rlm@141: <html>
rlm@141: 
rlm@141: <head>
rlm@141: <style type="text/css">
rlm@141: body {
rlm@141: font-size:18px;
rlm@141: line-height:1.5em;
rlm@141: margin:0;
rlm@141: padding:3em;
rlm@141: }
rlm@141: 
rlm@141: 
rlm@141: .ideas {
rlm@141: margin-top:4em;
rlm@141: }
rlm@141: 
rlm@141: blockquote {
rlm@141: text-align:center;
rlm@141: font-style:italic;
rlm@141: }
rlm@141: .project {
rlm@141: position:relative;
rlm@141: padding:0.15em;
rlm@141: min-height:4em;
rlm@141: }
rlm@141: 
rlm@141: .project:nth-child(even) {
rlm@141: /*background:#eee;*/
rlm@141: }
rlm@141: 
rlm@141: .project h2 {
rlm@141: position:absolute;
rlm@141: top:0.15em; bottom:0;
rlm@141: width:12em;
rlm@141: margin:0;
rlm@141: padding:0;
rlm@141: text-align:right;
rlm@141: text-transform:capitalize;
rlm@141: font-size:1.2em;
rlm@141: line-height:1.25;
rlm@141: }
rlm@141: 
rlm@141: .project .description {
rlm@141: padding-left:16em;
rlm@141: 
rlm@141: }
rlm@141: 
rlm@141: h1 {
rlm@141: font-size:3em;
rlm@141: line-height:1em;
rlm@141: text-align:center;
rlm@141: text-transform:uppercase;
rlm@141: }
rlm@141: 
rlm@141: .header h1 {
rlm@141: display:inline;
rlm@141: font-size:2em;
rlm@141: text-transform:none;
rlm@141: }
rlm@141: 
rlm@141: 
rlm@141: 
rlm@141: div.header {
rlm@141: 
rlm@141: background:#0F4D92;
rlm@141: color:#fff;
rlm@141: position:absolute;
rlm@141: top:0;
rlm@141: left:0;
rlm@141: right:0;
rlm@141: padding:1em 3em;
rlm@141: }
rlm@141: body {
rlm@141: padding-top:8em;
rlm@141: }
rlm@141: 
rlm@141: h1.title:before {
rlm@141: content:"\2022";
rlm@141: color:#d90;
rlm@141: border:0.1em double #d90;
rlm@141: border:0.1em solid #d90;
rlm@141: 
rlm@141: display:block;
rlm@141: width:1em;
rlm@141: height:1em;
rlm@141: border-radius:100%;
rlm@141: margin:0 auto;
rlm@141: margin-bottom:0.5em;
rlm@141: 
rlm@141: }
rlm@141: </style>
rlm@141: </head>
rlm@141: 
rlm@141: <body>
rlm@141: <div class="header">
rlm@141: <h1><em>aurellem</em>.org</h1>
rlm@141: </div>
rlm@141: 
rlm@141: 
rlm@141: <h1 class="title">Ideas</h1>
rlm@141: <p>
rlm@141: This is a list of all the ideas I've had that I felt like writing down
rlm@141: for the past ~ 8 years. Some of them could be practical inventions and
rlm@141: are "just" waiting the that 95% perspiration to bring them to
rlm@141: fruition, some are ideas for science fiction, and some are simple
rlm@141: observations. Some are really only for my own personal notes and are
rlm@141: not meant to be comprehensible. They are arranged roughly in reverse
rlm@141: chronological order, with the most recent ideas at the top of the
rlm@141: list. The ones at the bottom of the list are heavily influenced by my
rlm@141: time at MIT.
rlm@141: </p>
rlm@141: <p>
rlm@141: If you find some of these interesting and would like to collaborate on
rlm@141: them with me or discuss them in more detail, I'd love to hear from
rlm@141: you. You can email me at <a href="mailto:ideas@aurellem.org">ideas@aurellem.org</a>.
rlm@141: </p>
rlm@141: <p>
rlm@141: If you want to use one of these ideas as your own and run with it,
rlm@141: please feel free. I'd love
rlm@141: to <a href="mailto:ideas@aurellem.org">hear about it</a> if you do.
rlm@141: </p>
rlm@141: <blockquote>
rlm@141: 
rlm@141: <p>There's no end to what a man can accomplish if he doesn't care about
rlm@141: getting credit.
rlm@141: </p>
rlm@141: </blockquote>
rlm@141: 
rlm@141: <hr/>
rlm@141: 
rlm@141: <div class="ideas">
rlm@141: <div class="project"><h2>the great computing slow-down</h2><div class="description">In general, our computers are
rlm@141:      getting faster and faster. However, eventually our brains will be
rlm@141:      made of the same stuff our computers are made of! This has very
rlm@141:      interesting consequences &ndash; I can add 2+2 and get four in about a
rlm@141:      second. Since my neurons actually work at around 10-60 hertz in
rlm@141:      parallel, this means that it takes me around 10-30 operations to
rlm@141:      do this addition. That's actually not bad in terms of computing
rlm@141:      time. If my neurons were as fast as the latest transitors, then
rlm@141:      most calculators would be SLOWER than me at adding numbers. Only
rlm@141:      the newest, most optimized calculators would be faster, and then
rlm@141:      only about 10 times faster! This means that once we begin to
rlm@141:      think at the speed of our technology, that technology will
rlm@141:      suddenly seem pitifully slow in comparison to how it seems
rlm@141:      now. And no amount of technical progress will remedy it, because
rlm@141:      that same progress will also make us all think faster. We'll
rlm@141:      either have to settle with living in "slow time" to do some
rlm@141:      computations, or learn to make smarter hardware with special
rlm@141:      optimizations. But this is actually really hard, because we'll be
rlm@141:      working with machines that will appear to us about as fast as
rlm@141:      MECHANICAL computers. So, in the future, all the cool parties
rlm@141:      will be in cyperspace at vastly accelerated speeds compared to
rlm@141:      how we exist now. But at these parties, the computers will SUCK!
rlm@141:      Of course, this is one of the few things that can save us from AI
rlm@141:      risk, because those AI's won't seem so scary when the're build
rlm@141:      out of rickety mechanical parts form our perspective. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>unitary reverse evolution of chaos+minds</h2><div class="description">Chaotic systems diverge
rlm@141:      exponentially in state space. Do you get anything interesting
rlm@141:      when part of the physical system associated with the chaotic
rlm@141:      system is a object that performs some sort of computation? Is it
rlm@141:      possible for the computational system to play a
rlm@141:      percision-enabling role in determining the final/initial
rlm@141:      conditions of the chaotic system, just by tracing out thoughts in
rlm@141:      its decision paths? This is probably too vague of an idea right
rlm@141:      now, I just wanted to write it down.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>microwave time</h2><div class="description">the cooking time you enter on most microwaves is
rlm@141:      insane. It's expressed in what I call a "hybrid base", a
rlm@141:      combination of base 10 and base 60. You can get absurd things
rlm@141:      like 100 &lt; 61, and 120 == 80! I wonder if these hybrid base
rlm@141:      systems could be very useful for some purposes!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>three-eyes</h2><div class="description">if you had three eyes, would you still draw cubes like
rlm@141:      we currently draw them? Or would all 2D-representations of 3D
rlm@141:      space always look hopelessly fake?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>visual taste/smell assay</h2><div class="description">get a grid of bacteria, each expressing
rlm@141:      a human taste/smell receptor linked to some sort of fluorscent
rlm@141:      activity or ion pump. Use a camera / electrical grid to transduce
rlm@141:      the smell / taste signal into bits!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>carabiner mushroom lock</h2><div class="description">you can take a trapazodial carabiner and
rlm@141:      make it so that a chain link is caught between the wide end of
rlm@141:      the carabiner and another chain attached to the carabiner. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>children's tool shop</h2><div class="description">I think that kids should be provided with
rlm@141:      tool shops &ndash; these would be nice sheds with a good collection of
rlm@141:      tools to do various things &ndash; circuit components and soldering
rlm@141:      irons, wires, a small lathe, drill press, belt sander, a
rlm@141:      centrifuge, microscope, and telescope, etc. The idea is that the
rlm@141:      kid can now think, "I could use X to do this thing that I'm
rlm@141:      thinking about" &ndash; the building becomes an extension of the kid's
rlm@141:      body &amp; mind.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>fluid display</h2><div class="description">like the previous idea about matching refractances
rlm@141:                    between glass and liquid, except you make a lot of
rlm@141:                    switchable glass tubes in various patterns in the
rlm@141:                    glass, and actively pump colored liquid through the
rlm@141:                    tubes (the tubes have glass-like fluid in them by
rlm@141:                    default.) The result is that you can cause the
rlm@141:                    tubes to appear and dissappear, and vary their
rlm@141:                    colors as well!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>immunoincompatibility</h2><div class="description">take the human genome, and refactor it so
rlm@141:      that it doesn't use a particular codon at all. Then remove the
rlm@141:      support from our ribosomes for that codon. What does this do for
rlm@141:      us? It makes us immune to almost all viruses!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>life cycle</h2><div class="description">it's called a cycle, right? So, the thing that repeats
rlm@141:                 itself over and over, right? Not much of a cycle if
rlm@141:                 you don't come back after you die, if you ask me!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>car with no blind spots</h2><div class="description">use some cameras in the back of the car
rlm@141:      to augment the rear-view mirror so that you never have to turn
rlm@141:      around in order to lane change.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>partial cell death</h2><div class="description">you freeze a set of cells using some cryo
rlm@141:      protocol and 60% survive. How can this be explained? It seems to
rlm@141:      me that if the cells are the same, and the conditions
rlm@141:      homogoneous, then all the cells should either die or
rlm@141:      live. However, suppose that there is a metabolic cycle that needs
rlm@141:      to be in a certain phase for the cell to survive. If the cells
rlm@141:      are asynchronous, then you might end up with some cells dying
rlm@141:      because there were in the wrong part of their cycle. This implies
rlm@141:      that you might be able to cryoprotect cells by causing them to
rlm@141:      enter a certain metabolic mode before freezing.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>cryonics color appeal</h2><div class="description">perfusate used by cryonics companies should
rlm@141:      have red food coloring in it. It's just a nice touch so that the
rlm@141:      cryonics patient looks more life-like than with clear CPAs, and
rlm@141:      hopefully might get treated with more respect.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>paramagnetic CPA</h2><div class="description">you take a CPA that can be influenced by
rlm@141:      magnetic fields so that its degrees of freedom are limited. Then,
rlm@141:      you release the field, instantaly increasing the size of the
rlm@141:      state space of the system and dramatically decreasing the
rlm@141:      temperature enough to plunge the system past homogenous
rlm@141:      nucleation temperature and directly to the glass transition
rlm@141:      temperature, creating a doubly unstable glass at much lower CPA
rlm@141:      concentrations than possible at conventional CPA concentrations. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>room temp noodles</h2><div class="description">how does the physics of cooking noodles work?
rlm@141:      Could you use a vacuum instead of heat to force water into the
rlm@141:      noodle?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>personal carbon offset</h2><div class="description">feel bad about contribuiting to global
rlm@141:      warming by using electricity / driving a car? Forget trying to
rlm@141:      "conserve" or "minimize your carbon footprint". Follow the
rlm@141:      Platinum rule &ndash; make the world BETTER off than you found it!
rlm@141:      This would be a small, self contained system that sucks C02 out
rlm@141:      of the air. It uses electricity, but it's so efficient at
rlm@141:      removing CO2 that it more than offsets the CO2 produced by even a
rlm@141:      coal plant to produce that electricity. This way, you can still
rlm@141:      drive even a gas guzzler, but have a net negative carbon
rlm@141:      footprint! Maybe something cool could be done with the carbon as
rlm@141:      well. Use as much electricity as you want, but negate the damage
rlm@141:      to the enviroment with more technology. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>undoing spermogenesis</h2><div class="description">with enough sperm, you can derive the
rlm@141:      donor's entire genome. You gain more confidence in the alleles
rlm@141:      for a particular gene the more sperm you have. Each additional
rlm@141:      sperm gives you the same sort of information you'd get flipping a
rlm@141:      coin and trying to decide whether the coin is H/T of H/H. Is
rlm@141:      there enough sperm in the the average load for you to be as
rlm@141:      confident as mitosis?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>mars life</h2><div class="description">we could engineer life that could survive on mars
rlm@141:                (probably some non-vascular photosynthetic
rlm@141:                poikilohydric creature like a lichen) by taking an
rlm@141:                extremophile from Antarctica and evolving it in
rlm@141:                increasingly Martian conditions. This could be an easy
rlm@141:                start to a terraforming process. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>problem with Aubrey de Grey's ideas</h2><div class="description">Aubrey de Grey says that we
rlm@141:      might be able to live forever by continually repairing our bodies
rlm@141:      at the cellular level &ndash; he details 7 different mechanisms of
rlm@141:      damage and says that if all of them are dealt with <i>together</i>
rlm@141:      that it would stop aging. (You can't miss even one because
rlm@141:      they're all fatal.)  However, it doesn't take into account that
rlm@141:      we are also beings of information and that there is a very real
rlm@141:      software component to our existence. Even if our biological
rlm@141:      chassies can be maintained forever, I think it is unlikely that
rlm@141:      our minds will operate well far outside of the design constraints
rlm@141:      that we've evolved to handle. Say I programmed a webserver with
rlm@141:      the express goal of it being able to serve webpages for month on
rlm@141:      some stock server. I'll do fairly rigorous testing to make sure
rlm@141:      that it can handle the expected load then then some. Now say that
rlm@141:      you want to keep a particular instance of this webserver running
rlm@141:      indefinitely. (The program instance is like your mind and the
rlm@141:      computer it's running on is like your body). You might very well
rlm@141:      be able to keep the physical computer infrastructure running for
rlm@141:      forever by replacing hard drives / ram / CPUs, etc. However,
rlm@141:      since I designed the webserver to work for a month, it probably
rlm@141:      has memory leaks, rare stochastic bugs, or other built in limits
rlm@141:      / constraints (think log files or some date rollover shenanigans)
rlm@141:      that will ultimately kill the webserver even with eternally
rlm@141:      perfect hardware. Do you really expect that a webserver
rlm@141:      engineered to work for 1 month will run for 10 years without
rlm@141:      catastrophically crashing? Not even Apache can do this! In fact,
rlm@141:      if I put in the extreme effort to make it that robust, I've
rlm@141:      wasted time that I could have spent on other projects by pursuing
rlm@141:      an unnecessary engineering goal. Likewise, human minds have only
rlm@141:      ever run for at most 122 years before they are destroyed due to
rlm@141:      hardware degradation. Fixing the hardware doesn't change any
rlm@141:      software bugs that are almost certainly present in the human
rlm@141:      mind. Think of all the pathological things that can go wrong with
rlm@141:      a webserver, multiply it by a million, and that likely how
rlm@141:      evolution has designed our minds. For example, consider memory :
rlm@141:      why should you expect that we have evolved the ability to
rlm@141:      coherently organize memories past say 150 years? There's been
rlm@141:      absolutely no selective pressure for this ability, so you can bet
rlm@141:      that if there's any fitness to be gained from not having
rlm@141:      unlimited memory potential (such as better metabolic efficiency),
rlm@141:      we have it! You might think that maybe we would just forget
rlm@141:      things the same way that we sort of forget things that happen
rlm@141:      earlier in our lives, but complicated information processing
rlm@141:      systems don't have to fail gracefully when they're pushed far
rlm@141:      past their design constraints. A 150 year old person is just as
rlm@141:      likely to suffer a catastrophic psychosis due to software
rlm@141:      limitations associated with memory as he is to do something with
rlm@141:      all those memories we might consider reasonable. More likely, in
rlm@141:      fact, since there are so very many ways for a complicated
rlm@141:      software system to break and so few ways for it to run
rlm@141:      successfully. Therefore, I think Aubrey de Grey's "hardware-only"
rlm@141:      approach is missing a very important component of longevity
rlm@141:      science, and any successful effort to make people live orders of
rlm@141:      magnitude longer than they do naturally will need to deal with
rlm@141:      people's software as well as their hardware.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>validating neurocryopreservation</h2><div class="description">Problem : you want to test
rlm@141:      whether a brain is functionally preserved through vitrification,
rlm@141:      but you don't want to figure out how to preserve all the other
rlm@141:      organs in the animal. It might be possible to keep the rest of
rlm@141:      the body at almost 0C and vitrify just the head for only a few
rlm@141:      minutes. Induce hypothermia, then separate out the head's blood
rlm@141:      supply from the rest of the body, then just cryoptotect and
rlm@141:      vitrify the head. Might need some sort of thermal guard to keep
rlm@141:      the outer head / neck from becoming too cold. You leave the
rlm@141:      spinal cord intact! Then you devitrify to 0C, remove
rlm@141:      cryoprotectant, and then reattach the blood supply. You can
rlm@141:      determine brain preservation using behavioral assays!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>freezing water purifier</h2><div class="description">you slowly freeze water, but also run
rlm@141:      liquid water over the frozen mass. This takes away basically all
rlm@141:      impurities and creates "washed ice" then you melt the ice. Maybe
rlm@141:      you could re-use the heat from creating the ice to melt the ice?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>ultra strength</h2><div class="description">allow a person to visualize their muscle
rlm@141:                     recruitment patterns. Give them adrenaline and let
rlm@141:                     them feel what it's like to have the normal limits
rlm@141:                     removed. See if they can replicate the effects.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>phone names</h2><div class="description">make a PX record for domain names that's like the MX
rlm@141:                  record, except that it is a phone number instead of
rlm@141:                  an IP address. That way, you can use the domain name
rlm@141:                  registration system to provide names for phone
rlm@141:                  numbers. Then, as long as you control the domain, you
rlm@141:                  can point people to your current phone number by
rlm@141:                  updating that record. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>edible flowers</h2><div class="description">Edible white flowers that you put in a colored
rlm@141:                     solution with flavor. When the flower turns the
rlm@141:                     right color, it is also flavored and ready to eat! 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>suicide cryonics</h2><div class="description">according to <a href="http://www.nytimes.com/2008/07/06/magazine/06suicide-t.html">this</a>, people who commit impulsive
rlm@141:      suicides have a newfound sense of the importance of life. Perhaps
rlm@141:      they are good cryonics targets.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>lead bone</h2><div class="description">Could you fill in all the empty spaces in a bone with
rlm@141:                lead? Might be cool.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>the quest for life </h2><div class="description">Every stupid story has the "immortal who
rlm@141:      wants to become mortal" or some other such idea. I want to story
rlm@141:      where the protagonist loses their immortality and feels <i>angry</i>
rlm@141:      and ashamed about losing something that's so absolutely crucial
rlm@141:      to their identity. A reverse of "death makes life worth living",
rlm@141:      they feel that living forever is what makes life worth
rlm@141:      living. Now they've "lost their sunrise" or their "connection to
rlm@141:      the timeless universe" or something. So they go on a quest to get
rlm@141:      it back, learning about themselves along the way, and regaining
rlm@141:      the precious thing they lost in the beginning.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>world-map</h2><div class="description">take a small table and paint the continents in
rlm@141:                toothpaste on the table. Make a slightly raised barrier
rlm@141:                around the table. Slowly pour water onto the table, and
rlm@141:                it will form the oceans!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>stage magic rituals</h2><div class="description">rituals should incorporate elements of stage
rlm@141:      magic. Foe example in Teller's tempest, they have a scene where
rlm@141:      they levitate a crown in front of someone, then put it on his
rlm@141:      head. They also have a wedding ceremony where they levitated the
rlm@141:      bride as well. Actual weddings and other ceremonies should
rlm@141:      incorporate stage magic as an enhancement. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>isotope time dilation</h2><div class="description">use a cyclotron to speed up rare isotopes
rlm@141:      developed in nuclear fusion experiments. The relativistic time
rlm@141:      dilation will stop the isotopes from decaying, and allow time to
rlm@141:      study them. This is based on radioactive isotopes that fall
rlm@141:      through the earth's atmosphere that take hundreds of times
rlm@141:      longer to decay than normal.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>marsupial stimulation</h2><div class="description">You take a freshly pouched marsupial baby,
rlm@141:      and show it videos and other interactive things while it matures
rlm@141:      in the pouch. What mental effects would this have?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>dynamic re-keying</h2><div class="description">Some older ways of tuning instruments sound
rlm@141:      better, but we use the even-tempered scale today because it makes
rlm@141:      it easier to switch keys. With electronic music, why not make
rlm@141:      key-annotations and dynamically re-tune the piece to sound good
rlm@141:      in the current key? Could be done as a midi+annotation -&gt; midi
rlm@141:      compiler for experimentation.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>death always implies damage</h2><div class="description">is is possible for a corpse to differ
rlm@141:      from a living person only in the fact that one is dead and the
rlm@141:      other is alive? NO! A corpse must always have some sort of
rlm@141:      molecular damage which causes the loss of function!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>inner eye</h2><div class="description">Surgically install a bunch of tiny cameras inside a
rlm@141:                person. Then, you can activate them all and get a
rlm@141:                picture of your internal organs for diagnostic
rlm@141:                purposes. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>chaos rails</h2><div class="description">should make a visualization of the homoclinic tangle,
rlm@141:                  it's truly beautiful.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>context gobbler</h2><div class="description">this would be in "inside-out macro" that takes
rlm@141:      the context (like you use for things like error, continuations,
rlm@141:      and friends) and transforms it to something else. Maybe useful?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>cryonics middle ages</h2><div class="description">some people say that cryonics is an
rlm@141:      experiment and that it is foolish to wait until we have revived a
rlm@141:      human. There is a middle ground where the procedure has a dismal
rlm@141:      success rate on humans, say 1 in 20, so that you'd be a fool to
rlm@141:      try revival. Nonetheless, this very risky procedure could be the
rlm@141:      legal proof of concept needed to create a new class of life
rlm@141:      between "living" and "dead": "stasis".
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>philosophy of the mirror</h2><div class="description">neat thought experiment &ndash; if you take a
rlm@141:      mirror of someone by actually reversing a person's chirality
rlm@141:      molecule by molecule, then will the only be able to read mirror
rlm@141:      writing? The answer is yes, by analogy to a purely mechanical
rlm@141:      scan-tron device. This is one of the only interesting transforms I
rlm@141:      know that can take a human brain and change it in subtle,
rlm@141:      non-destructive ways. It's also an argument against dualism.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>biosphere in a bottle</h2><div class="description">There are around 15 million species. 15
rlm@141:      million stem cells will fill only a tiny size, far less than a cubic
rlm@141:      inch. Preserve a single cell from every species on earth in this
rlm@141:      small space, and you will have a record of our current biosphere
rlm@141:      that can be protected. "Hold the genetic data of all species in
rlm@141:      your hand!"
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>chaos lock</h2><div class="description">The "arrow of time" points in the direction of
rlm@141:                 increasing entropy. The time evolution of chaotic
rlm@141:                 systems depend exquisitely on their initial state. If
rlm@141:                 you take a measurement of a chaotic system at any
rlm@141:                 given point of time, you can evolve that system
rlm@141:                 backwards or forwards based on your measurement. So
rlm@141:                 let's say you start the chaotic system in a VERY low
rlm@141:                 entropy state, then let it run for a while, then take
rlm@141:                 a measurement with some uncertainty. Your
rlm@141:                 measurement is pretty good, but obviously not
rlm@141:                 PERFECT. If you evolve the chaotic system back in
rlm@141:                 time, then you will see that you don't really reach
rlm@141:                 a state with low entropy an hour before (the entropy
rlm@141:                 is easy to measure with surrogates like alignment,
rlm@141:                 etc). So use this technique to SEARCH for a more
rlm@141:                 accurate measurement! This potentially can give you
rlm@141:                 many more orders of magnitude than you could get alone
rlm@141:                 just using an instrument. Sometimes it will give you
rlm@141:                 bad results, the the odds of it doing that are
rlm@141:                 infinitesimal, and you can just measure a couple of
rlm@141:                 times. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>cryo-evolution</h2><div class="description">perhaps there would be a way to rapidly evolve a
rlm@141:                     symbiotic bacterial organism that could protect
rlm@141:                     human tissues from freezing damage.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>suicide parasite</h2><div class="description">sometimes, people kill themselves for no good
rlm@141:      reason. We often explain this with things like "hidden
rlm@141:      depression" or we say that they had something like chronic jaw or
rlm@141:      back pain. I think that smells of rationalization. I don't buy
rlm@141:      it. I propose that in many suicide cases there is a disease that
rlm@141:      causes the suicidal behavior. We already know that certain
rlm@141:      parasites have mind-bending properties in other animals, even
rlm@141:      mammals like mice. It's not much of a stretch to imagine a
rlm@141:      parasite that causes suicides in humans. Some problems:
rlm@141: 
rlm@141: <dl><dt>What does the suicide parasite get out of it?</dt>
rlm@141: <dd>This might be answered by the whole thing being a glitch caused by cross-species contamination. Toxoplasma works this way.
rlm@141: </dd>
rlm@141: <dt>What predictions does a disease model make</dt><dd>suicide should
rlm@141:           be more common among people who share a contagion
rlm@141:           vector. There should be suicides that don't make any
rlm@141:           sense : people who weren't really depressed, who had no
rlm@141:           reason to kill themselves. People who have killed themselves
rlm@141:           should have a higher incidence of some unknown parasite in
rlm@141:           their brains. 
rlm@141: 
rlm@141: </dd>
rlm@141: </dl>
rlm@141: </div>
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>domestic insects</h2><div class="description">People should eat more bugs because they're much
rlm@141:      more efficient, so why not do some major domestication research
rlm@141:      to make very appealing bugs? Beetles, in particular, seem to be
rlm@141:      excellent targets for domestication because they have extreme
rlm@141:      levels of genetic malleability.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>birth-clones</h2><div class="description">What if each person was intentionally split at birth
rlm@141:                   into a normal embryo and a few "backup" cells which
rlm@141:                   is then frozen. The backup cells are created just
rlm@141:                   the same way as natural identical twins. The backups
rlm@141:                   can be used to regenerate organs. etc. Also, it
rlm@141:                   would be a good sci fi concept, because you could
rlm@141:                   have a culture where people reward people who were
rlm@141:                   especially awesome are "reborn" from their
rlm@141:                   backups. Imagine having a young Bach every
rlm@141:                   generation, etc.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>pronunciation guide</h2><div class="description">a simple webpage where you type in a word and
rlm@141:      it returns a simple, English sentence describing exactly how to
rlm@141:      pronounce the word. For people who don't want to learn IPA. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>cortex-search</h2><div class="description">use the repertoire of actions learned to limit the
rlm@141:                    search space of possible actions.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>learning to teleport</h2><div class="description">scifi idea, this is a story about a person
rlm@141:      who is struggling with his/her society's ideas about
rlm@141:      teleportation. It's considered a fundamental part of being a
rlm@141:      member of that society (after all, the difference between animals
rlm@141:      and humans is that humans are creatures of pure information while
rlm@141:      animals are burdened with base matter, "that's how you travel the
rlm@141:      stars, etc") Humans are born normally, grow up, and then
rlm@141:      eventually transcend via destructive upload. Analogies to jumping
rlm@141:      off a diving board into a pool (which I simply <i>could not do</i> for
rlm@141:      a long time), etc.     
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>no-float-ice</h2><div class="description">cup that has cross beams at the bottom where ice
rlm@141:                   forms. Then when you drink liquid from the glass,
rlm@141:                   the ice stays at the bottom and doesn't hit your
rlm@141:                   lips. For bars and fancy things.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>bitcoins for immigrants</h2><div class="description">A common case with Mexican immigrants
rlm@141:      (illegal or not) is that they want to send money they've earned
rlm@141:      in the US back to their families in Mexico. They currently do this
rlm@141:      through things like Money Gram or Western Union, and they get
rlm@141:      fleeced in the process with fees. Bitcoin could greatly reduce
rlm@141:      the cost of sending money from America to Mexico, but I don't
rlm@141:      believe that it's currently used for that among Mexican
rlm@141:      immigrants currently due to lack of knowledge. I bet you could
rlm@141:      set up physical locations like those obnoxious Western Union huts
rlm@141:      in places like Texas, Arizona, etc, and greatly undercut
rlm@141:      them. Or, perhaps some educational seminars about bitcoin might
rlm@141:      be in order. There's some money to be made there because there is
rlm@141:      great demand, and it's a good thing to boot!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>reverse eye-tracking</h2><div class="description">A painting that is actually a digital screen
rlm@141:      with a camera. It records people's eye tracks permanently. It's
rlm@141:      "artistic" because paintings are normally these things that you
rlm@141:      look at without changing, but this one is changed the second you
rlm@141:      look at it, recording where <i>you</i> looked forever for others to
rlm@141:      see. Make it be a painting of a woman and see the trolling as the
rlm@141:      breasts and groin area light up with interest from all the males
rlm@141:      passing by.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>smart toilets</h2><div class="description">Instead of using indirect measures like infrared
rlm@141:                    detectors of the presence of a person, use computer
rlm@141:                    vision to directly measure whether the toilet needs
rlm@141:                    to be flushed. I think a lot of things will end up
rlm@141:                    going this way as we get better computer vision.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>validate chemopreservation</h2><div class="description">chemopreservation is difficult to
rlm@141:      validate because it destroys the functionality of a brain, and
rlm@141:      brain simulation will take a long time to mature as a
rlm@141:      technology. However, one very powerful way to validate
rlm@141:      chemopreservation would be to have a person/animal learn
rlm@141:      something with high complexity such as a number or the solution
rlm@141:      to a maze, or a flashbulb memory. Then you preserve their brain
rlm@141:      chemically, slice it up, and read <i>that specific memory</i> from the
rlm@141:      detailed brain scan. Much more difficult, but much more doable.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>candy screw</h2><div class="description">edible candy screw with candy nuts that you can screw
rlm@141:                  as well. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>better bibliography</h2><div class="description">when writing a thesis or paper, have the
rlm@141:      bibliography not just be an opaque list of resources, but have it
rlm@141:      be a list of <i>summaries</i> and <i>qualities</i> that each paper has in
rlm@141:      the context of the paper being written. When examining a
rlm@141:      bibliography, I want to know if reading the papers in the
rlm@141:      bibliography are worth my time, and I also am probably also
rlm@141:      interested in exactly the things that are being discussed in the
rlm@141:      paper I'm reading. The bibliography is the perfect place to
rlm@141:      provide information about the referenced papers from the
rlm@141:      author's perspective. I will use this biographic form in my own
rlm@141:      thesis. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>digital inter-library loan</h2><div class="description">libraries at universities already do
rlm@141:      inter-library loans for books, so why not do the same for access
rlm@141:      to stupid paywalled digital papers? All the universities could
rlm@141:      allow access to articles for registered students to all the files
rlm@141:      available through any participating university. This could be
rlm@141:      achieved by sending requests through proxies at participating
rlm@141:      universities. Each university would decide who at the university
rlm@141:      can access the proxy network. Access to the proxy network could
rlm@141:      be made easy using something like <a href="http://libx.org/">http://libx.org/</a>.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>chess visual</h2><div class="description">to show the vast size of the game trees considered
rlm@141:                   by computers, show two people playing chess in a
rlm@141:                   void. They are floating in space, and there is a
rlm@141:                   simple chess board between them. Then, as they play,
rlm@141:                   the game tree's they are considering are drawn
rlm@141:                   behind him. The root of the tree starts centered in
rlm@141:                   their heads or whatever they use to think, and the
rlm@141:                   tree grows out from behind, never crossing the
rlm@141:                   dividing plane between the two players. Each
rlm@141:                   player's tree is a different color. As they grow,
rlm@141:                   there are animations for pruning, etc. Eventually,
rlm@141:                   they look like the hemispheres of a brain, wings,
rlm@141:                   etc. A human's tree might occasionally have a long
rlm@141:                   chain, while the computer tree would be more
rlm@141:                   uniform. You could compare deep blue and a modern
rlm@141:                   laptop. Use actual data when fighting two computers! 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>time verification</h2><div class="description">some standard way to verify that some piece of
rlm@141:      data was recorded at a specific time. Might involve a time
rlm@141:      server, a key for each time period, something like that.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>tamper proof gold bars</h2><div class="description"><a href="http://www.tungsten-alloy.com/gold-plated-tungsten-alloy-bar.html">this site</a> offers gold plated tungsten bars
rlm@141:      as "novelty" items. One reason to prefer coins is because they
rlm@141:      are much harder to counterfeit because there is less surface area
rlm@141:      to mass ratio. However, gold bars are still a great design
rlm@141:      because they can hold a lot of value in a small space. A gold bar
rlm@141:      could be given the same protections (and more) that gold coins
rlm@141:      have to offer by changing it into a "gold book", which would have
rlm@141:      hundreds of "pages" of gold bound together. This could be
rlm@141:      implemented with multiple steel rods going through the book which
rlm@141:      can be removed, or some more classier mechanism for holding the
rlm@141:      pages. The point is that the bar can be EASILY subdivided (and
rlm@141:      people would perform this test before buying), thus guaranteeing
rlm@141:      it's authenticity.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>aurellem shirt</h2><div class="description">I should make an aurellem star symbol tee-shirt. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>touch vision</h2><div class="description">inspired by GelSight, I want to reexamine cortex and
rlm@141:                   see if I could implement touch as a very low range
rlm@141:                   form of vision.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>high school science</h2><div class="description">this is a lesson in scientific ethics. The
rlm@141:      goal is to calculate <i>g</i>, the local gravitational
rlm@141:      acceleration. The students are told that the textbook says it's
rlm@141:      <i>exactly</i> 9.81 before they start the experiment. See how they
rlm@141:      doctor their results to get closer to the textbook value. It's
rlm@141:      neat because for any given school, <i>g</i> is probably <b>not</b> exactly
rlm@141:      equal to 9.81, because that is just an average!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>opencourseware subtitles</h2><div class="description">there are ladies who type up lectures
rlm@141:      while they are being given. These recordings should be kept and
rlm@141:      given to OCW for subtitles. If the timestamps of keys are
rlm@141:      recorded, then it is easy to make subtitles.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>screen locking timing</h2><div class="description">you use your computer camera to see if you
rlm@141:      are sitting in front of the computer. If you are, then the screen
rlm@141:      will never lock. If you are, then the screen will lock with a
rlm@141:      30-40 second timeout. It's an extension of using inactivity to
rlm@141:      initiate the countdown, just with more information.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>mirror toilet</h2><div class="description">a toilet with a square basin made or mirror instead
rlm@141:                    or porcelain. That way, you can see how good of a
rlm@141:                    wipe job you have done / watch how your excretion
rlm@141:                    system works.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>test dummies</h2><div class="description">why don't we clone anencephalic humans and use then to
rlm@141:                   test <i>in vivo</i> human organ systems and drugs? It
rlm@141:                   would be ethical as long as there are women who are
rlm@141:                   willing to host the clones, and it would be a
rlm@141:                   tremendous resource for studying the human body. I
rlm@141:                   see nothing wrong with it morally, since no one is
rlm@141:                   suffering, and it stands to save many lives throughout
rlm@141:                   more advanced technology.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>X-ray telepresence</h2><div class="description">given that a doctor is operating on a patient
rlm@141:      via telepresence, one cool things you can do is shine X-rays into
rlm@141:      the patient to view the insides during real time. If the system
rlm@141:      was coupled with a Bayesian model of the layout of the structure,
rlm@141:      and the x-rays were only fired whenever the uncertainty of the
rlm@141:      model reached a certain threshold, then the radiation damage
rlm@141:      and surgery risk could be minimized.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>superfluid vascular system</h2><div class="description">I wonder what would happen if you
rlm@141:      replaced the blood in a human with a superfluid. What would the
rlm@141:      physical dynamics be? Would the superfluid flow through the
rlm@141:      vasculature, or would it ignore it and travel through the cells,
rlm@141:      or something else entirely. Since superfluids need to be cold to
rlm@141:      retain their superfluidity, how would the dynamics change during
rlm@141:      perfusion of a superfluid, where the fluid gains and looses
rlm@141:      superfluidity as it goes deeper into the body and is cooled by
rlm@141:      superfluid from upstream. In summary there are two things to
rlm@141:      simulate 1.) replace all blood in human with superfluid
rlm@141:      instantly. 2.) perfuse superfluid into human.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>projective guessing</h2><div class="description">I think that we read and see things by
rlm@141:      making a really good guess about what we're expecting to see,
rlm@141:      and then searching for our guess in what we see. If it really
rlm@141:      doesn't match, then we start to make more guesses / analyze the
rlm@141:      image from first principles, but most stuff is projective
rlm@141:      guessing.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Intestinal flora maintenance</h2><div class="description">why not inoculate babies at birth
rlm@141:      with "ideal" gut flora instead of whatever bullshit they
rlm@141:      naturally get, thus giving them optimal digestive/nutrient
rlm@141:      extraction capabilities. Might also be able to make their farts
rlm@141:      not stink for life, too. MORE IMPORTANTLY, might help to
rlm@141:      preventatively stop some forms of <i>colic</i>, which affects 1 in 5
rlm@141:      babies and causes constant screaming and pain for about 5 weeks.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>server culture : mirrors</h2><div class="description">make a distributed system where people
rlm@141:      can mirror the websites of people they like &ndash; essentially cover
rlm@141:      the server costs of favored websites. This could make popular
rlm@141:      websites run at no cost. The system would require that the
rlm@141:      mirrored content be the same as the official source. Sort of like
rlm@141:      bit-torrent for websites.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>map programming</h2><div class="description">one problem with functional programming is that
rlm@141:      in order to remain functional, you have to pass up arguments up
rlm@141:      into each calling function to get the full range of behavior
rlm@141:      from the lower level functions. Normally people come to a
rlm@141:      compromise involving abstraction and sparing use of dynamic
rlm@141:      variables to configure runtime behavior. What would be the
rlm@141:      advantages of making a programming language where every function
rlm@141:      receives one argument, a map, which contains all the symbol
rlm@141:      bindings it would ever need? This map is passed on to all
rlm@141:      subordinate functions. This way, you could replace functions on
rlm@141:      the fly, and arrange for there to be sensible defaults,
rlm@141:      etc. Might cause more harm than good but is an interesting idea.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>rest nest</h2><div class="description">a small EEG device you would attach to your head when
rlm@141:                you go to sleep at night. ML algorithms would determine
rlm@141:                your particular sleep cycles. This would mostly be an
rlm@141:                alarm clock that you could give a time range, say
rlm@141:                7:00AM - 7:15AM, and it would wake you up during an
rlm@141:                ideal time corresponding to then end of one of your 90
rlm@141:                min sleep cycles. You would feel much more rested upon
rlm@141:                waking up, and would wake up faster. There might be
rlm@141:                some other uses for the EEG data as well.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>image compression</h2><div class="description">use a library like gimp or opencv to process an
rlm@141:      image to make it have less entropy, then store the reverse of
rlm@141:      those operations along with the compressed simpler image as a
rlm@141:      super-compressed image file (possibly accepting some
rlm@141:      losses). Trades file size for decompression time, and allows one
rlm@141:      to cheat by using information in gimp/opencv to compress the
rlm@141:      image. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>fixed cryopreservation</h2><div class="description">why not use a fixative to buy enough time
rlm@141:      to ramp up cryoprotectants to an acceptable level at room
rlm@141:      temperature? Then, the whole system can be rapidly cooled and
rlm@141:      vitrified. This method "severs the biological link" in that the
rlm@141:      fixatives are highly toxic, but current vitrification procedures
rlm@141:      do this anyway since there can be a lot of freezing damage.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>dilated security camera</h2><div class="description">a security camera that would capture
rlm@141:      full video footage of everything at 60fps but then decide to keep
rlm@141:      only every 1 frame every 5 seconds unless there's something
rlm@141:      "interesting" happening.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>bitcoin wallet</h2><div class="description">Part of "server culture", this would be something
rlm@141:                     like "coin.your-domain.com" which would serve as
rlm@141:                     your personal trusted access to your own bitcoins
rlm@141:                     from anywhere.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>libpay</h2><div class="description">this would be a free library which would enable
rlm@141:             micro-donations to software projects and other projects,
rlm@141:             so that you could donate a penny to "emacs" and it would
rlm@141:             be automatically split up to every person who has ever
rlm@141:             contributed to emacs in proportion to the amount of
rlm@141:             community esteem, code quantity, bugs fixed, whatever the
rlm@141:             community decides. This might make it possible for
rlm@141:             programmers to live entirely off of free programming.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>distributed graphics</h2><div class="description">Browser based graphics-card accelerated
rlm@141:      distributed computing API.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>pronouns</h2><div class="description">use capital letters A-Z instead of pronouns. They solve
rlm@141:               pronoun referents and gender neutrality, are short to
rlm@141:               say, and you can encode useful information into the
rlm@141:               choice of letter. For example, instead of "Meetings
rlm@141:               shall be presided over by the president, unless she is
rlm@141:               absent." USE "Meetings shall be presided over by the
rlm@141:               president, unless P is absent." We already use this a
rlm@141:               little, since I and U are reserved for the subject and
rlm@141:               object respectively.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>phone DSP</h2><div class="description">software app that inserts an audio DSP between the
rlm@141:                input to a phone and the output. The DSP is delicious
rlm@141:                and configurable, and can allow men to make their
rlm@141:                voices deeper, etc. The app would allow you to hear
rlm@141:                your own voice as others hear it. Most people hate how
rlm@141:                their own voice sounds. The app would also allow one to
rlm@141:                immediately change the parameters of the DSP using good
rlm@141:                presets.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>predestined body learning</h2><div class="description">a good example of predestined learning
rlm@141:      might be the mirror neurons.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>restaurant receipt</h2><div class="description">use a carbon copy receipt instead of two stupid
rlm@141:      copies. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>anti google glass</h2><div class="description">glasses with mounted lasers and computer vision
rlm@141:      that targets the cameras in google glass and destroy them.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>wearable towel</h2><div class="description">towel with clasp, velcro, whatever, that allows
rlm@141:                     one to wear the towel more securely than just
rlm@141:                     wrapping it tightly and hoping for the best.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>crossdressing</h2><div class="description">Easiest way to disguise oneself as a woman is to
rlm@141:                    wear a burka.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>book-mode</h2><div class="description">intelligent color highlighting for books and
rlm@141:                articles. It would disambiguate pronouns and involved
rlm@141:                references. For example, if "Rachael" was assigned the
rlm@141:                color red, and "the blonde haired girl" refers to
rlm@141:                "Rachael", then "the blonde haired girl" would be
rlm@141:                colored red. Also, you could disambiguate multi part
rlm@141:                run-on sentences by highlighting each
rlm@141:                subcomponent. Maybe would also have applications to
rlm@141:                scientific reading.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Handheld light Rain measurement</h2><div class="description">this would be a clear, teflon
rlm@141:      coated plastic disk with a camera underneath the disk. You would
rlm@141:      be able to hold the device out and it would measure the rate of
rlm@141:      accumulation of water droplets from fine mists and light rain by
rlm@141:      using computer vision to measure the diameters of the drops.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Big Brother Farming</h2><div class="description">This would be a vision system that would
rlm@141:      individually monitor each plant and turn on water, etc to ensure
rlm@141:      maximum/uniform growth for each plant. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Discrete Faucet</h2><div class="description">A faucet with discrete ticks instead of
rlm@141:      continuous. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Laser Circle</h2><div class="description">take a glass microfiliment and shine a laser at one
rlm@141:                   end at an oblique angle. It will make a perfect,
rlm@141:                   large circle on the wall, converting a laser beam
rlm@141:                   into a laser cone, preserving most of the energy of
rlm@141:                   the laser.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Invisible Glass</h2><div class="description">Take a container of liquid and embed a
rlm@141:      glass sculpture made out of glass that has exactly the same index
rlm@141:      of refraction and color of the liquid. Then the sculpture will be
rlm@141:      totally invisible in the container, and will only be revealed
rlm@141:      when the liquid is drained. The container might be a fancy
rlm@141:      wine/spirit bottle or an hourglass.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Caterpillar people</h2><div class="description">A race of caterpillar like creatures gains
rlm@141:      intelligence after eons of predation by birds, etc. These
rlm@141:      caterpillar creatures still undergo metamorphosis into a large
rlm@141:      butterfly-like creature. The metamorphosis process turns the
rlm@141:      caterpillar's brain into mush and reforms it into a minimal,
rlm@141:      dumb, truly insect-like mind, completely destroying the person
rlm@141:      the caterpillar was. The society develops all sorts of customs and
rlm@141:      religious interpretations of the metamorphosis. It is viewed as
rlm@141:      good and natural by some since it is part of their life cycle and
rlm@141:      necessary to propagate the species, as only the butterflies can
rlm@141:      mate. Some think that the butterflies are still the same person
rlm@141:      because they have the same soul, even they no longer posses the
rlm@141:      memories or personality of the original caterpillar. Some see the
rlm@141:      butterfly form as the "true form" of the species, since the
rlm@141:      butterflies can fly, mate, and are beautiful. Many make a big
rlm@141:      deal out of the fact that 1-2% of the caterpillar's mind is
rlm@141:      actually preserved in the butterfly. Some see it as a terrible
rlm@141:      tragedy and argue that the caterpillars should try to stop the
rlm@141:      metamorphosis by technology. Practically, some very important
rlm@141:      members of society undergo hormone therapy and/or surgery to
rlm@141:      prevent metamorphosis so that they can live longer as themselves.
rlm@141: 
rlm@141: <p>
rlm@141:      This is a continuation of Marvin Minsky's ideas about pain being
rlm@141:      something that preserves our bodies while destroying our minds,
rlm@141:      something that is a remnant from our too harsh animal days that
rlm@141:      hasn't caught up to the fact that we have very complex brains
rlm@141:      now. It's a worst-case scenario about a maladaptive genetic
rlm@141:      legacy. Also, it's inspired by "There She Is!!!", which makes a
rlm@141:      compelling point about homosexuality by introducing a second
rlm@141:      gender characteristic (bunny/cat, male/female), which makes
rlm@141:      homophobia look very silly. Here, our own biological legacy of
rlm@141:      pain and death is made to look like the tragedy it is through the
rlm@141:      lens of the the caterpillar people.
rlm@141: </p>
rlm@141: </div></div>
rlm@141: <div class="project"><h2>relationships as a business</h2><div class="description"><a href="http://thedailywtf.com/Articles/Up-or-Out-Solving-the-IT-Turnover-Crisis.aspx">Turnover-Crisis</a> is an excellent talk
rlm@141:      about the "culture of quitting," which is about better business
rlm@141:      by letting people go instead of keeping them around past their
rlm@141:      "apex". Focuses on information transfer. Cool idea of an alumni
rlm@141:      network, which for relationships would be a group of satisfied
rlm@141:      ex-lovers, who would recommend new people your way, and who might
rlm@141:      consider coming to you again, refreshed from their time away with
rlm@141:      new stories/experiences. I should look for examples of this and
rlm@141:      how they worked out.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>coffee with tea</h2><div class="description"><i>rlm-tea</i> contains 2% sugar, 10% cream, and 20%
rlm@141:                     dylan coffee. <i>dylan coffee</i> contains 5% sugar,
rlm@141:                     20% cream, and 10% rlm-tea. Start your mornings
rlm@141:                     with recursion!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>psychic crystal</h2><div class="description">in a science fiction story, this would be an
rlm@141:      object that is very easy to move physically but is extremely
rlm@141:      difficult to move with telekinesis.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2><a href="http://betsofbitco.in/">http://betsofbitco.in/</a></h2><div class="description">what a great place for an AI/person to
rlm@141:      prove themselves as a good predictor. I wish this could be
rlm@141:      automated. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>true reflection</h2><div class="description">don't forget about that mirror in the student
rlm@141:      center!, it's two mirrors at right angles, like staring at a
rlm@141:      corner of a room. The light reflects so that it shows you what
rlm@141:      you actually look like, instead of your mirror image.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>remote control wasp</h2><div class="description">use computer to drive wings with remote
rlm@141:   power/logic.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>encrypted email phone book</h2><div class="description">public (distributed?) database of
rlm@141:      email-&gt;private-key pairs, to enable automatic encryption.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>universal eye color</h2><div class="description">every equivalent creature will see each
rlm@141:      others' eyes as black &ndash; it's universal. Even if the creatures
rlm@141:      see in radio waves, and their eyes are 2m long pieces of jagged
rlm@141:      metal, when those creatures look at each other, they will see
rlm@141:      black, the absence of light and color (since it's being absorbed
rlm@141:      by the sensor array).
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>intelligent microwave</h2><div class="description">it learns where the hot nodes of its fields
rlm@141:   are, and uses them to evenly heat any food item. It has an infrared
rlm@141:   camera or something to keep track of how hot the food is. That way,
rlm@141:   you don't get bowls where the edges are boiling, while the center is
rlm@141:   still frozen. Requires a little bit of intelligence/vision, since
rlm@141:   the exact pattern of heating totally depends on the exact shape of
rlm@141:   the food.  Wouldn't need a carousel, and wouldn't need a timer,
rlm@141:   just a desired temperature. Could also detect ice, and automatically
rlm@141:   defrost the parts which are frozen. Might be able to work much
rlm@141:   faster since it can avoid overheating; might have problems with
rlm@141:   heating the insides of thick things, might need a weight sensor too.
rlm@141: 
rlm@141: <ul>
rlm@141: <li>Would be much cleaner than other microwaves, since food would
rlm@141:     "sputter" and splash liquid much less.
rlm@141: 
rlm@141: </li>
rlm@141: <li>Throw in some SIFT+R processing to match previously cooked foods
rlm@141:     and learn the exact heating profiles for things that have been
rlm@141:     cooked before &ndash; it can get faster the more it's used.
rlm@141: 
rlm@141: </li>
rlm@141: </ul>
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>compression</h2><div class="description">brain-aware image compression algorithm
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Credit card proxy</h2><div class="description">would be a company which works like paypal
rlm@141:      except for real world transactions
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Flesh pillow</h2><div class="description">a pillow like the arm or torso of a human, complete
rlm@141:                   with simulated temperature, bones, and heartbeat.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>super screw</h2><div class="description">a screw which has only one or two threads and instead
rlm@141:                  uses compression to fit into a hole (the whole shank
rlm@141:                  of the screw is split into multiple pieces to
rlm@141:                  accomplish this; the tip is a point, then the middle
rlm@141:                  bulges out and gets compressed when screwed in.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>light filter</h2><div class="description">(like light tweezers) to mechanically separate
rlm@141:                   fluids with different index of refraction
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>chalk eraser project</h2><div class="description">maybe make a directional eraser, for easy
rlm@141:      release of chalk dust, like fur, and how it likes to rest in a
rlm@141:      certain direction.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>silver socks</h2><div class="description">socks laced with silver for the antimicrobial
rlm@141:                   properties.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>UROP</h2><div class="description">magnet gear/metal teeth tape 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Rod of Moses</h2><div class="description">device to distill urine through evaporation and
rlm@141:                   easily dispose of urea crystals for use in desert --
rlm@141:                   produce drinkable water.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>UROP</h2><div class="description">Make the LED in line with the flow for the micro injector, so
rlm@141:           that it may transmit maximum flow.  Motor that changes
rlm@141:           distance of internal magnet from windings depending on
rlm@141:           desired speed so as to obtain maximum power efficiency.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>lottery scraper</h2><div class="description">web scraper which monitors various lotteries,
rlm@141:      looking for "special" gimmick changes in the rules (like 4x
rlm@141:      winnings on Wednesdays) and computes expected value&hellip;
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Memristiors novel design</h2><div class="description">make an evolutionary algorithm to make
rlm@141:   old stuff using all four basic circuit elements.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Conductive concrete</h2><div class="description">concrete that has embedded metal fibers so
rlm@141:      that it can conduct electricity.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>little bitty melting pot</h2><div class="description">might be useful for some types of
rlm@141:      manufacturing/3D printing &ndash; how small can an induction melter be
rlm@141:      made, for example.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>power strip/timer programmable combination</h2><div class="description">meh
rlm@141: </div></div>
rlm@141: 
rlm@141: <div class="project"><h2>algorithms...</h2><div class="description">which learn what their inputs are and in what order,
rlm@141:                and can adapt to changing circumstances &ndash; they
rlm@141:                remember previous arguments and adapt so as to respond
rlm@141:                to different connections.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>true pure tones</h2><div class="description">hear a true pure tone by direct stimulation of the
rlm@141:      nerves of the ear
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>mechanical analogue to the electrical op-amp</h2><div class="description">would be an object
rlm@141:      with two levers &ndash; you pull on one lever and the other moves the
rlm@141:      same way, no matter what's in the way or what it is driving. This
rlm@141:      analogy could be useful to teach op amps to people.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>paper folding device</h2><div class="description">make it convenient to fold lots of papers in
rlm@141:      various ways.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>concrete epoxy</h2><div class="description">epoxy with sand/ some other solid material.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>light capacitor</h2><div class="description">suspend some ball of material with a high index
rlm@141:      of refraction and shine light into it so it gets stuck &ndash; would
rlm@141:      the light stay trapped forever? Could you build up unlimited
rlm@141:      quantities of light inside the sphere (which could then be
rlm@141:      released slowly by frustrated internal reflection?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>movie screening</h2><div class="description">Movies always are too long at first. One way to
rlm@141:      shorten them ``scientifically" is to record blink rate during the
rlm@141:      move and then remove / shorten the frames of the parts in which
rlm@141:      there are a lot of blinking (average this over multiple people)
rlm@141:      better yet, put it online and do it across thousands of people. I
rlm@141:      got this from youtube in which there is an episode of kill bill
rlm@141:      which is composed entirely of the parts in which people had their
rlm@141:      eyes closed. slogan: want to make a movie people can't take their
rlm@141:      eyes off of? Just take those parts out!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>optimize an article</h2><div class="description">capture reading of a scientific article via
rlm@141:      screen capture while people read it, then use it to make the
rlm@141:      article better. like the movie-pruning idea.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>super reading program</h2><div class="description">teaches people the ideal mental mask to
rlm@141:      apply during reading so as to read very fast.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>explosive thermite epoxy putty</h2><div class="description">one part would contain the rust,
rlm@141:      one part the aluminum.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>reading comprehension</h2><div class="description">use the above screen capture routine to
rlm@141:      make a quiz program that constructs questions about the content
rlm@141:      you seemed to gloss over while reading. could be easy if the pdf
rlm@141:      came with embedded questions.  Dylan: automatically generate
rlm@141:      word-cloud about the parts you found most interesting; help
rlm@141:      others who read the same stuff by drawing attention to the
rlm@141:      interesting parts.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>hard sword</h2><div class="description">make a samurai sword, but use osmiridum instead of
rlm@141:                 martensite for the cutting part; it should be a better
rlm@141:                 sword.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>close range wireless</h2><div class="description">use the induction technology used to
rlm@141:      recharge electric toothbrushes with no metal links to send data
rlm@141:      without any metal at all!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>reading</h2><div class="description">is a form of synsethesia
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>DNA printer</h2><div class="description">A machine which translates the text eg, "ACTGAC" into
rlm@141:                   actual DNA
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>black generator</h2><div class="description">ferro-fluid magnetic field suspended micro
rlm@141:      generator to make electricity
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>alcohol battery</h2><div class="description">alcohol/fluid flow powered battery
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>folding razor blade sword</h2><div class="description">
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>perfect pitch</h2><div class="description">learn perfect pitch using another sense in
rlm@141:                    combination (sight or touch)
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>kaleidoscope projector</h2><div class="description">
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>razor blade de-sharpener</h2><div class="description">for guilt free disposal
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>bricks</h2><div class="description">filled with luminescent plant material
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>bio metallic structure</h2><div class="description">metal grids with seeds inside, which grow
rlm@141:      together and form a durable biological matrix. The metal
rlm@141:      substrate delivers water. (maybe use plastic instead of metal?)
rlm@141:      Dylan: enrich plants with inorganic compounds; electrical
rlm@141:      interfaces in cellular plant matter =&gt; remote-controlled
rlm@141:      photosynthetic/bioluminescent structures.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>conducting extracellular matrix</h2><div class="description">to allow better control of
rlm@141:      organic systems and an enhanced nervous system.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>cross-modal memory hashing</h2><div class="description">a way to retrieve memories more
rlm@141:      robustly. 
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>flossing thimble-guards</h2><div class="description">(these actually exist)
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>rules + lattice learning</h2><div class="description">integrate lattice learning with rules by
rlm@141:      generating hypothetical examples
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>wooden refrigerator</h2><div class="description">to give food a better taste Dylan: like
rlm@141:      barrels for wine, or planks for salmon. Maybe just have "flavor
rlm@141:      planks" for your pre-existing fridge. Need to mitigate effect of
rlm@141:      temperature on volatility?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>radioactive transmutation molecule by molecule</h2><div class="description">create precious
rlm@141:      metals or something else economically advantageous.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>crowd preservation</h2><div class="description">inoculate food with tons of harmless
rlm@141:      bacteria so that there's no room for bad bacteria as a method of
rlm@141:      preservation
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>old school preservation</h2><div class="description">Pasteur - style holding jar with siphon
rlm@141:      as a way to store liquids at room temperature indefinitely w/o
rlm@141:      refrigeration.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>restaurant policy</h2><div class="description">Throw rude people out of restaurant as a matter
rlm@141:      of course &ndash; make ambiance much better.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>clean windows</h2><div class="description">make something that mixes soap with fire hydrant
rlm@141:                    water (and reduces the pressure a bit) and use it
rlm@141:                    to clean windows of buildings.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>ocarina</h2><div class="description">make an ocarina out of pure silver
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>fire pen</h2><div class="description">pen which burns words on to the page, thus never needing
rlm@141:               any ink. Is there a way to make it runnable from the
rlm@141:               human's energy?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>website to design your own soda</h2><div class="description">and label, and have it mailed to
rlm@141:      you / sell it from your own online store.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>solar panels</h2><div class="description">that float on the ocean
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>handcuffs with more than two cuffs (3?)</h2><div class="description">great for daisy chaining
rlm@141:      people, binding them to environment, etc.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>vector based SOUND files</h2><div class="description">like the pictures but with SOUND. codify
rlm@141:      sound in a language with enough symbols so that it can describe
rlm@141:      everything and encode it in that. would be like going from speech
rlm@141:      to text or smtg. Could also store sound as an image of the
rlm@141:      wavefront encoded as a vector image.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Mouse</h2><div class="description">with a horizontal scroll wheel in addition to the vertical
rlm@141:            scroll wheel
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>logic maintenance system for big institutions</h2><div class="description">to make sure the
rlm@141:      things they are thinking about doing are not retarded
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2><a href="http://www.regulations.gov/">http://www.regulations.gov/</a></h2><div class="description">cool site
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>genetically engineered glowing fruit</h2><div class="description">sell seeds?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>memory slide</h2><div class="description">IF memories are encoded using particular sensory
rlm@141:                   impressions, what happens if the sensory organ
rlm@141:                   itself changes? those memories would become
rlm@141:                   inaccessible. maybe this is why we can't remember
rlm@141:                   much from our childhoods. also, could this happen
rlm@141:                   throughout life as well? Could S remember stuff from
rlm@141:                   his childhood?
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>make a completely indestructible phone</h2><div class="description">no moving parts or display
rlm@141:      you should be able to slam it around all you want, and it will
rlm@141:      just work. brutally simple. aerogel around the battery, minimal
rlm@141:      interface - never gets too hot, and can be dropped into water. no
rlm@141:      holes &ndash; uses field effects for everything from the buttons to
rlm@141:      inductive charging and data transfer.
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>midi to ocarina "tabs" program</h2><div class="description">(online website? buy ocarinas from
rlm@141:      it too)
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>3d printing with sound pulses (or just patterns)</h2><div class="description">like the 8.03
rlm@141:      lecture
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>lighter flint on spring</h2><div class="description">make hot, throw it at something, and it
rlm@141:      makes sparkles!
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>nuclear energy</h2><div class="description">Rebranding New+Clear Energy with informational
rlm@141:                     campaign and public debate forum to enforce its
rlm@141:                     transparent and open nature. France needn't be the
rlm@141:                     world leader in nuclear energy. (Dylan)
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>bubbles</h2><div class="description">Engineer a material which has both ductility and high
rlm@141:              surface tension to make the "third"
rlm@141:              minimal-surface-energy solution to a bubble suspended
rlm@141:              between two equal-diameter rings. (Solutions are
rlm@141:              cylindrical catenary curve, two separated half-bubbles,
rlm@141:              and a double-cone)
rlm@141: 
rlm@141: </div></div>
rlm@141: <div class="project"><h2>Textbook whose content can be varied continuously</h2><div class="description">alter level of
rlm@141:      difficulty, rigor, diction, emphasize crossover with certain
rlm@141:      other discipline, etc. Content generated dynamically from
rlm@141:      knowledge base, along with questions that are moreover altered to
rlm@141:      guide knowledge acquisition. Motivation: One book of
rlm@141:      knowledge. <i>One.</i>
rlm@141: </div></div>
rlm@141: </div>
rlm@141: 
rlm@141: 
rlm@141: 
rlm@141: </div>
rlm@141: 
rlm@141: <div id="outline-container-1-1" class="outline-3">
rlm@141: <h3 id="sec-1-1"><span class="section-number-3">1.1</span> From Jacob's idea list</h3>
rlm@141: <div class="outline-text-3" id="text-1-1">
rlm@141: 
rlm@141: 
rlm@141: <ul>
rlm@141: <li>Roommate-canceling headphones: uses roommate's laptop mic to seed
rlm@141:   noise cancellation alg in your headphones (would this
rlm@141:   work?). -Update on sound canceling headphones that take feed from
rlm@141:   tv: how about ones that cancel people talking on the phone by
rlm@141:   receiving the phone signals and playing inverse sound
rlm@141:   waves. #signalprocessing ~jcole@mit.edu
rlm@141: 
rlm@141: </li>
rlm@141: <li>ClackerAlert &ndash; tells if you slam the keys too hard using sound data
rlm@141:   (and speed/jerkiness data)!.Prevents RSI ~jcole@mit.e
rlm@141: 
rlm@141: </li>
rlm@141: <li>separate pin that you can tell someone if forced to
rlm@141:                identify your PIN (idea from idea about credit cards)
rlm@141: </li>
rlm@141: </ul>
rlm@141: 
rlm@141: </div>
rlm@141: </div>
rlm@141: </div>
rlm@141: </div>
rlm@141: 
rlm@141: <div id="postamble">
rlm@141: <p class="date">Date: 2015-02-04 23:52:02 EST</p>
rlm@141: <p class="author">Author: Robert McIntyre</p>
rlm@141: <p class="creator">Org version 7.7 with Emacs version 23</p>
rlm@141: <a href="http://validator.w3.org/check?uri=referer">Validate XHTML 1.0</a>
rlm@141: 
rlm@141: </body>
rlm@141: </html>