I guess Donald Normand's "The Design of Everyday Things" is on the border of "technical", but I'd say it's on the "non-technical" side of the border. It's a fascinating book that goes into detain on how we perceive information, store things in our memory, and interact with the world. I'm enjoying it a lot.
The actual study says that 25% of accidents are caused by distracted driving, not by gadgets per-se. Their list of distractions include (among other things):
1) Vehicle controls/displays.
2) Food.
3) Scenery / roadside features.
4) Daydreaming.
I must be a bit confused about how exactly a magnet link works. In order for me to access the magnet stream for a file I wanted to download, I would need a magnet link pointing to somebody who was serving the file, right?
What happens when the person who the magnet link references turns off their computer? If the magnet link needs to "check in" and update itself on a regular basis to prevent against this, what makes it functionally different than hosting a
What I took from that presentation was essentially:
1) Apple wants to build a monument to itself.
2) Apple is intending to bully the city of Cupertino into allowing it, using their continued presence in the town as leverage.
Another promising reactor design is the pebble-bed reactor. Its reaction has a negative temperature coefficient, meaning that the reaction self-moderates if it gets too hot, rather than requiring an external control system to prevent meltdown. This means that if the cooling system were to fail, the reactor would just sit in a mostly-dormant state until cooling was re-established.
Does anybody really expect that the TSA would admit that their scanners are dangerous and then remove them? No way. Not after the hundreds of millions of dollars they've spent buying them. I guarantee their tests will show that everything is A-OK regardless of what the truth might actually be.
I would assume the big advantage comes in new applications benefiting from their flexibility, such as the pipe pressure sensors mentioned in the OP. While I agree there's no need to worry about saving space, there could be significant cost reduction and easier complexity if you could actually build your microprocessor on the same flexible substrate as your pressure sensor.
The Deepwater Horizon oil spill released over 6 times as much oil as the Valdez did. Very little of it was actually cleaned up. Of the oil which even made it to the surface, mostly BP just dumped enough dispersants down to cause the oil to sink back down to the ocean floor. Where exactly did people think it was going to go anyways?
For the record, in the embedded world it's pretty common to see malloc() used without free(). The use case is providing pointers with memory blocks during program initialization.
In no medium until the CD has it been possible to store a dynamic range compressed audio without giving up something. On vinyl, a loud mix means less audio can be stored, while it doesn't matter on a CD.
This is a bit off base. The "loudness" of a vinyl recording has to do with groove depth. "Louder" audio has nothing to do with how much audio could be stored, which is a function of groove width. Also, the whole point of dynamic range compression is to improve audio fidelity in lossy systems such as tape decks or radio broadcasts. That's why u-law compression is built into our telephone standards, with a compressor on the transmit side and an expander on the receive side.
The other effect is what makes tube amps "better" as well - what happens when you overdrive them. A vinyl record when clipped doesn't hit a hard stop - it hits a soft stop and ends up distorted. Ditto a tube amp - overdrive them and the waveform distorts. However, do that to a CD or transistor amp, and you get clipping. The harmonics induced by clipping the audio are far more harsh to most people's ears than the soft-clip distortion you get with vinyl/tube.
Also why some of the best guitar FX pedals use tubes in their final stages - you want that nice distortion, tubes are really the only way. The alternative is to waste a lot of ADC/DAC and DSP processing power by not using the full dynamic range so there's no possible way to clip, and then process the signal to add soft-clip effects.
There's nothing magical about vacuum tubes. They amplify and clip according to physical processes, and can be described as mathematics just like anything else. An ADC -> DSP -> DAC system is more power efficient, more easily reproducible, much more reliable, and more rugged than any tube-based anything. Ponder this - if you were looking for a specific sound to add to your guitar playing, wouldn't you rather have something _specific_ and consistent from use to use than something which is temperature-dependent, age-dependent, and tube-brand dependent?
I could be mistaken here, but I think that's probably an urban legend. Even assuming that you were using a 300-baud modem that could show a nibble at a time on 4 parallel LEDs and that the LEDs were updated on every single bit, that would still be a potential flicker rate of 75Hz. That would be impossible to catch on any consumer-grade camera, although some specialized equipment could capture it. At 14.4kbps, it would be completely impossible with any video equipment that I'm aware of. At 56kbps forget about it.
Depending on directionality, one could certainly plant an IR bug if desired. However, that's not any different than wifi now.
This is near technology, but I wouldn't want to use it for more than a few minutes at a time. Sliding my hand around on a table for 8 or 10 hours a day would almost certainly give me blisters, and possibly a rash depending on the material. It would callus the heck out of whatever skin was touching the table as well.
Contrast this with a mouse where your hand isn't sliding around on anything, and the winner is pretty clear in my mind.
But when you buy stock from a company during its IPO, what do you get as the result? Stock that you can sell to somebody who wants to buy it. Not dividends, not tangible benefits, not even the promise of tangible benefits. Only the ability to sell it at a higher price when the company does well. You're not addressing the issue of _why_ a company doing well has direct impact on stock prices. I'm actually genuinely curious about this if you have an answer.
It is easier to write an incorrect program than understand a correct one.
