you don't need to "do" anything with the data itself but selling it to companies that do have a purpose for it... mostly SEO and marketing.
Buy, my point is that Google doesn't sell data to other companies, except in aggregate, anonymized sets (and even that is a trivial, negligible part of Google's business).
The Ads aren't manager by Google but by a bunch of semi-literate imbeciles called marketeers that buy the data delivered by Google.
This is wrong, top to bottom.
First, Google does all of the ad display selection. Advertisers ("marketeers") have little to no control over who Google displays their ads to. Google runs a real time auction taking into account the advertiser's per-click bid, the estimated probability that a given user will click a given ad, and a measure of ad "quality" which attempts to quantify the experience the user will have if they click on the ad (if users are generally happy with what they find when they click Google ads, they're more likely to click more of them).
Second, Google does not provide any of the user data to advertisers. There is some sale of aggregated, anonymized statistics, but Google keeps the individual data to itself, partly because Google doesn't trust advertisers not to be irresponsible with user privacy, and partly because Google is much better at ad targeting, so it's actually worth more money if Google keeps it.
I was working until last month for one of the big players in web analysis... and you would cry like I did with the type individuals that are doing all the "smart advertising" thing.
I have no doubt there are a lot of idiots out there trying to do this stuff... but they don't have Google's data to do it with.
Let me rephrase how I understand AC's point: It causes the programmer to rely on a proprietary automated build system instead of a free, portable automated build system.
Meh.
If you want to find something to complain about with respect to IDEs and integrated build systems, here's a legitimate one, at least for new programmers: it can prevent them from learning how their code is assembled into a functioning binary. With Makefiles, it's almost impossible not to understand the distinction between an editor, a compiler and a linker (and you might even get exposed to an assembler from time to time).
That said, once you do understand how the pieces fit together and work, there's no reason to create additional work for yourself. In addition, there's plenty to be said for simplified environments for new programmers, so they can focus on the code. They can learn the nuts and bolts later. Once you're beyond that, if portability matters then you may have a good reason to stick with make or another portable build system. Or if you have concerns about being tied to a proprietary system which may not be around long enough, or may tie you to tools you might wish to replace (though, generally, switching to a new build system is the least difficult part of that process), then it might make sense.
Otherwise, use the tool that makes you most productive.
QNX is clearly a better choice for a system that should just work, all of the time. However, I doubt it's really all that unreliable, and the bigger problem with Sync is that the UI is horrible, among the worst I've ever seen. I've had a couple of rental cars with it, and the last time Hertz offered me a Ford, I told them I wanted a different car, it's that bad. And the UI isn't Microsoft's fault, I don't think.
Except for the eco-proud, nobody wants a car that looks like a Leaf, or a Prius, or anything like an economy car.
Meh. I couldn't care less what my car looks like. It's for transportation.
I want it to be comfortable, reliable, cost-efficient and to be able to do what I need. So I have a LEAF for commuting and a Durango for hauling my family around, driving in the mountains, pulling the boat or camp trailer, etc. I don't wash either of them. I'm sure there are lots of people to whom the appearance of their vehicle is important, but I'm equally sure that there are lots like me who just don't give a damn.
He has had to have 4 software updates done on the car
That doesn't ring true. There has only been one required software update in the last two years. And it was free. And I didn't even bother with it (it's supposed to make the transition between regenerative and friction braking smoother, but I've already adapted to the sharp transition).
as well as a couple of electric issues
I suppose that's possible, but it's rather bizarre. The LEAF is quite popular where I work, and among the 8 vehicles there none of us has has any issues at all.
Mine has been in 0 times, I can stop to get an oil change done in 10 minutes on my way home
Your oil changes are maintenance, so you've been in every 4K miles (or whatever). My LEAF, in comparison, has gone 20,000 miles with no maintenance other than tire rotations at 7,500 and 15,000. I haven't spent a single penny on maintenance, of any sort, and I don't expect to until I have to replace the tires.
Oh, sorry, I guess there is one thing I've had to spend money on: wiper fluid.
Price difference, I will be ahead of him for 10 years at $4.00 a gallon gas before the Leaf saves him enough money to reduce its TCO to my civic.
Nope. I did the math before I bought my LEAF, and against the most cost-efficient ICE vehicle I could find (the $18,000 Honda Insight hybrid), the LEAF broke even at 6 years -- without considering the tax credits. With the tax credits the LEAF beat out every ICE and hybrid in 2-3 years for my driving patterns.
That is if he has no major service needed.
The thing about electrics is that there is basically nothing to go wrong. The battery will wear out eventually, but it's warrantied for 8 years. Other than that... electric motors are extremely reliable, there is no transmission, no clutch, there are no fluids to change, no exhaust system... the car is just so darned simple that there's hardly anything that can go wrong.
You buy a leaf for other reasons, less service and operating cost are not them.
I bought mine for low total cost of ownership. Period. I actually didn't factor in service costs because I didn't have any way to estimate them, so I just looked at purchase price and fuel costs. Now that I've had it for a couple of years, I know that my calculations were pessimistic.
Most other electric cars are firmly in the D category. They are both worse and more expensive. None of this is a game breaker by itself, but the combination is.
Nonsense. I bought a LEAF specifically because it's less expensive than a comparable gasoline-powered car. The range is limited, but I have to have another car anyway, so we just use that one for long trips. The LEAF is unbeatable as a commuter vehicle.
Here's a better idea: Scale the US Army down to about 100,000, or less. Retain a small full-time force to man the equipment and technology-heavy portions of an army (e.g. armor, artillery and highly-specialized forces), though even most of those can be turned over to national guard forces (especially artillery), and use the rest to form a training and logistics cadre whose job it is to prepare to train and equip an actual army, should we need one.
To make that easier, encourage the unorganized militia to self-train and equip. Expand the Civilian Marksmanship Program, funding more rifle ranges and more competitive shooting events, especially dynamic events like three-gun competitions and others that attempt to simulate the level of fitness and the skills required for combat. Make competitive shooting a widespread high school and college sport.
Of course, a tiny professional army designed to be filled out by quickly recruiting and equipping a large pool of semi-trained civilians is great for defending the nation against invasion, as per the apocryphal quote attributed to Admiral Yamamoto "You cannot invade the mainland United States, there would be a rifle behind every blade of grass", but terrible for projecting force around the world. I happen to think that's a good thing.
This, of course, is a design closer to that which was intended by the authors of the US Constitution.
Strictly speaking, you DON'T have to have a Gmail account for any Android phone, only the ones with stock ROMs.
This is, perhaps, a bit pedantic, but you don't need a Gmail account even then. You need a Google account, but you can set up a Google account using any e-mail address. In general this is a distinction without a difference, since the only difference is Google is handling the e-mail, and you can always create a Gmail account that you don't use for e-mail.
With a 50% divorce rate in the US, if you have children and you're the primary wage earner, it is likely you...
Or you could choose to take care of your marriage. Keeping a relationship strong is a lot of hard work, and requires a lot of sacrifice, but it can be done. Note that it's important that your spouse understand this and be willing going in to invest the effort required, but your willingness can and will affect your spouse's willingness.
Some specific recommendations, though this list is more descriptive than prescriptive. The specifics may vary; it's the attitude and intent that matters:
1. Keep the courtship going. Establish a standard that, barring some extreme circumstances, you and your spouse go on a date at least once per week. On your date, make a concerted effort not to talk about kids, house, etc. You need to continue finding and building common ground, because both of you will gradually change over time and it's very, very easy to let your shared responsibilities become the whole of your topics of discussion. Don't let that happen. People don't "fall out of love", they allow themselves to grow apart. My wife and I go on a date every Friday night -- generally just dinner & a movie, but something every week, without fail (or close to it).
2. Find/build some shared interests. Both of you will likely have to give a little and do things that you don't really love in order to do something that the other does really love. For example, my wife has taken up Scuba diving because I love it; I have taken ballroom dance courses because she loves it. She likes diving and I like dancing, and we both love doing both together.
3. Maintain the attitude that your spouse is the most important thing in your life. More important than your kids, more important than your job, more important than anything else.
4. Take care that you do not allow anyone else to get close enough to you that they begin to encroach on what should be your spouse's territory. If you find yourself sharing your very personal feelings and beliefs extensively with another person (especially of the opposite gender, if you're heterosexual), STOP. Allowing those sorts of relationships to grow risks supplanting your relationship with your spouse. It's not a bad idea to take it a step further and simply refuse to spend time alone with other women (or men, depending). Sometimes work demands that you, for example, meet with a female subordinate or manager, and you do what you have to, but keep it strictly professional and limit the frequency and duration as much as possible.
5. Don't be afraid to avail yourself of professional help, and to do it at the first sign of discontent. A good marriage counselor can help you both understand things that you can do better to maintain and improve your relationship.
In every way, focus on first and foremost defending your marriage and keeping it strong. If this seems like too much work and/or too limiting, then accept that you're likely to get divorced, and plan accordingly.
Personally, I've been married for 23 years and I don't expect ever even to contemplate divorce.
Some people actually believe everything you wrote.
For any ridiculous idea you care to name you can find some people who believe it. That doesn't change the fact that the GP is making strawman arguments.
Have you tried putting a slash on the end? We use a lot of single-word URLs internally and putting a slash on the end always convinces Chrome that I mean it as a URL not a search keyword.
I don't think "controversial UI" counts as "evil."
Not even all that controversial, since the UI change happened about a year ago and Chrome market share has continued climbing.
The only way to avoid annoying some percentage of your user base with UI changes is to never change the UI. Even clear improvements will generate screams of outrage from a few percent of the users, just because they don't like change.
In any case, if people don't like Chrome's UI, there are plenty of other options. If you really dislike this change, just use a different browser that you like better. If enough people do that, Google will get the message.
I used to think so too, then they reduced the highlighting of the ads that surround the search results (so that people are more likely to click on them)
They're still clearly marked, and studies show no change in the ability of users to distinguish them from the search results.
then started pushing people to G+
That has been handled poorly, I agree. G+ is the single Google account system, which makes a lot of sense, and it's also a social network. Unfortunately, they were introduced in the wrong order. If Google had consolidated the accounts first and then added a social network later people would have understood the goals better.
In any case, neither of those things indicate any kind of a shift in Google's culture or attitude, not if you understand them and the rationale behind them.
as Upton Sinclair pointed out, when your salary depends on something you tend to start to decide it's OK.
My salary doesn't depend on my agreeing with Google's actions. For that matter, if I were too annoyed getting another job would be trivial.
That may be true financially, but it's not the way Google sees it. Seriously, the perspective within Google is that users are customers.
its the same way as the postal service in the US re: spam. I once asked the letter carrier if we could put a sign up on our mailboxes saying 'no junk mail or UCE please'. he said that this is how they make most of their money these days and that they are paying to have their 'junk' put in my inbox. you can see who works for whom; and its not the recipient!
Google doesn't make any money from e-mail spam. Exactly the opposite. Google (like any mail service provider) spends huge amounts of resources receiving, processing and discarding spam Google does make money on the ads in gmail, and those are most effective if you only get real mail, because then the ads can be targeted based on stuff you really care about.
google also makes money on their search and have you noticed that when you search for something that has a tech nature and also a salesy nature, the sales stuff comes first and you have to trawl thru many pages to find the actual tech info?
That would be because most people who search for that salesy stuff are looking to buy. Outside of the links marked explicitly as sponsored, Google does not play games with the results.
have you noticed that there are PAGES of fake sites that serve only ebay ads, trying to fool you into clicking on them? google does nothing (zero, nada, zilch) to stop this even though the Powerful Google could easily fix this if they wanted.
It's actually not so easy. Google puts a lot of effort into eliminating crap results like this, but it's a gigantic game of Whack-a-Mole to do it site by site, so Google doesn't do that. Instead, Google adjusts ranking algorithms to identify and eliminate the crap, but changing the ranking algorithm is a Big Deal that requires vast amounts of analysis and testing, so it only happens occasionally. Generally, when the algorithm is changed most of the garbage sites drop out of the top results and things are better for a while, but then the SEO people learn how to game the new algorithm and they start to rise again.
(I'm a Google engineer, though I don't work on search, or gmail.)
Slashdot Mirror
The private data linked to individuals is of no use for anybody not even for Google.
You really don't know what you're talking about.
s/Buy/But/
you don't need to "do" anything with the data itself but selling it to companies that do have a purpose for it... mostly SEO and marketing.
Buy, my point is that Google doesn't sell data to other companies, except in aggregate, anonymized sets (and even that is a trivial, negligible part of Google's business).
It's naive to assume that advertisers will remain google's largest customers.
Google doesn't sell user data, to advertisers or anyone else.
The Ads aren't manager by Google but by a bunch of semi-literate imbeciles called marketeers that buy the data delivered by Google.
This is wrong, top to bottom.
First, Google does all of the ad display selection. Advertisers ("marketeers") have little to no control over who Google displays their ads to. Google runs a real time auction taking into account the advertiser's per-click bid, the estimated probability that a given user will click a given ad, and a measure of ad "quality" which attempts to quantify the experience the user will have if they click on the ad (if users are generally happy with what they find when they click Google ads, they're more likely to click more of them).
Second, Google does not provide any of the user data to advertisers. There is some sale of aggregated, anonymized statistics, but Google keeps the individual data to itself, partly because Google doesn't trust advertisers not to be irresponsible with user privacy, and partly because Google is much better at ad targeting, so it's actually worth more money if Google keeps it.
I was working until last month for one of the big players in web analysis... and you would cry like I did with the type individuals that are doing all the "smart advertising" thing.
I have no doubt there are a lot of idiots out there trying to do this stuff... but they don't have Google's data to do it with.
(I work for Google, though not on advertising.)
Let me rephrase how I understand AC's point: It causes the programmer to rely on a proprietary automated build system instead of a free, portable automated build system.
Meh.
If you want to find something to complain about with respect to IDEs and integrated build systems, here's a legitimate one, at least for new programmers: it can prevent them from learning how their code is assembled into a functioning binary. With Makefiles, it's almost impossible not to understand the distinction between an editor, a compiler and a linker (and you might even get exposed to an assembler from time to time).
That said, once you do understand how the pieces fit together and work, there's no reason to create additional work for yourself. In addition, there's plenty to be said for simplified environments for new programmers, so they can focus on the code. They can learn the nuts and bolts later. Once you're beyond that, if portability matters then you may have a good reason to stick with make or another portable build system. Or if you have concerns about being tied to a proprietary system which may not be around long enough, or may tie you to tools you might wish to replace (though, generally, switching to a new build system is the least difficult part of that process), then it might make sense.
Otherwise, use the tool that makes you most productive.
QNX is clearly a better choice for a system that should just work, all of the time. However, I doubt it's really all that unreliable, and the bigger problem with Sync is that the UI is horrible, among the worst I've ever seen. I've had a couple of rental cars with it, and the last time Hertz offered me a Ford, I told them I wanted a different car, it's that bad. And the UI isn't Microsoft's fault, I don't think.
if you want to compete in the American market you have to sacrifice some function for form.
Sacrifice function for form and I'll buy the car that doesn't.
Car Manufacturers need to take a look at the Chevy Volt for ideas on how to style their green machines
I compared the Volt and the LEAF, and bought a LEAF.
Except for the eco-proud, nobody wants a car that looks like a Leaf, or a Prius, or anything like an economy car.
Meh. I couldn't care less what my car looks like. It's for transportation.
I want it to be comfortable, reliable, cost-efficient and to be able to do what I need. So I have a LEAF for commuting and a Durango for hauling my family around, driving in the mountains, pulling the boat or camp trailer, etc. I don't wash either of them. I'm sure there are lots of people to whom the appearance of their vehicle is important, but I'm equally sure that there are lots like me who just don't give a damn.
He has had to have 4 software updates done on the car
That doesn't ring true. There has only been one required software update in the last two years. And it was free. And I didn't even bother with it (it's supposed to make the transition between regenerative and friction braking smoother, but I've already adapted to the sharp transition).
as well as a couple of electric issues
I suppose that's possible, but it's rather bizarre. The LEAF is quite popular where I work, and among the 8 vehicles there none of us has has any issues at all.
Mine has been in 0 times, I can stop to get an oil change done in 10 minutes on my way home
Your oil changes are maintenance, so you've been in every 4K miles (or whatever). My LEAF, in comparison, has gone 20,000 miles with no maintenance other than tire rotations at 7,500 and 15,000. I haven't spent a single penny on maintenance, of any sort, and I don't expect to until I have to replace the tires.
Oh, sorry, I guess there is one thing I've had to spend money on: wiper fluid.
Price difference, I will be ahead of him for 10 years at $4.00 a gallon gas before the Leaf saves him enough money to reduce its TCO to my civic.
Nope. I did the math before I bought my LEAF, and against the most cost-efficient ICE vehicle I could find (the $18,000 Honda Insight hybrid), the LEAF broke even at 6 years -- without considering the tax credits. With the tax credits the LEAF beat out every ICE and hybrid in 2-3 years for my driving patterns.
That is if he has no major service needed.
The thing about electrics is that there is basically nothing to go wrong. The battery will wear out eventually, but it's warrantied for 8 years. Other than that... electric motors are extremely reliable, there is no transmission, no clutch, there are no fluids to change, no exhaust system... the car is just so darned simple that there's hardly anything that can go wrong.
You buy a leaf for other reasons, less service and operating cost are not them.
I bought mine for low total cost of ownership. Period. I actually didn't factor in service costs because I didn't have any way to estimate them, so I just looked at purchase price and fuel costs. Now that I've had it for a couple of years, I know that my calculations were pessimistic.
Most other electric cars are firmly in the D category. They are both worse and more expensive. None of this is a game breaker by itself, but the combination is.
Nonsense. I bought a LEAF specifically because it's less expensive than a comparable gasoline-powered car. The range is limited, but I have to have another car anyway, so we just use that one for long trips. The LEAF is unbeatable as a commuter vehicle.
Here's a better idea: Scale the US Army down to about 100,000, or less. Retain a small full-time force to man the equipment and technology-heavy portions of an army (e.g. armor, artillery and highly-specialized forces), though even most of those can be turned over to national guard forces (especially artillery), and use the rest to form a training and logistics cadre whose job it is to prepare to train and equip an actual army, should we need one.
To make that easier, encourage the unorganized militia to self-train and equip. Expand the Civilian Marksmanship Program, funding more rifle ranges and more competitive shooting events, especially dynamic events like three-gun competitions and others that attempt to simulate the level of fitness and the skills required for combat. Make competitive shooting a widespread high school and college sport.
Of course, a tiny professional army designed to be filled out by quickly recruiting and equipping a large pool of semi-trained civilians is great for defending the nation against invasion, as per the apocryphal quote attributed to Admiral Yamamoto "You cannot invade the mainland United States, there would be a rifle behind every blade of grass", but terrible for projecting force around the world. I happen to think that's a good thing.
This, of course, is a design closer to that which was intended by the authors of the US Constitution.
Strictly speaking, you DON'T have to have a Gmail account for any Android phone, only the ones with stock ROMs.
This is, perhaps, a bit pedantic, but you don't need a Gmail account even then. You need a Google account, but you can set up a Google account using any e-mail address. In general this is a distinction without a difference, since the only difference is Google is handling the e-mail, and you can always create a Gmail account that you don't use for e-mail.
With a 50% divorce rate in the US, if you have children and you're the primary wage earner, it is likely you...
Or you could choose to take care of your marriage. Keeping a relationship strong is a lot of hard work, and requires a lot of sacrifice, but it can be done. Note that it's important that your spouse understand this and be willing going in to invest the effort required, but your willingness can and will affect your spouse's willingness.
Some specific recommendations, though this list is more descriptive than prescriptive. The specifics may vary; it's the attitude and intent that matters:
1. Keep the courtship going. Establish a standard that, barring some extreme circumstances, you and your spouse go on a date at least once per week. On your date, make a concerted effort not to talk about kids, house, etc. You need to continue finding and building common ground, because both of you will gradually change over time and it's very, very easy to let your shared responsibilities become the whole of your topics of discussion. Don't let that happen. People don't "fall out of love", they allow themselves to grow apart. My wife and I go on a date every Friday night -- generally just dinner & a movie, but something every week, without fail (or close to it).
2. Find/build some shared interests. Both of you will likely have to give a little and do things that you don't really love in order to do something that the other does really love. For example, my wife has taken up Scuba diving because I love it; I have taken ballroom dance courses because she loves it. She likes diving and I like dancing, and we both love doing both together.
3. Maintain the attitude that your spouse is the most important thing in your life. More important than your kids, more important than your job, more important than anything else.
4. Take care that you do not allow anyone else to get close enough to you that they begin to encroach on what should be your spouse's territory. If you find yourself sharing your very personal feelings and beliefs extensively with another person (especially of the opposite gender, if you're heterosexual), STOP. Allowing those sorts of relationships to grow risks supplanting your relationship with your spouse. It's not a bad idea to take it a step further and simply refuse to spend time alone with other women (or men, depending). Sometimes work demands that you, for example, meet with a female subordinate or manager, and you do what you have to, but keep it strictly professional and limit the frequency and duration as much as possible.
5. Don't be afraid to avail yourself of professional help, and to do it at the first sign of discontent. A good marriage counselor can help you both understand things that you can do better to maintain and improve your relationship.
In every way, focus on first and foremost defending your marriage and keeping it strong. If this seems like too much work and/or too limiting, then accept that you're likely to get divorced, and plan accordingly.
Personally, I've been married for 23 years and I don't expect ever even to contemplate divorce.
Very clever :-)
Really, private companies do not serve the interests of the public. They never have. They never will.
Private companies are great at the luxuries of life, not the basics.
Yeah, because government-run farms have been so successful, historically. Well, if by "successful" you mean "starved millions".
Some people actually believe everything you wrote.
For any ridiculous idea you care to name you can find some people who believe it. That doesn't change the fact that the GP is making strawman arguments.
I don't see any problem with that. It makes perfect sense to used different codecs in different contexts.
Er, Firefox? Safari? Opera? For that matter, Chromium?
And I hear that IE isn't so awful these days.
Have you tried putting a slash on the end? We use a lot of single-word URLs internally and putting a slash on the end always convinces Chrome that I mean it as a URL not a search keyword.
I don't think "controversial UI" counts as "evil."
Not even all that controversial, since the UI change happened about a year ago and Chrome market share has continued climbing.
The only way to avoid annoying some percentage of your user base with UI changes is to never change the UI. Even clear improvements will generate screams of outrage from a few percent of the users, just because they don't like change.
In any case, if people don't like Chrome's UI, there are plenty of other options. If you really dislike this change, just use a different browser that you like better. If enough people do that, Google will get the message.
I used to think so too, then they reduced the highlighting of the ads that surround the search results (so that people are more likely to click on them)
They're still clearly marked, and studies show no change in the ability of users to distinguish them from the search results.
then started pushing people to G+
That has been handled poorly, I agree. G+ is the single Google account system, which makes a lot of sense, and it's also a social network. Unfortunately, they were introduced in the wrong order. If Google had consolidated the accounts first and then added a social network later people would have understood the goals better.
In any case, neither of those things indicate any kind of a shift in Google's culture or attitude, not if you understand them and the rationale behind them.
as Upton Sinclair pointed out, when your salary depends on something you tend to start to decide it's OK .
My salary doesn't depend on my agreeing with Google's actions. For that matter, if I were too annoyed getting another job would be trivial.
Sorry, about the messed up quotation... It's early morning..
If only there were some way to preview posts...
You know it’s coming...
Actually, I'm quite certain it's not. Not without a dramatic change in Google's culture, which I don't see happening.
google does NOT work for us.
That may be true financially, but it's not the way Google sees it. Seriously, the perspective within Google is that users are customers.
its the same way as the postal service in the US re: spam. I once asked the letter carrier if we could put a sign up on our mailboxes saying 'no junk mail or UCE please'. he said that this is how they make most of their money these days and that they are paying to have their 'junk' put in my inbox. you can see who works for whom; and its not the recipient!
Google doesn't make any money from e-mail spam. Exactly the opposite. Google (like any mail service provider) spends huge amounts of resources receiving, processing and discarding spam Google does make money on the ads in gmail, and those are most effective if you only get real mail, because then the ads can be targeted based on stuff you really care about.
google also makes money on their search and have you noticed that when you search for something that has a tech nature and also a salesy nature, the sales stuff comes first and you have to trawl thru many pages to find the actual tech info?
That would be because most people who search for that salesy stuff are looking to buy. Outside of the links marked explicitly as sponsored, Google does not play games with the results.
have you noticed that there are PAGES of fake sites that serve only ebay ads, trying to fool you into clicking on them? google does nothing (zero, nada, zilch) to stop this even though the Powerful Google could easily fix this if they wanted.
It's actually not so easy. Google puts a lot of effort into eliminating crap results like this, but it's a gigantic game of Whack-a-Mole to do it site by site, so Google doesn't do that. Instead, Google adjusts ranking algorithms to identify and eliminate the crap, but changing the ranking algorithm is a Big Deal that requires vast amounts of analysis and testing, so it only happens occasionally. Generally, when the algorithm is changed most of the garbage sites drop out of the top results and things are better for a while, but then the SEO people learn how to game the new algorithm and they start to rise again.
(I'm a Google engineer, though I don't work on search, or gmail.)