I'm the co-founder of Banksimple. We are still a few months away from launch, but our entire front end is built atop and API that we intend to expose to the public. I personally started working on this project when I came to the realization that banks in American fundamentally make money by keeping customers confused. Banks make more money from fees than interest margin. Because of this, banks have very little incentive to let their customers better understand their own finances. We think that sucks, and our API is just one of the ways that we are going about fixing this.
Each year that the competition was running NetFlix awarded a progress prize to the best ranked team at the end of the year. Part of the requirements of winning this prize is publication of scientific papers describing key elements of their algorithms. BellKor's Yehuda Koren presented a paper at SIGKIDD in July describing the improvements they made to their algorithm to take advantage of predictable temporal dynamics of ratings. Check out the paper here
From an article about the paper:
While movies themselves stay the same, the humans who rate them are anything but static. As Koren puts it, "The way I rate movies today can be very different from how I rate them even tomorrow." To the frustration of Netflix Prize contenders, a four-star rating can mean "great" or merely "so-so" depending on the user's current mood or comparisons with other recently seen movies. Besides such erratic shifts in the rating scale, people's actual tastes tend to change over time -- as when someone tires of action films, for example, and develops a yen for screwball comedies. There's an overall rise in ratings over the years, as well: for various reasons, a typical movie's ratings become more favorable as the DVD ages.
If you read the entire blog post you'll see that they describe that option fully. When enabled, it lets your friends see whether you have joined a Fan or Group page. Completely tame and clearly explained.
In the past couple of days, a rumor has begun spreading that claims we have changed our policies for third-party advertisers and the use of your photos. These rumors are false, and we have made no such change in our advertising policies.
If you see a Wall post or receive a message with the following language or something similar, it is this false rumor:
FACEBOOK has agreed to let third party advertisers use your posted pictures WITHOUT your permission.
The advertisements that started these rumors were not from Facebook but placed within applications by third parties. Those ads violated our policies by misusing profile photos, and we already required the removal of those deceptive ads from third-party applications before this rumor began spreading.
We are as concerned as many of you are about any potential threat to your experience on Facebook and the protection of your privacy. That's why we prohibit ads on Facebook Platform that cause a bad user experience, are misleading, or otherwise violate our policies. Along with removing ads, we've recently prohibited two entire advertising networks from providing services to applications on Facebook Platform because they were not compliant with our policies and failed to correct their practices.
Maybe you haven't spent enough years to recognize the author of the article Dominic Connor - a highly regarded individual in these circles.
I found $2.34x10^x dollars yesterday when I worked out that one of our manual data entry people forgot to put a minus sign in front of a trade. Happens all the time.
I think I suffered from a series of Type III errors (rtfa). After merging lots of poorly maintained backups of my/home file system I decided to write a little script to look for duplicate files (using file size as a first indicator, then md5 for ties). The script would identify duplicates and move files around into a more orderly structure based on type, etc. After doing this i noticed that a small number of my mp3's now contain chunks of other songs in them. My script was only working with whole files, so I have no idea how this happened. When I refer back to the original copies of the mp3s the files are uncorrupted.
Of course, no one believes me. But maybe this presentation is on to something. Or perhaps I did something in a bonehead fashion totally unrelated.
Heh. It appears that there is a little bug in the presentation video. Once the presenter has loaded everything onto the phone, the iTunes space utilization bar shows that the phone has 74GB of capacity. WTF?
see http://blog.i2pi.com/2007/06/26/iphone-74gb-model/ for photos.
Duh. Its clearly a chamber where one is exposed to extremes amounts of hyperbole:
hyperbole/haprbli/
1. obvious and intentional exaggeration.
2. an extravagant statement or figure of speech not intended to be taken literally, as "to wait an eternity."
You can read a review here. If you don't feel like jumping the link, then welcome to slashdot, the home of hyperbole.
Don't knock the CSIRO. At one of their 'Double Helix' club meetings I learned how to program my calculator to generate a Mandelbrot set. Might not be so much of a feat to you TI fanboys, but this was on an HP-42S (which I still own & use) - a non-graphing calculator.
Later I was placed in a summer program where I learned matlab whilst working at a steel testing lab.
Check out Attention Trust. These guys are onto something. Recently Bezos from Amazon was asked (I'm paraphrasing) "So, who owns my purchase history?", the answer being "Well, you own a copy. And we own a copy.". Think about that.
Direct marketers have developed the art and science of buying and selling personal data. But when you think about it, you should really be able to control that flow. If you want, you should be able to set your price, if you are interested in selling at all.
Disclaimer: I work for a company that upholds the Attention Trust principles.
Most conversion processes occur in under 14 days. While 3rd party permenant cookies are an easy way to track this, there are methods that are as reliable over this time frame.
For conversions taking longer than 7 days, you are generally looking at products with high 'consideration' (to use marketing speak), such as expensive consumer products or travel. These people do have problems when relying on cookies. Not to downplay their pain, but they hardly make up the majority (in number, not dollars) of online advertising.
Now to contradict myself...:)
Travel is a very large segment of online advertising spend. But it is so large that the 60% of people who keep cookies (worst case scenario) provide oodles of valuable business information for determining future media budgets.
What I am trying to say is that user tracking is generally only a concern for small to mid sites that don't have the volume to infer from large samples that do keep cookies. These smaller sites, in general sell low consideration products - thus don't really have to worry about long lead time purchases.
In effect, the problem is not as large as people in the press say it is.
A far larger problem (from my recent conversations at AdTech and the web analytics conference) is that 90% of people are pretty clueless in how to determine what the appropriate ROI metric is for their web site. This worries me far more than cookie issues.
The Jupiter report stating that 37% of cookies are being deleted has not really been accepted wholly by the web analytics community. See the recent NY Times article that was linked from Slashdot a few weeks back.
The main reason that companies are not willing to try this new paradigm shift UIP technology is that most people in the industry are already doing it.
The paradigm shift is simply using a bundle of already known tricks and throwing them in a big soup. There is nothing amazing here.
The big problem is with 3rd party permenant cookies. There are easier solutions, such as:
Serving all tracking cookies 1st party. (Most good analytics tools alow you to host the 1x1 tracking GIF's to do this)
Using 1st party session cookies + user agent + IP (or a subset of). This is fairly accurate
I have ditched using unique visitor metrics as they are terribly database intensive and really are not important for true measures of ROI. We track large media buys and have no reports based on UV's as they don't lend themselves to any 'actionable' optimization or other techniques for improving our business.
On the most part user behaviour is variable & only a small subset of analysis is significant for analyzing user behaviour over periods longer than 20 days. I have seen no academic research stating that session cookie + IP + uagent is invalid over such timeframes.
but afaik macromedia released an upgrade to flash soon after this technique was launched so that it would pop up a dialog asking users if they wanted to store these.
Hmm.. As an Australian living in the US, I don't understand how Americans can put up with the terrible remake of the otherwise hillarious British comedy - The Office.
On the other hand, I wish I could find torrents of Kath & Kim... I guess there isn't much international demand for extraordinarily localizsed Australian comedy.
The beauty of the UNIX development model is that the bulk of your applications are 'command line' applications, so that things can be plugged together to get the job done.
Whilst GUI applications require lots of knowledge to implement correctly, the innards usually need a good understanding of pipes, regex etc.
Slashdot Mirror
I'm the co-founder of Banksimple. We are still a few months away from launch, but our entire front end is built atop and API that we intend to expose to the public. I personally started working on this project when I came to the realization that banks in American fundamentally make money by keeping customers confused. Banks make more money from fees than interest margin. Because of this, banks have very little incentive to let their customers better understand their own finances. We think that sucks, and our API is just one of the ways that we are going about fixing this.
Three years ago - see http://www.felixsalmon.com/000763.html
Oh welp. History repeats itself.
Each year that the competition was running NetFlix awarded a progress prize to the best ranked team at the end of the year. Part of the requirements of winning this prize is publication of scientific papers describing key elements of their algorithms. BellKor's Yehuda Koren presented a paper at SIGKIDD in July describing the improvements they made to their algorithm to take advantage of predictable temporal dynamics of ratings. Check out the paper here
From an article about the paper:
If you read the entire blog post you'll see that they describe that option fully. When enabled, it lets your friends see whether you have joined a Fan or Group page. Completely tame and clearly explained.
the odd link is a rickroll now.
as i said.. they should have changed their default password. duh.
Woops...
:)
Maybe you haven't spent enough years to recognize the author of the article Dominic Connor - a highly regarded individual in these circles.
I found $2.34x10^x dollars yesterday when I worked out that one of our manual data entry people forgot to put a minus sign in front of a trade. Happens all the time.
Of course the fa that I was referring to is here. Much more informative than AC's post if I may say...
I think I suffered from a series of Type III errors (rtfa). After merging lots of poorly maintained backups of my /home file system I decided to write a little script to look for duplicate files (using file size as a first indicator, then md5 for ties). The script would identify duplicates and move files around into a more orderly structure based on type, etc. After doing this i noticed that a small number of my mp3's now contain chunks of other songs in them. My script was only working with whole files, so I have no idea how this happened. When I refer back to the original copies of the mp3s the files are uncorrupted.
Of course, no one believes me. But maybe this presentation is on to something. Or perhaps I did something in a bonehead fashion totally unrelated.
Heh. It appears that there is a little bug in the presentation video. Once the presenter has loaded everything onto the phone, the iTunes space utilization bar shows that the phone has 74GB of capacity. WTF? see http://blog.i2pi.com/2007/06/26/iphone-74gb-model/ for photos.
You can read a review here. If you don't feel like jumping the link, then welcome to slashdot, the home of hyperbole.
No Problem! Keep up the good work.
Now if only I could find my gold double helix card.....
RTFP. Read the f-ing patent. This is hardly a submarine patent, they go into great detail about to to design an implement such a system.
Don't knock the CSIRO. At one of their 'Double Helix' club meetings I learned how to program my calculator to generate a Mandelbrot set. Might not be so much of a feat to you TI fanboys, but this was on an HP-42S (which I still own & use) - a non-graphing calculator.
Later I was placed in a summer program where I learned matlab whilst working at a steel testing lab.
Cool stuffs.
The standard way of doing this is to use the Marching Cubes algorithm. Google it --> http://www.exaflop.org/docs/marchcubes/
Disclaimer: I work at Root.
Root Vaults currently uses the Open Source firefox extension from attentiontrust. http://attentiontrust.org/
Imagine a formalized marketplace for that.
Check out Attention Trust. These guys are onto something. Recently Bezos from Amazon was asked (I'm paraphrasing) "So, who owns my purchase history?", the answer being "Well, you own a copy. And we own a copy.". Think about that.
Direct marketers have developed the art and science of buying and selling personal data. But when you think about it, you should really be able to control that flow. If you want, you should be able to set your price, if you are interested in selling at all.
Disclaimer: I work for a company that upholds the Attention Trust principles.
For conversions taking longer than 7 days, you are generally looking at products with high 'consideration' (to use marketing speak), such as expensive consumer products or travel. These people do have problems when relying on cookies. Not to downplay their pain, but they hardly make up the majority (in number, not dollars) of online advertising.
Now to contradict myself... :)
Travel is a very large segment of online advertising spend. But it is so large that the 60% of people who keep cookies (worst case scenario) provide oodles of valuable business information for determining future media budgets.
What I am trying to say is that user tracking is generally only a concern for small to mid sites that don't have the volume to infer from large samples that do keep cookies. These smaller sites, in general sell low consideration products - thus don't really have to worry about long lead time purchases.
In effect, the problem is not as large as people in the press say it is.
A far larger problem (from my recent conversations at AdTech and the web analytics conference) is that 90% of people are pretty clueless in how to determine what the appropriate ROI metric is for their web site. This worries me far more than cookie issues.
but afaik macromedia released an upgrade to flash soon after this technique was launched so that it would pop up a dialog asking users if they wanted to store these.
:)
see you at work on monday
Hmm.. As an Australian living in the US, I don't understand how Americans can put up with the terrible remake of the otherwise hillarious British comedy - The Office.
On the other hand, I wish I could find torrents of Kath & Kim... I guess there isn't much international demand for extraordinarily localizsed Australian comedy.
The beauty of the UNIX development model is that the bulk of your applications are 'command line' applications, so that things can be plugged together to get the job done.
Whilst GUI applications require lots of knowledge to implement correctly, the innards usually need a good understanding of pipes, regex etc.