I just had the strongest feeling that Google is going to eventually acquire LinkedIn. For once, I can’t explain why. It just seems to fit, somehow…
A completely random intuition…
Leave a reply
How to write a sonata like Beethoven (first movement)
- Create a simple introductory theme.
- Play it really really fast.
- Make a lot of noise that has nothing to do with the theme but makes the piece sound demanding, including tremolos, arpeggios, etc.
- About 1/4 of the way through, lead into a beautiful 20 second theme that makes the entire piece worth listening to.
- Now’s a good place for a repeat!
- Create an “out-of-place” chord for a modulation in the second ending of the repeat. Might as well play it really loudly too, just to make the audience go “what!?” (this is definitely Haydn’s influence on the composer).
- Get really quiet as you introduce the new key (possibly repeating the introductory theme in the new key).
- Make more noise.
- Repeat beautiful theme again, usually twice.
- End either by quietly fading out or striking a simple cadence as loud as possible.
There’s your first movement!
This is a joke, of course, but listen to enough of his sonatas and this is indeed the pattern Beethoven followed.
Thunderstorms
I bet that the most prevalent type of lightning in a thunderstorm is a property of the relative charges within the stormclouds themselves, rather than independent phenomena! This would make particular storms more likely to carry cloud-to-ground lightning than others.
My intuition is introverted, but every so often I do manage to pull something from the external environment, rather than my own mind. This is one of those times.
Cross-compression: Compressing one file based on others
If taking many photos of the same scene, for example, it would be a good idea to compress these photos in terms of each other, using something like differential encoding (though probably more sophisticated; maybe building an LZW dictionary on one file then applying it to all files?). Domain-specific methods could be used as well. The files would necessarily be paired (don’t even think about changing one without altering the other as well), but it would offer significant savings over solely file-by-file compression.
Something like ZIP or TAR+GZIP may already do this. I’ll have to check the specifics of the algorithms employed. Regardless, I think I see a new coding opportunity coming my way.
Edit: Zip does not compress files together, but this approach is more or less exactly what tar+gzip does (including the idea of using a common LZW dictionary).
Work and schedules
When asked to show up at the lab three days a week, I quickly noticed a change in my work patterns: after about the first week, I noticed I was “saving” work for the three days in the lab. Prior to this, any time was fair game.
The change was brought about by simple boredom. By being coerced into the lab, I was being withheld from pursuing other creative activities (such as finishing the Treatise on the Objective Reality of Ideas, which is running far behind schedule thanks to my general unavailability). As a result, I had three or four hours of time to focus on solely research work each day (a 7 hour day minus 3 hours for commuting and about a half hour for lunch). However, because I was finishing this work at home, I found myself staring at the walls for hours, exactly as had happened in industry. Since this is something I wish to avoid, I’ve been saving research work for my “lab days”.
It is also ironic that I was asked to be in the lab for the purpose of communication, yet I am frequently the only person there. When my advisor is present, he usually either has no advice to give me (do I really need to be there just to receive pats on the back?) or is too busy to speak with me. This completely negates any communication benefit that may have arisen from being there.
The moral of this story? When time is partitioned into discrete scheduling units and ceases to be a continuum, less work gets done. And a commute time that dominates total work time is the equivalent of thrashing in a virtual memory system: lots of work gets done, but none of it is useful.
Score another point for society for depriving me of my independence and usefulness. At this rate, perhaps I can be stopped from making any meaningful contributions at all!
Our latest research project
We’re studying fMRI images of patients who have been exposed to cocaine in-utero (this succeeded in dropping the mean IQ of the group by an entire standard deviation from the general population).
I think I’m going to begin referring to that as the “this is your brain on drugs” research project. It’d be great if we could publish a paper that uses that phrase 🙂
Figures
Not two weeks after coming up with it, my idea was just scooped by Microsoft.
Ah, well; I couldn’t build it anyway.
Learning traffic lights
Inductors under the road can be used to detect traffic at an intersection. Consequently, it should be possible to use machine learning techniques to optimize a strategy to minimize traffic at a particular intersection (and especially to avoid buildup beyond the intersection due to poor timing).
I'm an idiot
I brought an SD card for my camera. It only takes CF cards. Whoops.
Theory of Synchronized Spontaneity
I hypothesize that both small and large-scale sociological behavior (that is, the behavior of individuals and of organizations; the connection being formalized through my 9th psychological postulate: the “linking principle”) are governed by a sort of circadian rhythm, by which various relationships will form and various behaviors will manifest at specific times each year. For example, this is the fourth May in a row that freelance work has begun streaming in (I should note that I no longer actively search for work of any type; the process is entirely passive and yet I’m still being found), yet it’s usually fairly quiet during the rest of the year.
One of my friends brought up a very interesting point: this may be the result of prior conditioning; after all, this is about the time that the school year traditionally ends. I suppose the question is then whether this sort of behavior manifests in response to the targets (in the case of freelance work, the workers) or conditioning of the sources (the clients).
…Or both.