Great sift.

*clap..clap..clap..clap..clap*

Hilariously enough with the right "product" every single hot female celebrity (and some male ones) can take advantage of this google trend to say/sell whatever they want and it would get hits.

I swear wikileaks was 10th in google trends hot topics 30 mins ago and now I refreshed the page and it's not on the list.

It's really trendy to fear Google these days, isn't it?

In a world with companies directly and indirectly exploiting their customers, secretly collecting data and constantly reorganizing their business to maximize profit, the company that comes up front and tells you "we're gathering data" gets demonized for it.

Never mind that they're not specifically collecting data about you. Never mind that they make some of the best products out there, and they're usually free, with a few exceptions. Never mind that many other companies have identical services and collect the same data, and just neglect to tell you about it.

Never mind that they actually release their own data, free of charge, so for example, researchers can do things like track Swine Flu (H1N1) by using Google Trends.

------------

In my line of work I use Google products on a daily basis.

My company emails are on Gmail, which saves us absurd amounts of money not having to house a couple of Exchange servers in-house, be it physical server cost, electricity cost, Exchange and Windows Server costs, backup utility and storage costs and the list goes on. Free.

Every website we develop uses Google Analytics to track their statistics. It used to cost us in licensing, storage and server load with our old Analytics software. Now it's just a few lines of code and all the actual work is done on Google's servers. Videosift uses it as well, FYI. Just look through the source code and search for "ga.js". Free.

Most of our commercial websites have Google ads in them. They send us money, so Less than Free. It's not exactly a service they provide, but it does help pay those server-hosting bills.

I use Google maps whenever I need to go somewhere new, since it's hands down the best mapping service available to me. Free.

----------

I sort of went on and on there, but to me, this video is very very close to saying "Death Panels".

I agree with Farhad about infrastructure spending being absolutely necessary, and about productivity sometimes being hard to measure. Public Schools need to be improved, but throwing money at it isn't necessarily the solution, when sometimes lower-funded schools perform better than higher funded schools.

One could argue that public school students in better-performing, richer districts are performing better because of their parents' genes, prenatal environment, and lifestyle habits. The parents on average had some characteristics that increased their propensity to get a house in the richer district. Or you could argue that the performance of the school caused the high property values. Or you could argue that the extra money for the school from high property values accounts for some performance. All three are in play and positive feedback sustains the disparity.

The best way to help personal profit motives coincide with actual productivity is to tax negative externalities and subsidize positive externalities.

For example, all fossil fuels should be taxed at a very high rate, and the income from this tax should be rebated to everyone equally. Thus it provides the incentive to conserve without harming the poor in the way that has previously caused people to oppose such taxes.

The government should also subsidize the placement of copyrighted work into the public domain voluntarily by copyright holders.

Health computerization is a nice goal. Such a database would be a gold mine for research that could save millions of lives if only the privacy paranoia could be dealt with by anonymizing the data like Google Trends.

All subsidies and safety nets of the form "If x>y you get nothing, otherwise you get something greater than epsilon" (also, atomic tax credits and atomic scholarships) should be eliminated because they create local maxima where there is no incentive to earn slightly more. They should be replaced with smooth linear subsidies such as "take home income = 10k plus 3/5 of earned income".

One reason for the decrease in demand for goods is that most of the money is in the hands of those who use it least, and not in the hands of those who would use it. Supply side economics is crap when clearly there's an excess supply and not enough demand in most industries. The amount of money they've already spent on stimulus and bailouts is more than enough to send a flat $10,000 rebate to every adult citizen (perhaps split up into a smoother schedule), which would surely increase consumer spending back to pre-recession levels without negatively impacting anyone's profit motive. Mortgage-default bailouts conditioned upon being in default should be avoided because of the perverse incentives they create. More generally, any policy of help being conditioned upon something bad happening should be eliminated because of the perverse incentive to create the bad occurrence. Unconditional yet meager help is better because it keeps the incentives straight. Many forms of insurance should be abolished because of the perverse incentives they create.

People can only be really counted on to do the right thing when the the situation is such that doing the right thing is in their own personal interest. Primary purpose of all governance is to change the way the game is played so that individual interests become more congruent with each other (metaphorically, congruent with the public good). This includes punishment, which can only be justified as a disincentive to future crime.

The stimulus package contains less than $150 billion in infrastructure spending, out of over 3 trillion that has been spent on stimulus and bailouts. Most of the stimulus package consists of reckless tax cuts.

Hey I posted a google trends Sift Talk but people dismissed it as silly!

Ooh good to know. Another boredom cure! Checked out compete.com. The layout is weird but the info is good.

In a way comparison is better than numbers, 'cause the blind ranking isn't gonna tell you much. Even if a billion people visit goatse, it doesn't make it a cool, informative site. Plus general purpose sites like google are gonna always rank up higher because of the simple fact that they can appeal to anyone.

>> ^dag:
^ and me.
Alexa is helpful, but maybe not the best ranker out there.
We use compete.com increasingly - as the stats seem to be a bit more accurate. The absolute BEST one is Google Trends - because they're google - but the downside is that they don't give any real numbers, just comparisons. http://trends.google.com/websites?q=videos
ift.com%2C+popurls.com&geo=all&date=all&sort=0

Comment hidden because you are ignoring dag. (show it anyway)

^ and me.

Alexa is helpful, but maybe not the best ranker out there.

We use compete.com increasingly - as the stats seem to be a bit more accurate. The absolute BEST one is Google Trends - because they're google - but the downside is that they don't give any real numbers, just comparisons. http://trends.google.com/websites?q=videosift.com%2C+popurls.com&geo=all&date=all&sort=0

There are several flaws in this I see:

1. Google searches does not show support. If you check the words "Nazi, Democracy" in google trends, Nazi continuously has more hits. Does this mean more people want Nazism than democracy?

2. Media coverage does not reflect public interest. What should be compared here is how well viewed Ron Paul coverage was compared to any other candidate. If Ron Paul pieces receive very few hits, then it's little wonder. It is interesting to note that Ron Paul hits do still follow the same curve as coverage. He may have a point here, but even so, it's hard to tell which is the cause and which is the effect.

3. Youtube does NOT represent the voting population. That's like doing a political pole at an elementary school. Over 50% of the voting population in the US is over 50. Most people I know who are that old probably don't even know what Youtube is, let alone post content to it. He also didn't compare hits, just posted videos.

4. Last, but certainly not least, he fails to acknowledge that the internet may not be a reasonable gauge for the public opinion. Many people still don't use the internet for finding information (after all, you can believe every thing you see written on a webpage, right?). It's a good concept, but still has some very flawed reasoning.

Clearly a media bias towards Ron Paul can be seen in this Google Trends comparison:

http://google.com/trends?q=%28ron+paul%29%2C+%28naruto%29&ctab=0&geo=US&geor=all&date=all&sort=0

Naruto for President!

What I would ask:

1) Is the Internet crowd representative of the US population, in terms of age, gender, political views, etc.?
2) Can a Google search be considered as an endorsement?

The other conclusion could be that Ron Paul's high "popularity" on the web was an anomaly and that he had the media coverage he deserved considering his chances to win the GOP primaries. I think the point here is that Ron Paul did suffer from media bias, but you need more than Google Trends and YouTube to prove it.

Take a look at the 'Regions' section in this Google Trends page that was briefly shown in the video:
http://www.google.com/trends?q=hillary%2C+obama&ctab=0&geo=all&date=all&sort=0

Obama has had twice the hits coming from US when compared to Hillary, while the latter obviously has her PR bureau in New Zealand. Seriously, I'm not even trying to kid; can anyone come up with a better explanation for these stats?

Isn't Google Trends tracked globally not just US local. There was more interest in Ron Paul outside of US.