone could take it down (AFAIK it is not). BUT if AT&T is truly insistent on making this argument, then I believe a full investigation by the FCC is mandated. The self-admitted fragile state of their network means that their stewardship of a public resource (radio spectrum) is being poorly managed and truly endangering national security.
While you are correct, in the PARTICULAR case of the iphone, most of the "jailbreaking" has gone hand-in-hand with modifying the cellular baseband to carrier-unlock the phone. The same software pretty much always does both (jailbreak and carrier unlock)!
Chill. It's a cell phone, not a religion. Half the people on here have posted about FIOS and cell phones. I've also posted a ton of "knowledgeable" stuff about physics, computers, electronics, etc. etc. What does THAT tell you about me? I've never worked for a cell company. I don't even own stock in a cell company. I just happen to use the phone (and recently data) a lot, and have traveled pretty extensively in the US. When I see a discussion I can contribute info to, I post a comment.
besides... you're saying I'm TOO knowledgeable to give my opinion on the OP's comment about poor GSM coverage in Maine?!!??! What exactly is the cutoff for being too smart here? A chimp could identify that one magical talking box works while the other does not. Granted, I was only there for a total of approx 3-4 days recently, but it was very quickly evident which phone was the paperweight. Especially once you ventured off the main highways. I hope someone else who actually lives in Maine and has used both GSM and CDMA chimes in here.
Anyway, good job cherry picking posts to highlight. If you even bothered to look at half the comments I've made (here and on howardforums), you would see that I routinely slam Verizon (wireless) pretty hard. Especially on all of the shit they pull with the funny firmware in their phones and more recently their mandatory data plan policies for smart phones. However, I won't deny that I have always had nothing but praise for their voice/data network coverage and reliability.
Some phones already have dual radios. For example, the Blackberry storm comes in a hybrid GSM/CDMA design. When sold by verizon, it has a sim card already installed.
HOWEVER, the phone will not currently roam on domestic GSM carriers. The SIM card and GSM radio are only used for international roaming in countries without a CDMA network. However, that seems to be a contractual limitation rather than a technical one. It likely just doesn't make any business sense right now to offer GSM roaming within the US.
That being said, the future of US cellular is in LTE. Right now it looks like all of the major networks are going down that path. I predict we will be seeing a LOT of hybrid radio designs with bizarro frequency combinations in the coming years (e.g. GSM/UMTS/HSDPA/LTE, CDMA/1X/EVDO/LTE, etc. at all sorts of TV/850/PCS/2100 combos)
All this does is allow infighting for handsets but doesn't solve the problem of crappy service over the US. If the war torn middle east and mount everest can get cell coverage why can't we get decent coverage in maine. Mount everest has people on it 1 month a year, there are over a million people in maine at any given time! I can't use my phone is 1/2 the counties here and that's with the AT&T.
DING DING DING DING... There's your problem! GSM service in North America is a complete joke in my experience. ESPECIALLY once you venture out of any major city or highway! Just look at the coverage maps for each carrier!
I've had both a CDMA and GSM work phone for many years. Traveled through much of the US. I always chuckle when I see some reviewer favorably comparing the two, ESPECIALLY on coverage.
I was actually up in Maine (Bangor and Bar Harbor) just last week. I had my personal verizon phone with me, and a GSM work phone. The GSM phone had a t-mobile sim but all of the carriers seem to mutually roam in Maine. The phone could associate with banner (company) : Cingular (AT&T), US-890 (Unicel), and T-mobile (T-mobile). It autoregistered to any one of those networks depending on the strongest signal. All THREE of those GSM networks combined were completely dwarfed by Verizon's native CDMA coverage. I mean it wasn't even remotely close! Hell, I had full EVDO revA coverage in areas that couldn't even get a regular GSM/GPRS signal.
In my experience, GSM in Canada is no different. For example, I continued up to Cape Breton after Maine. At one point, the closest GSM tower (Rogers) was a hundred miles away! Full CDMA coverage almost all the way up there, and many spots with EVDO!
So... In my opinion, the easiest way fix to your problem with coverage in the boonies is to go visit a verizon store, and just bite the bullet on the BS craptacular locked-down handset they will give you. At least you'll be able to use your phone to... you know... make phone calls...
Just look at the manner in which JPEGs are encoded for your answer!
Take the DCT (discrete cosine transform) of blocks of pixels throughout the image. Examine the frequency content of the each of these blocks and determine the amount of spatial frequency suppression. This will correlate with the quality factor used during compression!
Maybe I'm missing something here... but if you process data in 2GB chunks shouldn't your software just keep it all in memory. Once processing is complete writing it out to one of those SSD arrays should take 10 seconds (which is nothing for 2 hours worth of processing time!!!). If you don't have access to the source code, a quick fix is to just mount a RAM drive.
Furthermore, OCR is stupidly easy to parallelize. The results of each page do not depend on previous pages. You can process each page independently on independent cpus/memory in a cluster of computers WITHOUT needing a supercomputing interconnect. It should pretty much scale linearly with cpu count! Hell for 1GB data/hour even 10base-T is overkill by a factor of 36!
To me, it sounds like you are limited by shitty programming, not disk I/O. Throwing more expensive hardware and faster arrays of expensive SSD's at the problem won't really fix it!
#1) Most e-mails are still sent in plaintext and unencrypted. Not only are the contents unencrypted, but the communications channel is unencrypted as well. Hell, in my experience most e-mail is still sent using text/html over unencrypted smtp!
#2) Most blogs and e-mails do not have a PGP signature... Those that do usually don't have the signing key stored in any repository, much less any repository that you could establish trust to.
#3) Very few people encrypt their hard drive contents... (hell, not even those that store personal/credit card info). The prevalence of TPM chips and built-in partition encryption in the newer operating systems is just starting to make a teeny tiny dent in this.
I agree that PGP was a revolutionary contribution to computer security. However, it is dishonest to claim that it is as widely used as you claim!
I think all brother inkjets do this now. I agree that it's a HUGE waste of ink if you don't print color regularly. On the flip side, I've never had a clogged head with this printer. hmmm....
It's simply called adaptive optics (AO). In AO, a guidestar is a natural isolated point-like star that is close to your science object (what you are trying to look at). If a laser is used to excite the sodium layer to create an artificial reference, it's called a "laser guidestar".
Anyway, this "trick" is completely different from adaptive optics in both the mathematics and implementation.
Slashdot Mirror
Then why have them on by default without informing the user???
Think zippy, think.
Why doesn't windows *default* to 640x480 256 color anymore?
Technology progresses, and so should the defaults. Multi-core computers and fast internet connections have been prevalent for several years now. Defaulting these new features to *on* certainly uses more resources, but ultimately makes my e-mail client faster (caching) and more productive (full-text search, offline capability). Who cares if my sixth core is spending a few idle cycles indexing my e-mail, and I'm using a few more kbits on a 20mbit connection.
IMHO, software defaults should target the median machine available at the time of its release. If you choose to run the latest software on a dinosaur, the burden is on YOU to optimize it by tweaking the default settings!
The two proposed changes in the article are to :
- disable the global indexer
- disable caching of messages to the local computer
It should come as no surprise that these two features increase cpu load and bandwidth consumption respectively...
That is one hell of a distributed computer... I don't think this universe is going to be big enough... hmm...
Anand shows that the iphone 4 is SUBSTANTIALLY worse than the 3GS in terms of signal attenuation in four different situations (in one case, by as much as 18db !) This is contrary to Apple's claims that the revolutionary antenna design gives the iphone4 superior RF performance to previous generations.
You are correct that the iphone4 is better than the HTC nexus one in a SINGLE test by 0.5db. It is, however, worse in all remaining tests by 2.5 to 9.1db.
As for signal strength indicators, it has been my casual observation that GSM carriers typically over-estimate the idle "bars" they report. A GSM phone with 1-2 bars is rarely usable for an actual conversation, especially if you are in motion. In contrast, pretty much every CDMA phone I have had could place and receive acceptable calls with 0 bars showing while idle.
Additional thoughts :
- I think it's comical that apple will now make the first 3 bars "taller" to make them "easier to see"
- I believe the fact that they are still selling the 3GS, and specifically re-state that they are willing to take any undamaged phone back within 30 days for a full refund absolves them of all responsibility for this bruhaha... if you keep or purchase an iphone4 at this point, despite the well-publicized reception problems, you have no right to complain, IMHO.
As long as we are going with "things the original author specifically discounted in his post", I think he should purchase VPN service...
But what about the other way? Hearing someone when I've got a lot of background noise. My problem with "in-the-ear" has been that, because the distance is so short and the driver so small, the rate of volume increase / decrease is sharp. In short, I quickly go from "can't hear you" to "too loud, but still can't understand".
The jawbone does a bit of passive isolation in the ear it's inserted, provided you pick the right earpiece gel. It's plenty for use in a car, but I'm not sure if it will be enough for your particular application.
I can second the Jawbone Prime suggestion. The background (non-wind) noise cancellation is fantastic. I use one in the car all the time. The party on the other end of the call is never able to discern that I am in a car, even at highway speed. The Prime's noise cancellation is substantially better than any other BT headset I have ever tried.
I can clarify the verizon thing for you. They were only testing for speed, and only where everyone had coverage. Given those restrictions, the results aren't surprising. I have had data cards and smartphones from AT&T, Verizon, and T-mobile data through various jobs (oftentimes simultaneously). AT&T is generally the fastest when it works. Verizon is by far the most consistent (500kbps~1 mbit almost anywhere in the country in my experience) with the largest 3G network by a HUGE margin. T-mobile is just a disaster all-around (virtually no coverage, so 99% of your time is spent on EDGE).
Verizon is pretty much the only carrier that consistently has 3G coverage outside of the major metro areas. Even in major cities, I usually find that the verizon card has better signal inside buildings. Outside of major cities, it is my experience that it's easier to find a verizon 3G signal than an AT&T edge signal.
NOW if they modified the test by driving around the country, and averaging in zeros when they had no coverage on a particular carrier (or EDGE results when they had no 3G coverage), then I anticipate verizon would win this contest by a HUGE margin.
Personally, for mobile computing I find 1mbit is plenty for me... so I'll take reliability over speed any day.
I don't have any anecdotal sources of my own, sorry. That's why I ask you.
And the measles vaccine was introduced in 1963, which seems to go against your claim that it helped the outspread? If it helped, then... it wasn't by much, or was it?
On a side note, I hate taking it inductively because it always comes down to "who has the most numbers" which doesn't add anything to my understanding...
Yes... don't let pesky things like "numbers" get in the way of your "understanding"...
I've actually read every one of your posts with great interest. You are either one of the densest human beings on the planet, or one of the most skillful trolls I have ever come across. I really can't decide which it is.
Vaccines could be a big scam, I don't know enough to assert. However, I do remain unconvinced of their effectiveness.
What would it take to convince you? It might involve numbers.
To my layman's understanding, I feel sanitation and nutrition play a much bigger role in the immune system than exposing your body to certain proteins ever could. The vaccines can certainly help in an alarmist way, to tell the body to "get ready" for infection, but if your whole system hasn't structured itself right in the first place, it's pointless.
Your layman's understanding is wrong...
A) Would you consider sanitation and nutrition in the United States to be acceptable? (based on your previous posts, I presume "yes", since you assert that this is the principle reason why certain infectious disease that we erroneously vaccinate against are on a decline)
B) Are you aware that millions of people catch the flu annually despite our "modern" sanitation and nutrition? How can you explain this? What are your "feelings" on flu trends? Are more people getting the flu every year, less, or does it stay about the same... compare this to 100 years ago. Do you feel more people got the flu then? now how about smallpox?
C) Are you aware that not all vaccines are universally "mandatory" or even "recommended" in all parts of the country (e.g. chickenpox, HPV, etc. etc. etc.)? How do you explain the gross disparity in infection rates between those who are "vaccinated" versus those that are not. On average, we all share the same "modern" sanitation and nutritional benefits.
It seems that you assumed that I wouldn't suggest first to use wpa2, etc. Seeing as the article is about cracking advanced encryption, I would hope that this is already in place. Poor advice? I think not. It adds additional roadblocks. I also said that it 'helps'. Not that it's a foolproof plan. It just makes it more of a pain to break in.
For example, using a MAC address filter would mean that they would have to spoof a MAC address that you have whitelisted. This requires additional effort and information gathering.
Using a SSID that is not broadcasted, and also not easily guessable (not a dictionary word, and a certain length, etc), makes it harder for SSID crackers to pick it up as well.
You may be happy with just using strong encryption, but I very much prefer enabling these additional security features to harden it even further, even if it is just a little bit further.
I will second what the other two people replying to you have said :
#1) SSID just requires a single deauth to any client. This literally takes 2 seconds to do.
#2) Your clients are broadcasting their MAC addresses in the clear, and it's a fair assumption that any associated client is on your MAC whitelist... Anyone hacking your wireless network is literally staring at these MACS (and probably continuously typing them back into the console).
Anyone with the technical sophistication to go after WPA already knows this (and can bypass both your MAC and SSID measures in LITERALLY 10 seconds)
If enabling these two features makes you FEEL safer then by all means keep them on. But they offer NO additional protection (not even a teeny tiny bit), and are probably a bit of a hassle for you
- Have to add each legit client to MAC table
- Some clients barf on the hidden SSID
Your ONLY effective consumer-level protection at the moment is to pick a completely random long WPA PSK! (even then, there are a few attacks that allow a hacker to decypt WPA packets without knowing the key)
See If you can find some moca bridges (like the ones used in early verizon FIOS installs). I got a pair of motorola NIMS, and can pull 100mbit over the coax in my house all day long (internal signaling is 250mbits, but the port on the device is only 100mbit) . WAY more reliable than the wireless bridge it replaced, and WAY faster than a powerline or phoneline network!
Yeah, my old AT&T GSM (non 3G) phone would not only drive speakers crazy, but also my CRT monitor when it was close enough. I've never directly observed interference from a CDMA phone.
Pick the laptop that you travel with (perhaps between home and work). Store all shared data on it. Samba mount that directory on the OS you are currently using (all of the OS's you list support samba).
You can work on your data on that laptop while on the move... but when you are using another computer (e.g. desktop) that laptop just effectively serves as an external disk. As long as you use GigE, there is no performance drop over a single locally-attached disk.
As an added benefit, you can backup this data whenever you mount the share (i use rsync). This ensures that I automatically have redundant, geographically diverse, versioned backups of my important files!
Except that part of establishing secure communications is authenticating the other end, or else you are vulnerable to a man-in-the-middle attack; that is why SSL has a certificate system. If the only intention was to have a means of encrypting communication, then there would be no reason for SSL to have such a complicated protocol that includes identification and capabilities management.
I agree with you in principle, but practically speaking this system is hobbled by the lowest common denominator. A decade+ ago, I remember having to pay many dollars, talk to a rep over the phone, fax over government issued ID's, and incorporation documents in order to get a signed SSL cert that was included in the browser root. This process took several days, and you paid for the verification step! Nowadays I just add one to my shopping cart, pay ~$5, specify any domain name I want, and hit checkout. If you want a signed SSL cert for a MITM attack of paypal.com you can go out and buy one within 5 minutes! IMHO, practically speaking, the signatures means absolutely nothing anymore!!!
Just tried this on my CDMA HTC touch with the following string sent from google voice %n \%n "%n" '%n' %n . As claimed in the article it did indeed crash TouchFlo! The phone needed a reboot!
This is IDIOTIC. How can any reasonable person possibly buy this argument.
Anyone that wants to bring down a cell phone tower or cell network IS NOT GOING TO CARE whether or not it's LEGAL to screw with the cell radio baseband software. They are ALREADY attempting to do something much worse.
Let's be honest here, the "security" aspect of this argument is a smokescreen. It's blatantly all about the profit!
Furthermore, the cellular network should NOT be so fragile that a single rogue cell phone could take it down (AFAIK it is not). BUT if AT&T is truly insistent on making this argument, then I believe a full investigation by the FCC is mandated. The self-admitted fragile state of their network means that their stewardship of a public resource (radio spectrum) is being poorly managed and truly endangering national security.
While you are correct, in the PARTICULAR case of the iphone, most of the "jailbreaking" has gone hand-in-hand with modifying the cellular baseband to carrier-unlock the phone. The same software pretty much always does both (jailbreak and carrier unlock)!
Chill. It's a cell phone, not a religion. Half the people on here have posted about FIOS and cell phones. I've also posted a ton of "knowledgeable" stuff about physics, computers, electronics, etc. etc. What does THAT tell you about me? I've never worked for a cell company. I don't even own stock in a cell company. I just happen to use the phone (and recently data) a lot, and have traveled pretty extensively in the US. When I see a discussion I can contribute info to, I post a comment.
besides... you're saying I'm TOO knowledgeable to give my opinion on the OP's comment about poor GSM coverage in Maine?!!??! What exactly is the cutoff for being too smart here? A chimp could identify that one magical talking box works while the other does not. Granted, I was only there for a total of approx 3-4 days recently, but it was very quickly evident which phone was the paperweight. Especially once you ventured off the main highways. I hope someone else who actually lives in Maine and has used both GSM and CDMA chimes in here.
Anyway, good job cherry picking posts to highlight. If you even bothered to look at half the comments I've made (here and on howardforums), you would see that I routinely slam Verizon (wireless) pretty hard. Especially on all of the shit they pull with the funny firmware in their phones and more recently their mandatory data plan policies for smart phones. However, I won't deny that I have always had nothing but praise for their voice/data network coverage and reliability.
Some phones already have dual radios. For example, the Blackberry storm comes in a hybrid GSM/CDMA design. When sold by verizon, it has a sim card already installed.
HOWEVER, the phone will not currently roam on domestic GSM carriers. The SIM card and GSM radio are only used for international roaming in countries without a CDMA network. However, that seems to be a contractual limitation rather than a technical one. It likely just doesn't make any business sense right now to offer GSM roaming within the US.
That being said, the future of US cellular is in LTE. Right now it looks like all of the major networks are going down that path. I predict we will be seeing a LOT of hybrid radio designs with bizarro frequency combinations in the coming years (e.g. GSM/UMTS/HSDPA/LTE, CDMA/1X/EVDO/LTE, etc. at all sorts of TV/850/PCS/2100 combos)
All this does is allow infighting for handsets but doesn't solve the problem of crappy service over the US. If the war torn middle east and mount everest can get cell coverage why can't we get decent coverage in maine. Mount everest has people on it 1 month a year, there are over a million people in maine at any given time! I can't use my phone is 1/2 the counties here and that's with the AT&T.
DING DING DING DING... There's your problem! GSM service in North America is a complete joke in my experience. ESPECIALLY once you venture out of any major city or highway! Just look at the coverage maps for each carrier!
I've had both a CDMA and GSM work phone for many years. Traveled through much of the US. I always chuckle when I see some reviewer favorably comparing the two, ESPECIALLY on coverage.
I was actually up in Maine (Bangor and Bar Harbor) just last week. I had my personal verizon phone with me, and a GSM work phone. The GSM phone had a t-mobile sim but all of the carriers seem to mutually roam in Maine. The phone could associate with banner (company) : Cingular (AT&T), US-890 (Unicel), and T-mobile (T-mobile). It autoregistered to any one of those networks depending on the strongest signal. All THREE of those GSM networks combined were completely dwarfed by Verizon's native CDMA coverage. I mean it wasn't even remotely close! Hell, I had full EVDO revA coverage in areas that couldn't even get a regular GSM/GPRS signal.
In my experience, GSM in Canada is no different. For example, I continued up to Cape Breton after Maine. At one point, the closest GSM tower (Rogers) was a hundred miles away! Full CDMA coverage almost all the way up there, and many spots with EVDO!
So... In my opinion, the easiest way fix to your problem with coverage in the boonies is to go visit a verizon store, and just bite the bullet on the BS craptacular locked-down handset they will give you. At least you'll be able to use your phone to... you know... make phone calls...
Just look at the manner in which JPEGs are encoded for your answer!
Take the DCT (discrete cosine transform) of blocks of pixels throughout the image. Examine the frequency content of the each of these blocks and determine the amount of spatial frequency suppression. This will correlate with the quality factor used during compression!
Maybe I'm missing something here... but if you process data in 2GB chunks shouldn't your software just keep it all in memory. Once processing is complete writing it out to one of those SSD arrays should take 10 seconds (which is nothing for 2 hours worth of processing time!!!). If you don't have access to the source code, a quick fix is to just mount a RAM drive.
Furthermore, OCR is stupidly easy to parallelize. The results of each page do not depend on previous pages. You can process each page independently on independent cpus/memory in a cluster of computers WITHOUT needing a supercomputing interconnect. It should pretty much scale linearly with cpu count! Hell for 1GB data/hour even 10base-T is overkill by a factor of 36!
To me, it sounds like you are limited by shitty programming, not disk I/O. Throwing more expensive hardware and faster arrays of expensive SSD's at the problem won't really fix it!
Wow... what world do yo live in?
#1) Most e-mails are still sent in plaintext and unencrypted. Not only are the contents unencrypted, but the communications channel is unencrypted as well. Hell, in my experience most e-mail is still sent using text/html over unencrypted smtp!
#2) Most blogs and e-mails do not have a PGP signature... Those that do usually don't have the signing key stored in any repository, much less any repository that you could establish trust to.
#3) Very few people encrypt their hard drive contents... (hell, not even those that store personal/credit card info). The prevalence of TPM chips and built-in partition encryption in the newer operating systems is just starting to make a teeny tiny dent in this.
I agree that PGP was a revolutionary contribution to computer security. However, it is dishonest to claim that it is as widely used as you claim!
I think all brother inkjets do this now. I agree that it's a HUGE waste of ink if you don't print color regularly. On the flip side, I've never had a clogged head with this printer. hmmm....
It's simply called adaptive optics (AO). In AO, a guidestar is a natural isolated point-like star that is close to your science object (what you are trying to look at). If a laser is used to excite the sodium layer to create an artificial reference, it's called a "laser guidestar".
Anyway, this "trick" is completely different from adaptive optics in both the mathematics and implementation.