A few years ago at Infocomm I remember seeing a Samsung demonstration of a digital signage system with front facing cameras to track viewer metrics. Using real-time facial tracking the system was able to very accurately determine from a crowd of viewers looking towards the display how many were male and female. If the majority of viewers were female then the system would switch to ads more likely to appeal to women, and vice versa for men.

Take this one step further… if the system were able to identify individuals using facial recognition then it could tailor ads in real time based on the personal preferences of who it identifies. With current technology, these systems are already capable of reading enough facial biometrics to identify individuals in a crowd. However, they are missing the key component which is the data that can match faces with identity. This is where facebook comes in.

Facebook is one of only a few companies in the world that could provide this information to advertisers. For years, we have been uploading pictures and tagging ourselves and friends, training Facebook’s machine learning algorithms to identify us. When you combine facial recognition technology to what Facebook already knows about us, you get the ultimate ad personalization service. This type of service would be hugely valuable to advertisers as it would allow them to tailor very specific ads to their audiences. The holy grail of advertising.

From this perspective it is clear why the Instagram acquisition was so valuable. Not only does the acquisition give them access to even more picture sharing services from which they can continually improve their facial recognition technology but, maybe even more importantly, it takes a major resource away from potential competitors.

A few years ago along came a project called DD-WRT. From the website:

DD-WRT is a Linux based alternative OpenSource firmware suitable for a great variety of WLAN routers and embedded systems. The main emphasis lies on providing the easiest possible handling while at the same time supporting a great number of functionalities within the framework of the respective hardware platform used.

Basically, DD-WRT is an open source firmware that can overwrite the factory firmware on inexpensive routers to unlock more advanced features and configuration capabilities. The DD-WRT website maintains a database of compatible routers. One of the more popular models is the Linksys WRT54GL available for about $50. Once you’ve chosen a router all you need to do is download the appropriate file, and follow the instructions.

Obviously this new firmware will void the warranty and the factory instructions no longer apply but considering the cost savings it is well worth it.

Here is the system and guidelines I follow to keep my passwords secure, while at the same time not inconveniencing myself.

Always Use Unique Passwords

This is a tough rule to follow because obviously its easier to remember only one password and just use it for everything. However, the consequences for doing this can be devastating. All it takes is one mistake logging in to one of your accounts from a system running keylogging software, falling for a phishing attack, or logging in over an insecure wifi connection to expose your universal password. From there, an attacker may be able to identify other accounts of yours and login with the same credentials. You may never even know it happened until it is too late, and even then you will never know just how many accounts were compromised.

The only way to avoid this domino effect is to use a unique password for every account. And by unique, I mean truly unique so that if one were compromised it wouldn’t be obvious to an attacker that a particular pattern was used to generate the password. It may seem like an impossible task to keep track of every password but I will explain how you can do it with relative ease and convenience.

The Last Password You Will Ever Need to Remember

Obviously, nobody can reliably keep track of every unique password in their head so they must be stored someplace. That someplace would be the password jackpot if it were ever discovered so a spreadsheet named Passwords.xls is probably not the best idea. (Note – even with password protection, office documents can be easily cracked).

KeePass

I prefer a handy utility called KeePass. Its free, open source software for storing encrypted contents. The full feature list is available here but basically it uses 256 bit encryption which is pretty much impossible to crack. Even if someone gets access to the keepass file, they wont be getting in without the right password.

Another good thing about KeePass is that it is widely supported. There are numerous Android and iOS apps that utilize it such as  KeePassDroid and KyPass. Using something like Google Drive or Dropbox, you could share the same file across all your devices and have access anywhere.

LastPass

Another great tool for storing passwords is LastPass. Once you have created an account, the LastPass plugin automatically stores passwords and retrieves them when you visit the same site. It saves you the hassle of having to look up passwords when browsing the web. Just keep in mind that anyone who uses your computer (specficially, your user account on your computer) would have access to your passwords if you stay logged in to LastPass.

Getting a system like this setup only takes a few minutes and once you get into the habit of actually using it for storing passwords you will find its not so inconvenient. And by doing so you will have mitigated a major catastrophe in the event one of your passwords gets stolen.

I can’t remember the last time I was so eager to read an autobiography. When the Steve Jobs book came out I could not wait to start it and once I did I couldn’t put it down.

Its a very well written story about a fascinating person who led an extraordinary life. By now most everyone has learned the juicy bits from the book but what I found most surprising were the character traits that Jobs exhibited throughout his entire life.

His appreciation for design goes back at least to his early college days where he audited a calligraphy class. Design was such an important part of the entire product, not just the outward presentation. Jobs even insisted that the circuits be laid out in a way that was aesthetically pleasing. This level of detail is proof that Jobs commitment to design was more than what was visible on the surface.

But for all the great things he did, he was also quite an ass. He regularly derided people and their work. He screwed over many people that were nothing but loyal to him, including Wozniak. He practically denied the existence of his first daughter while at the same time building a product named after her. If you take away all his accomplishments (that he achieved with some help of course) you get a picture of a master manipulator who was loyal to nobody but himself.

The thing that worries me is that Jobs will go down in history as one of the greatest revolutionaries/visionaries of all time. And while I think he deserves credit for his visions and the quality of the products he produced, I’m not sure that the end justified the means. There is an entire generation that idolize Jobs for what he has done but I hope they realize some of the choices he made were probably no the best ones.

If only there were a historical revolutionary figure whose legacy had such a major impact on the world without the trail of devastation and broken relationships – that would be someone to idolize.

This past weekend I participated in the most demanding endurance event I have ever done. CrossFit South County hosted its first ever 24 in 24 on Friday Oct 7 2011. Starting at 7pm we performed 1 girl WOD every hour for 24 hours.

Here’s the full workout schedule:

I really had no idea how I would hold up coming into the event. I’m only a few months recovered from a shoulder impingement that pretty much made it impossible for me to train for about 6 months. I wasn’t sure if that would come back to haunt me halfway through the competition. And I had never done more than 2 WODs in the same day, let alone 24 so I had no idea what my endurance level was for this kind of thing. I’ve only done a few of the official WODs so most of these were my first attempt. My goal was simply to come out the other end alive. Finishing was all that mattered so in an effort to conserve energy I scaled everything down as much as I could. Plus I’m not yet able to do all the movements so I had to substitute some things. For example, we could substitute 2.5 pull-ups and push-ups for every muscle-up.

But I made it. And it really wasn’t as tough as I thought it would be. Maybe I just psyched myself up too much beforehand. I expected the worst so anything less was a relief. There wasn’t any particular point where it got tough or I considered quitting. As time progressed our stamina decreased but we were ever closer to the end so there was an even trade off of energy and optimism. I just concentrated on getting through each WOD one at a time. After having completed it all and knowing what I know now I probably could have gone harder. Next time I will definitely try to Rx as much as possible.

We weren’t going for stellar times or PRs but we did keep track of our performance. My final tally for all the reps and work performed is:

Not bad for a days work.

3D Stereo Basics

The primary 3D (stereoscopic) video viewing modes are: Active Stereo, Passive Stereo and Auto Stereo.

Active Stereo displays left/right eye images in alternating frames. Shutter glasses actively blank out each eye in alternating sequence. Synchronization is handled with an IR pulse from an emitter connected to the display or 3D source. DLP chips used in many projectors can support active stereo which is partly why it has become the de facto standard for consumer 3D.

Passive Stereo uses two display sources (projectors) to show left/right eye images simultaneously. Polarization filters are used to polarize the light from each projector in opposite directions. The viewer wears polarized glasses to filter the light allowing only the left eye image to pass through the left eye lens and only the right eye image to pass through the right eye lens. Common types of polarization are linear and circular. Circular polarization allows for some head movement, whereas tilting the head off-axis with linear polarized glasses causes a loss of polarization resulting in a double image. Generally speaking, circular polarization is a better way to go.

Passive stereo projection requires both images to be in near perfect alignment. I say “near” because with glasses on its difficult for the viewer to notice any minor misalignment but without glasses the image will appear blurry and will probably not be acceptable for non-3D viewing. It is very difficult to align two projectors without either digital image warping or a very sophisticated mechanical setup. You also need to make sure you use projectors with enough lens shift to completely overlay both images, depending on how close the lenses are.

Auto Stereo uses no glasses and relies on the convergence of left/right eye images at the viewers eyes. This is typically done with lenticular lenses overlaying an LCD display. Currently, the major problem with auto stereo is that it has a very narrow viewing angle and a very low parallax, resulting in a limited viewing range with a weak stereoscopic effect. This technology is quite immature which is why auto stereo is not very prevalent despite the demand for viewing 3D without glasses. For the sake of this discussion I will omit auto stereo, as the current technology has very limited applications.

Now that you know the difference between the two primary viewing modes: Active and Passive, the question is when to use either one. Here are the three primary factors to keep in mind anytime you design a 3D Stereoscopic system:

Audience

Active Stereo is useful for small audiences of trusted individuals. The reason for this is the cost of shutter glasses. Prices are falling but even at about $100 a pair, they aren’t cheap. Even if its for a smaller audience, make sure the viewers being given a pair of $100 glasses can be trusted not to break or steal them.

Passive Stereo is ideal for large audiences or venues where glasses may be easily lost or broken. Decent passive stereo (circular polarized) glasses go for under $20/pr. Or you can get custom, disposable paper polarized glasses even cheaper in bulk.

Environment

Active Stereo displays can be easily retrofitted to just about any existing projection system. A single projector may also be necessary if the installation environment has strict space limitations where only a single projector will fit. For multi-screen displays, all projectors and sources must be genlocked (typically).

Passive Stereo requires screens made specifically for 3D with metallic properties to keep light polarized. This could be as simple as using some brands of metallic paint although professional 3D screens are always recommended.

Because passive projection requires precise image alignment, they are particularly sensitive to any movement including vibrations that could cause the images to become misaligned.

Content

Active Stereo is the only option if you plan to show 3D Blu-Ray content. Most consumer level 3D sources are designed to work with active 3D displays (primarily based on DLP 3D technology). HDMI 1.4 3D specification supports many different 3D formats, but only to one display.

Passive Stereo is typical for high resolution visualization systems, typically from PC sources. Having two separate connections effectively doubles the amount of bandwidth available to each display, making it ideal for very high resolution imaging.

Conclusion

I hope these quick tips help. They were learned through years of toiling and sometimes costly mistakes. For even more useful information check out some of these links:

3D Film
3D In ProAV – Part 1
Planning for 3D

Dean Kamen on A Culture of Distraction

The first half of the book is really good. The first Chapter is about Why Nerds Are Unpopular, which pretty much sets the tone for the rest of the book. The second half of the book reads more like a pitch for Lisp but still offers some useful tidbits about programmers and programming languages.

Overall, I really enjoyed the book. I actually charged through it in 3 days which is pretty fast for me to finish a book. In fact, it has inspired me to start blogging more in the hopes of putting some thoughts out there that others might find useful someday.

All the founders are fairly young, most of which started their startup during or right after school. In fact, most actually dropped out of college to pursue their startup. Most of the companies (if not all) are from Y Combinator batches and most speak about their experience getting accepted. I found this bit of information very informative and helpful in understanding the kind of criteria Y Combinator looks for in the companies they invest in.

I highly recommend this book to anyone thinking of starting a business or just wanting to know what its like inside the mind of a young entrepreneur.

The Epiphany I got from this book was probably something different than what was intended but which is about the idea of building a company specifically for the purpose of selling it. Steve talks about the numerous companies he built and sold during the dotcom boom and bust. When you stop and think about it this way, the company is more than just a vehicle for building and selling products to customers. The company is itself, a product. This idea really opened my eyes to see entrepreneurship not as starting a lifestyle business but as the business of building and selling companies.

Steve is currently a professor of entrepreneurship at Stanford University. He does a podcast which I think is just a narrated version of his blog but I highly recommend it.