We can't keep pumping shit into the atmosphere and water supplies thinking it won't have some major cumulative effect down the road.
Sure you can. If you don't pump it in faster than it is cleared it doesn't accumulate.
In particular, take the case of shit into water: That's where it's been going since early ocean life invented shit. (Before that it was things like dead bodies of bacteria...) One life form's shit is another's dinner.
Meanwhile, destroying one's own habitat is normal for many higher animals. One example is the three-way cycle of migrating elephants, hippos, and (i forget the third beast), where each destroys the local area for itself (while making it suitable for the next) and migrates on, coming back many years later after the other two species have rehacked things and made the location suitable again. (That's one we get to manage now that we've broken the normal migration pattern. Similar to having to cull and drive deer around now that we've eliminated many of the predators that used to chase them to fresh food and kept their population in check.)
= = = =
The issue with the environment isn't whether we're affecting it. Of COURSE we are. The issue is a stack of stuff related to the Global Warming Catastrophe claims and prescriptions.
We recently got instrumentation and techniques in place capable of coming up with a reasonably good set of data for global temperature for the last several hundred years. And of course we discovered that the temperature is different in different years. Well, duh! We already knew that. We already knew about ice ages and that human civilization rose during the warmup at the end of one. No surprise there.
But the Global Warming mongers jumped on it and started promulgating a complete kit of sky-is-falling predictions, prescriptions for fixing it that involve a massive transfer of power from populations to governments and wealth from populations to new institutions - set up by the same people (example: Al Gore's new companies to trade carbon credits and manufacture and sell carbon offsets), and cries that this must be done RIGHT NOW or we're ALL GOING TO DIE! And of course claims that it's "settled science" (an oxymoron) and demonization of anyone who wants to check the work as a "denier" (as in "holocaust denier" with all the genocidal NAZI references that implies.)
We've heard that before. Look up Malthus and the "Club of Rome" simulations for one example. Or the "new ice age imminent" predictions from the mid 20th century. So before we enslave and impoverish ourselves we need to check the claims - ALL of them:
Is the temperature really climbing as a result of human action? Some other possibilities:
- We're still coming out of that ice age.
- It's an honest artifact from things like cities growing out around locations of the long-term temperature measurement instruments.
- There's a long-term oscillation around a stable or slow temperature change trend, the measurements got it during an upcurve, and this was extrapolated with an exponential, turning a gentle wave into a discontinuous "hockey stick".
- It's just errors in the model.
- The data was faked.
- Maybe we've been holding off the next ice age with our carbon emissions and once we throttle them back (or run out of fossil fuel) we'll freeze over - and all this cutback does is start it earlier (and push us off a "snow reflects solar heat" positive-feedback cliff).
If it's really happening, is it bad?
- How much will it warm up, and how fast?
- Does this kill things off? Or does it just mean that animal habitat moves a few hundred miles toward the poles over a couple centuries and farmers switch to crops that used to be grown a couple hundred miles farther south.
- Maybe global warming IMPROVES things food-wise: Growing grapes in England and veggies in Iceland like in the Medieval Warm Period, turning the permafrost tundra into anothe
I wonder how long it will actually take to phase out bluetooth. I mean, that tech has been around forever and never really caught on outside of phones.
Bluetooth passes the 8KHz network timing natively, by timing its frame rate to the network clock and having built-in provisions for picking a good clocking master. This is very handy for cellphone peripherals because it makes them cheap: The phone provides an accurate and (if appropriate) network-synchronized clock to the the A/D converters in microphones, which only have to synchronize to the frame rate from the phone's bluetooth signal rather than have a stratum-III or better clock built in.
With WiFI any solutions to timing-transfer issues (other than those of the link itself) are add-on kludges.
When the organization itself contributes to the Republican party, all "news" is suspect.
Does that apply to the Democratic party and news operations other than Fox, too?
Be careful how you answer: Reporters and news operation executives are NEARLY unanimous in contributing to Ds and not to Rs. >80% typically. (Plenty of documentation on that is available, thanks to campaign finance reporting laws.)
Unfortuntely, here in the U.S. [the N900] only works with one carrier, T-Mobile, whose coverage is a joke.
You mean it's only SOLD WITH THE SERVICE by T-Mobile, right?
When I signed up for GSM with AT&T (because only AT&T had a cell covering my Nevada place), they told me I could pull the SIM, put it in any other (US bands) GSM phone, and the service would work. The N900 has GSM on all the bands used in the US, according to its specs.
I haven't tried this yet (with a N900 or any other). But perhaps someone here has and can tell us if it works?
Silicon Valley may have proficient coders oozing out of every condo complex, but it lacks--and isn't likely to develop--large numbers of engineers with the right mix of automotive mechatronics and high-voltage systems skills.
But it misses a point: Silicon Valley has the wrong kind of PROGRAMMERS, too. In particular, the valley's levels of software reliability and bug density are far too poor.
I started my programming career in Southeastern Michigan, and spent 15 of the first 20 years of it in the auto industry, so I know whereof I speak. ANY bit of software written for the auto industry is almost certainly life-critical. Some examples, from my own experience (mainly keeping the nightmare scenarios from happening):
- A bug in the idle speed control results in a line of cars that tends to stall after a car length or two when accelerating from a stop sign.
- A bug in the airbag testing software fires a proof-sample airbag while the worker is leaning over it on the test fixture (rather than after he's out of the chamber, the doors are closed, and the alarm has sounded for the required time).
- A bug in the plant energy management system blacks out all the lights in the factory while the workers are interacting with the still-operating machinery.
- A bug in the alarm system doesn't signal when the "flame curtain" over one end of the annealing oven fails. With no warning the plant soon fills (starting near the ceiling) with hot, carbon-monixide laden, "reducing atmosphere" gas, poisoning hundreds of workers before reaching lower-explosive-limit at an ignition source and blowing acres of roof into the next county.
And so on.
When I moved to Silicon valley I was ASTOUNDED at the low level of software reliability here. Design-for-reliability and even debugging subordinated to "feature velocity". Product shipped with hundreds, or thousands, of bugs. Business models that MONETIZED bugs - by selling contracts to fix them (creating the incentive to ship them for fixing later). And so on. (And open source isn't a cure for this: While it doesn't ship until the original programmer or team is happy with it, it mostly gets its reliability by accelerating the fixes, not by annealing the code into crystalline perfection BEFORE it first ships.)
Ship a bug in a car's software and you incur the cost of a RECALL.
At the first place I worked here in the valley one of my colleagues said I was the only guy he'd trust to program his pacemaker. Another said "["Rod"] takes three times as long to write code - but his stuff usually works the first time." (Which is not true: When you do it right - which involves getting the bugs out right away - you can code and debug blazingly fast. I would only deliver when something was finished to my satisfaction - after hundreds of debugging iterations. But my delivery of a completed project would be compared to single iterations of the others' debugging.)
Thus I gravitated (back) to "the hard side of the force" - moving into chip design. (It's of comparable complexity to a large application these days. And it's about the only function in the valley where Detroit-level reliability is valued: Eliminating a silicon spin is about equivalent to eliminating a recall in cost to the company, but it shows up in time-to-market savings.)
So while there are some other programmers like me available here, an auto company attempting to staff-up in Silicon Valley won't be looking for the sort of programmer that constitutes the bulk of the Valley's programming culture. (They'll do well to hire from "back home" in the rust belt or people transplanted from there, hardware designers, or programmers of medical, telecom, or MIL products.) Worse, the middle-managers here who administer the programmers are steeped in - actually the creators of - this software-unreliability culture. If the new auto company's personnel execs don't figure this out in time you can imagine the debacle when the product hits market - or the delays and cost during the delicate venture-funded stage as they try to retrofit quality into their firmware - or rip it out and replace it.
It's a bicycle, not a... hanging offense. Believe me, I have a set of hanging offenses.
Never said stealing the bicycle WAS a hanging offense - or in a class with the other crimes you mentioned.
But dismissing it as "easily replaced property" isn't appropriate, either. Maybe YOUR wealth is such that the cost of a bicycle is trivial and you have cars in the garage so the opportunity cost from loss of the transport is negligible. But that isn't necessarily true for everyone.
Money, and the property it buys, is "crystalized labor". It represents a PIECE of the LIFE of its owner. THAT is what is stolen. How much is a day of your life worth? A month? A year?
"It's just property" is the cry of affluent eliteists for whom such a stolen object represents a drop in their financial bucket - along with criminals attempting to avoid the consequences of their crimes.
And, dude, seriously?... it's a bicycle, and you want to get on with the old-testament wrath, like branding someone's forehead or hacking off appendages?
I think you may need some perspective.
Perhaps you need perspective.
So "it's a bicycle", eh? What's that represent?
How much did it cost? How much labor went into earning that money? Suppose the thief had enslaved the owner and put her to hard labor for that amount of time? What would be an appropriate punishment for that? (Suppose the thief had done that to YOU. Would that change your estimate?)
But the value of something is normally higher than its price (or it would never have been bought). What is the actual value of the bicycle which was lost by the owner when it was stolen? Was it transportation to school? How many classes were missed? What will be the effect on the gradepoint? On the ability to get into a good college? On the future lifetime income? Was it transportation to work? What will be the effect on tardiness, job performance, paid hours? Will the owner lose the job? How much of her own free time will the owner lose by walking to work or using slower transport? How much extra cost to use public transit?
In the old west a horse was the car and the farm tractor. Stealing a horse could end up killing the owner and perhaps his family - by stranding in a hostile environment, crop failure, loss of access to markets for necessities or to medical help, etc. So horsetheft was a hanging offense. Similarly with cattle rustling. (Even today, cattle rustling is big business - and the rustlers often kill any chance witnesses, resulting in the deaths of kids riding out of the supervision of armed adults from time to time.)
Now loss a bicycle in an urban setting MIGHT not rate quite that level of penalty. But when assessing what punishment fits the crime you need to look at the actual costs to the victim. Dismissing it as "just a bicycle" or "just property that's (allegedly) easily replaced" doesn't cut it.
I didn't get that part from the Clarke(?) story: It's used for signaling mirrors, so you can reliably flash sunlight at a viewer in an airplane or a neighboring mountaintop. That's why camping/hiking emergency signaling mirrors - and often the mirrors in camping shaving kits - are double-sided and have a hole in the middle. B-)
Archimedes could have figured out the geometry of that - or one of several related methods of aiming. He was quite the geometer. Or he could have just used about three times as many soldiers. B-)
They have busted it twice now, and both times it was with two guys and small mirrors.
Yep: They busted it because they didn't do it right.
They should try it this way:
- "Pave" the battlements or hills with "soldiers" armed with double-sided flat mirrors with a small hole in them, as large as is convenient for each to hold for significant periods, i.e. the size of a Greek shield. (The back sides only need to be shiny around the hole.)
- Have them hold up their shields at partial arms-length in front of them, look through the hole at the target, and adjust the tilt until the bright spot on their own reflection where the sun shines through the hole is also aligned with the hole.
That lets you build an arbitrarily large solar concentrator with very good focus.
(I think it was Arthur C. Clarke who used this trick in a short story. The army of a small South American country, seated as a block in the stands at a soccer game, uses mirror-fronted program books to ignite the head referee after the first bad call against their team.)
What about blanketing large tracts of land in solar-cells? Is that still okay?
That ought to heat things up a whole bunch. (Ditto solar heat concentrators, such as the solar power plants using parabolic-cylinder reflectors to make process steam for generation.)
Solar panels are a pretty good black color and reflect very little light. Maybe a fifth of it is turned into electric power (to make heat elsewhere when it's used). The rest makes heat locally. Both become infrared radiation which is largely blocked by greenhouse gasses - unlike the unchanged sunlight that would have been reflected by whatever was covered by the panels. (Similarly with the power and waste heat from concentrators powering heat engines.)
So solar panels erected over anything that is lighter than a similar black (i.e. over anything but, say, shale or an exposed coal seam) increase global warming. Horrors!
1. Set up a couple of shell company Wholesalers 2. Add clause to contract to Wholesalers which states that if they utter the word "Peekaboo!" their license becomes invalid 3. Wholesalers all say "Peekaboo!" 4. Wholesalers happily continue to sell products to retailers, who sell them to end users 5. Sue end users and retailers after 3. for copyright infringement (as their license is revoked) 6. Money pours in from retailers and end-users after lawyers crush them like bugs 7. Retailers and end-users try to retaliate against Wholesalers which are shell companies and go "poof" 8. Profit!
9. Retailers and end-users sue the manufacturer, the shell wholesale companies, and the officers of each, individually for damages due to fraud. They also a civil RICO suit. They also ask for invalidation of the copyrights due to their use to further a criminal scheme.
10. On the fraud case the court happily "pierces the corporate veil" due to the willful violation of law by the conspirators, making the personal assets of the officers involved available to pay the judgment.
11. On the civil RICO case the personal assets of the conspirators are also subject to seizure as fruits of the continuing criminal enterprise.
12. Because it was used to further a crime, the copyright(s) on the software MAY be invalidated (though this is {very} unlikely).
13. So the retailers and end users end up with much of the manufacturers's assets and the bad guys' money, houses, and cars. And just maybe the software product goes into the public domain (or the retailers and customers end up with the copyrights on it).
14. Goodbye profit.
Somehow I don't think the officials of a typical software company will play that game - especially since it doesn't NEED to do such a thing to fix the retail price of its product.
If I've got this right: British Empire slang since WWII or so for physicist, engineer. medic, or other technical type. Nearest American English equivalent is "egghead", although "geek" and "nerd" are also in the general vicinity.
It is a compromise between the need to have very small bandwidth per channel (so you can multiplex a lot of channels,..
It predates multiplexing and long distance by a bunch.
It was chosen partly because it was all that was needed and cut through ambient noise better.
The first issue with lower bandwidth for transmission occurred with rural local wiring. Wire has lots of stray capacitance and resistance, much less stray inductance. So transmission lines attenuate high frequencies more with distance than lower ones. By a couple miles from the switchboard you're getting enough low-pass filtering to be annoying.
So the telcos installed "loading coils" to add inductance and flatten the frequency response - at the cost of drastically attenuating any frequencies above the resonance of the loading coils with the line capacitance. Also: The loading coils had to be spaced no more than a quarter wave from each end of the wire and a half wave from each other along it. So the higher that resonance the more loading coils you need for running a line out to the farms surrounding the town with the switchboard. 3.4kHz-ish was picked as a good spot to quit and still have intelligible calls on the farmers' phones.
(By the way: The high-frequency attenuation and phase distortion near the resonance is why loading coils must be removed from phone lines to get decent data rates on POTS modems. And don't even THINK of trying to push DSL through one - the entire DSL signal is above the resonance. B-) )
... they are accustomed to hearing some level of white noise in the background and are prone to misinterpret a period of silence as call disconnection.
Some systems compensate for that by inserting (too much IMHO) white noise in intervals when there is silence from the other end but the connection is still up. (Sometimes they do it poorly by noticeably switching it on and off. For instance: My AT&T cell service starts inserting noise about midway between the first and second ring during call connection, too. It's very annoying.)
The term-of-art is "comfort noise".
It's there specifically to head off an "Are you there?" exchange (which can become confusing and protracted if there's significant latency in the line and the question interrupts the other speaker mid-sentence at his end.)
The latency they're talking about is delay between the speaker at one end saying something and the earphone at the other end reproducing an approximation of the sound.
It looks like you have that confused with time-to-product-release.
Also: Phone systems work on standards to exchange signals. A big one is the TDM CODECs.
For about a half century the telcos have been digitizing at 8,000 samples per second using one of two 8-bit codings called "A-law" and "U-law" (where "U" is actually "mu"). (Think 8-bit signed floating point numbers with smoothing tweaks around the values where the exponent changes.) 8k samples/sec has a Nyquist frequency limit of 4,000 Hz (and you have to low-pass filter it somewhat below that to keep higher frequencies from being "folded back" around 4kHz and fouling the signal you're after.
While digital cellphones and VoIP have been driving adoption of other CODECs, the push has been to reduce the data rate required to carry a "voice-quality" call (either on wires or through radio noise on limited cellular bandwidth) and to survive transcoding to and from A/U-law for interchange with the telcos' installed base, rather than taking advantage of higher wire/fiber data rates to improve audio fidelity.
For some reason I thought of farm implements when I saw "tractor." Didn't make a whole lot of sense.
Sure it does. "Tractor beams" and "tractors" are named that because they pull. (Same root as "traction".)
However the beam in TFA is, in the science fiction vernacular, a "pressor beam", the "tractor beam"'s other-direction counterpart, because it pushes. The hollow cylindrical beam pushes inward, while the beam-down-the-middle pushes along the "tube".
Everything must be as boring as possible or it's not suitable for "professionals".
According to the PHBs, that's damned right.
Companies are paying their workers to use their attention on work. If that work happens to include Google searches and Google does things to distract them, it costs productivity.
A SMALL amount of distraction may actually improve performance by breaking fixation and creating a moment of stress relief. But there's a happy medium and an excess is counterproductive.
I know folks in the defense industry - all the critical stuff has not physical path to the internet. To access that information means switching machines.
Same goes for other industries. I mean, network admins aren't stupid - it's pretty obvious that if it's really critical you don't connect it to the internet. Even the PHBs get that.
Actually, the PHBs DON'T get it. They outsource the administration of their networks , including fileservers containing their critical IT (both technical and business-secret), to contractors. They mandate the use of notoriously security-bug-ridden software products in business-critical infrastructure roles. Then they depend on firewalls and signature-driven commercial anti-malware products to try to keep the mice out of the resulting swiss cheese.
Last month Datamation magazine estimated that 75% of the major tech and high-end businesses had already had their IP compromised by spear-phishing attacks on their IT infrastructure. (Spear-phishing tools go right past signature-based antimalwre products, which typically don't know about them until they get to "general-availability" or a customer of the malware vendor (typically a low-tier customer who gets the tool late) gets caught. This is estimated as occurring about two YEARS after a given exploit is built into an attack tool.)
It's widely thought that early forms of life were based on RNA rather than DNA, so there you go.
In particular, TFA (at least as characterized by the slashdot article) seems to be hunting for a "missing link" programmable catalyst bridging a perceived gap between DNA/RNA and proteins for chemical synthesis, and proposing transition element complexes to fill this gap. But RNA works just fine as self-folding charge/shape/leverage-based molecular machinery, as well as self-copying genetic information "tapes". So it can be the whole ball of wax for the initial startup. There IS no gap to be filled.
Of course it's less effective than protein for molecular machinery, and less stable than DNA as a data repository, etc. But those can be evolved-on upgrades later.
Now we just need to figure out how to make a ribosome.
Ribosomes are a case in point: They're the bulk of a very complex chemical factory for building proteins. But when you tear them apart you'll find that MOST of the pieces are RNA enzymes. And other parts of the machinery are RNA as well - notably the T-RNA that gets bound to the various amino acids, carries them into the ribosome, and lines them up in the sequence specified by the M-RNA "tape" being transcribed.
Seems to me the logical sequence is for the whole protein synthesis mechanism to initially be built out of RNA, then (once directed protein synthesis is up and running), some pieces of it - such as chunks of the ribosome and aminoacyl tRNA synthetases - would eventually be replaced by proteins that would do the job better.
Ultimately the spacecraft's goals are to help scientists understand why the sun's outer atmosphere so much hotter than the sun's visible surface and what propels the solar wind...
I thought they'd figured that out (recently): Vibrations of the solar magnetic field line loops pump energy into the plasma fraction of the gas above the visible "surface", heating it. Reconnection of the lines cause the new loops to expand like released springs, catapulting the entrapped plasma outward.
Didn't that work out once they finished the math on the details?
theoretically, ECL would trounce any CMOS tech fabricated with current accuracy. It's just that it's so horribly power-hungry, that nobody will do this.
CMOS beats ECL in integrated circuits because you can put a LOT of gates VERY CLOSE TOGETHER without cooking them. This means that, though you lose on gate speed, you gain more back in short propagation time on the interconnects.
ECL is inherently hot because each gate starts with an analog current regulator, followed by logic that "steers" the current into one or the other of the two complimentary outputs. That current source dissipates a lot of power. Then the rest of the power is mostly dissipated in the terminator on the other end of the two interconnect "balanced transmission line" wires.
Up to the Cray 1, Seymour designed with discrete components to get better speed. With the Cray 1 he switched to ICs. But he only used four types: Three small-scale integration multi-gate chips with different counts of gates and inputs, and a 16-bit flop array (if I recall correctly).
The cute thing about this - avoiding complex functions like adders - is that the ICs didn't have any (unbalanced) internal signal nodes. Their interconnects were by current steering and the differential outputs were terminated to each other at the far end. So, though they were frantically switching, they presented a resistive load to the power supply, drastically reducing the need for power supply filtering. (A downside is that the system pullss the same amount of power when sitting idle as when running full-bore.)
CMOS is low power because it only draws current (other than leakage current) when it is switching the level of the output (and charging or discharging the stray capacitance of the interconnect wiring) - which means the power supply load varies drastically with switching rate. TTL - the main contemporary competitor for ECL at the Cray 1's time - has a node with a pull-up resistor internally, which draws different current in one and zero state, and also pulls a big current spike when switching its output state because the pull-up and pull-down transistors are both on momentarily.
I built a P[D]P11 "supercluster" and started with Xilinx. The hardware is great, but their software toolset is horrendous.
Ran into a similar problem a few years ago with a FPGA programming tool. (I think it was with Xilinx but I'd have to check.)
We were trying an FPGA implementation of a heavily-channelized ASIC, using the same trick Cray used for the peripheral processors in his CDC-6x00 series: One set of fast logic switching between N sets of states stored in a N-word memory to handle N channels (in his case, to produce 10 independent "peripheral processors" running at the memory speed, which was 1/10th the logic speed, using only one set of logic and ten memories). You have to MUX a lot of stuff every clock, depending on how each channel is configured. So you have a LOT of MUXes / a very wide MUX to be driven by the sequencer, requiring a lot of fanout on the signals driving the MUX selects.
Turns out the FPGA compiler assumed MUXes were only switched occasionally and for configuration. So it didn't honor the drive strength modifiers and add extra drivers or distribute the MUX selection inputs across multiple buffers to get the MUXes to switch fast enough to do a different logic configuration every clock.
Couldn't get 'em to fix the compiler in a timely fashion (or at all) so the project was shelved.
Slashdot Mirror
We can't keep pumping shit into the atmosphere and water supplies thinking it won't have some major cumulative effect down the road.
Sure you can. If you don't pump it in faster than it is cleared it doesn't accumulate.
In particular, take the case of shit into water: That's where it's been going since early ocean life invented shit. (Before that it was things like dead bodies of bacteria...) One life form's shit is another's dinner.
Meanwhile, destroying one's own habitat is normal for many higher animals. One example is the three-way cycle of migrating elephants, hippos, and (i forget the third beast), where each destroys the local area for itself (while making it suitable for the next) and migrates on, coming back many years later after the other two species have rehacked things and made the location suitable again. (That's one we get to manage now that we've broken the normal migration pattern. Similar to having to cull and drive deer around now that we've eliminated many of the predators that used to chase them to fresh food and kept their population in check.)
= = = =
The issue with the environment isn't whether we're affecting it. Of COURSE we are. The issue is a stack of stuff related to the Global Warming Catastrophe claims and prescriptions.
We recently got instrumentation and techniques in place capable of coming up with a reasonably good set of data for global temperature for the last several hundred years. And of course we discovered that the temperature is different in different years. Well, duh! We already knew that. We already knew about ice ages and that human civilization rose during the warmup at the end of one. No surprise there.
But the Global Warming mongers jumped on it and started promulgating a complete kit of sky-is-falling predictions, prescriptions for fixing it that involve a massive transfer of power from populations to governments and wealth from populations to new institutions - set up by the same people (example: Al Gore's new companies to trade carbon credits and manufacture and sell carbon offsets), and cries that this must be done RIGHT NOW or we're ALL GOING TO DIE! And of course claims that it's "settled science" (an oxymoron) and demonization of anyone who wants to check the work as a "denier" (as in "holocaust denier" with all the genocidal NAZI references that implies.)
We've heard that before. Look up Malthus and the "Club of Rome" simulations for one example. Or the "new ice age imminent" predictions from the mid 20th century. So before we enslave and impoverish ourselves we need to check the claims - ALL of them:
Is the temperature really climbing as a result of human action? Some other possibilities:
- We're still coming out of that ice age.
- It's an honest artifact from things like cities growing out around locations of the long-term temperature measurement instruments.
- There's a long-term oscillation around a stable or slow temperature change trend, the measurements got it during an upcurve, and this was extrapolated with an exponential, turning a gentle wave into a discontinuous "hockey stick".
- It's just errors in the model.
- The data was faked.
- Maybe we've been holding off the next ice age with our carbon emissions and once we throttle them back (or run out of fossil fuel) we'll freeze over - and all this cutback does is start it earlier (and push us off a "snow reflects solar heat" positive-feedback cliff).
If it's really happening, is it bad?
- How much will it warm up, and how fast?
- Does this kill things off? Or does it just mean that animal habitat moves a few hundred miles toward the poles over a couple centuries and farmers switch to crops that used to be grown a couple hundred miles farther south.
- Maybe global warming IMPROVES things food-wise: Growing grapes in England and veggies in Iceland like in the Medieval Warm Period, turning the permafrost tundra into anothe
I wonder how long it will actually take to phase out bluetooth. I mean, that tech has been around forever and never really caught on outside of phones.
Bluetooth passes the 8KHz network timing natively, by timing its frame rate to the network clock and having built-in provisions for picking a good clocking master. This is very handy for cellphone peripherals because it makes them cheap: The phone provides an accurate and (if appropriate) network-synchronized clock to the the A/D converters in microphones, which only have to synchronize to the frame rate from the phone's bluetooth signal rather than have a stratum-III or better clock built in.
With WiFI any solutions to timing-transfer issues (other than those of the link itself) are add-on kludges.
When the organization itself contributes to the Republican party, all "news" is suspect.
Does that apply to the Democratic party and news operations other than Fox, too?
Be careful how you answer: Reporters and news operation executives are NEARLY unanimous in contributing to Ds and not to Rs. >80% typically. (Plenty of documentation on that is available, thanks to campaign finance reporting laws.)
Unfortuntely, here in the U.S. [the N900] only works with one carrier, T-Mobile, whose coverage is a joke.
You mean it's only SOLD WITH THE SERVICE by T-Mobile, right?
When I signed up for GSM with AT&T (because only AT&T had a cell covering my Nevada place), they told me I could pull the SIM, put it in any other (US bands) GSM phone, and the service would work. The N900 has GSM on all the bands used in the US, according to its specs.
I haven't tried this yet (with a N900 or any other). But perhaps someone here has and can tell us if it works?
The article says:
Wrong kind of engineers ...
Silicon Valley may have proficient coders oozing out of every condo complex, but it lacks--and isn't likely to develop--large numbers of engineers with the right mix of automotive mechatronics and high-voltage systems skills.
But it misses a point: Silicon Valley has the wrong kind of PROGRAMMERS, too. In particular, the valley's levels of software reliability and bug density are far too poor.
I started my programming career in Southeastern Michigan, and spent 15 of the first 20 years of it in the auto industry, so I know whereof I speak. ANY bit of software written for the auto industry is almost certainly life-critical. Some examples, from my own experience (mainly keeping the nightmare scenarios from happening):
- A bug in the idle speed control results in a line of cars that tends to stall after a car length or two when accelerating from a stop sign.
- A bug in the airbag testing software fires a proof-sample airbag while the worker is leaning over it on the test fixture (rather than after he's out of the chamber, the doors are closed, and the alarm has sounded for the required time).
- A bug in the plant energy management system blacks out all the lights in the factory while the workers are interacting with the still-operating machinery.
- A bug in the alarm system doesn't signal when the "flame curtain" over one end of the annealing oven fails. With no warning the plant soon fills (starting near the ceiling) with hot, carbon-monixide laden, "reducing atmosphere" gas, poisoning hundreds of workers before reaching lower-explosive-limit at an ignition source and blowing acres of roof into the next county.
And so on.
When I moved to Silicon valley I was ASTOUNDED at the low level of software reliability here. Design-for-reliability and even debugging subordinated to "feature velocity". Product shipped with hundreds, or thousands, of bugs. Business models that MONETIZED bugs - by selling contracts to fix them (creating the incentive to ship them for fixing later). And so on. (And open source isn't a cure for this: While it doesn't ship until the original programmer or team is happy with it, it mostly gets its reliability by accelerating the fixes, not by annealing the code into crystalline perfection BEFORE it first ships.)
Ship a bug in a car's software and you incur the cost of a RECALL.
At the first place I worked here in the valley one of my colleagues said I was the only guy he'd trust to program his pacemaker. Another said "["Rod"] takes three times as long to write code - but his stuff usually works the first time." (Which is not true: When you do it right - which involves getting the bugs out right away - you can code and debug blazingly fast. I would only deliver when something was finished to my satisfaction - after hundreds of debugging iterations. But my delivery of a completed project would be compared to single iterations of the others' debugging.)
Thus I gravitated (back) to "the hard side of the force" - moving into chip design. (It's of comparable complexity to a large application these days. And it's about the only function in the valley where Detroit-level reliability is valued: Eliminating a silicon spin is about equivalent to eliminating a recall in cost to the company, but it shows up in time-to-market savings.)
So while there are some other programmers like me available here, an auto company attempting to staff-up in Silicon Valley won't be looking for the sort of programmer that constitutes the bulk of the Valley's programming culture. (They'll do well to hire from "back home" in the rust belt or people transplanted from there, hardware designers, or programmers of medical, telecom, or MIL products.) Worse, the middle-managers here who administer the programmers are steeped in - actually the creators of - this software-unreliability culture. If the new auto company's personnel execs don't figure this out in time you can imagine the debacle when the product hits market - or the delays and cost during the delicate venture-funded stage as they try to retrofit quality into their firmware - or rip it out and replace it.
See the week or so starting here a few days after the strip begins in June of 2000.
It's a bicycle, not a ... hanging offense. Believe me, I have a set of hanging offenses.
Never said stealing the bicycle WAS a hanging offense - or in a class with the other crimes you mentioned.
But dismissing it as "easily replaced property" isn't appropriate, either. Maybe YOUR wealth is such that the cost of a bicycle is trivial and you have cars in the garage so the opportunity cost from loss of the transport is negligible. But that isn't necessarily true for everyone.
Money, and the property it buys, is "crystalized labor". It represents a PIECE of the LIFE of its owner. THAT is what is stolen. How much is a day of your life worth? A month? A year?
"It's just property" is the cry of affluent eliteists for whom such a stolen object represents a drop in their financial bucket - along with criminals attempting to avoid the consequences of their crimes.
And, dude, seriously? ... it's a bicycle, and you want to get on with the old-testament wrath, like branding someone's forehead or hacking off appendages?
I think you may need some perspective.
Perhaps you need perspective.
So "it's a bicycle", eh? What's that represent?
How much did it cost? How much labor went into earning that money? Suppose the thief had enslaved the owner and put her to hard labor for that amount of time? What would be an appropriate punishment for that? (Suppose the thief had done that to YOU. Would that change your estimate?)
But the value of something is normally higher than its price (or it would never have been bought). What is the actual value of the bicycle which was lost by the owner when it was stolen? Was it transportation to school? How many classes were missed? What will be the effect on the gradepoint? On the ability to get into a good college? On the future lifetime income? Was it transportation to work? What will be the effect on tardiness, job performance, paid hours? Will the owner lose the job? How much of her own free time will the owner lose by walking to work or using slower transport? How much extra cost to use public transit?
In the old west a horse was the car and the farm tractor. Stealing a horse could end up killing the owner and perhaps his family - by stranding in a hostile environment, crop failure, loss of access to markets for necessities or to medical help, etc. So horsetheft was a hanging offense. Similarly with cattle rustling. (Even today, cattle rustling is big business - and the rustlers often kill any chance witnesses, resulting in the deaths of kids riding out of the supervision of armed adults from time to time.)
Now loss a bicycle in an urban setting MIGHT not rate quite that level of penalty. But when assessing what punishment fits the crime you need to look at the actual costs to the victim. Dismissing it as "just a bicycle" or "just property that's (allegedly) easily replaced" doesn't cut it.
I did wonder about aiming. I like that idea.
I didn't get that part from the Clarke(?) story: It's used for signaling mirrors, so you can reliably flash sunlight at a viewer in an airplane or a neighboring mountaintop. That's why camping/hiking emergency signaling mirrors - and often the mirrors in camping shaving kits - are double-sided and have a hole in the middle. B-)
Archimedes could have figured out the geometry of that - or one of several related methods of aiming. He was quite the geometer. Or he could have just used about three times as many soldiers. B-)
They have busted it twice now, and both times it was with two guys and small mirrors.
Yep: They busted it because they didn't do it right.
They should try it this way:
- "Pave" the battlements or hills with "soldiers" armed with double-sided flat mirrors with a small hole in them, as large as is convenient for each to hold for significant periods, i.e. the size of a Greek shield. (The back sides only need to be shiny around the hole.)
- Have them hold up their shields at partial arms-length in front of them, look through the hole at the target, and adjust the tilt until the bright spot on their own reflection where the sun shines through the hole is also aligned with the hole.
That lets you build an arbitrarily large solar concentrator with very good focus.
(I think it was Arthur C. Clarke who used this trick in a short story. The army of a small South American country, seated as a block in the stands at a soccer game, uses mirror-fronted program books to ignite the head referee after the first bad call against their team.)
What about blanketing large tracts of land in solar-cells? Is that still okay?
That ought to heat things up a whole bunch. (Ditto solar heat concentrators, such as the solar power plants using parabolic-cylinder reflectors to make process steam for generation.)
Solar panels are a pretty good black color and reflect very little light. Maybe a fifth of it is turned into electric power (to make heat elsewhere when it's used). The rest makes heat locally. Both become infrared radiation which is largely blocked by greenhouse gasses - unlike the unchanged sunlight that would have been reflected by whatever was covered by the panels. (Similarly with the power and waste heat from concentrators powering heat engines.)
So solar panels erected over anything that is lighter than a similar black (i.e. over anything but, say, shale or an exposed coal seam) increase global warming. Horrors!
1. Set up a couple of shell company Wholesalers
2. Add clause to contract to Wholesalers which states that if they utter the word "Peekaboo!" their license becomes invalid
3. Wholesalers all say "Peekaboo!"
4. Wholesalers happily continue to sell products to retailers, who sell them to end users
5. Sue end users and retailers after 3. for copyright infringement (as their license is revoked)
6. Money pours in from retailers and end-users after lawyers crush them like bugs
7. Retailers and end-users try to retaliate against Wholesalers which are shell companies and go "poof"
8. Profit!
9. Retailers and end-users sue the manufacturer, the shell wholesale companies, and the officers of each, individually for damages due to fraud. They also a civil RICO suit. They also ask for invalidation of the copyrights due to their use to further a criminal scheme.
10. On the fraud case the court happily "pierces the corporate veil" due to the willful violation of law by the conspirators, making the personal assets of the officers involved available to pay the judgment.
11. On the civil RICO case the personal assets of the conspirators are also subject to seizure as fruits of the continuing criminal enterprise.
12. Because it was used to further a crime, the copyright(s) on the software MAY be invalidated (though this is {very} unlikely).
13. So the retailers and end users end up with much of the manufacturers's assets and the bad guys' money, houses, and cars. And just maybe the software product goes into the public domain (or the retailers and customers end up with the copyrights on it).
14. Goodbye profit.
Somehow I don't think the officials of a typical software company will play that game - especially since it doesn't NEED to do such a thing to fix the retail price of its product.
But never overestimate the criminally inclined.
Is a boffin some kind of sea-going bird?
If I've got this right: British Empire slang since WWII or so for physicist, engineer. medic, or other technical type. Nearest American English equivalent is "egghead", although "geek" and "nerd" are also in the general vicinity.
It is a compromise between the need to have very small bandwidth per channel (so you can multiplex a lot of channels,..
It predates multiplexing and long distance by a bunch.
It was chosen partly because it was all that was needed and cut through ambient noise better.
The first issue with lower bandwidth for transmission occurred with rural local wiring. Wire has lots of stray capacitance and resistance, much less stray inductance. So transmission lines attenuate high frequencies more with distance than lower ones. By a couple miles from the switchboard you're getting enough low-pass filtering to be annoying.
So the telcos installed "loading coils" to add inductance and flatten the frequency response - at the cost of drastically attenuating any frequencies above the resonance of the loading coils with the line capacitance. Also: The loading coils had to be spaced no more than a quarter wave from each end of the wire and a half wave from each other along it. So the higher that resonance the more loading coils you need for running a line out to the farms surrounding the town with the switchboard. 3.4kHz-ish was picked as a good spot to quit and still have intelligible calls on the farmers' phones.
(By the way: The high-frequency attenuation and phase distortion near the resonance is why loading coils must be removed from phone lines to get decent data rates on POTS modems. And don't even THINK of trying to push DSL through one - the entire DSL signal is above the resonance. B-) )
... they are accustomed to hearing some level of white noise in the background and are prone to misinterpret a period of silence as call disconnection.
Some systems compensate for that by inserting (too much IMHO) white noise in intervals when there is silence from the other end but the connection is still up. (Sometimes they do it poorly by noticeably switching it on and off. For instance: My AT&T cell service starts inserting noise about midway between the first and second ring during call connection, too. It's very annoying.)
The term-of-art is "comfort noise".
It's there specifically to head off an "Are you there?" exchange (which can become confusing and protracted if there's significant latency in the line and the question interrupts the other speaker mid-sentence at his end.)
The latency they're talking about is delay between the speaker at one end saying something and the earphone at the other end reproducing an approximation of the sound.
It looks like you have that confused with time-to-product-release.
Also: Phone systems work on standards to exchange signals. A big one is the TDM CODECs.
For about a half century the telcos have been digitizing at 8,000 samples per second using one of two 8-bit codings called "A-law" and "U-law" (where "U" is actually "mu"). (Think 8-bit signed floating point numbers with smoothing tweaks around the values where the exponent changes.) 8k samples/sec has a Nyquist frequency limit of 4,000 Hz (and you have to low-pass filter it somewhat below that to keep higher frequencies from being "folded back" around 4kHz and fouling the signal you're after.
While digital cellphones and VoIP have been driving adoption of other CODECs, the push has been to reduce the data rate required to carry a "voice-quality" call (either on wires or through radio noise on limited cellular bandwidth) and to survive transcoding to and from A/U-law for interchange with the telcos' installed base, rather than taking advantage of higher wire/fiber data rates to improve audio fidelity.
For some reason I thought of farm implements when I saw "tractor." Didn't make a whole lot of sense.
Sure it does. "Tractor beams" and "tractors" are named that because they pull. (Same root as "traction".)
However the beam in TFA is, in the science fiction vernacular, a "pressor beam", the "tractor beam"'s other-direction counterpart, because it pushes. The hollow cylindrical beam pushes inward, while the beam-down-the-middle pushes along the "tube".
Everything must be as boring as possible or it's not suitable for "professionals".
According to the PHBs, that's damned right.
Companies are paying their workers to use their attention on work. If that work happens to include Google searches and Google does things to distract them, it costs productivity.
A SMALL amount of distraction may actually improve performance by breaking fixation and creating a moment of stress relief. But there's a happy medium and an excess is counterproductive.
Or at least so sayeth the PHBs.
I know folks in the defense industry - all the critical stuff has not physical path to the internet. To access that information means switching machines.
Same goes for other industries. I mean, network admins aren't stupid - it's pretty obvious that if it's really critical you don't connect it to the internet. Even the PHBs get that.
Actually, the PHBs DON'T get it. They outsource the administration of their networks , including fileservers containing their critical IT (both technical and business-secret), to contractors. They mandate the use of notoriously security-bug-ridden software products in business-critical infrastructure roles. Then they depend on firewalls and signature-driven commercial anti-malware products to try to keep the mice out of the resulting swiss cheese.
Last month Datamation magazine estimated that 75% of the major tech and high-end businesses had already had their IP compromised by spear-phishing attacks on their IT infrastructure. (Spear-phishing tools go right past signature-based antimalwre products, which typically don't know about them until they get to "general-availability" or a customer of the malware vendor (typically a low-tier customer who gets the tool late) gets caught. This is estimated as occurring about two YEARS after a given exploit is built into an attack tool.)
This would be useful in a lot of large cities, as pest control. Even better, would be if they could be powered by the dead rats that they kill.
So why use robots? Just import some snakes. Wild biotech has been optimized for this job for years.
It's widely thought that early forms of life were based on RNA rather than DNA, so there you go.
In particular, TFA (at least as characterized by the slashdot article) seems to be hunting for a "missing link" programmable catalyst bridging a perceived gap between DNA/RNA and proteins for chemical synthesis, and proposing transition element complexes to fill this gap. But RNA works just fine as self-folding charge/shape/leverage-based molecular machinery, as well as self-copying genetic information "tapes". So it can be the whole ball of wax for the initial startup. There IS no gap to be filled.
Of course it's less effective than protein for molecular machinery, and less stable than DNA as a data repository, etc. But those can be evolved-on upgrades later.
Now we just need to figure out how to make a ribosome.
Ribosomes are a case in point: They're the bulk of a very complex chemical factory for building proteins. But when you tear them apart you'll find that MOST of the pieces are RNA enzymes. And other parts of the machinery are RNA as well - notably the T-RNA that gets bound to the various amino acids, carries them into the ribosome, and lines them up in the sequence specified by the M-RNA "tape" being transcribed.
Seems to me the logical sequence is for the whole protein synthesis mechanism to initially be built out of RNA, then (once directed protein synthesis is up and running), some pieces of it - such as chunks of the ribosome and aminoacyl tRNA synthetases - would eventually be replaced by proteins that would do the job better.
Ultimately the spacecraft's goals are to help scientists understand why the sun's outer atmosphere so much hotter than the sun's visible surface and what propels the solar wind ...
I thought they'd figured that out (recently): Vibrations of the solar magnetic field line loops pump energy into the plasma fraction of the gas above the visible "surface", heating it. Reconnection of the lines cause the new loops to expand like released springs, catapulting the entrapped plasma outward.
Didn't that work out once they finished the math on the details?
theoretically, ECL would trounce any CMOS tech fabricated with current accuracy. It's just that it's so horribly power-hungry, that nobody will do this.
CMOS beats ECL in integrated circuits because you can put a LOT of gates VERY CLOSE TOGETHER without cooking them. This means that, though you lose on gate speed, you gain more back in short propagation time on the interconnects.
ECL is inherently hot because each gate starts with an analog current regulator, followed by logic that "steers" the current into one or the other of the two complimentary outputs. That current source dissipates a lot of power. Then the rest of the power is mostly dissipated in the terminator on the other end of the two interconnect "balanced transmission line" wires.
Up to the Cray 1, Seymour designed with discrete components to get better speed. With the Cray 1 he switched to ICs. But he only used four types: Three small-scale integration multi-gate chips with different counts of gates and inputs, and a 16-bit flop array (if I recall correctly).
The cute thing about this - avoiding complex functions like adders - is that the ICs didn't have any (unbalanced) internal signal nodes. Their interconnects were by current steering and the differential outputs were terminated to each other at the far end. So, though they were frantically switching, they presented a resistive load to the power supply, drastically reducing the need for power supply filtering. (A downside is that the system pullss the same amount of power when sitting idle as when running full-bore.)
CMOS is low power because it only draws current (other than leakage current) when it is switching the level of the output (and charging or discharging the stray capacitance of the interconnect wiring) - which means the power supply load varies drastically with switching rate. TTL - the main contemporary competitor for ECL at the Cray 1's time - has a node with a pull-up resistor internally, which draws different current in one and zero state, and also pulls a big current spike when switching its output state because the pull-up and pull-down transistors are both on momentarily.
I built a P[D]P11 "supercluster" and started with Xilinx. The hardware is great, but their software toolset is horrendous.
Ran into a similar problem a few years ago with a FPGA programming tool. (I think it was with Xilinx but I'd have to check.)
We were trying an FPGA implementation of a heavily-channelized ASIC, using the same trick Cray used for the peripheral processors in his CDC-6x00 series: One set of fast logic switching between N sets of states stored in a N-word memory to handle N channels (in his case, to produce 10 independent "peripheral processors" running at the memory speed, which was 1/10th the logic speed, using only one set of logic and ten memories). You have to MUX a lot of stuff every clock, depending on how each channel is configured. So you have a LOT of MUXes / a very wide MUX to be driven by the sequencer, requiring a lot of fanout on the signals driving the MUX selects.
Turns out the FPGA compiler assumed MUXes were only switched occasionally and for configuration. So it didn't honor the drive strength modifiers and add extra drivers or distribute the MUX selection inputs across multiple buffers to get the MUXes to switch fast enough to do a different logic configuration every clock.
Couldn't get 'em to fix the compiler in a timely fashion (or at all) so the project was shelved.