That's an AdWords arbitrage site, one that buys cheap clicks to get traffic and sells expensive clicks to its own advertisers. Such sites are just another form of webspam. When Google raised their minimum bid for ads on search, many of those bottom-feeders dropped out, and ad clutter was reduced. Google revenue went down, too, but may recover in time.
First, this is basically Demo '97 technology. The CALTRANS PATH people have been fooling around with this for years. I saw this around 1990 or so up at the CALTRANS Richmond test facility. Automated lane following was demonstrated in 1959 by General Motors with Firebird III.
About the only justification for this is to improve stop accuracy at bus stops so the bus can
get close to the curb without scraping the tires. A bit of automated parking assistance there
might be helpful. A neat trick would be to use rear wheel steering so that
when the bus pulls up to the curb, the bus ends up parallel to the curb. Let the driver drive the front end, and put the back end on autopilot. This would be a big help for articulated buses, which tend to stop with the trailer hanging out in an adjacent lane, and might allow for smaller bus stop zones.
This is far more primitive than DARPA Grand Challenge technologies.
It's no longer acceptable for games to need administrator privileges to install. That's a holdover from the Windows 95/98/ME era, and should have disappeared with Windows XP. With Vista, there's no excuse for it.
If Google uses user comments to affect search, massive attempts would be made by the "search engine optimization" people to game the system. If you thought link farms were bad, phony user farms would be worse. Google won't be able to identify the phonies; they can't even
More fundamentally, there's a scaling problem. As I've pointed out before, the number of raters per site has to be large for rating to work. Rating for movies and TV shows works fine. Hotels might get enough ratings to be useful. Joe's Plumbing will be rated only by Joe, Joe's relatives, and Joe's employees.
CustomizeGoogle and GiveMeBackMyGoogle have some good ideas, although GiveMeBackMyGoogle is probably violating Google's terms of service by redistributing Google search results as a web site. Google lets you annotate their search results via their AJAX API, but you're not allowed to add or delete from their results list.
If you want to delete items from Google search results, you have to do that via a browser plug-in.
(Note, by the way, that Google's Chrome doesn't allow non-Google browser plug-ins. That's a form of DRM, when you think about it.)
With our SiteTruth SiteTruth system, we're addressing the problem by looking at off-web sources of legitimacy. The first question is always "can we find a name and address for the business behind the web site"? We have about four ways to do that. If none of them work, and they're selling something, they get moved down in our search results. If they do have an address, we look them up in various business databases. Considerable data is available about a business, once you can identify it. Ultimately, we want to make the business's credit rating affect their search results. It's necessary to reach out to those hard off-web data sources to separate the real companies from the bottom-feeders. Yes, the "affiliate" crowd will scream. Tough.
As for bottom-feeders, I really like this site, where someone in Brooklyn, NY, took pictures of the storefronts of every Brooklyn photo company he could find that advertised online. It's very funny. Now that's what Google should be doing with StreetView.
They're not encoding video. They're transcoding it. They're starting from one compressed representation and outputting another compressed representation. (Now, with twice the artifacts!)
The good test for this is football. The players, ball, and field are all moving in different directions. If the motion compensation gets that right, it's doing a very good job.
Where's the launch party? One of the big nightclubs, like Ruby Skye or the DNA Lounge? No. Lucasfilm's facility in the Presidio? No. The Metreon, where Sony still has the Playstation store?
No.
It's at the Best Buy on Harrison at 101. That's not a launch party. It's not even a good retail outlet.
At the moment its like saying "this will be really useful for when I genetically engineer a dragon".
Agreed. One of the problems to be solved if fusion reactors ever work is that the inner walls will, over time, become radioactive, which is a maintenance headache. If they could be made of some material that didn't become radioactive when exposed to heavy gamma radiation, that would be helpful, because then you could just turn the thing off and work on it, without having to use remote manipulators and robots.
So it's been suggested that running iron through an isotopic separation plant, like the ones used to enrich uranium, would be useful. The stable isotopes would be used to make the steel for the fusion reactor.
This idea might be helpful in reducing the operating costs of future fusion reactors, but right now, it's just a footnote.
From the article: "The system is composed by software installed on standard computer equipment running on commercial-of-the-shelf Microsoft Windows operating systems."
And that's the problem. It's not running on QNX, or VxWorks, or LynxOS, or MonteVista, or even Windows XP Embedded. With those systems, the system is usually built to have just the components needed to do the job, not with every gimmick Microsoft puts in their desktop OS.
And, of course, it's yet another buffer overflow, part of the defective-by-design semantics C and C++ use for arrays. (And yes, I know what I'm talking about.)
Wow. Take a look at Monday's trading history for UAUA. Look at that drop. And notice that it happened on huge volume; several hundred million dollars changed hands within fifteen minutes. It wasn't just a few traders running the price down in light trading.
The stock hasn't come back all the way. It's still down 20% for the week.
I'm currently dealing with a hosting service that wants a copy of my driver's license before allowing me SSH access. Because their support operation is outsourced (not clear to where), I asked if the information would be transmitted outside the United States, and was verbally told "no".
So I sent an agreement to them for signature, addressed to their general counsel:
Data protection Any personal identity information disclosed under this agreement to "APlus.net" shall be held in confidence. Said information shall not be disclosed to anyone other than a direct APlus.net employee based within the United States. At no time shall said information be transferred outside the borders of the United States, by physical, electronic, or other means. The security measures used to protect such information shall be comparable to those used by the Legal Department of APlus.net to protect the company's own proprietary information.
Indemnification. APlus.net shall indemnify, defend by counsel reasonably accepted by..., protect and hold... harmless from and against any and all claims, liabilities, losses, costs, damages, expenses, including consultants' and attorneys' fees and court costs, demands, causes of action, or judgments directly or indirectly arising out of or related to disclosure of the personal identity information disclosed under this agreement.
We'll have to see what happens.
The "indemnification" clause, by the way, uses the same language they use in their terms of service.
Right now, they're not hosting the live site. So I can pull the plug on them at any time, and if they don't sign, I will.
One of the big-name televanglists (Billy Graham?) had an early computerized system for answering his fan mail. A staff of people read the mail, and used highlighter to mark phrases that contained relevant keywords. Data entry operators keyed in the address and the highlighted phrases. A program used the phrases to select an appropriate canned reply, filled in keywords, added bible citations, and printed out a letter.
If the "fingerprint" is a cryptographic checksum...
That wouldn't be useful. Most modern attacks have at least some variation from copy to copy. "Polymorphic" viruses vary considerably. That's why signature-based recognition doesn't really work any more.
Guy/girl does something goofy in 70s as a teenager.
Gets covered by local news (at that time).
I've seen that already. I looked up an executive, and Google returned a hit from a student newspaper from the 1960s that they'd digitized from microfilm. The story mentioned the guy being
a member of the Socialist Workers Alliance.
A user receives a file that the scan agent deems suspicious (for example, an encrypted or packed file) and for which there is no signature in the local.DAT database.
Using McAfee Artemis Technology, the agent sends a fingerprint of the file for instant lookup to the comprehensive database at McAfee Avert® Labs.
In less than a second, if the fingerprint is identified as known malware, an appropriate response is sent to the user to block or quarantine the file.
In other words, every time you download a binary file, McAfee HQ knows about it and logs it. Was this dreamed up by the RIAA, the NSA, or the anti-child-porno people?
Question. What was ISD? (what does that stand for?)
ISD was Information Systems Design, a time-sharing service bureau for engineering computation.
UCC was University Computing Company, a larger company in the same business. Both used UNIVAC mainframes. ISD was originally in Oakland (and shared a building with the Oakland Raiders), then moved to Santa Clara. UCC had its headquarters in Dallas, with facilities in LA and in Oakland.
Interesting that he had to pay $305,000 for a plotting program in 1971.
One of ISD's competitive advantages in the early 1970s is that they offered remote plotting,
using CALCOMP pen plotters, when almost nobody else did. Engineering companies liked this.
The remote plotting was implemented by emulating a UNIVAC 1004 on a very small minicomputer, then
hooking up a plotter which was fed from the "output card punch" stream. Since the printer/plotter
message protocol had checking and retransmit, this could produce clean plots, unlike
competing systems that used async modems of the period, which had no checking.
All this stuff was much harder back then. The mainframes were 1.2 MIPS machines; the remote minicomputers were something like 0.1 MIPS with 8K of memory.
They missed Jeffery Ward, the first person to do jail time for computer crime.
This was the stone age of computer crime. Ward was convicted of grand theft for stealing a proprietary plotting program from ISD for the benefit of his employer, UCC. One of UCC's customers. Shell, was also an ISD customer, and they had a remote terminal, a UNIVAC 1004, with a card reader, printer, (optional) card punch, and 2400 baud synchronous modem. The customer used the same terminal ID (wired into a plugboard; there weren't really passwords then) to use both UCC and ISD. Ward used a similar terminal at UCC to impersonate the customer's terminal and connect to ISD. Then he submitted a job (on punched cards!) to request that the binary for the plotting program be sent to his terminal and punched on the card punch.
And that's his plan started looking like "America's Dumbest Criminals". The customer terminal he was impersonating didn't have a card punch. So the ISD computer instead punched the desired card deck on a punch in ISD's computer room, and printed a message for the operator indicating who wanted the card deck. The card deck was then packaged up by ISD staff and mailed to Shell.
The package was received at Shell. Since they hadn't ordered it, they sent it back to ISD with a request for a refund. The ISD staff took a look at the card deck, and after some puzzlement, someone realized what it was.
It took a while to figure out what was going on, but the Alameda County DA's office and the Oakland police were brought in, and the first search warrant ever for the search of a computer was issued, to be served on UCC. Nobody was really sure how to do this, but an outside consultant with UNIVAC experience was brought in for the search.
So the big day came. Oakland cops, an assistant DA, and the UNIVAC expert show up at the front door of UCC in Oakland. It's not clear that a search would have found anything; most data
back then was on magnetic tape, and the UCC data center had thousands of reels of tape.
However, Ward was in the building at the time, and he decided to grab all the incriminating material and duck out the back door.
Big mistake for Ward. Cops know about covering the back door. Ward was quickly arrested, and since he had all the incriminating data, the search was unnecessary and Ward was carted off to jail.
There was a later civil settlement between UCC and ISD. ISD got four tape drives and a "CTMC", a 32-line async port controller. (This was a truckload of 1970s technology.)
I worked for ISD when that gear arrived, and it was not in good shape, but we got it working.
The company is unique in that it has enough resources to "indefinitely" maintain a lab of 60 researchers without making any profit.
Now that's promising. The big lesson of the DARPA Grand Challenge is that mobile robotics takes about 10x the resources that typical academic groups had previously been applying to the problem, and with sufficient resources, the problems start to yield. Previous automatic driving efforts had been a professor and a few grad students. Once the efforts were scaled up to NASCAR team size, things started to work.
Willow Garage has had a few projects. They did an autonomous model boat.
They started on a driverless car, but never got very far in that direction.
They showed the Stanford PR1 robot at RoboDevelopment two years ago, but
their own second generation version is still at the parts-prototyping stage.
Anybots is probably further along. Take a look at their pictures. I've seen that machine in operation. Balance is automatic, but manipulation and movement are teleoperated.
The Internet appears to be slowing down because of issues at the endpoints, not the network.
Many web sites are pulling in a huge number of files just to display a simple page. Javascript libraries and CSS files require extra network transactions to load, and if there's any delay at the servers for any of the components, page display stalls.
Then there are ads. It's not uncommon for a page to be hitting ten different servers related to advertising content and other dreck. Often, the page won't load until some ad comes in, and some of the ad-serving services aren't that quick.
These effects can cause browser stalls when the browser hits its connection limit or some operation with a lock set stalls halfway through waiting for a server. Then you get a brief browser freeze. Firefox seems to have this problem.
When you see delays of seconds to load a web page, that's usually the reason.
The Samsung Jitterbug and the Owasys 22C (screenless) are useful for this.
Current phones tend to be terrible at screenless operation. For use while driving, it should be possible to do everything important with voice, through a wireless headset. But that's unnecessarily hard with many phones. Reviews don't address this issue well. Things like the speed of voice recognition are important. Samsung phones seem to have voice recognition that takes 5-8 seconds to load, then about as long to recognize a name on the speed dial list. Motorola was doing better than that five years ago, and Wildfire (a centralized service) did even better eight years ago.
"Don't be evil"
on
Google Turns 10
·
· Score: 5, Interesting
The problem with Google is that their "don't be evil" claim is hard to take seriously any more.
Ads at the right of search results weren't too bad, but then it went downhill. They created the "content-related ad" industry, which resulted in a vast number of "made for AdWords" junk sites and blogs, the "domaining" industry, and a vast amount of crap. Even real advertisers don't like it; the smarter ones opt out of the Google Content Network and stick with the search result ads.
From there it went downhill. Google doesn't do much to qualify their advertisers, and as we point out occasionally, about 35% of them are "bottom feeders", where you can't even identify the real business behind the ad.
Then there's Google Checkout. They accept very marginal businesses. They ought to be doing the kind of validation a bank does of its clients, but clearly, they don't.
Google's real problem is that they went public at the top of their game. Google was #1 in search when they went public, so they couldn't grow in their main business area. They had to expand to justify their high P/E ratio, and none of their expansion areas (YouTube, GMail, etc.) made money. So they had to figure out how to get more revenue per search result.
At that point they started to turn to the dark side.
Slashdot Mirror
Is this the "Yahoo BOSS API" being relaunched, or something else?
That's an AdWords arbitrage site, one that buys cheap clicks to get traffic and sells expensive clicks to its own advertisers. Such sites are just another form of webspam. When Google raised their minimum bid for ads on search, many of those bottom-feeders dropped out, and ad clutter was reduced. Google revenue went down, too, but may recover in time.
First, this is basically Demo '97 technology. The CALTRANS PATH people have been fooling around with this for years. I saw this around 1990 or so up at the CALTRANS Richmond test facility. Automated lane following was demonstrated in 1959 by General Motors with Firebird III.
About the only justification for this is to improve stop accuracy at bus stops so the bus can get close to the curb without scraping the tires. A bit of automated parking assistance there might be helpful. A neat trick would be to use rear wheel steering so that when the bus pulls up to the curb, the bus ends up parallel to the curb. Let the driver drive the front end, and put the back end on autopilot. This would be a big help for articulated buses, which tend to stop with the trailer hanging out in an adjacent lane, and might allow for smaller bus stop zones.
This is far more primitive than DARPA Grand Challenge technologies.
don't games require direct interaction with the video hardware to run?
1994 called. It wants VGA mode 13h back.
It's no longer acceptable for games to need administrator privileges to install. That's a holdover from the Windows 95/98/ME era, and should have disappeared with Windows XP. With Vista, there's no excuse for it.
If Google uses user comments to affect search, massive attempts would be made by the "search engine optimization" people to game the system. If you thought link farms were bad, phony user farms would be worse. Google won't be able to identify the phonies; they can't even More fundamentally, there's a scaling problem. As I've pointed out before, the number of raters per site has to be large for rating to work. Rating for movies and TV shows works fine. Hotels might get enough ratings to be useful. Joe's Plumbing will be rated only by Joe, Joe's relatives, and Joe's employees.
CustomizeGoogle and GiveMeBackMyGoogle have some good ideas, although GiveMeBackMyGoogle is probably violating Google's terms of service by redistributing Google search results as a web site. Google lets you annotate their search results via their AJAX API, but you're not allowed to add or delete from their results list. If you want to delete items from Google search results, you have to do that via a browser plug-in. (Note, by the way, that Google's Chrome doesn't allow non-Google browser plug-ins. That's a form of DRM, when you think about it.)
With our SiteTruth SiteTruth system, we're addressing the problem by looking at off-web sources of legitimacy. The first question is always "can we find a name and address for the business behind the web site"? We have about four ways to do that. If none of them work, and they're selling something, they get moved down in our search results. If they do have an address, we look them up in various business databases. Considerable data is available about a business, once you can identify it. Ultimately, we want to make the business's credit rating affect their search results. It's necessary to reach out to those hard off-web data sources to separate the real companies from the bottom-feeders. Yes, the "affiliate" crowd will scream. Tough.
As for bottom-feeders, I really like this site, where someone in Brooklyn, NY, took pictures of the storefronts of every Brooklyn photo company he could find that advertised online. It's very funny. Now that's what Google should be doing with StreetView.
Here's our master plan for cleaning up the Web.
They're not encoding video. They're transcoding it. They're starting from one compressed representation and outputting another compressed representation. (Now, with twice the artifacts!)
The good test for this is football. The players, ball, and field are all moving in different directions. If the motion compensation gets that right, it's doing a very good job.
Where's the launch party? One of the big nightclubs, like Ruby Skye or the DNA Lounge? No. Lucasfilm's facility in the Presidio? No. The Metreon, where Sony still has the Playstation store? No.
It's at the Best Buy on Harrison at 101. That's not a launch party. It's not even a good retail outlet.
This article is from 2007. Didn't we cover this once before?
If you want a really cheap wind turbine, the usual answer is to chop an old oil drum in half and mount the halves on the end of an old auto alternator. Then you can charge a car battery.
At the moment its like saying "this will be really useful for when I genetically engineer a dragon".
Agreed. One of the problems to be solved if fusion reactors ever work is that the inner walls will, over time, become radioactive, which is a maintenance headache. If they could be made of some material that didn't become radioactive when exposed to heavy gamma radiation, that would be helpful, because then you could just turn the thing off and work on it, without having to use remote manipulators and robots.
So it's been suggested that running iron through an isotopic separation plant, like the ones used to enrich uranium, would be useful. The stable isotopes would be used to make the steel for the fusion reactor.
This idea might be helpful in reducing the operating costs of future fusion reactors, but right now, it's just a footnote.
From the article: "The system is composed by software installed on standard computer equipment running on commercial-of-the-shelf Microsoft Windows operating systems."
And that's the problem. It's not running on QNX, or VxWorks, or LynxOS, or MonteVista, or even Windows XP Embedded. With those systems, the system is usually built to have just the components needed to do the job, not with every gimmick Microsoft puts in their desktop OS.
And, of course, it's yet another buffer overflow, part of the defective-by-design semantics C and C++ use for arrays. (And yes, I know what I'm talking about.)
Wow. Take a look at Monday's trading history for UAUA. Look at that drop. And notice that it happened on huge volume; several hundred million dollars changed hands within fifteen minutes. It wasn't just a few traders running the price down in light trading.
The stock hasn't come back all the way. It's still down 20% for the week.
Here's the newspaper page that started it all, as archived by Google.
I'm currently dealing with a hosting service that wants a copy of my driver's license before allowing me SSH access. Because their support operation is outsourced (not clear to where), I asked if the information would be transmitted outside the United States, and was verbally told "no". So I sent an agreement to them for signature, addressed to their general counsel:
Data protection Any personal identity information disclosed under this agreement to "APlus.net" shall be held in confidence. Said information shall not be disclosed to anyone other than a direct APlus.net employee based within the United States. At no time shall said information be transferred outside the borders of the United States, by physical, electronic, or other means. The security measures used to protect such information shall be comparable to those used by the Legal Department of APlus.net to protect the company's own proprietary information.
Indemnification. APlus.net shall indemnify, defend by counsel reasonably accepted by ..., protect and hold ... harmless from and against any and all claims, liabilities, losses, costs, damages, expenses, including consultants' and attorneys' fees and court costs, demands, causes of action, or judgments directly or indirectly arising out of or related to disclosure of the personal identity information disclosed under this agreement.
We'll have to see what happens. The "indemnification" clause, by the way, uses the same language they use in their terms of service.
Right now, they're not hosting the live site. So I can pull the plug on them at any time, and if they don't sign, I will.
One of the big-name televanglists (Billy Graham?) had an early computerized system for answering his fan mail. A staff of people read the mail, and used highlighter to mark phrases that contained relevant keywords. Data entry operators keyed in the address and the highlighted phrases. A program used the phrases to select an appropriate canned reply, filled in keywords, added bible citations, and printed out a letter.
If the "fingerprint" is a cryptographic checksum...
That wouldn't be useful. Most modern attacks have at least some variation from copy to copy. "Polymorphic" viruses vary considerably. That's why signature-based recognition doesn't really work any more.
Guy/girl does something goofy in 70s as a teenager. Gets covered by local news (at that time).
I've seen that already. I looked up an executive, and Google returned a hit from a student newspaper from the 1960s that they'd digitized from microfilm. The story mentioned the guy being a member of the Socialist Workers Alliance.
Here's McAfee's explanation of how it works:
In other words, every time you download a binary file, McAfee HQ knows about it and logs it. Was this dreamed up by the RIAA, the NSA, or the anti-child-porno people?
ISD was Information Systems Design, a time-sharing service bureau for engineering computation. UCC was University Computing Company, a larger company in the same business. Both used UNIVAC mainframes. ISD was originally in Oakland (and shared a building with the Oakland Raiders), then moved to Santa Clara. UCC had its headquarters in Dallas, with facilities in LA and in Oakland.
Interesting that he had to pay $305,000 for a plotting program in 1971.
One of ISD's competitive advantages in the early 1970s is that they offered remote plotting, using CALCOMP pen plotters, when almost nobody else did. Engineering companies liked this. The remote plotting was implemented by emulating a UNIVAC 1004 on a very small minicomputer, then hooking up a plotter which was fed from the "output card punch" stream. Since the printer/plotter message protocol had checking and retransmit, this could produce clean plots, unlike competing systems that used async modems of the period, which had no checking.
All this stuff was much harder back then. The mainframes were 1.2 MIPS machines; the remote minicomputers were something like 0.1 MIPS with 8K of memory.
They missed Jeffery Ward, the first person to do jail time for computer crime.
This was the stone age of computer crime. Ward was convicted of grand theft for stealing a proprietary plotting program from ISD for the benefit of his employer, UCC. One of UCC's customers. Shell, was also an ISD customer, and they had a remote terminal, a UNIVAC 1004, with a card reader, printer, (optional) card punch, and 2400 baud synchronous modem. The customer used the same terminal ID (wired into a plugboard; there weren't really passwords then) to use both UCC and ISD. Ward used a similar terminal at UCC to impersonate the customer's terminal and connect to ISD. Then he submitted a job (on punched cards!) to request that the binary for the plotting program be sent to his terminal and punched on the card punch.
And that's his plan started looking like "America's Dumbest Criminals". The customer terminal he was impersonating didn't have a card punch. So the ISD computer instead punched the desired card deck on a punch in ISD's computer room, and printed a message for the operator indicating who wanted the card deck. The card deck was then packaged up by ISD staff and mailed to Shell.
The package was received at Shell. Since they hadn't ordered it, they sent it back to ISD with a request for a refund. The ISD staff took a look at the card deck, and after some puzzlement, someone realized what it was.
It took a while to figure out what was going on, but the Alameda County DA's office and the Oakland police were brought in, and the first search warrant ever for the search of a computer was issued, to be served on UCC. Nobody was really sure how to do this, but an outside consultant with UNIVAC experience was brought in for the search.
So the big day came. Oakland cops, an assistant DA, and the UNIVAC expert show up at the front door of UCC in Oakland. It's not clear that a search would have found anything; most data back then was on magnetic tape, and the UCC data center had thousands of reels of tape. However, Ward was in the building at the time, and he decided to grab all the incriminating material and duck out the back door.
Big mistake for Ward. Cops know about covering the back door. Ward was quickly arrested, and since he had all the incriminating data, the search was unnecessary and Ward was carted off to jail.
There was a later civil settlement between UCC and ISD. ISD got four tape drives and a "CTMC", a 32-line async port controller. (This was a truckload of 1970s technology.) I worked for ISD when that gear arrived, and it was not in good shape, but we got it working.
The company is unique in that it has enough resources to "indefinitely" maintain a lab of 60 researchers without making any profit.
Now that's promising. The big lesson of the DARPA Grand Challenge is that mobile robotics takes about 10x the resources that typical academic groups had previously been applying to the problem, and with sufficient resources, the problems start to yield. Previous automatic driving efforts had been a professor and a few grad students. Once the efforts were scaled up to NASCAR team size, things started to work.
Willow Garage has had a few projects. They did an autonomous model boat. They started on a driverless car, but never got very far in that direction. They showed the Stanford PR1 robot at RoboDevelopment two years ago, but their own second generation version is still at the parts-prototyping stage.
Anybots is probably further along. Take a look at their pictures. I've seen that machine in operation. Balance is automatic, but manipulation and movement are teleoperated.
The Internet appears to be slowing down because of issues at the endpoints, not the network. Many web sites are pulling in a huge number of files just to display a simple page. Javascript libraries and CSS files require extra network transactions to load, and if there's any delay at the servers for any of the components, page display stalls.
Then there are ads. It's not uncommon for a page to be hitting ten different servers related to advertising content and other dreck. Often, the page won't load until some ad comes in, and some of the ad-serving services aren't that quick.
These effects can cause browser stalls when the browser hits its connection limit or some operation with a lock set stalls halfway through waiting for a server. Then you get a brief browser freeze. Firefox seems to have this problem.
When you see delays of seconds to load a web page, that's usually the reason.
The Samsung Jitterbug and the Owasys 22C (screenless) are useful for this.
Current phones tend to be terrible at screenless operation. For use while driving, it should be possible to do everything important with voice, through a wireless headset. But that's unnecessarily hard with many phones. Reviews don't address this issue well. Things like the speed of voice recognition are important. Samsung phones seem to have voice recognition that takes 5-8 seconds to load, then about as long to recognize a name on the speed dial list. Motorola was doing better than that five years ago, and Wildfire (a centralized service) did even better eight years ago.
The problem with Google is that their "don't be evil" claim is hard to take seriously any more. Ads at the right of search results weren't too bad, but then it went downhill. They created the "content-related ad" industry, which resulted in a vast number of "made for AdWords" junk sites and blogs, the "domaining" industry, and a vast amount of crap. Even real advertisers don't like it; the smarter ones opt out of the Google Content Network and stick with the search result ads.
From there it went downhill. Google doesn't do much to qualify their advertisers, and as we point out occasionally, about 35% of them are "bottom feeders", where you can't even identify the real business behind the ad.
Then there's Google Checkout. They accept very marginal businesses. They ought to be doing the kind of validation a bank does of its clients, but clearly, they don't.
Google's real problem is that they went public at the top of their game. Google was #1 in search when they went public, so they couldn't grow in their main business area. They had to expand to justify their high P/E ratio, and none of their expansion areas (YouTube, GMail, etc.) made money. So they had to figure out how to get more revenue per search result. At that point they started to turn to the dark side.