Innovations of the year by the 2009 American Business
> Whozat™, The People Search Engine™, named one of the 100 hottest start-ups in the world by famed technology blog TechCrunch, a semantic search engine with machine vision that garnered tens of thousands of pageviews within a week of launch. Whozat beat Google and every other search engine tested 3 to 1 or more in results relevance in a blind people search comparison. Prof. Harry Lewis, endowed chair in CS at Harvard; former Dean of Harvard College, wrote in his blog:“Try whozat.com. I learned things about myself I didn’t know. Seriously. And now I’m checking on you.” Whozat has since launched SocialDiligence.com, pioneering the first search engine that can search based on a resume (or any document).
> PREPROVE™, a statistically sound behavioral way of measuring brand power and testing marketing messages which is orders of magnitude cheaper and more accurate than traditional focus groups and surveys.
> TheSEMExperts™, for online marketers looking for their products to be found.
Under Alex's leadership, abinventio has consistently grown revenues every quarter since the launch of its first product, at an average 98% quarterly (1535% annual) and median 80% quarterly (1057% annual) growth.
California
He has been an invited speaker at numerousconferences on diverse science and technology topics around the world. Alex is always lookingforworld-classentrepreneurialpeopletohelpbringaboutinnovationthroughtechnology. Alex likes to spend his free time traveling with his family, kitesurfing, filmmaking, or kayaking.
In addition to this wiki, Alex keeps a blogoflifeatab|inventio™,TheInventionFactory™.
]]>
Visual Invariance in Children: Understanding Early Vision
Christopher Wetzel and Alex Bäcker
Sandia National Labs & California Institute of Technology
When children begin to write, some have difficulty correctly reproducing letters; often, they sketch the reflection of a letter or number. We hypothesized that the problem lies not in their writing skills, but in their perception: that children perceive a symbol or picture as the set of visual features in the images they see, and that, early in development, a child does not perceive the relationships between visual features that differentiates a symbol from its reflection or rotation. This hypothesis is inspired by the hierarchical nature of visual processing: assuming that elements higher in the visual hierarchy, which detect confluences of sets of elements lower in the hierarchy, can only properly develop after their component elements lower in the hierarchy have been formed, one should be able to find a developmental stage at which the features represented by the lower echelons of the hierarchy are represented, but their higher-order combinations are not. To test this hypothesis, children from Caltech’s Children’s Center were shown a series of sets of three cards: first a reference card, bearing a symbol or picture, and then two other cards, one bearing the exact same symbol, and the other bearing a different symbol, or a reflection or rotation of the reference symbol. The children were asked to point to the symbol which was the same as the reference symbol. The data collected show that children have difficulty distinguishing reflections, and, to a lesser degree, rotations of symbols, supporting our hypothesis.
Examining the development of drawing skills in young children provides insight into and raises questions concerning the development of child vision. Generally, children begin to scribble around age two. At this time, a child may have developed the motor skills necessary to hold a crayon or other drawing instrument, but will generally not have any purpose in drawing other than the enjoyment gained from motor movement and seeing marks appear on paper (Crosser). After about three years, however, children discover symbolism and begin to associate symbols, often very simple ones, like circles, with objects in their environment. Soon, children begin to draw simple shapes—lines, circles, squares, etc. Initially, children will notice that a scribble resembles an object without first intending to symbolically represent the object (Crosser). At some point children begin to recognize other symbols shown to them and begin to copy them. Here something interesting happens. Consider the following drawing:
Here, the child, a two-year-old, has responded to the text already present (Veronica) by inversion of the characters over a horizontal axis. This raises an interesting question: how are visual symbols, such as the letters of the alphabet, represented in a child’s brain? We know that the brain can break down an image by component shapes. A certain neuron may fire when a series of photoreceptors in a vertical line all fire, thus detecting a vertical line. Another neuron may fire when a series of photoreceptors in an L shape all fire, thus detecting a right angle. We know that this information is given to the brain because of the hierarchical arrangement of neurons in the brain; we don’t know exactly how the information processed by this system is translated into vision, especially in children. Consider the symbols L and Γ. Both have two lines and a right angle. It is possible that the similarity in representation of both L and Γ account for drawings like the one above. Each symbol has the same relative relationships (angles and distances) between each of its key features. In these symbols are identical if thought of as three dimensional objects.
Although vision is, to some extent, hard-wired into the brain at birth, children still must complete a journey of discovery that enables them to use their vision system. Just as an infant waves its limbs randomly at first and then with increasing degrees of purpose later, so also does an infant explore and develop its sight. We hypothesize that children perceive a symbol or picture as the set of relationships among the visual features in the images they see, and that, early in development, a child does not perceive the relationships between visual features that differentiates a symbol from its reflection or rotation. This hypothesis is inspired by the hierarchical nature of visual processing. Assuming that elements higher in the visual hierarchy, which detect confluences of sets of elements lower in the hierarchy, can only properly develop after their component elements lower in the hierarchy have been formed, one should be able to find a stage of development at which the features represented by the lower echelons of the hierarchy are represented, but their higher-order combinations are not. For example, while a developing brain may be able to detect right angles, it may not be able to distinguish between L and Γ.
To test this hypothesis, we performed two experiments on 27 children from three to five years old from the Caltech Children’s Center. In each experiment, children were shown sets of three cards. Each set contained one reference card (the first card shown to the child) and two test cards, one of which was a copy of the reference card. The other card contained a different symbol, a reflection of the reference symbol, or a rotation of the reference symbol. The child, after briefly viewing the reference card, was required to choose which test card looked like the reference card. We performed two experiments, the first using a combination of ASCII symbols and the second using combinations of only three simply figures (Figure 3). We performed two versions of the second experiment, one “Card Up,” where the child could see the reference card while choosing between the test cards, and “Card Down,” where, as in the first experiment, the child could not see the reference card.
 |
Results
The data collected from our experiments indicate that children have difficulty distinguishing between rotations and reflections and that this effect is not linked to memory. On average the children scored 19.7% worse when asked to compare reflections of symbols and 15.6% worse when asked to compare rotations of symbols than when asked to compare different symbols. Children scored 14% worse on average in the card down experiments compared to the card up experiments (Figures 5 and 6). Because the card down experiment required the children to remember what the reference card looks like whereas the card up experiment did not, the fact that the results of both experiments are correlated (r2 = .6328) indicates that memory is not a significant factor in the differences seen among “different,” “reflection,” and “rotation” scores (See Figure 7).
Further Work
In order to further understand the development of the vision system of the brain, it is important to demonstrate the physiological implications of our hypothesis, namely, that the neurons that are high in the visual hierarchy which detect confluences of sets of elements lower in the hierarchy should display less activity, or actively not correlated to visual discrimination tasks in developing brains when compared with developed brains performing the same tasks. In addition, it is important to investigate the relationship between classes of visual elements and to correlate this with neural anatomy. Given that lines are lower in the visual hierarchy than intersections, and that points are lowing than lines, figures composed of points should be easier to distinguish than figures composed of lines for a developing nervous system. Finally, it is important to investigate this hypothesis in some way that does not involve children, as age could certainly be a confounding factor.
Methods
The first experiment consisted of 39 sets of 3 symbols, Each set contained a reference symbol and two test symbols, one which looked exactly like the reference symbol, and one which was a reflection of the reference symbol, a rotation of the reference symbol, or a completely different symbol. Of these 39 sets, 6 are sets of pictures, 6 are simple figures created using four straight lines, and 27 are a mix of letters, numbers, Greek letters, and symbols such as &, ), and @. I began by showing the child the first symbol from the reference deck for approximately five seconds, while telling the child to remember exactly what he or she sees. I then put the first symbol face down and held up the first cards from the other two decks. I then asked the child to point to the symbol that was the same as the first one. I then recorded his or her selection and proceed with the next set of three symbols, repeating this procedure until I finished going through the set of cards. After I finished with each child, I reordered the deck to prevent the experience the child gained during the experiment from biasing the results due to learning curve effects.
The second experiment contained two versions of the first experiment: “Card up” and “Card down”. For each child, I would run through a new set of cards twice, first leaving the first card, the reference card, face up and then a second time, leaving the reference card face down. The new deck contained only 18 sets of cards, with 6 sets of each type: different, reflection, and rotation. Only the three symbols in Fig. 3 were used in creating these cards to reduce the influence of the symbols on the results of the experiments. I encountered few problems in my research. Most children were reasonably interested in the experiment and cooperated. Some have even asked to play the “game” again. A few children, however, left in the middle, telling me that they did not want to play anymore.
 |
References
Sandra Crosser. “When Children Draw.” Earlychildhood.com (2002). 1 March 2004 http://www.earlychildhood.com/Articles/index.cfm?A=130&FuseAction=Article.
Acknowledgements
We would like to thank the Caltech Children’s Center and the parents of the children who participated in my experiment for allowing me to work at the Center over the summer; and the SURF Program for providing this opportunity to participate in undergraduate research.
An early version of this manuscript was presented in a 2005 Caltech SURF report.
]]>
Suggestions for Powerpoint
Did anybody at Microsoft give Windows Vista to regular users to test before launching?
How to judge if a UI is really intuitive
]]>
- Do not assume by default that a user wants slides to move automatically after a number of seconds! This is the most annoying thing I've ever had software do to me. You are in a hurry in front of an audience and have to cut the presentation to tell Microsoft not to play like it knows what you want better than you do.
- The whole point of rehearsing timings is to know if the show fits in your schedule. So for God's sake, show the total, and not just the individiual slides' timing, after you're done rehearsing, not just while you're rehearsing.
- Have a way for someone to re-rehearse starting at any slide, not just #1.
--Alex
Up to Technology
]]>Get an alert when the swine flu shot is available in your area --and perhaps even get to skip the lines.
Read about how QLess eliminates standing in line athttp://digg.com/tech_news/Qless_Changes_the_Way_You_Wait_in_Line/. 107 diggs and counting!
Join my free newsletter with stories, tips and resources on entrepreneurship and science
]]>
The money is not in until it's in the bank.
Take the money when offered.
Take advantage of the expectation of something great, because sometimes expectations are not fulfilled.
Always align incentives.
He empowered the people who worked with him. I will always remember the day in which Micah Siegel, he and I went in to pitch a VC fund for Adapt Technologies, and we were told an entrepreneur in residence in an unrelated field would be in attendance. When we asked this entrepreneur what he was working on, he said enough we immediately recognized him as a potential competitor. When we asked him to say some more about it so we could determine whether it truly was competitive, he refused. We were all uncomfortable about being in the room with someone who thought it dangerous to give us a one-minute elevator pitch yet tried to make us feel it was safe to give them an entire one-hour presentation on our company. We were more uncomfortable about the fact that this had come up through my question during chit-chat, whilst he had been introduced as someone in another field, using his credentials from a prior life. Without hesitating, Craig asked for a few minutes alone before the presentation began and asked me if I still wanted to go ahead with the presentation. There were two significant things about this: 1, he taught me that it was our decision to present or not to a VC, and that their having the money did not give them a license for unethical behavior, 2, despite his immensely larger experience, he told me to make that call, teaching me to make important decisions on the spot as a newly minted CEO. I decided I would not want to work with that firm, and we kindly thanked them for the appointment but left without making the pitch.
Another anecdote I remember is the day that Micah introduced me to him at a dinner in Palo Alto. After I was done talking the technologies I had developed, Micah asked him if he was interested in having a follow up discussion sometime. Craig, full with excitement, replied: "No, I want to start working with him tomorrow!" And we did.
I also sealed my deal with Craig as a partner with a handshake at the Ritz Carlton in Pasadena, and I have always honored a hand-shake. That first handshake taught me that with good people, a handshake is as good as a contract.
Craig was also the founder of Venture Law Group, Virtual Law Partners, and a co-founder of Concept2Company.
I enjoyed working with Craig very much and valued his advice greatly. He will be missed.
--Alex Backer
Alex is the founder of Adapt Technologies, abInventio, Whozat and QLess. Read about him and his companies here. Or read about entrepreneurship.
]]>
<p>Alex is the founder of Adapt Technologies, abInventio, <a target="_blank" href="http://www.whozat.com">Whozat</a> and <a target="_blank" href="http://qless.com">QLess</a>. Read about him and his companies <a href="/Who-Alex-B%C3%A4cker-is">here</a>. Or read about <a href="/Entrepreneurship">entrepreneurship</a>.</p>
]]>
The money is not in until it's in the bank.
Take the money when offered.
Take advantage of the expectation of something great, because sometimes expectations are not fulfilled.
Always align incentives.
He empowered the people who worked with him. I will always remember the day in which Micah Siegel, he and I went in to pitch a VC fund for Adapt Technologies, and we were told an entrepreneur in residence in an unrelated field would be in attendance. When we asked this entrepreneur what he was working on, he said enough we immediately recognized him as a potential competitor. When we asked him to say some more about it so we could determine whether it truly was competitive, he refused. We were all uncomfortable about being in the room with someone who thought it dangerous to give us a one-minute elevator pitch yet tried to make us feel it was safe to give them an entire one-hour presentation on our company. We were more uncomfortable about the fact that this had come up through my question during chit-chat, whilst he had been introduced as someone in another field, using his credentials from a prior life. Without hesitating, Craig asked for a few minutes alone before the presentation began and asked me if I still wanted to go ahead with the presentation. There were two significant things about this: 1, he taught me that it was our decision to present or not to a VC, and that their having the money did not give them a license for unethical behavior, 2, despite his immensely larger experience, he told me to make that call, teaching me to make important decisions on the spot as a newly minted CEO. I decided I would not want to work with that firm, and we kindly thanked them for the appointment but left without making the pitch.
Another anecdote I remember is the day that Micah introduced me to him at a dinner in Palo Alto. After I was done talking the technologies I had developed, Micah asked him if he was interested in having a follow up discussion sometime. Craig, full with excitement, replied: "No, I want to start working with him tomorrow!" And we did.
I also sealed my deal with Craig as a partner with a handshake at the Ritz Carlton in Pasadena, and I have always honored a hand-shake. That first handshake taught me that with good people, a handshake is as good as a contract.
Craig was also the founder of Venture Law Group, Virtual Law Partners, and a co-founder of Concept2Company.
I did not always agree on everything with Craig, but I enjoyed working with him and valued his advice greatly. He will be missed.
--Alex Backer
Alex is the founder of Adapt Technologies, abInventio, Whozat and QLess. Read about him and his companies here. Or read about entrepreneurship.
]]>
- The money is not in until it's in the bank.
- Take the money when offered.
- Take advantage of the expectation of something great, because sometimes expectations are not fulfilled.
- Always align incentives.
He empowered the people who worked with him. I will always remember the day in which Micah Siegel, he and I went in to pitch a VC fund for Adapt Technologies, and we were told an entrepreneur in residence in an unrelated field would be in attendance. When we asked this entrepreneur what he was working on, he said enough we immediately recognized him as a potential competitor. When we asked him to say some more about it so we could determine whether it truly was competitive, he refused. We were all uncomfortable about being in the room with someone who thought it dangerous to give us a one-minute elevator pitch yet tried to make us feel it was safe to give them an entire one-hour presentation on our company. We were more uncomfortable about the fact that this had come up through my question during chit-chat, whilst he had been introduced as someone in another field, using his credentials from a prior life. Without hesitating, Craig asked for a few minutes alone before the presentation began and asked me if I still wanted to go ahead with the presentation. There were two significant things about this: 1, he taught me that it was our decision to present or not to a VC, and that their having the money did not give them a license for unethical behavior, 2, despite his immensely larger experience, he told me to make that call, teaching me to make important decisions on the spot as a newly minted CEO. I decided I would not want to work with that firm, and we kindly thanked them for the appointment but left without making the pitch.
Another anecdote I remember is the day that Micah introduced me to him at a dinner in Palo Alto. After I was done talking the technologies I had developed, Micah asked him if he was interested in having a follow up discussion sometime. Craig, full with excitement, replied: "No, I want to start working with him tomorrow!" And we did.
I also sealed my deal with Craig as a partner with a handshake at the Ritz Carlton in Pasadena, and I have always honored a hand-shake. That first handshake taught me that with good people, a handshake is as good as a contract.
Craig was also the founder of Venture Law Group, Virtual Law Partners, and a co-founder of Concept2Company.
I did not always agree on everything with Craig, but I enjoyed working with him and valued his advice greatly. He will be missed.
--Alex Backer
Alex is the founder of Adapt Technologies, abInventio, Whozat and QLess. Read about him and his companies here. Or read about entrepreneurship.
]]>
This causes abuse of minority stockholders, and could use a legislative solution, but until then, beware.
Also, check the requirements to change the bylaws of the company, as I have seen these secretly changed to reduce quorum requirements.
Disclaimer: This is not legal advice.
]]>
A loophole in stockholder rights every entrepreneur must be aware of
I come from a family of scientists and entrepreneurs. My great-great-great-great-great-grandfather, Hayum Moos, was Albert Einstein's great-great-grandfather. My grandfather, Ernst Moos, had to quit medical school in Berlin due to Nazi oppression and fled to Argentina, where he started two small companies as a penniless Jewish refugee. My uncle RonBaecker is the founder of several companies, and Professor of Computer Science, Bell Universities Laboratories Chair in Human-Computer Interaction, and founder and Chief Scientist of the Knowledge Media Design Institute at the University of Toronto, was named a Pioneer of Computer Graphics by ACM SIGGRAPH, has been elected to the CHI Academy by ACM SIGCHI, and has been given the Canadian Human Computer Communications Society Achievement Award. My mother, Silvia Moos, is the founder of Centralab, Argentina's most important clinical chemistry lab and a state of the art automated facility, and Klik, Argentina's first mental fitness center. My Dad started the Argentina offices of Heidrick & Struggles and currently heads Korn Ferry South South America.
My own first entrepreneurial venture was making and selling ornamental candles with my brothers in the neighborhood, when we were all children --although my brother Nico was the spirit principally behind the candle factory. It got a little more serious in college, designing, making and selling T-shirts and sweatshirts with the university name and logo; we did very well until both partners left the country to pursue scientific studies, sacrificing the venture to science. In grad school, I took a break from earthly profits and founded the Caltech Filmmaking Club, the Caltech Kiteboard Club, the CNS Journal Club (also with Gabriel Kreiman) and the Da Vinci Club (with Dan Lieberman), all of which got funded and exist to this day. It was during this time that I started writing up some of my first inventions. Initially, I thought established companies such as Google would soon catch up to my ideas, so I did nothing. When several years passed and they had still not come up with any of my inventions, I decided to start Adapt. I have not stopped since. Read TheInventionFactory'sblog for up to date news on my latest ventures.
]]>
The law should punish those who infringe on others' rights, not try to tell us how to live our own lives. Legislators can hardly claim to know what's good for each of hundreds of millions of individuals better than the individuals themselves.
]]>
Multiple-choice Legislation
Data-driven Traffic Citations
]]>
The Times' Conserva-care opinion section this week made it clear that both Democrats and Conservatives believe healthcare could be improved and that both parties have good ideas about healthcare reform. So why has no reform happened after decades of attempts? The problem lies with a more general fault with our legislative system: by asking legislators to vote for or against each bill, no bill is likely to gather a majority of votes when dealing witha topic where there are many more than 2 competing proposals.The solution would be simple: switch voting from yes/no on individual bills to voting for the best among several competing solutions to the same problem. Each legislator could then vote for his/her favorite bill or for none (keeping the status quo), and the winner would become law.By revealing the true nature of a vote against a bill as a vote for a different solution to the same problem, this system would make it easier to pass laws where a majority of legislators agree a new law is needed but disagrees on the details of the solution.A related alternative system would send the two winners of any first round where the combined votes for competing bills outnumber votes for the status quo to a second round or ballotage for a final face-off. Our oversimplistic legislative system prevents progress in the face of many competing solutions just like a pig starves to death when unable to decide with of two equidistant piles of food to go toward.
How to group bills into competing proposals? Surely there are many ways. Here's one: Any time a bill gets introduced by a legislator, give the rest a period (30 days?) to come up with alternative proposals, which get submitted as an alternative solution to the same problem, or an alternative bill. The more alternative bills, the more competition, but also the lower the number of votes usually needed for one to win and become law, so there appears to be no disincentive for legislators to submit their bill as an alternative one.
]]>
How to group bills into competing proposals? Surely there are many ways. Here's one: Any time a bill gets introduced by a legislator, give the rest a period (30 days?) to come up with alternative proposals, which get submitted as an alternative solution to the same problem, or an alternative bill. The more alternative bills, the more competition, but also the lower the number of votes usually needed for one to win and become law, so there appears to be no disincentive for legislators to submit their bill as an alternative one.
Alex Backer, Ph.D.
For a related proposal to solve another problem derived from the same binary voting system, namely how to pass a balanced budget, read the author'shttp://alexbacker.pbworks.com/A-Better-Way-to-Balance-the-State-and-Federal-Budgets.
]]>
Did anybody at Microsoft give Windows Vista to regular users to test before launching?
How to judge if a UI is really intuitive
The ridiculous security features in Windows Vista
]]>
How to group bills into competing proposals? Surely there are many ways. Here's one: Any time a bill gets introduced by a legislator, give the rest a period (30 days?) to come up with alternative proposals, which get submitted as an alternative solution to the same problem, or an alternative bill. The more alternative bills, the more competition, but also the lower the number of votes usually needed for one to win and become law, so there appears to be no disincentive for legislators to submit their bill as an alternative one.
]]>Data-driven Traffic Citations
A Better Way to Balance the State and Federal Budgets
]]>
How to judge if a UI is really intuitive
The ridiculous security features in Windows Vista
American Airlines' website is useless
]]>
- Trying to resize or move images in the slide master of a PPT 2003 file in PPT 2007? Save it as a 2007 file, and magically Powerpoint will stop making your life miserable in exchange for you having forced every reader of it to perform an expensive upgrade in order to read it.
- Looking for the Insert menu option in Excel? Right click your mouse.
How to get Powerpoint 2007 to let you move or resize images in the master slides of a PPT 2003 file
If CGIs do not run in Apache server on Windows XP
Fixing a laptop screen: Try this before buying a new one
]]>
var sc_project=1777361; var sc_invisible=1; var sc_partition=16; var sc_security="c84ed9d0";
]]>
1. A few days into using Windows Vista, I have yet to detect an advantage of it over XP, other than the years-in-waiting "do this for every file" option.
2. Vista incessantly "protects" you by asking for your intervention every time a program tries to run or access the Internet --basically, most every time you do something. The problem is that all the information it gives you to decide is the name the program claims for itself. So a malicious program would just need to call itself with a trustworthy name to gain most users' trust. This could be easily improved by having Windows check the Internet for any of many digital fingerprint of the program trying to run or connect to the Internet to find if it is associated with any malicious threats or not.
Up to Technology.
--Alex Backer
]]>The ridiculous security features in Windows Vista
American Airlines' website is useless
Is the range of RangeMax NEXT Wireless-N routers really 10X that of Wireless-G?
]]>
urchinTracker();
var sc_project=1777361;
var sc_invisible=1;
var sc_partition=16;
var sc_security="c84ed9d0";
]]>
Theworld'smostdangerouscommutetoschool
Gallegos
]]>
Follow me on Twitter
Recent favorite posts:
Recent posts with the highest number of pageviews:
]]>
How not to write a resume and cover letter, how to get hired (and how not to), and how to screen job candidates
I come from a family of scientists and entrepreneurs. My great-great-great-great-great-grandfather, Hayum Moos, was Albert Einstein's great-great-grandfather. My grandfather, Ernst Moos, had to quit medical school in Berlin due to Nazi oppression and fled to Argentina, where he started two small companies as a penniless Jewish refugee. My uncle RonBaecker is the founder of several companies, and Professor of Computer Science, Bell Universities Laboratories Chair in Human-Computer Interaction, and founder and Chief Scientist of the Knowledge Media Design Institute at the University of Toronto, was named a Pioneer of Computer Graphics by ACM SIGGRAPH, has been elected to the CHI Academy by ACM SIGCHI, and has been given the Canadian Human Computer Communications Society Achievement Award. My mother, Silvia Moos, is the founder of Centralab, Argentina's most important clinical chemistry lab and a state of the art automated facility, and Klik, Argentina's first mental fitness center. My Dad started the Argentina offices of Heidrick & Struggles and currently heads Korn Ferry South South America.
My own first entrepreneurial venture was making and selling ornamental candles with my brothers in the neighborhood, when we were all children --although my brother Nico was the spirit principally behind the candle factory. It got a little more serious in college, designing, making and selling T-shirts and sweatshirts with the university name and logo; we did very well until both partners left the country to pursue scientific studies, sacrificing the venture to science. In grad school, I took a break from earthly profits and founded the Caltech Filmmaking Club, the Caltech Kiteboard Club, the CNS Journal Club (also with Gabriel Kreiman) and the Da Vinci Club (with Dan Lieberman), all of which got funded and exist to this day. It was during this time that I started writing up some of my first inventions. Initially, I thought established companies such as Google would soon catch up to my ideas, so I did nothing. When several years passed and they had still not come up with any of my inventions, I decided to start Adapt. I have not stopped since. Read TheInventionFactory'sblog for up to date news on my latest ventures.
]]>
- Candidates looking to write a great resume and cover letter.
- People looking to build a career that will *allow* them to write a truthful great resume.
- Hiring managers looking to make an initial screen of the best talent based on resumes and cover letters.
- Myself when I'm older and forgot how I used to do everything I do.
I just finished personally reading through each of more than 150 resumes and cover letter for an Executive Assistant job I advertised. Took me from 9 PM to 4 AM last night, and from 4 PM to 7 PM today. That's 10 hours. 15 resumes per hour, or 4 min per resume. That includes replying to those who made the cut. So if your hiring manager is anything like me, you have 3 minutes to impress him.
Here are some of the things that impressed me:
- Candidates who took the time and interest to research my company/ies.
- Candidates who expressed some passion about what we do.
- Candidates with a sense of humor. This is BIG.
- Candidates which managed to convey concisely several unusual skills that I thought I could use. The most important question I ask myself when looking at a resume is, what has this person done well before, and what can he/she do for me?
- Candidates who stayed with one interesting job for a long time --typically means they can't be that bad.
- Candidates with GPAs of 3.6 or more (I'm being facetious about the cut-off, but higher is better).
- Candidates with degrees from good colleges.
- Candidates who stayed for a meaningful period of time doing a relevant job at a well-respected company.
- Candidates who listed relevant admirable specific accomplishments. This was very rare.
- Candidates that showed a clear understanding of my business.
- Candidates that had worked for a long time for people who clearly have a wide choice of whom to hire.
- Candidates who admitted to a weakness (and pointed out a way to overcome it).
Here are some of the things that did NOT impress me:
- Typos. Especially those with two typos in an email w/a single short sentence. Especially those with two typos in an email w/a single short sentence telling me that a URL with a typo did not work.
- Those whose descriptions are too long and fail to highlight the important. 3 minutes is not enough to read something long if you don't first find a good reason to.
- Generic cover letters that showed no interest in the job I was offering.
- Resumes and cover letter that were so vanilla that I could not find a single special reason why I'd want to hire that candidate over the other 149. You are looking to differentiate yourself from the rest!
- No cover letter.
- Candidates asking if I could provide a website to research my company, when the website was provided in the job posting.
- Resumes submitted by a woman for a male candidate (I'd say or viceversa, except that did not occur). It does not bode well for you if your significant other is trying to get you off your ass!
- Candidates who never stayed with any employer longer than a year.
- Resumes which indicated a college without indicating degree. That smacks of not having finished a degree, which is not a great sign unless you dropped off to start Microsoft.
- Candidates that expressed anger about things they should not have been angry about.
- Candidates who forgot to attach their resume. Although if this was their only shortcoming, I would overlook it --everybody is allowed one mistake even in this highly competitive context.
- Lists of references with two friends (described as such, no less) among the three references.
So if you want to get a great job next, make sure you:
- Try to get in to the best college you can get into. College admissions is one of only two stages in a career where candidates are ranked by a very clear pecking order, so which one you get in to is a powerful indicator that future employers will look for in the absence of outstanding achievements.
- Finish college.
- Get a great GPA. College GPAs is the second stages in a career where candidates are ranked by a very clear pecking order.
- Make sure you accomplish specific outstanding achievements at every one of your jobs, that you list them in your resume, and that you have references willing to speak about them.
Good luck!
Alex Backer
9/18/2009
]]>
Candidates who took the time and interest to research my company/ies.
Candidates who expressed some passion about what we do.
Candidates with a sense of humor. This is BIG.
Candidates which managed to convey concisely several unusual skills that I thought I could use. The most important question I ask myself when looking at a resume is, what has this person done well before, and what can he/she do for me?
Candidates who stayed with one interesting job for a long time --typically means they can't be that bad.
Candidates with GPAs of 3.6 or more (I'm being facetious about the cut-off, but higher is better).
Candidates with degrees from good colleges.
Candidates who stayed for a meaningful period of time doing a relevant job at a well-respected company.
Candidates who listed relevant admirable specific accomplishments. This was very rare.
Candidates that showed a clear understanding of my business.
Candidates that had worked for a long time for people who clearly have a wide choice of whom to hire.
Candidates who admitted to a weakness (and pointed out a way to overcome it).
Here are some of the things that did NOT impress me:
Typos. Especially those with two typos in an email w/a single short sentence. Especially those with two typos in an email w/a single short sentence telling me that a URL with a typo did not work.
Those whose descriptions are too long and fail to highlight the important. 3 minutes is not enough to read something long if you don't first find a good reason to.
Generic cover letters that showed no interest in the job Iwas offering.
Resumes and cover letter that were so vanilla that I could not find a single special reason why I'd want to hire that candidate over the other 149. You are looking to differentiate yourself from the rest!
No cover letter.
Candidates asking if I could provide a website to research my company, when the website was provided in the job posting.
Resumes submitted by a woman for a male candidate (I'd say or viceversa, except that did not occur). It does not bode well for you if your significant other is trying to get you off your ass!
Candidates who never stayed with any employer longer than a year.
Resumes which indicated a college without indicating degree. That smacks of not having finished a degree, which is not a great sign unless you dropped off to start Microsoft.
Candidates that expressed anger about things they should not have been angry about.
Candidates who forgot to attach their resume. Although if this was their only shortcoming, I would overlook it --everybody is allowed one mistake even in this highly competitive context.
Lists of references with two friends (described as such, no less) among the three references.
So if you want to get a great job next, make sure you:
Try to get in to the best college you can get into. College admissions is one of only two stages in a career where candidates are ranked by a very clear pecking order, so which one you get in to is a powerful indicator that future employers will look for in the absence of outstanding achievements.
Finish college.
Get a great GPA. College GPAs is the second stages in a career where candidates are ranked by a very clear pecking order.
Make sure you accomplish specific outstanding achievements at every one of your jobs, that you list them in your resume, and that you have references willing to speak about them.
Good luck!
Alex Backer
9/18/2009
]]>
- Candidates looking to write a great resume and cover letter.
- People looking to build a career that will *allow* them to write a truthful great resume.
- Hiring managers looking to make an initial screen of the best talent based on resumes and cover letters.
- Myself when I'm older and forgot how I used to do everything I do.
I just finished personally reading through each of more than 150 resumes and cover letter for an Executive Assistant job I advertised. Took me from 9 PM to 4 AM last night, and from 4 PM to 7 PM today. That's 10 hours. 15 resumes per hour, or 4 min per resume. That includes replying to those who made the cut. So if your hiring manager is anything like me, you have 3 minutes to impress him.
Here are some of the things that impressed me:
Here are some of the things that did NOT impress me:
Alex Backer
9/18/2009
]]>
On Profit Maximization as the Social Responsibility of Business
The Role of Marketing in a Start-Up
]]>
Getting funded means getting fired
I come from a family of scientists and entrepreneurs. My great-great-great-great-great-grandfather, Hayum Moos, was Albert Einstein's great-great-grandfather. My grandfather, Ernst Moos, had to quit medical school in Berlin due to Nazi oppression and fled to Argentina, where he started two small companies as a penniless Jewish refugee. My uncle RonBaecker is the founder of several companies, and Professor of Computer Science, Bell Universities Laboratories Chair in Human-Computer Interaction, and founder and Chief Scientist of the Knowledge Media Design Institute at the University of Toronto, was named a Pioneer of Computer Graphics by ACM SIGGRAPH, has been elected to the CHI Academy by ACM SIGCHI, and has been given the Canadian Human Computer Communications Society Achievement Award. My mother, Silvia Moos, is the founder of Centralab, Argentina's most important clinical chemistry lab and a state of the art automated facility, and Klik, Argentina's first mental fitness center. My Dad started the Argentina offices of Heidrick & Struggles and currently heads Korn Ferry South South America.
My own first entrepreneurial venture was making and selling ornamental candles with my brothers in the neighborhood, when we were all children --although my brother Nico was the spirit principally behind the candle factory. It got a little more serious in college, designing, making and selling T-shirts and sweatshirts with the university name and logo; we did very well until both partners left the country to pursue scientific studies, sacrificing the venture to science. In grad school, I took a break from earthly profits and founded the Caltech Filmmaking Club, the Caltech Kiteboard Club, the CNS Journal Club (also with Gabriel Kreiman) and the Da Vinci Club (with Dan Lieberman), all of which got funded and exist to this day. It was during this time that I started writing up some of my first inventions. Initially, I thought established companies such as Google would soon catch up to my ideas, so I did nothing. When several years passed and they had still not come up with any of my inventions, I decided to start Adapt. I have not stopped since. Read TheInventionFactory'sblog for up to date news on my latest ventures.
]]>
average
In 2008, Alex was appointed by the President to serve in the California Insitute of Technology's Information Sciences and TechnologyBoard of Advisors together with a distinguished cadre of scientists, technologists and philantropists including Carver Mead and Phil Neches.
He has been an invited speaker at numerousconferences on diverse science and technology topics around the world. Alex is always lookingforworld-classentrepreneurialpeopletohelpbringaboutinnovationthroughtechnology. Alex likes to spend his free time traveling with his family, kitesurfing, filmmaking, or kayaking.
]]>
saved
> Whozat™, The People Search Engine™, named one of the 100 hottest start-ups in the world by famed technology blog TechCrunch, a semantic search engine with machine vision that garnered tens of thousands of pageviews within a week of
> PREPROVE™, a statistically sound behavioral way of measuring brand power and testing marketing messages which is orders of magnitude cheaper and more accurate than traditional focus groups and surveys.
for online marketers looking for their products to be
Under Alex's leadership, abinventio has consistently grown revenues every quarter since the launch of its first product, at an average 89% quarterly (1282% annual) and median 47% quarterly (462% annual) growth.
In 2008, Alex was appointed by the President to serve in the California Insitute of Technology's Information Sciences and TechnologyBoard of Advisors together with a distinguished cadre of scientists, technologists and philantropists including Carver Mead and Phil Neches.
He has been an invited speaker at numerousconferences on diverse science and technology topics around the world. Alex is always lookingforworld-classentrepreneurialpeopletohelpbringaboutinnovationthroughtechnology. Alex likes to spend his free time traveling with his family, kitesurfing, filmmaking, or kayaking.
]]>
]]>]]>
- Put the board facing 90 degrees from teh wind direction.
- Put the sail downwind from the board.
- Stand in the middle of the board, with a foot on each side of the mast, looking downwind.
- Raise the sail by the rope holding the mast to the board, slowly, sliding one hand over the other in succession.
- Grab the mast with your front hand, being careful to balance so as not to fall backwards.
- Move your back foot to the back of the board, facing downwind, while your front foot faces forward (toward the front of the board), as you put your hands on the boom and close the sail (pull inwards with the back arm).
To waterstart:
- Point the board in the direction you want to go.
- Hold the sail over the board and your head downwind from the board.
- Put your feet in the straps to climb on the board, while closing the sail and lifting yourself up with the push of the wind generated by this. Closing the sail is key.
2. The
2. The attack lines are passed under the chicken loop line on the corresponding side.
3. For
3. For more power, tie the lines so as to make the inside (bottom) shorter and the attack lines longer, and viceversa.
Getting the kite up
2. Fold
2. Fold the point of the kite that’s closest to the bar inward on top of the kite.
3. Put
3. Put sand on it to keep it there.
4. Pull
4. Pull from the top side of the bar to get it to grab wind.
5. Pull
5. Pull from the bottom to get the sand released.
Flying the kite
2. Keep
2. Keep the kite within the wind window.
Waterstarting
Turning around in the water
Signals
• Touch
• Touch your head to ask for help in getting it down.
Lesson 2 (September 19-20, 2003):
• Lines
• Lines must not cross if the kite is placed with the front toward the bar, as it will be in the air.
• In
• In good wind, inflate the kite by holding it high with the front facing the wind.
• To
• To waterstart, first fly the kite downwind so that it is downwind of you, otherwise it does not have enough power. Then make it fall in the direction you want to travel to gain power and lift off.
Lesson 3 (based loosely on instruction by Bill Kraft, Best kite instructor at St. Croix, USVI):
o Do
o Do not use kite leash; dangerous and you can fetch the board w/o a leash.
o Big
o Big U on bar can be removed for beginners, it’s for tricks.
o Leading
o Leading edge is the main big rib.
o Always
o Always hold kite by main rib at the center.
o To
o To inflate, kite should be concave: ribs facing downwind first, then curving up.
o Inflate
o Inflate main rib last.
o To
o To inflate, use Inflate hole; make sure Deflate one is closed.
o While
o While inflating, attach inflator ball to loop in the middle of the main rib to prevent kite from flying away.
o To
o To rest kite on beach, kite should be convex: place the main rib on the sand, other ribs facing up first then curving downwind.
o To
o To get kite to that position when kite is concave, hold main rib center with one hand, using that point as a pivot, and push the main rib end over and above the center, walking with it.
o Attach
o Attach lines to kite before completely unwinding lines from bar? If there is no knot/mess to begin with, this will help avoid one.
o Red
o Red lines go from middle of bar (chicken loop) to the leading edge. Black lines go from ends of bar to the trailing edge. Best logo goes to the left of the bar (looking at it from the kitesurfer).
o Black
o Black line goes under the red line and the kite (when the kite is in the resting, convex position).
o Tie
o Tie lines to the outermost knot in each attachment cord.
• Starting
• Starting the kite:
o To
o To start kite, have helper hold kite on the side of the window, 90 degrees from wind direction wrt kitesurfer, with leading edge facing the wind, and both ends facing the kitesurfer. Kite should look like a C facing the kitesurfer when watched from downwind.
o Always
o Always start kite toward sea. If you get dragged, you’ll get dragged out to sea, not to beach.
• Getting
• Getting up on the board:
o Board’s
o Board’s letters should not be readable by kitesurfer; the board is asymmetric due to footpads.
o Start
o Start by flying the kite ever so little in the direction opposite from that you want to launch in, then diving the kite quite a bit (but not completely), then pulling it back up and keeping it there, or if there is little wind do one or two eights before.
o Start
o Start with the board facing slightly downwind.
• Riding:
o Do
• Riding:
o Do slight and slow movements!
o Always
o Always look at the kite.
o Eights’
o Eights’ turning is always in the direction you are going, i.e. the leading edge always keeps facing in the direction you’re going.
o Sit
o Sit more (push ass back, not forward) while on board riding. Always lean back.
o Distribute
o Distribute weight equally on both feet.
• Fetching
• Fetching the board in the water:
o To
o To fetch board to your right, put your right arm in water, hold bar with left, move kite to the right of the window, leave legs behind you and go fetch it. If board is behind you, you may have to do multiple zigzags trying to go as far upwind as possible in each. Never put legs forward.
o When
o When the kite is overhead, you stand vertically instead of putting legs behind you.
• Water
• Water relaunching:
o If
o If the kite falls on water, swim toward the kite. The loose line will make the kite turn from neutral (convex) to concave (pulling) position, the kite will then at some point pick a side to start going toward, and once it’s on a side, you can start it normally.
• Jumping
o While
• Jumping
o While riding, let the kite go up to the overhead position, and you will jump.
o While
o While coming back down, move the kite back toward the direction you are going to land and resume normal flight.
• Turning
• Turning around
o Simply
o Simply let the board come overhead and back toward the opposite side.
• At
• At the end of the day:
o Tie
o Tie lines as an 8 along long side of the bar, then wind ends around the bar shaft circumference and tie.
o Do
o Do not rinse seawater; freshwater can corrode board, so dry after freshwater use.
Lesson 4 (Fernando, Denia):
• To
• To start w/o helper, put kite in same C position a helper would put it in, tip the bottom end upward toward the center of the kite, place sand on top of that tip, make sure lines go on top of the tip so they don’t tangle, then run to the bar, pull the bottom lines to take sand off, and lift kite.
• To
• To depower a kite on water, pull on the brake line on the wind side, or drive it to the edge of the window.
• Buy
• Buy a ball to keep the bar fixed so it does not go up and down the lines, to save you from
Self-lesson 5 (Long Beach, CA):
•To get a Peter Lynn ARC kite up, face the front of the kite toward the wind, then alternatively pull one and the other end of the bar, one side to get the kite facing the wind and the other to get it to start going up across the wind window. Make sure the brake is off, i.e. that the bar is as far as possible from you, making the central (front) lines shorter relative to the outside (back) ones. This makes the kite go up.
_uacct = "UA-524523-1";
urchinTracker();
]]>
After being more than 85 days late for the last budget, the California Legislature and the Gubernator are once again at a stalemate in discussions to approve a budget. This threatens to bring the state government to a screeching halt, and prompts the question,doesn't California deserve a mechanism to set the budget that isn't subject to stalemates?
In California, the magnitude of the budget is set by The California Balanced Budget Act’s requirement that the budget be balanced. All that’s needed, then, is to decide how to spend it. The candidates to receive money are all the programs and offices that have previously been voted into existence by the legislature. The question of how much to apportion to each program or office can best be thought of not as how much to fund any given program or office, but rather which program or office is the best to spend the next dollar on for each of the dollars in the budget. In other words, once a budget is subject to a cap, the process of devising a balanced budget can be broken into the process of figuring out the highest priority for each dollar available.
]]>
submitted
After being more than 85 days late for the last budget, the California Legislature and the Gubernator are once again at a stalemate in discussions to approve a budget. This threatens to bring the state government to a screeching halt, and prompts the question,doesn't California deserve a mechanism to set the budget that isn't subject to stalemates?
In California, the magnitude of the budget is set by The California Balanced Budget Act’s requirement that the budget be balanced. All that’s needed, then, is to decide how to spend it. The candidates to receive money are all the programs and offices that have previously been voted into existence by the legislature. The question of how much to apportion to each program or office can best be thought of not as how much to fund any given program or office, but rather which program or office is the best to spend the next dollar on for each of the dollars in the budget. In other words, once a budget is subject to a cap, the process of devising a balanced budget can be broken into the process of figuring out the highest priority for each dollar available.
]]>
QLess
• Founder and CEO. 3/07 - Present
Whozat, The People Search Engine
• Founder and CEO. 12/06 - Present
abInventio, the invention factory
• Founder and CEO. 9/06 - Present
• Co-Founder.
Adapt
• Founder and Director. 12/04 –
• Founder, President and Chief Technology Officer. 11/05 – 07/06
• Founder, President and Chief Executive Officer. 12/04 – 10/05
• Teaching Assistant, Neurobiology, California Institute of Technology, 1995-96.
• Teacher in the Center for Talented Youth Program, MIT, 1994 and Tutor in Genetics, MIT, 1994-95.
"UA-524523-1";
urchinTracker();
var sc_project=1777361;
var
var
var
]]>
submitted to the Op-Ed pages of the LATimes
After being more than 85 days late for the last budget, the California Legislature and the Gubernator are once again at a stalemate in discussions to approve a budget. This threatens to bring the state government to a screeching halt, and prompts the question,doesn't California deserve a mechanism to set the budget that isn't subject to stalemates?
In California, the magnitude of the budget is set by The California Balanced Budget Act’s requirement that the budget be balanced. All that’s needed, then, is to decide how to spend it. The candidates to receive money are all the programs and offices that have previously been voted into existence by the legislature. The question of how much to apportion to each program or office can best be thought of not as how much to fund any given program or office, but rather which program or office is the best to spend the next dollar on for each of the dollars in the budget. In other words, once a budget is subject to a cap, the process of devising a balanced budget can be broken into the process of figuring out the highest priority for each dollar available.
]]>
By Alex Bäcker, Ph.D.
Manuscript submitted the Op-Ed pages of the LATimes
After being more than 85 days late for the last budget, the California Legislature and the Gubernator are once again at a stalemate in discussions to approve a budget. This threatens to bring the state government to a screeching halt, and prompts
It is. The government ought to start with
In California, the
Once the magnitude of
To decide how best to spend each dollar, the marginal benefit of a dollar invested in each program needs to be known.
While critics ridicule assigning a value to a human life, the point of the process is not to value a life, but rather to decide how best to spend limited resources. Clearly, if your goal is to save lives, and you have limited resources, and one way of spending it saves a thousand lives and the others saves five hundred, the former will usually be
So how
How do we get a large set of legislators with competing agendas to arrive at one
The budget thus set would be much less likely to be objectionable to the Governor (or the President, if discussing the Federal budget). A similar process could be used to settle disputes between the Legislature and the Executive branches of government.
Perhaps it is time to complement the classic wisdom instilled in our legal system by the Founding Fathers, schooled in the liberal arts, with some mathematics.
Altadena, CA, February 2, 2009 (revised June 25)
References:
]]>
This Is Your Brain on Music: The Science of a Human Obsession by Daniel J. Levitin comes closer. Yet Levitin provides no real explanation for what constitutes music or why we like it: he mentions predictability seems to play a role, but he also mentions that the best songs don't seem all that predictable. Music has a Goldilocks problem: we like music that's not too predictable, nor too unpredictable, but just right. We need a better definition of just what just right is.
If music was simply about predictability, then we'd like uniform sequences of tones, or even uniform tones. But we don't. Yet things start making sense when you recall that brains are sensitive to change. Music, then, is predictable change. In other words, music is recursive pattern of patterns, a pattern of patterns of patterns of...Music is P, where P=a pattern of P.
]]>
If music was simply about predictability, then we'd like uniform sequences of tones, or even uniform tones. But we don't. Yet things start making sense when you recall that brains are sensitive to change. Music, then, is predictable change. In other words, music is recursive pattern of patterns, a pattern of patterns of patterns of...Music is P, where P=a pattern of P.
Our brain has evolved to predict our environment. It is a hierarchical pattern predictor. To this end, it has evolved to derive pleasure out of correct predictions. The more predictable patterns there are, the more of our brain that will experience prediction confirmation, and the more likely we are to like the music.
This theory assumes innate positive reinforcement from seeing predictions confirmed. This can be tested by measuring crying cessation or other behavioral manifestations of comfort and discomfort in infants during presentation of predictable rhythmic tunes versus arhythmic unpredictable tunes which skip beats, etc.
July 16, 2003; expanded October 28, 2007
urchinTracker();
]]>
This Is Your Brain on Music: The Science of a Human Obsession by Daniel J. Levitin comes closer. Yet Levitin provides no real explanation for what constitutes music or why we like it: he mentions predictability seems to play a role, but he also mentions that the best songs don't seem all that predictable. Music has a Goldilocks problem: we like music that not too predictable, nor too unpredictable, but just right. We need a better definition of just what just right is.
If music was simply about predictability, then we'd like uniform sequences of tones, or even uniform tones. But we don't. Yet things start making sense when you recall that brains are sensitive to change. Music, then, is predictable change. In other words, music is recursive pattern of patterns, a pattern of patterns of patterns of...Music is P, where P=a pattern of P.
]]>
by
Alex Backer, Ph.D.
Altadena, CA, February 2, 2009 (revised June 25)
]]>
office.
To decide how best to spend each dollar, the marginal benefit of a dollar invested in each program needs to be known. Fortunately, there is an office in charge of computing just that --theOffice of Information and Regulatory Affairs. That office could estimate, for example, that the first 80 million spent on healthcare save a life for every million dollars, but that the next 20 million save a life only every 3 million dollars, whilst the first 20 million dollars spent on firefighting programs saves a life every 2 million dollars. Those estimates would dictate that, if legislators decide that saving lives is their first priority, among those two programs, the first 80 million be spent on healthcare and the next 20 on firefighting.
While critics ridicule assigning a value to a human life, the point of the process is not to value a life, but rather to decide how best to spend limited resources. Clearly, if your goal is to save lives, and you have limited resources, and one way of spending it saves a thousand lives and the others saves five hundred, the former will usually be preferrable. Not knowing the marginal efficacy of investing an additional dollar in each program is tantamount to making those important decisions blindly.
So how do we get a large set of legislators with competing agendas to arrive at one budget, or an entire set of dollar numbers? Again, a yes/no voting process is inappropriate, but there are voting mechanisms that fit the bill. The first constraint, as we saw above, is that the budget must add up to the total, so what is being voted for is what fraction of the budget to spend on each program or office. What we seek is a distribution that best matches the legislators' perceived priorities. To this end, each legislator could be asked to submit a distribution of the agreed-upon total budget into the programs and offices he sees fit to fund. These distributions could then be combined to arrive at the final outcome, not by consensus (which can be virtually impossible when you ask hundreds of legislators to agree on each of hundreds of numbers, as we have seen repeatedly over the past year), but by finding out the most moderate budget among all the ones proposed by legislators. This can be achieved in many ways, but a simple one is ask each to look for the budget whose most extreme allocation is least extreme. In other words, you takeeach line item in the budgetone at a time and rank each legislator by how extreme their proposed allocation of the budget to that line item is. A legislator who proposed the median allocation to that line item, i.e. the one who had as many legislators proposing more as legislators proposing less, would get an 'extremism' score of zero. Conversely, the legislators proposing the largest and smallest allocations to that line item would get the highest extremism score. This process is repeated for each line item, and each legislator receives a 'total extremism' score equal to the maximum extremism score he got for each line item. The budget proposed by the legislator with the lowest total extremism score is adopted. This process will arrive at the "least objectionable" budget.Alternatively, the process above can be used to discover the two (or more) least objectionable budgets, which can then be put up for a vote where each legislator must select one of the two, with the budget receiving the greatest number of votes adopted.
budget).
Perhaps it is time to complement the classic wisdom instilled in our legal system by our Harvard- and Yale-educated Founding Fathers with some of the mathematics developed over the last two centuries.
Alex Backer, Ph.D.
]]>
Once the magnitude of the budget is set, the next step is to decide how to spend it. The candidates are easy: all the programs and offices that have previously been voted into
To decide how best to spend each dollar, the marginal benefit of a dollar invested in each program needs to be known. Fortunately, there is an office in charge of computing just that --theOffice of Information and Regulatory Affairs. That office could estimate, for example, that the first 80 million spent on healthcare save a life for every million dollars, but that the next 20 million save a life only every 3 million dollars, whilst the first 20 million dollars spent on firefighting programs saves a life every 2 million dollars. Those estimates would dictate that, if legislators decide that saving lives is their first priority, among those two programs, the first 80 million be spent on healthcare and the next 20 on firefighting.
While critics ridicule assigning a value to a human life, the point of the process is not to value a life, but rather to decide how best to spend limited resources. Clearly, if your goal is to save lives, and you have limited resources, and one way of spending it saves a thousand lives and the others saves five hundred, the former will usually be preferrable. Not knowing the marginal efficacy of investing an additional dollar in each program is tantamount to making those important decisions blindly.
]]>
It is. The government ought to start with the most basic notion of budget setting: that you can't spend what you don't have. So the first step ought to be to set the level of overall expenditure, never more than what is available, less than that in good times to save for rainy days. How do you get a large number of legislators to agree on one number with a voting system that was designed to agree on a binary yes/no decision?Well, there are voting processes that can arrive at a "consensus"number. For example, legislators could each vote for a budget level (in dollars), and the median selected (the median is the number that falls in the middle of a distribution:for example, the median of {2,2,4,6,7} is 4). The beauty of the median is that the actual number gets set by the most moderate
Once the magnitude of the budget is set, the next step is to decide how to spend it. The candidates are easy: all the programs and offices that have previously been voted into law by the legislature (i.e. those that have passed the test of the binary yes/no vote of the legislature). The difficult question is how much of the budget to apportion to each program or office. It is critical to note here that the decision sought is not really how much to fund any given program or office, but rather which program or office is the best to spend the next dollar on. In other words, because the budget is subject to the constraint of not adding up to more than what is available, the process of devising a balanced budget can be broken into the process of figuring out the highest priority for each dollar available. So, for example, the legislature could decide that the top priority for the first 100 million dollars is to have hospitals work, the top priority for the next 100 million is to have public schools, etc., until the budget has been exhausted.
To decide how best to spend each dollar, the marginal benefit of a dollar invested in each program needs to be known. Fortunately, there is an office in charge of computing just that --theOffice of Information and Regulatory Affairs. That office could estimate, for example, that the first 80 million spent on healthcare save a life for every million dollars, but that the next 20 million save a life only every 3 million dollars, whilst the first 20 million dollars spent on firefighting programs saves a life every 2 million dollars. Those estimates would dictate that, if legislators decide that saving lives is their first priority, among those two programs, the first 80 million be spent on healthcare and the next 20 on firefighting.
While critics ridicule assigning a value to a human life, the point of the process is not to value a life, but rather to decide how best to spend limited resources. Clearly, if your goal is to save lives, and you have limited resources, and one way of spending it saves a thousand lives and the others saves five hundred, the former will usually be preferrable. Not knowing the marginal efficacy of investing an additional dollar in each program is tantamount to making those important decisions blindly.
So how do we get a large set of legislators with competing agendas to arrive at one budget, or an entire set of dollar numbers? Again, a yes/no voting process is inappropriate, but there are voting mechanisms that fit the bill. The first constraint, as we saw above, is that the budget must add up to the total, so what is being voted for is what fraction of the budget to spend on each program or office. What we seek is a distribution that best matches the legislators' perceived priorities. To this end, each legislator could be asked to submit a distribution of the agreed-upon total budget into the programs and offices he sees fit to fund. These distributions could then be combined to arrive at the final outcome, not by consensus (which can be virtually impossible when you ask hundreds of legislators to agree on each of hundreds of numbers, as we have seen repeatedly over the past year), but by finding out the most moderate budget among all the ones proposed by legislators. This can be achieved in many ways, but a simple one is ask each to look for the budget whose most extreme allocation is least extreme. In other words, you takeeach line item in the budgetone at a time and rank each legislator by how extreme their proposed allocation of the budget to that line item is. A legislator who proposed the median allocation to that line item, i.e. the one who had as many legislators proposing more as legislators proposing less, would get an 'extremism' score of zero. Conversely, the legislators proposing the largest and smallest allocations to that line item would get the highest extremism score. This process is repeated for each line item, and each legislator receives a 'total extremism' score equal to the maximum extremism score he got for each line item. The budget proposed by the legislator with the lowest total extremism score is adopted. This process will arrive at the "least objectionable" budget.Alternatively, the process above can be used to discover the two (or more) least objectionable budgets, which can then be put up for a vote where each legislator must select one of the two, with the budget receiving the greatest number of votes adopted.
The budget thus set would be much less likely to be objectionable to the Governor (or the President, if discussing the Federal budget). But a similar process could be used to settle disputes between the Legislature and the Executive branches of government.
Perhaps it is time to complement the classic wisdom instilled in our legal system by our Harvard- and Yale-educated Founding Fathers with some of the mathematics developed over the last two centuries.
Alex Backer, Ph.D.
]]>
Read more of Alex Backer's Proposals for Government.
]]>
Once the magnitude of the budget is set, the next step is to decide how to spend it. The candidates are easy: all the programs and offices that have previously been voted into law by the legislature (i.e. those that have passed the test of the binary yes/no vote of the legislature). The difficult question is how much of the budget to apportion to each program or office. It is critical to note here that the decision sought is not really how much to fund any given program or office, but rather which program or office is the best to spend the next dollar on. In other words, because the budget is subject to the constraint of not adding up to more than what is available, the process of devising a balanced budget can be broken into the process of figuring out the highest priority for each dollar available. So, for example, the legislature could decide that the top priority for the first 100 million dollars is to have hospitals work, the top priority for the next 100 million is to have public schools, etc., until the budget has been exhausted.
To decide how best to spend each dollar, the marginal benefit of a dollar invested in each program needs to be known. Fortunately, there is an office in charge of computing just that --theOffice of Information and Regulatory Affairs. That office could estimate, for example, that the first 80 million spent on healthcare save a life for every million dollars, but that the next 20 million save a life only every 3 million dollars, whilst the first 20 million dollars spent on firefighting programs saves a life every 2 million dollars. Those estimates would dictate that, if legislators decide that saving lives is their first priority, among those two programs, the first 80 million be spent on healthcare and the next 20 on firefighting.
While critics ridicule assigning a value to a human life, the point of the process is not to value a life, but rather to decide how best to spend limited resources. Clearly, if your goal is to save lives, and you have limited resources, and one way of spending it saves a thousand lives and the others saves five hundred, the former will usually be preferrable. Not knowing the marginal efficacy of investing an additional dollar in each program is tantamount to making those important decisions blindly.
fund.
Perhaps it is time to complement the classic wisdom instilled in our legal system by our Harvard- and Yale-educated Founding Fathers with some of the mathematics developed over the last two centuries.
Alex Backer, Ph.D.
2,
References:
1. An alternative proposal with several good and complementary ideas to revise the state budget process, including multi-year planning, mid-year revisions, and accountability of existing programs can be found atcaliforniaforward.com/CA_Forward_Revised.pdf
]]>
-------------------------
P.S. If you want to read more:
With Ram Prayaga, Alex co-foundedCafe|Think™, which launchedChoizi™, an online personalized recommendation service to help users find restaurants of their liking (most of the credit should go to Ram, says Alex). He is also the founder ofTreeofBabel, a fledgling nonprofit collaborative web project to map out the genealogy of human languages.
]]>
In 2007, Alex founded abinventio™, an inventionfactory with a focus on challenging algorithmic problems, where he has since been practicing parallel entrepreneurship. abinventio has since released:
QLess™,
> Whozat™, The People Search Engine™, named one of the 100 hottest start-ups in the world by famed technology blog TechCrunch, a semantic search engine with machine vision that garnered tens of thousands of pageviews within a week of launch;
> HoundWire™, a location-aware community-powered news engine.
> PREPROVE™, a statistically sound behavioral way of measuring brand power and testing marketing messages which is orders of magnitude cheaper and more accurate than traditional focus groups and surveys.
> TheSEMExperts™, for marketers looking for their products to be found.
In 2008, Alex was appointed by the President to serve in the California Insitute of Technology's Information Sciences and TechnologyBoard of Advisors together with a distinguished cadre of scientists, technologists and philantropists including Carver Mead and Phil Neches.
He has been an invited speaker at numerousconferences on diverse science and technology topics around the world. Alex is always lookingforworld-classentrepreneurialpeopletohelpbringaboutinnovationthroughtechnology. Alex likes to spend his free time traveling with his family, kitesurfing, filmmaking, or kayaking.
In addition to this wiki, Alex keeps a blogoflifeatab|inventio™,TheInventionFactory™.
With Ram Prayaga, Alex co-foundedCafe|Think™, which launchedChoizi™, an online personalized recommendation service to help users find restaurants of their liking (most of the credit should go to Ram, says Alex). He is also the founder ofTreeofBabel, a fledgling nonprofit collaborative web project to map out the genealogy of human languages.
abInventio's companies under development include:
>REPcloud™, a reputation engine which garnered tens of thousands of users within weeks of launching;
>HoundWire™, a location-aware community-powered news engine.
>1OFTHESE.COM™, a better way to buy and sell in the world's most efficient marketplace --signuptobenotifieduponlaunch--, and
>mynew.TV, a better way to be entertained that's all about you.
P.S. 2:If you want to read even more, read Alex'sCV.
]]>
The Paradox of Monogamy
My readers' all-time favorite posts:
These are posts that 100% of readers found engaging enough that they kept reading afterwards:
]]>
my
See below for computational details.
--Alex Backer, Ph.D., Caltech, Pasadena, California, May 31st, 2002.
]]>
Rule 30, a pattern that grew more intricate and unpredictable with each step. It was stuffed with what mathematicians call "emergent effects": events that cannot be predicted in advance. From the simplest of parts, Wolfram had created infinite complexity. Theaha!moment had arrived. "The Rule 30 automaton is the most surprising thing I've ever seen in science," Wolfram told London'sDaily Telegraph. "'Even though it starts off from just one black cell, applying the same simple rule over and over again makes Rule 30 produce %5Ban%5D amazingly complex pattern.
"It took me several years to absorb how important this was. But in the end, I realized that this one picture contains the clue to what's perhaps the most long-standing mystery in all of science: where in the end, the complexity of the natural world comes from."
Now, Iadmire Richard Feynman like I admire nobody but Albert Einstein, yetI hypothesized based on preliminary visual inspection that two successive whites would predict the next one is white more often than not. This was indeed confirmed. The plot below shows the probability that a square in the centra column is white given that the preceding N squares are black, in rows 26,400 through 40,795:
As you can see, the distribution is far from a random homogeneous distribution. QED.
Alex Bäckerat 10:56am May 13
Wolfram wrote of this, which he called "probably the single most surprising scientific discovery" he has ever made: "none of the tests I have ever done on it %5Bthe sequence of colors directly below the initial black cell%5D show any meaningful deviation at all from perfect randomness" (op. cit., p. 28). He goes on to claim it can be used for anyapplication that requires random number generation. Now, if I can predict the probability of a given cell to be greater of being black or white, that's not "perfect randomness", and in particular, if online casinos used it to generate poker draws or anything else, they'd already be bankrupt
Any time that you have non-homogenous distributions you ought to be able to compress.
Alex Bäckerat 1:10pm May 13
Here is one algorithm you can use to compress that sequence: simply transmit the number of consecutive squares with the same color every time the sequence deviates from the prediction made by my discovery above, namely every time the sequence changes colors, instead of sending the color of every bit. Because the sequence changescolors less often than it stays the same, as proved above, transmitting the changes will result in compression.
In fact, as shown in the graph above, some positions in the sequence can be determined with about 90% accuracy, and I bet the accuracy goes up as you look at longer sequences.
In other words, Rule 30 is no more "perfectly random" than a biased coin is. Try telling the person you're betting against that your biased coin should be accepted as "perfectly random". :-)
Wolfram himself gives 8 tests of randomness, the first of which is "frequency or equidistribution test (possible elements should occur with equal frequency)" (op. cit., p. 1084). It is this first test that Rule 30's sequence fails to pass: sequences with many identical colors are more common than a quasi-identical sequence of the same length with the last color reversed.
Unless, of course, there is a bug in my programs.
See below for computational details.
--Alex Backer, Ph.D., Caltech, Pasadena, California, May 31st, 2002.
]]>
var sc_project=1777361; var sc_invisible=1; var sc_partition=16; var sc_security="c84ed9d0";
]]>
First, they don't let me select a city as origin or destination. I need to choose an airport. Which means that going from a city with 6 airports to one with 3, I need to try 18 combinations. Great!
They also don't allow me to narrow the time of arrival or departure that I can make, the way that Kayak does.
Then, the button for Submit/Search is labeled Go, and is a tiny spec placed right next to the Start Over button, making it real easy to lose everything after you've finished filling in all the information they ask for.
Finally, they give an error: This feature is currently unavailable. Please wait a few moments and try again.
It goes without saying that waiting a few moments does not magically solve the problem.
Conclusion: Once again, I won't fly American.
P.S. Why don't I use Kayak for this? Cause they won't find miles flights. God knows why.
]]>American Airlines' website is useless
Is the range of RangeMax NEXT Wireless-N routers really 10X that of Wireless-G?
What make Facebook so successful
]]>
Ph.D., Caltech, Pasadena, California, May 31st, 2002.
function %5Bcen,pval%5D=rule30lomem(Niter,start,cen)
% function %5Bcen,pval%5D=rule30lomem(Niter,start,cen)
]]>
As
See below for computational details.
--Alex Backer, Ph.D., May 31st, 2002.
function %5Bcen,pval%5D=rule30lomem(Niter,start,cen)
]]>
]]>Rule 30, a pattern that grew more intricate and unpredictable with each step. It was stuffed with what mathematicians call "emergent effects": events that cannot be predicted in advance. From the simplest of parts, Wolfram had created infinite complexity. Theaha!moment had arrived. "The Rule 30 automaton is the most surprising thing I've ever seen in science," Wolfram told London'sDaily Telegraph. "'Even though it starts off from just one black cell, applying the same simple rule over and over again makes Rule 30 produce %5Ban%5D amazingly complex pattern.
"It took me several years to absorb how important this was. But in the end, I realized that this one picture contains the clue to what's perhaps the most long-standing mystery in all of science: where in the end, the complexity of the natural world comes from."
The more Wolfram studied Rule 30, the more incredible it became. For example, though the black-and-white triangle, the product of 2 million calculations, seemed to exhibit a certain symmetry, it was, in fact, chaotic. In particular, following the single line of black and white tiles that ran vertically from the peak of the pyramid, Wolfram found perfect chaos—i.e., a pure random number generator. He showed it to his old Caltech physics mentor, the late Nobel laureate Richard Feynman. Feynman was convinced there had to be some regularity in Rule 30. He took off for Hawaii on vacation and, for fun, spent the time there bent on proving Wolfram wrong. When he returned, he admitted he'd failed to find any sign of order.
Now, Iadmire Richard Feynman like I admire nobody but Albert Einstein, yetI hypothesized based on preliminary visual inspection that two successive whites would predict the next one is white more often than not. This was indeed confirmed.
white fn len black string iter 26400-40795.emf
As you can see, the distribution is far from a random homogeneous distribution.
See below for details.
--Alex Backer, Ph.D., May 31st, 2002.
function %5Bcen,pval%5D=rule30lomem(Niter,start,cen)
]]>
--Alex Backer, Ph.D., May 31st, 2002.
function %5Bcen,pval%5D=rule30lomem(Niter,start,cen)
% function %5Bcen,pval%5D=rule30lomem(Niter,start,cen)
% See if I can predict central column better than chance (50-50)
% AB May 02
%
% Default START = 3; used as start of interval for prediction at bottom, end of interval=Niter
%
% Supersedes RULE30 - uses less memory xq does not store pattern matrix, and is newer
if nargin<2,
start=3; % Must be >=3 xq uses previous 2
end
saveeveryN=100;
cm
cd Automata
h=timebar(0);
if nargin<3,
% Calculate pattern necessary for central column up to Niter iterations:
% Inductive:
% Boundary cond:
c=zeros(2,Niter*2); % Initial conditions
c(1,Niter)=1; % c(lin,col); 1=black, 0=white
cen(1)=1;
%c=uint8(c);
% Induction:
% In order to calculate central column value at iteration Niter, you need 1 value each way of previous iteration, 2 values each way
% of the one before, etc.
% Because values more than iter-1 columns out of the central one are zero, we need to calculate a rhombus of values to get
% central column value at iteration Niter
nb=0;nw=0;
a=2;
lin=2;
for col=Niter-(lin-1):Niter+(lin-1),
if ~c(a-1,col) & ~c(a-1,col+1), % if above and above right are white,
c(a,col)=c(a-1,col-1); % keep color of above left
else,
c(a,col)=1-c(a-1,col-1); % else use opposite color
end
end
cen(lin)=c(a,Niter);
im=c;
colormap gray
c(1,:)=c(a,:);
for lin=3:floor(Niter/2)+1,
timebar(lin/Niter,h,'Calculating pattern');
for col=Niter-(lin-1):Niter+(lin-1),
if ~c(a-1,col) & ~c(a-1,col+1),
c(a,col)=c(a-1,col-1);
else,
c(a,col)=1-c(a-1,col-1);
end
end
cen(lin)=c(a,Niter);
im=%5Bim;c(a,:)%5D;
c(1,:)=c(a,:);
if ~cen(lin-1) & ~cen(lin-2),
if cen(lin),
nb=nb+1; % # black
else,
nw=nw+1; % # white
end
% nb & nw don't add to Niter because they only include elements for which previous two where white
end
%%5Bnb,nw%5D
if ~rem(lin,saveeveryN),
save rule30 lin cen c
imagesc(1-im(:,Niter-lin:Niter+lin))
end
end
width=Niter-lin;
for lin=floor(Niter/2)+2:Niter,
timebar(lin/Niter,h,'Calculating pattern');
for col=Niter-width:Niter+width,
if ~c(a-1,col) & ~c(a-1,col+1),
c(a,col)=c(a-1,col-1);
else,
c(a,col)=1-c(a-1,col-1);
end
end
cen(lin)=c(a,Niter);
c(1,:)=c(a,:);
width=width-1;
if ~cen(lin-1) & ~cen(lin-2),
if cen(lin),
nb=nb+1; % # black
else,
nw=nw+1; % # white
end
% nb & nw don't add to Niter because they only include elements for which previous two where white
end
%%5Bnb,nw%5D
if ~rem(lin,saveeveryN),
save rule30 lin cen c
end
end
% Recursive:
% recursion uses previous results more than once, so keep track
%c(Niter,Niter)=
else,
% Predict
%timebar(0,h,'Calculating frequencies...')
%5Bnb,nw%5D=predict(cen,start,Niter)
end
% X-corr:
XCORR=0;
if XCORR,
xc=xcorr(cen,'unbiased');
figure(2);plot(xc)
m=mean(xc(1:Niter-2))
xc1=xc(Niter-1)
pval=sum(xc(1:Niter-2)>xc1)/(Niter-2)
end
% Significance:
Nrand=100;
npos=0;
rand('state',sum(100*clock));
for i=1:Nrand,
timebar(i/Nrand,h,'Evaluating significance...');
r=floor(rand(1,Niter)*2);
%5Bcnb,cnw%5D=predict(r,start,Niter);
if abs(cnw-cnb)>=abs(nw-nb),
npos=npos+1;
end
end
cnb,cnw
pval=npos/Nrand
close(h)
function %5Bnb,nw%5D=predict(cen,start,Niter)
% Predict
nb=0;nw=0;
for i=start:Niter,
if ~cen(i-1) & ~cen(i-2),
if cen(i),
nb=nb+1; % # black
else,
nw=nw+1; % # white
end
% nb & nw don't add to Niter because they only include elements for which previous two where white
end
end
%%5Bnb,nw%5D
function pwhitefnn(maxn,white,cenrange,nv)
if nargin<1,
maxn=15;
end
if nargin<2,
white=1;
end
if nargin<4,
nv=1:maxn;
end
Nrand=0; %100;
cm;
cd Automata
load rule30iter40795 %31300 %26400
if nargin<3,
cenrange=1:length(cen);
end
cen=cen(cenrange);
Niter=length(cen);
start=%5B%5D;
h=timebar(0);
for n=nv,
timebar(n/maxn,h);
%5Bcen,pval(n),nb,nw%5D=rule30lomemn(Niter,start,cen,n,Nrand,white);
pwhite(n)=nw/(nb+nw);
end
close(h);
figure
plot(pwhite)
xlabel('N')
if white,
ylabel('p(white|N previous samples were white)')
else,
ylabel('p(white|N previous samples were black)')
end
figure
plot(pval)
xlabel('N')
ylabel('p-value')
function pwhitefnnfncen(maxn,white,cenrange,nv,cen)
% pwhitefnnfncen(maxn,white,cenrange,nv,cen)
if nargin<1,
maxn=15;
end
if nargin<2,
white=1;
end
if nargin<4,
nv=1:maxn;
end
Nrand=0; %100;
cm;
cd Automata
%load rule30iter40795 %31300 %26400
if nargin<3,
cenrange=1:length(cen);
end
cen=cen(cenrange);
Niter=length(cen);
start=%5B%5D;
h=timebar(0);
for n=nv,
timebar(n/maxn,h);
%5Bcen,pval(n),nb,nw%5D=rule30lomemn(Niter,start,cen,n,Nrand,white);
pwhite(n)=nw/(nb+nw);
end
close(h);
figure
plot(pwhite,'x')
xlabel('N')
if white,
ylabel('p(white|N previous samples were white)')
else,
ylabel('p(white|N previous samples were black)')
end
% figure
% plot(pval)
% xlabel('N')
% ylabel('p-value')
>> cen=rule30lomem(1000);
nb = 127
nw = 156
>> cen=rule30lomem(2000,1000);
nb = 121
nw = 144
>> cen1=rule30lomem(2000,3,cen);
nb = 248
nw = 300
pval = 0 (Nrand=100)
After long run that I lost:
%5Bnb,nw%5D= 2548 2505
pval<0.001 (using absolute values)
After long strings of white, white is much more probable than black. The longer the string, the more the asymmetry.
The reverse happens with long strings of black (see below).
Note that p-values below are double-sided, so real p-values are half of that shown.
0 below means white string:
>> pwhitefnn(14,0,26401:40200)
nb = 3377
nw = 3414
pval = 0.6600
nb = 1702
nw = 1675
pval = 0.6400
nb = 863
nw = 839
pval = 0.5500
nb = 454
nw = 409
pval = 0.1600
nb = 249
nw = 205
pval = 0.0300
nb = 128
nw = 121
pval = 0.5900
nb = 67
nw = 61
pval = 0.6500
nb = 37
nw = 30
pval = 0.4400
nb = 23
nw = 14
pval = 0.1200
nb = 14
nw = 9
pval = 0.1700
nb = 11
nw = 3
pval = 0.0500
nb = 9
nw = 2
pval = 0.0200
nb = 8
nw = 1
pval = 0
nb = 7
nw = 1
pval = 0
Black strings:
>> pwhitefnn(14,1,26401:40795,10:14)
nb = 10
nw = 15
pval = 0.2100
nb = 4
nw = 11
pval = 0
nb = 3
nw = 8
pval = 0.0300
nb = 2
nw = 6
pval = 0
nb = 1
nw = 5
pval = 0.0100
pwhitefnn(14,0,26401:40795)
To start rule30lomemcontfast using data from rule30lomemcont:
load rule30iter44100
newc=ones(1,Niter*2-1);
newc(1000000-110000:1000000-110000+size(c,2)-1)=c(1,:);
Niter=1000000;%5Bcen,pval%5D=rule30lomemcontfast(Niter,11,cen,newc,lin);
>> pwhitefnnfncen(20,1,cen)
nb = 35037
nw = 35092
nb = 17565
nw = 17527
nb = 8676
nw = 8851
nb = 4428
nw = 4423
nb = 2211
nw = 2212
nb = 1100
nw = 1112
nb = 558
nw = 554
nb = 267
nw = 287
nb = 139
nw = 148
nb = 61
nw = 87
nb = 28
nw = 59
nb = 22
nw = 37
nb = 13
nw = 24
nb = 9
nw = 15
nb = 6
nw = 9
nb = 3
nw = 6
nb = 3
nw = 3
nb = 2
nw = 1
nb = 1
nw = 0
nb = 0
nw = 0
>> pwhitefnnfncen(20,0,cen)
nb = 35133
nw = 35037
nb = 17620
nw = 17513
nb = 8840
nw = 8780
nb = 4484
nw = 4356
nb = 2323
nw = 2161
nb = 1183
nw = 1140
nb = 596
nw = 587
nb = 313
nw = 283
nb = 156
nw = 157
nb = 76
nw = 80
nb = 43
nw = 33
nb = 20
nw = 23
nb = 11
nw = 9
nb = 9
nw = 2
nb = 7
nw = 2
nb = 5
nw = 2
nb = 4
nw = 1
nb = 3
nw = 1
nb = 2
nw = 1
nb = 1
nw = 1
Vectorized form shown to produce same sequence as non-vectorized one for 1st 20 iterations. If I made a mistake, it could only have been in the sequence passed to the vectorized to continue after 43300.
All iterations after conversion to vectorized form feeding the last line inverted:
>> length(cen),pwhitefnnfncen(20,0,cen(43301:end))
ans = 140300
nb = 24191
nw = 24240
nb = 12102
nw = 12089
nb = 6025
nw = 6077
nb = 2995
nw = 3030
nb = 1510
nw = 1485
nb = 750
nw = 760
nb = 370
nw = 380
nb = 190
nw = 180
nb = 92
nw = 98
nb = 40
nw = 52
nb = 20
nw = 20
nb = 5
nw = 15
nb = 0
nw = 5
nb = 0
nw = 0
nb = 0
nw = 0
nb = 0
nw = 0
nb = 0
nw = 0
nb = 0
nw = 0
nb = 0
nw = 0
nb = 0
nw = 0
>> pwhitefnnfncen(20,1,cen(43301:end))
nb =
24241
nw =
24327
nb =
12192
nw =
12135
nb =
6024
nw =
6111
nb =
3076
nw =
3035
nb =
1536
nw =
1499
nb =
764
nw =
735
nb =
371
nw =
364
nb =
181
nw =
183
nb =
91
nw =
92
nb =
37
nw =
55
nb =
17
nw =
38
nb =
14
nw =
24
nb =
9
nw =
15
nb =
6
nw =
9
nb =
4
nw =
5
nb =
2
nw =
3
nb =
2
nw =
1
nb =
1
nw =
0
nb =
0
nw =
0
nb =
0
nw =
0
To start rule30lomemcontfast using data from rule30lomemcont: correctly:
load rule30iter44100
cen=1-cen;
c=1-c;
newc=ones(1,Niter*2-1);
newc(1000000-110000:1000000-110000+size(c,2)-1)=c(1,:);
Niter=1000000;%5Bcen,pval%5D=rule30lomemcontfast(Niter,11,cen,newc,lin);
--Alex Backer, Ph.D., May 31st, 2002.
]]>Wolfram was wrong: 'Random number generator' Rule 30 Shows Statistical Regularities
Neuroscience and Neural Networks:
How does the brain recognize faces?
]]>
A Better Way to Balance the State and Federal Budgets
A Better Democracy
]]>
Craig Johnson
]]>
ATribute to Robert Frost
The Rabbit Hole exists: My discovery of the Coffeehouse Backstage and Fairport Convention
]]>
abInventio, the invention factory
CafeThink, LLC.
Adapt Technologies, Inc.
• Founder,
• Founder, President and Chief Technology Officer. 11/05 – 07/06
• Founder,
• Founder, President and Chief Executive Officer.
Computational Biology
• Principal
• Principal Investigator, independent research program in linguistics, evolution, olfaction, computation, mathematics, neuroinformatics and bioinformatics. 6/02 – 05/05
Beckman Institute, Center for Neuromorphic Systems Engineering
• Visiting
• Visiting scientist. 06/02 – 05/05
Advanced Concepts Group, Sandia National Laboratories
Computation
• Sloan
• Sloan Fellow for Theoretical Neurobiology. Thesis: Pattern recognition in the olfactory system. Priming, gain control and coding issues. 4/96 – 4/02
Sensory Information Processing Laboratory, California Institute of Technology
Christof Koch’s Lab, California Institute of Technology
Cortical Development, Plasticity and Dynamics Laboratory, Brain and Cognitive Sciences Dept., MIT
MIT Center for Cancer Research
Immunogenetics Laboratory, Hospital of the University of Buenos Aires
Institute of Biochemical Investigations, Fundacion Campomar, L.F. Leloir
EDUCATION
and
6/97 – 4/02
M.S., California Institute of Technology, Computation and Neural
9/95 – 6/97
• GRE
• GRE scores: Analytical 790/800, Quantitative 770/800, Verbal (English) 640/800.
S.B., Massachusetts Institute of
9/93 - 6/95
• Cross-registered
• Cross-registered at Harvard University.
University of Buenos Aires, School of Natural and Exact
3/91 - 8/93
AWARDS
•
• Graduate
• Graduate Dean's Award for Outstanding Community Service, for “great contributions to Graduate Life and outstanding qualities of Leadership and Responsibility”, 2002.
• ISTA
• ISTA award from the American Association for the Advancement of Science (AAAS), 2001.
• Award
• Award for Contribution to Biology Undergraduate Student Association, MIT, 1995.
• Member
• Member of Argentina’s National Team in International Olympiad of Informatics, Greece, 1991.
• Second
• Second place, National Olympiad of Informatics, Argentina, 1991.
• National
• National Champion, National Olympiad of Informatics, Argentina, 1990.
• Third
• Third Millennium Prize from the National Ministry of Health and Welfare for performance in 1990 International Olympiad of
• Member
• Member of Argentina’s National Team in International Olympiad of Informatics, USSR, 1990.
• Dux
• Dux Medallist for best results ever in school’s 150-yr. history (maximum scores in seven subjects) in Cambridge University (UK) ‘O’, ‘S’ and ‘A’-level examinations, St. Andrew’s Scots School, Buenos Aires, 1990.
• Graduated
• Graduated first in class of 120, St. Andrew’s Scots School, Buenos Aires, 1990.
• Ian
• Ian Seyda Prize for Intellectual Curiosity, St. Andrew’s Scots School, Buenos Aires, 1990.
• Captain,
• Captain, Champion Team, ESSARP Inter-School General Knowledge Competition, Buenos Aires, 1990.
• Member
• Member of Argentina’s National Team selected for International Olympiad of Informatics, Bulgaria, 1989.
MENTORING
SELECTED PATENTS
• Bäcker,
• Bäcker, A.: Reputation system. U.S. Patent in course.
• Bäcker,
• Bäcker, A.: System to bid on a class of objects. U.S. Patent in course.
• Bäcker,
• Bäcker, A.: USING MACHINE LEARNING TO IDENTIFY SIMILAR ADVERTISERS AND/OR CAMPAIGN
• Bäcker,
• Bäcker, A.: ADVERTISING AGENT WITH INTERNAL BID AUCTION. U.S. Patent in course.
• Bäcker,
• Bäcker, A.: CHARGING ADVERTISERS FOR ESTABLISHED PERFORMANCE. U.S. Patent in course.
• Bäcker,
• Bäcker, A.: NORMALIZING BIDS FOR AND/OR ACROSS DIFFERENT ADVERTISING RESOURCE PROVIDERS. U.S. Patent in course.
• Bäcker,
• Bäcker, A.: KEY EXPRESSION GENERATOR AND OPTIMIZER. U.S. Patent in course.
• Bäcker,
• Bäcker, A.: PRICING METHOD FOR ONLINE ADVERTISING. U.S. Patent in course.
• Bäcker,
• Bäcker, A.: SYSTEM AND METHOD FOR ONLINE ADVERTISING. U.S. Patent in course.
• Gonzalez,
• Gonzalez, J. E. and Bäcker, A.: Queryless Search using a Computational Model of Curiosity and Learning from Text. U.S. and PCT Patent in course.
• Bäcker,
• Bäcker, A.: Query-specific Webpage Relevance Ranking. U.S. Provisional Patent in course.
• Bäcker,
• Bäcker, A.: Network-based trust evaluation. U.S. Provisional Patent in course.
• Bäcker,
• Bäcker, A.: System for Authentication of Electronic Communications. U.S. Provisional Patent in course.
• Bäcker,
• Bäcker, A.: Computational Method to Find Interactions between Genes Using Multiple Genomes. U.S. Provisional Patent.
• Bäcker,
• Bäcker, A.: Automatic Methods to find Associations between Expressions. U.S. Provisional Patent.
• Bäcker,
• Bäcker, A.: Method to Automatically Associate a Web Page or Document with an Expression And Summarize Documents. U.S. Provisional Patent.
• Bäcker,
• Bäcker, A.: Method to Direct Users to Relevant Portion of a Web Document. U.S. Provisional Patent.
• Bäcker,
• Bäcker, A.: User-Specific Adaptive Method to Sort Search Results by Relevance for the World Wide Web and Other Interconnected Data Structures. U.S. Provisional Patent.
• Bäcker,
• Bäcker, A, Peral, E.
SELECTED INVITED TALKS
• Bäcker, A. Search Engine Marketing. IMA annual meeting, 2008.
• Bäcker,
• Bäcker, A. Search Engine Marketing. IMA inaugural annual meeting, 2007.
• Bäcker,
• Bäcker, A. Entrepreneurship --A Primer. University of Toronto, 2006.
• Bäcker,
• Bäcker, A. Pay Per Click Strategies. Professional eBay Sellers Association, 2006.
• Bäcker,
• Bäcker, A. All Consumidores Are Not Created Equal. Search Engine Strategies Latino Inaugural Edition, 2006.
• Bäcker,
• Bäcker, A. and Hipp, J. An activity-dependent model for the development and maintenance of the mammalian ORN-glomerular mapping. European Symposium for Insect Taste and Olfaction (ESITO), 2005.
• Bäcker,
• Bäcker, A. The mathematics of understanding: A graph-theoretical study of meaning and the self-referential nature of knowledge. Institute for Computational and Mathematical Engineering, Stanford University, 2005.
• Bäcker,
• Bäcker, A. Neural coding in an insect brain. Freie Universität Berlin, Germany, 2005.
• Bäcker,
• Bäcker, A. Estimation of the topological dimensionality of a manifold. Center for Neuromorphic Systems Engineering Retreat, Caltech, 2004.
• Bäcker,
• Bäcker, A. Growth and Democratization of Science: Lowering Barriers to Entry. 4th International Conference on University Evaluation and Research Evaluation, China, 2004.
• Bäcker,
• Bäcker, A. The Digital Brain. Cognitive Systems: Human Cognitive Models in System Design, Santa Fe, 2004.
• Bäcker,
• Bäcker, A. The Impact of Sputnik and the Race to the Moon on Innovation. San José Tech Museum of Innovation, 2004.
• Bäcker,
• Bäcker, A. et al. Automated ontology generation. Sandia National Laboratories, 2004.
• European
• European Symposium for Insect Taste and Olfaction, Norway, 2003.
• Lab
• Lab d’Ethologie et Cognition Animale, Université Paul Sabatier-Toulousse III, France, 2003.
• Salk
• Salk Institute for Biological Studies, La Jolla, CA, 2002.
• Rockefeller
• Rockefeller University, NY, NY, 2002.
• Neural
• Neural Information Processing Systems (NIPS) Symposium: Information and Statistical Structure in Spike Trains: How can we calculate what we really want to know?, Whistler, BC, 2001 (unable to attend).
• Sloan
• Sloan Center for Theoretical Neurobiology Annual Meeting, Lake Tahoe, CA, 2001.
• Facultad
• Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Argentina, 2001.
• Sandia
• Sandia National Laboratories, Computational Biology Group, Albuquerque, NM, 2001.
• Celera
• Celera Corporation, Bioinformatics Research Group, Rockville, MD, 2001.
• Sloan
• Sloan Center for Theoretical Neurobiology Annual Meeting, Salk Institute, La Jolla, CA, 1999.
• Hebrew
• Hebrew University of Jerusalem, Eilat, Israel, 1999.
TECHNICAL PUBLICATIONS
• Martin,
• Martin, S.
• Bäcker,
• Bäcker, A. A Future for Scientific Discourse, 2004, http://alexbacker.pbwiki.com/A%20Future%20for%20Scientific%20Publishing.
• Boyack,
• Boyack, K. and Bäcker, A. The growth and memory of science, 2004.
• Hipp,
• Hipp, J. and Bäcker, A. An activity-dependent model for the development and maintenance of the ORN–glomerular mapping. Soc for Neurosci Abstr , 2004.
• Bäcker,
• Bäcker, A.: A Roadmap for Cognition. Whitepaper restricted to internal SNL circulation, 2004.
• Boyack,
• Boyack, K.
• Bäcker,
• Bäcker, A: Gain control and concentration coding in early olfactory circuits. Soc for Neurosci Abstr , 2002.
• Bäcker,
• Bäcker, A.: Pattern recognition in the olfactory system of the locust: Priming, gain control and coding issues. Ph. D. Thesis, California Institute of Technology, 2002.
• Bäcker,
• Bäcker, A.: A novel computerized odor delivery system, 2002.
• Bäcker,
• Bäcker, A.: The olfactory system exhibits invariance to volatility, 2002.
• Bäcker,
• Bäcker, A.: The probability that a single underlying binomial distribution yield two given success proportions: An exact hypothesis test, 2002.
• Bäcker,
• Bäcker, A.: Asymmetric sliding-window cross-correlation, 2002.
• Bäcker,
• Bäcker, A., Dean, H. and Caporale, N.: Hedonic valence of odors in the locust, Schistocerca americana, 2002.
• Bäcker,
• Bäcker, A.: Gain control and multiplexing in early olfactory circuits, 2002.
• Bäcker,
• Bäcker, A.: Neuronal priming in early olfactory circuits, 2002.
• Bäcker,
• Bäcker, A.: A simple formula to accurately subtract positivity of cytometric negative controls, 2001.
• Bäcker,
• Bäcker, A.
• Bäcker,
• Bäcker, A.: Inconclusiveness of evidence against memory in visual search, 1999.
• Bäcker,
• Bäcker, A.
• Bäcker,
• Bäcker, A.: Mutator genes and natural selection, 1999.
• MacLeod,
• MacLeod, K., Bäcker, A.
• Bäcker,
• Bäcker, A., MacLeod, K., Wehr, M.
• Bäcker,
• Bäcker, A., Wehr, M.
• Bäcker,
• Bäcker, A
• Bäcker,
• Bäcker, A.: To sleep, perchance to dream. The Harvard Brain I: 50-55, 1994.
• Contributor
• Contributor to Marcos, Y. et al. Two-locus involvement in the association of human leukocyte antigen with the extrahepatic manifestations of autoimmune chronic active hepatitis. Hepatology 19: 1371-1374, 1994.
E-prints available at http://www.its.caltech.edu/~alex/pub.html:
SCHOLARSHIPS
• Data
• Data Mining on Attributed Relationship Graphs. SNL LDRD. With B. Hendrickson and T. Kolda. 2004-2007.
• Building
• Building Cognition on Biology: Self-organization in the brain. SNL LDRD. P.I. 2004-2005.
• Metrics
• Metrics for Multidimensional Objects with Correlated Dimensions, Mathematics, Information and Computational Science, Office of Science, Department of Energy. P.I. 2003-present.
• Seismic
• Seismic Ear, SNL Senior Scientist LDRD grant, with K. Larson, 2003-present.
• Augmented
• Augmented Cognition: Next-Generation Intelligent Systems, SNL Grand Challenge grant, with J. Ganter, P. Reeves and K. Larson, 2002-present. Technical lead for Perception thrust, 2004-present.
• Firefox:
• Firefox: A Brain Machine Interface, DARPA Brain Machine Interface Program grant, with R. Asher and R. Albanese, 2002-present.
• IBM
• IBM Innovation Grant, 2000-present.
• Alfred
• Alfred P. Sloan Fellowship for Theoretical Neurobiology, 1995-2002.
• Grant
• Grant from German-American Academic Council to lead expedition to Kenya to study locust behavior, 1998.
• Fellowship
• Fellowship for the German and American Young Scholar’s 1996/97 Institute (youngest fellow).
• L.P.
• L.P. Markey Fellowship (declined).
• Stanford
• Stanford University CAM fellowship (declined).
• Harvard
• Harvard University research fellowship (declined).
• Columbia
• Columbia University research fellowship (declined).
• Undergraduate
• Undergraduate Research Opportunities Program Scholarship, MIT, 1994-1995.
COMMUNICATION
• Editorial
• Editorial Director of Apeiron Magazine (1990), Co-editor of The Harvard Brain (1995), MIT Biology Undergraduate Student Association News Editor (1994-95), published articles in The Los Angeles Times
LEADERSHIP
• Trustee,
• Trustee, The Children’s Center at Caltech, 2002-2006.
• Vice-President
• Vice-President and Director, Dove Creek Condominium, 2004-2005.
• President
• President of the Graduate Student Body and Chairman of the Board of Directors, Caltech Graduate Student Council, 2000-2001.
• Ideated &
• Ideated & implemented new graduate student recruitment effort for Caltech’s Biology
• Director
• Director of the Graduate Student Council, California Institute of Technology, 1998-2002.
• Elected
• Elected Officer of the Biology Undergraduate Student Association, MIT, 1994-95.
• Elected
• Elected to the Council of the Department of Biology, School of Sciences, University of Buenos Aires, 1994.
TEACHING EXPERIENCE
• Invited
• Invited lecturer, Neurobiology
• Teaching
• Teaching Assistant, Neurophysiology Lab, California Institute of Technology, 1998.
• Teaching
• Teaching Assistant, Neurobiology, California Institute of Technology, 1995-96.
• Teacher
• Teacher in the Center for Talented Youth Program, MIT, 1994 and Tutor in Genetics, MIT, 1994-95.
_uacct = "UA-524523-1";
urchinTracker();
]]>
urchinTracker();
]]>
at
He is the founder of AdaptTechnologies™, an emerging leader in search engine marketing technology which has boosted the reach of Search Engine Marketing campaigns by up to 580% while simultaneously reducing their cost per action by up to 78%, and raised $10M in an oversubscribed round of financing with top-tier Silicon Valley venture capitalits. Alex served as Adapt's first President, CEO and CTO, growing net revenues by more than 200% every quarter of his tenure as CEO, and
With Ram Prayaga, he co-founded Cafe|Think™, which launched Choizi™, an online personalized recommendation service to help users find restaurants of their liking (most of the credit should go to Ram, says Alex). He is also the founder of TreeofBabel, a fledgling nonprofit collaborative web project to map out the genealogy of human languages.
In 2007, Alex founded abinventio™, an inventionfactory with a focus on challenging algorithmic problems, where he has since been practicing parallel entrepreneurship. abinventio has since released:
>
> Whozat™, The People Search Engine™, named one of the 100 hottest start-ups in the world by famed technology blog TechCrunch, a semantic search engine with machine vision that garnered tens of thousands of pageviews within a week of launch;
>
> REPcloud™, a reputation engine which garnered tens of thousands of users within weeks of launching;
>
> HoundWire™, a location-aware community-powered news engine.
>
> PREPROVE™, a statistically sound behavioral way of measuring brand power and testing marketing messages which is orders of magnitude cheaper and more accurate than traditional focus groups and surveys.
>
> TheSEMExperts™, for marketers looking for
In 2008, Alex was appointed by the President to serve in the California Insitute of Technology's Information Sciences and TechnologyBoard of Advisors together with a distinguished cadre of scientists, technologists and philantropists including Carver Mead and Phil Neches.
Watch out for 1OFTHESE.COM™, a better way to buy and sell in the world's most efficient marketplace --signuptobenotifieduponlaunch--, and mynew.TV, a better way to be entertained that's all about you.
He has been an invited speaker at numerousconferences on diverse science and technology topics around the world. Alex is always lookingforworld-classentrepreneurialpeopletohelpbringaboutinnovationthroughtechnology. Alex likes to spend his free time traveling with his family, kitesurfing, filmmaking, or kayaking.
]]>
line
Join my free newsletter with stories, tips and resources on entrepreneurship and science
Recent favorite posts:
]]>
Read about how QLess eliminates standing in line athttp://digg.com/tech_news/Qless_Changes_the_Way_You_Wait_in_Line/
Join my free newsletter with stories, tips and resources on entrepreneurship and science
Recent favorite posts:
]]>
E. B. White
Humor is a deeply ingrained human behavior. It triggers reflex reactions across the globe. Clearly, it evolved for a reason. So what is humor for?
Abstract
]]>
E. B. White
Humor is a deeply ingrained human behavior. It triggers reflex reactions across the globe. Clearly, it evolved for a reason. So what is humor for?
Abstract
]]>
urchinTracker();
var sc_project=1777361;
var sc_invisible=1;
var sc_partition=16;
var sc_security="c84ed9d0";
]]>
Why would Google do this, and punish good CTR words together with bad CTR words? To provide an incentive to advertisers to not advertise on irrelevant words, which cost Google money until their system can determine that their CTR is low (and even then cost space and processing time in the system). Worse, they cost Google in relevance to searchers, which goes against Google's religion.
Up to Search Engine Marketing.
]]>
affirming that we examine, by showing that we look, by writing that we
think.
*** Henri Frederic Amiel ***
Quotations
TheCitizen'sGuidetoSurvivingPoliceEncounters -- Stumbled upon this on YouTube and reminded me of the orientation we got when we got to Caltech. A must see for any resident of the USA.
]]>
rangemax next wireless-n range versus wireless-g, or rangemax next wireless-n range wireless-g comparison), so here are the results of my testing?
From my desk upstairs using Netgear RangeMax NEXT Wireless-N router:
2058 Kb/s download
200 Kb/s upload
60 ms latency
2nd time:
1929 Kb/s down
410 Kb/s up
41 ms latency
With 2Wire HomePortal 1000HG same time same place:
Down: 870 Kb/s
Up: 410 Kb/s
Latency 34 ms
2nd time
Down: 977 Kb/s
Up: 204 Kb/s
Latency: 82 ms
W/ Wifi to D-Link WBR 1310 from my desk upstairs:
Latency 42 ms
2358 down
207 up
2nd time:
54 ms latency
213 Kb/s down
28 kb/s up
Sonic.net says it's supposed to be:
1.5mbps-3.0mbps Download
384kbps-512kbps Upload
W/cable to Netgear RangeMax:
64 ms latency
Down 2447 Kb/s
Up 415 Kb/s
D-Link WBR-1310
w/cable
37 ms latency
Down 1304 Kb/s
Up 414 Kb/s
W/Wifi to Netgear RangeMax from 3 feet away:
59 ms latency
2436 Kb/s down
403 Kb/s up
W/Wifi to D-Link WBR 1310:
89 ms latency
2193 Kb/s down
428 Kb/s up
]]>
Is the range of RangeMax NEXT Wireless-N routers really 10X that of Wireless-G?
What make Facebook so successful
The advantages of SMS over voice calls
The First Law of UI Design: The Principle of Least Action
ARemoteVideoEyeTracker - Bäcker & Peral, 1997
"UA-524523-1";
urchinTracker();
var sc_project=1777361;
var
var
var
]]>
1. Their news feed, or social passive information flow: The ability to find out what the people you care about are doing without them needing to think of you or you think of them.
2. Permission-based sharing:The ability to share media only with those you want in a permission-based environment.
3. Notification-based WWW:The ability to be notified every time a new piece involving you or anybody you care about appears.
4. A social platform for apps: A way for developers to easily expand their user bases along the social network of their users.
6. The world's first email directory --a way to find anyone and send them a message immediately.
This makes me think that if Ihad to put my money on one innovator today, it would be Mark Zuckerberg. Sparkling
Maybe one day I'll expand on these, but this will have to do for now. Breakfast
]]>
his
Maybe one day I'll expand on these, but this will have to do for now. Breakfast is calling.
]]>
Companies can become as successful as Google or Youtube just by getting one important thing right. But Facebook has got least 4 important and revolutionary things right, an impressive feat:
1. Their news feed, or social passive information flow: The ability to find out what the people you care about are doing without them needing to think of you or you think of them.
2. Permission-based sharing: The ability to share media only with those you want in a permission-based environment.
3. Notification-based WWW: The ability to be notified every time a new piece involving you or anybody you care about appears.
4. A social platform for apps: A way for developers to easily expand their user bases along the social network of their users.
This makes me think that if I had to put my money on one innovator today, it would be Mark Zuckerberg. Sparkling four revolutions is no small feat for anyone, much less for someone his age.
Maybe one day I'll expand on these, but this will have to do for now. Breakfast is calling.
]]>
What make Facebook so successful
The advantages of SMS over voice calls
How to convert MXF files from a P2 card to AVI using AVID
Find Nearby Friends with your Cellphone
The First Law of UI Design: The Principle of Least Action
Bäcker
_uacct = "UA-524523-1";
urchinTracker();
]]>
*** Charles H. Duell Patent Commissioner (1899) ***
A few of my public inventions:
Bubble-proof financial instruments
]]>
-Steven Hansen in http://seekingalpha.com/article/116410-misunderstanding-the-great-recession
I venture say that's true not only of large corporations.
]]>
Start-up executives need to understand their product
I come from a family of scientists and entrepreneurs. My great-great-great-great-great-grandfather, Hayum Moos, was Albert Einstein's great-great-grandfather. My grandfather, Ernst Moos, had to quit medical school in Berlin due to Nazi oppression and fled to Argentina, where he started two small companies as a penniless Jewish refugee. My uncle RonBaecker is the founder of several companies, and Professor of Computer Science, Bell Universities Laboratories Chair in Human-Computer Interaction, and founder and Chief Scientist of the Knowledge Media Design Institute at the University of Toronto, was named a Pioneer of Computer Graphics by ACM SIGGRAPH, has been elected to the CHI Academy by ACM SIGCHI, and has been given the Canadian Human Computer Communications Society Achievement Award. My mother, Silvia Moos, is the founder of Centralab, Argentina's most important clinical chemistry lab and a state of the art automated facility, and Klik, Argentina's first mental fitness center. My Dad started the Argentina offices of Heidrick & Struggles and currently heads Korn Ferry South South America.
My own first entrepreneurial venture was making and selling ornamental candles with my brothers in the neighborhood, when we were all children --although my brother Nico was the spirit principally behind the candle factory. It got a little more serious in college, designing, making and selling T-shirts and sweatshirts with the university name and logo; we did very well until both partners left the country to pursue scientific studies, sacrificing the venture to science. In grad school, I took a break from earthly profits and founded the Caltech Filmmaking Club, the Caltech Kiteboard Club, the CNS Journal Club (also with Gabriel Kreiman) and the Da Vinci Club (with Dan Lieberman), all of which got funded and exist to this day. It was during this time that I started writing up some of my first inventions. Initially, I thought established companies such as Google would soon catch up to my ideas, so I did nothing. When several years passed and they had still not come up with any of my inventions, I decided to start Adapt. I have not stopped since. Read TheInventionFactory'sblog for up to date news on my latest ventures.
]]>
- Steven Hansen,http://seekingalpha.com/article/116410-misunderstanding-the-great-recession
I have often criticized VCs for the very same reason: they ask early stage, pre-product start-ups for accurate financial forecasts that are irrelevant due to the lack of data to ensure accuracy. In addition to proving futile, these exercises can distract management from the important job of doing the things that make a difference, such as innovating, designing and building products and services, and selling them.
Up toEntrepreneurship.
]]>
- Steven Hansen,http://seekingalpha.com/article/116410-misunderstanding-the-great-recession
as
]]>
- Steven Hansen, http://seekingalpha.com/article/116410-misunderstanding-the-great-recession
I have often criticized VCs for the very same reason: they ask early stage, pre-product start-ups for accurate financial forecasts that are irrelevant due to the lack of data to ensure accuracy. In addition to proving futile, these exercises can distract management from the important job of doing the things that make a difference, such as innovating and selling.
]]>
Most of all, I think this is a classic tale of how audiences fall in love with the performer, when the most (yet not only, of course) talent belongs to the composer. I dreamed a dream is one of the best songs ever written for a musical, and at least half of what shines through in Susan's performance is, in my opinion, Sir Andrew Lloyd Webber's talent. Susan brought me to the verge of tears, but so did Andrea the schoolgirl, repeatedly --because they were both singing (well) Lloyd Webber's masterpiece.
The same was true with Susan's predecessor in instant Britan's got talent-induced fame, Paul Potts. Paul is an amazing singer, but he also chose as well as Susan did. The song he performed, Nessun Dorma, is one (TBC
If you look at professional pop singers, such as Britney Spears, who sing songs composed by others, you'll find that her best hits (including Oops I did it again and Baby One More Time) a pretty consistently written by the same amazingly talented composer, Max Martin --whom you've surely never heard of. Songs sung by the same famous singer but composed by less talented composers have tended not to become hits. Songs written by the same composer but sung by other, less famous artists (e.g. Hot N Cold by Katy Perry) are hits.
In sum, while Susan's performance was moving, there are a number of amateur singers at least as talented (Andrea Benvenuto, the former schoolgirl, among them) who never garner millions of youtube views or achieve fame. And while that may not diminish your or my expectation to hear her sing again, we may find that when she's no longer the underdog, audiences are no longer giving her a standing ovation and she's not singing some of the best songs ever written, her performance will lose some of the quality that made it heart-warming, to become one worth listening to, but possibly not enough to compel the average listener to forward to more than one friend --the requirement for viral explosion and fame.
Hey, fifteen minutes is all Andy Warhol promised, so while they last, let's enjoy them with Susan.