Skip the screen and go with VR glasses? Then the projectors are at a fixed distance from the eyes, and you can (in theory) go wild applying individual corrections for everyone, up to and including correcting for individual focus.
923 karma · joined April 9, 2010
Looks like >60% of Apple's revenue comes from iPhone/iPad/iPod/iTunes.
Although you could argue that a modern cellphone is technically a computer, I'm pretty sure that in the minds of most people it's a "SMS and email and web browsing appliance" rather than a computer.
The natural state of mind of healthy (i.e. not clinically depressed) people is to default to think that they are important and good at what they do. So you get things like 70% of drivers saying that they drive better than average. (And I saw a psychologist post somewhere online claiming that they had asked 500 people how good they were in bed, and exactly one person out of the 500 thought they were below average.)
At the level of subjective reality "I'm important" is obviously true in general, as you are presumably the most important person in your own life.
At the level of objective reality you have things like "you are one random individual among 8 billion, all of you living on a small rock on the outer edges of an insignificant galaxy, and the lifetime of the entire human species is hardly a blink of the eye in the timeframe of the Earth".
Although objective truth has the advantage of being objective, it is also somewhat depressing.
Terry Pratchett (of the Discworld series) has the concept of "knurd" ("drunk" spelled backwards). The idea is that people normally look at the world through rose-tinted glasses as if they were slightly drunk, being "knurd" is the state of being hyper-sober, stripped of all comforting illusions and seeing the world in all its harsh reality.
I certainly believe that this story is fairly close to being true, although I have no scientific studies to back it up :)
Although most people are slightly delusional, it may be more visible in people that are successful somehow somewhere, since they have an external story to collaborate the internal story: "I feel important, and this proves that I actually am important".
We also have hubris as one of the three virtues of a programmer ( http://en.wikipedia.org/wiki/Larry_Wall#Virtues_of_a_program... ), I have no reason to think that this is less true for entrepreneurs in general.
In short, I think a moderate amount of delusion is quite healthy, and possibly even essential :)
- make the chunks small enough that they are essentially static images (one or a few frames)
- use some form of dynamic chunking where a chunk lasts until the image is sufficiently different according to some metric.
In both cases the chunks are fairly resistent to changes in speed, and the worst you need to deal with is that multiple chunks in clip A may map to the same chunk in clip B.
I did a master degree on pattern matching on text in 1999 (same year Shazam started), and it was obvious then that much of the same concepts (many of them from textbooks stemming from the 1970s) could be used for pattern matching on music. (We even discussed implementing music matching, but couldn't see a market for it.)
There may very well be other parts of music fingerprint algorithms that are patentable, but I have a hard time believing that the parts described in the article could be.
You chunk a song into small segments and create a sort of "fingerprint" for each piece using Fourier analysis or similar. (So a song segment is represented by a frequency histogram, similar to this: http://www.flickr.com/photos/svartling/4229109164/ ) The fingerprints are stored in a DB.
To match a song fragment, chunk and analyze it in the same way. For each segment, find the closest match in the database. If most of the closest matches belong to the same song, and appear in the same order as in the song fragment you are trying to find, you have a match.
Creating fingerprints from a series of video segments is roughly similar, Fourier can be used for 2D images as well.
Doing it like this, you can reduce the volume of data you need to compare against with several orders of magnitude. Scaling to YouTube volumes is still hard, but that's the sort of scaling a company like Google already has plenty of experience in.
[We will be] taking the heat outside of the building and
using cold [sea] water to remove it
I asked him once why he didn't switch to PCs, he couldn't see any reason to do so - the minicomputer was rock solid, still running strong and had the software he needed.
I imagine the oldest machines still in production would be with companies that regard computers the way a factory would regard expensive machinery - switching costs are high, and there is no sense in switching when the old one still works.
And I guess they would be mainframe class computers rather than PCs.
A "lucky break" is more likely to come along if you keep at it for a decade than if you give up after the first non-successful year.
I think skill comes in two forms: The skill of making something that people want to use, but also the skill of recognizing and making the most out of "lucky breaks".
But neither "mobile" nor "cell" says anything about voice in particular, so I would think they are suitable terms for "portable communication device connected to a cellular network" even when the device primarily deals with non-voice services.
We are a bunch of friends... some of us are writers,
programmers, artists, .... Among our writers are two guys
whom you may have heard of: Neal Stephenson and Greg Bear.
So it sounds like they are a whole bunch of people and Neil Stephenson is just one of many contributors.
Not sure whether there is a market for "buy one chapter per week for a year" books - Stephen King tried that a few years back, but gave it up due to lack of interest.
Still, it's an interesting (and natural) experiment. I don't think this particular implementation has much chance for success, but if they keep it up they may find some variation that does.
+1 for duck typing, a custom non-Date object can look like a date for operations that only use a subset of the methods.
I don't know why their version is so slow, I can only imagine that the conversion to Astronomical Julian Day (the Ruby Date internal representation) is expensive, inefficient or both.
I used year/month/day ints as internal representation, which was fast for my use cases (parse date string, format date string, compare two dates).
This reminds me of when I tried to switch to a Dvorak
keyboard. The trouble was, I don't use one computer. I
use all kinds of computers. I use other people's
computers. I use three computers fairly regularly at
home and three at work. I use computers in the test lab
at work. The trouble with customizing your environment
is that it just doesn't propagate, so it's not even
worth the trouble.
Which is my experience too. Dvorak would need to be at least 2-3 times faster than Qwerty to outweigh the disadvantages, a "possibly 10% faster" doesn't cut it.
She complained that the interviewers were asking stupid and irrelevant questions, along the lines of "what is the current stock price for company X?"
When she told them she had been away on vacation in France they asked her to describe the French stock market instead. (She: "How am I supposed to know, I was on holiday!")
Programming is apparently not the only field where "interest beyond 9-5" is a job requirement...
Or more precisely: There are plenty of jobs in both programming and finance where 9-5 interest is enough, but in my experience the jobs that are considered "good" tend to ask more than that.
If the pixel density of the output device is very
different from that of a typical computer display, the
user agent should rescale pixel values. [...]
It is recommended that the reference pixel be [...] one
pixel on a device with a pixel density of 96dpi
"How Google routed around Sun's IP-based licensing restrictions on Java ME": http://www.betaversion.org/~stefano/linotype/news/110/
It's apparently not a complete surprise, this guy in 2007 predicted that
... in the future expect a shit-load of IP-related lawsuits
on this.It is obviously true, but it still takes a while getting used to :)
The current world record is ~600 mph. http://en.wikipedia.org/wiki/Terminal_velocity#Examples
I got more interested in my early twenties, a friend of mine was a programmer and the computers an individual could afford became more powerful. (I was never very interested in Commodore 64 and its ilk, they seemed too constrained to be useful.)
I started programming in my late twenties, when I attended university and got a CS degree.
I'm not still not entirely sure that I have a passion for the hacker lifestyle. My view today is the same as when I was a teen - computers and programming can be very useful tools, but I find tools interesting only to the extent that they can help me solve an interesting problem.
The Netflix prize is interesting. According to a summary post on the Netflix forum
the early results were mainly by individuals...
the team members began to coalesce and combine, and
in the end, entire teams coalesced and recombined.
I can only speculate that the combination of a deadline, a problem that was too hard for any individual and a common forum for exchanging ideas helped encourage team formation. (I.e when you realize that you can't do this alone, teaming up is a rational thing to do.)
I assume that the problems we (as in society, humanity) want to tackle will grow ever more complex in the future, and may easily outgrow the capacity of individuals.
So yes, perhaps we could/should learn from how the Netflix prize played out...
It is easier to crowdsource once the main outline is in place, it structures a vague problem into a set of more specific subtasks that can (more or less) easily be distributed across many people.
Like in this case: Deconstructing an existing proof can work very well with crowdsourcing, since a proof contains a relatively small set of specific claims, and each "crowd" contributor can focus on one particular issue.
But this works because the crowd now has a common focus point and task list created by the paper being published. It's not clear to me how (or whether) the process could be reversed, that the same crowd and effort could be coordinated to cowrite an original paper with an alternative proof.
Except perhaps that only the successful startups get to make that dent.
P≠NP is exactly the ‘expected’ answer! But proving that
expected answer has been the central goal of the field
for 40 years—not so much (in my opinion) because the
answer itself is in serious doubt, as because of how much
will need to be learned about computation on the way to
the proof. If P≠NP is proved, then to whatever extent
theoretical computer science continues to exist at all,
it will have a very different character. [my novel] is being Printed again and again, by Pyrates
It's fairly easy to get a simpler implementation by shifting complexity from the parser to the user. But from a "total society productivity" standpoint, having a handful of developers spend a few months to implement a resilient parser is a cheap price to pay to have umpteen million people save hours of debugging for every page they create.
The basic idea may be simple, but by the time you have covered a reasonable fraction of real-world use cases and error conditions for a sizeable population, the resulting code is anything but.
And 10k lines still counts as "moderately simple" - 10k lines is well within scope for one developer working alone.