The continuous roll paper makes it simple to determine how each elector voted. There is a record of who voted in what order (signatures on a ledger, for example) and the paper record is in the same order. Line them up, and all traitors to the party can be identified and scheduled for "re-education".
The Subaru Justy was available with a Van Doorne CVT in the early '90s. This transmission did hold constant RPM: 2500 for standard economy/cruise operation, or 4000 in the "sport" mode/passing gear. This set-up was so effective you would never know the car had a tiny (1.2L?) 3-cyliner engine.
It works great until the CFOs see fourth-quarter projections below their MBO bonus targets. Then you'll get laid off from all those jobs and need to find another. About your resume, sir. How did you get laid off six times in the last three months?... We'll be in touch.
Spam will exist as long as those who outfit the spammers can make money doing it. Mailing lists, bulk mail software, web hosts who will ignore any AUP for money, email laundering through cracked PCs: these guys make money from spam, lots of it. They don't care if the spammer makes money or not, as long as they get paid. If you want a real solution to spam go after the wholesale suppliers, not the small fry dealers.
Absorption doesn't have that much to do with the wavelength. The dye molecules in printing ink are a factor 1000 smaller than the wavelength of visible light, but that does not hinder them from absorbing light quite effectively.
Dye color is a different phenomenon. The energy carried by one photon of visible light (h*nu) is sufficient to move electrons from one orbital to another. The energy in one photon of 1.5MHz RF is far less; there is eight and a half orders of magnitude difference: 5.3e-20J for light at 600nm (480THz), 1.6e-28J for RF at 200m (1.5MHz). Compare this to Boltzmann thermal energy: kT = 4.1e-21J at 300K. A photon of light exceeds thermal energy by an order of magnitude; the 1.5MHz RF photon is seven and a half orders of magnitude less than thermal energy. One visible light photon is sufficient to change chemical bonds stable at room temperature (thus extended exposure to bright light fades the dye). But one RF photon has negligible probability of causing such a change (a 1K temperature increase is much more significant). Thus a large flux of RF photons (high power density) is needed to accumulate enough energy to have a measurable effect, whether by thermal or postulated non-thermal means.
The reason wavelength matters is because the human body is a more effective "receiving antenna" for wavelengths where body features are near 1/4 wavelength, 1/2 wavelength, etc. That's why the FCC exposure limits http://a257.g.akamaitech.net/7/257/2422/14mar20010 800/edocket.access.gpo.gov/cfr_2002/octqtr/pdf/47c fr1.1310.pdf are lowest between 30MHz (10m) and 300MHz (1m). Higher exposure is allowed at frequencies where the body is a poorer antenna (549V/m at 1.5MHz vs. 27.5V/m at 30-300MHz). Current, voltage, and resulting heat within the body are higher (for a given field strength) at wavelengths where body parts form resonant antennas; bulk absorption is not the only mechanism at work at those wavelengths.
I don't think that standing next to a 50 kW transmitter is very healthy.
Agreed! The usual textbook formula (R=sqrt(30*EIRP)/E, MKS units) isn't valid in the near field of the transmitting antenna, and may provide a false sense of security. For AM broadcast transmitters, the near field is about 30-100m radius. 100m also is about as close as you should get to a 50KW (EIRP) FM broadcast antenna.
The trouble with the examples is that the spacing is too small for the point-source (1/r^2) model. The size of radiating antenna is significant compared to the separation. In RF jargon, you're in the "Near Field". The actual exposure (electric field strength and magnetic field strength) can be either higher or lower, depending on the current and voltage distribution in the antenna.
In any case, the amount of power the human body absorbs from a 1500kHz AM signal is phenomenally small. The body is small compared to the signal wavelength (2m/200m=0.01 wavelength), which means it absorbs almost none of the radiated power. The only way it is likely to be a hazard is if you touch a conductor with considerable RF voltage on it. That could give you an RF burn.
The re-allocation of spectrum world wide is constrained by treaty. All nations follow the agreements of the International Telecommunications Union (ITU), a UN agency. What flexibility the nations have is limited to signals that stay within their borders. In practice, this means that any spectrum use that overlaps an authorized satellite link or that could propogate over the horizon (i.e., all below 30MHz) is subject to the ITU plans. Until and unless the ITU adopts the "commons" principle, it will see limited use.
Ask those who were coding a decade or so earlier. More than 80 characters won't fit on a punch-card. That leaves you 72 usable columns, after taking out the continuation character and reserved line number columns. Teletypes and early VDTs had the same limitation. I'm not saying there's anything sacred about 80 columns, just that it originally was a hardware limitation.
Security is a good reason for using the minimum connection range, but there's another that's just as important. The available spectrum is limited, prospects for increased spectrum are long range at best (not to mention that your existing system can't use it), and the user base is growing. For broadband wireless to remain useful, all users must copperate so channels can be re-used a short distance away, just as cell phones do. The alternative is a chaotic war zone where productive activity is impossible: observe what happened to the 27MHz CB radio spectrum. Making a long distance link is an impressive achievement, but comes at the expense of many others who also need to use the channel. Make your range sufficient for the job, no more. Don't hog shared resources!
That points to the big problem with software patents. They became available only after the technology had been developed for many decades. Because prior art searches center on previously issued patents, patents are mistakenly issued for old ideas. If OS/360, MULTICS, early unix, TOPS-10, CP/M, and MS/DOS is considered, how much of the kernel was not anticipated by these works? Every one of them is at least 20 years old, and therefore would be in the public domain, had they been patentable in their day.
The late change in patentability also destroys the most affordable protection available to financially limited small businesses: expired patents. The least risky path is to select a good, expired patent and follow its claims exactly. This strategy is not perfect, but it is strong: find prior art in the public domain, then copy it faithfully!
Gasoline engines for street cars already run at cooler combustion temperatures than they can withstand. This is to reduce nitrogen oxide (Nox) emissions, which increase greatly at high combustion temperatures. The emissions control system uses combinations of intake dilution (exhuast gas recirculation), reduced compression ratios, and retarded ignition timing to limit peak temperatures, and thus control Nox emissions. So while ceramic would allow higher temperatures, that isn't an issue for emissions-controlled engines.
It's more correct to say that some vacuum tube amplifiers employ only local negative feedback, in the form of cathode degeneration resistors. But many well-regarded designs also use feedback all the way from the output transformer secondary to the input stage.
The most common solid state designs employ feedback around the entire amplifier. But some use only local feedback, and some a combination of both methods (I recall the SWTP "Universal Tiger" being one of these).
In any case, arguing too long and loud over amplifier distortion distracts you from the goal, which is getting the best performing system for your money. The big sonic variables are the electromechanical elements: loudspeakers, microphones, and their interactions with the environment you put them in. Today's amplifiers range from decent to superb; there is much more variation in loudspeaker quality. Speakers are your most important audio purchase.
Microsoft's "midlife crisis" is no surprise. The days of double-digit compound growth couldn't go on forever. Essentially everybody in western countries who needs or wants a computer already has one (or two, or a garage full). The upgrade market and third-world market penetration aren't large enough to maintain the exponential growth they enjoyed during the glory days, when companies in a booming economy bought PCs loaded with Windows and Office for every employee.
Now we have a comparatively stagnant economy, where companies are postponing the computer and software upgrades. The ROI on computer systems must be demonstrated; these days bean counters don't automatically approve any purchase request with "computer" written on it.
Microsoft has had a longer winning streak than any other PC business. But once you take the top spot in the market, it's difficult to find enough new market opportunity to maintain that momentum. Passive, defensive business practices (cost cutting) gain favor over active, productive ones ("betting the company" on an all-new product). Microsoft probably won't go away, but they're already not the company we once knew. People grow old; companies do too. The hard part is doing so gracefully.
It also can negate the call-the-cops option. In some places, using "fighting words" meets the legal definition of "disorderly conduct". If you call in the law at that point, both you and the employee may be needing bail bonds!
If you agree the extended warranty is an insurance policy, then you can make a case for some policies, some of the time. Insurance means paying money to reduce your risk exposure. On average it is a loss, but you may be willing to pay extra to avoid a catastrophic expense. It makes the most sense on high cost items with significant risk of very expensive repairs.
For an X-Box or standard TV set, it makes very little sense. The service contract is a high percentage of the replacement cost, you can live without one for a while, and the out-of-warranty failure rate is very low.
For automobiles, it may be worth it (depending on complexity and that make's durability). In the case of my wife's FWD V6 mini-van, one look under the hood convinced me. There's no way I would attempt anything beyond routine maintenance at home. Access to major engine components (heads, manifolds, etc.) requires dropping the entire power train, which means dismantling the front suspension. It took roughly 8 hours of shop time to replace a leaky intake gasket. Required tools included a coil spring compressor and a wheel alignment rack. The insurance company lost money on that bet.
But for my 4 cylinder RWD pick-up, it would be silly. Everything is easily accessible for DIY repairs, the parts don't cost much, and it has none of the expensive-to-fix options. I can afford to assume the foreseeable repair costs, so I won't pay the insurance company to assume that risk.
The manufacturers anticipated this, and use (admittedly less-than-perfect) countermeasures. While the details vary from make to make, the following scheme is typical:
The key fob and auto receiver use "rolling codes". Once the vehicle accepts a code as valid, it increments to the next code in the sequence. Likewise, each time the fob is pressed it increments to the next code in the sequence. The vehicle allows, say, +5/-0 codes in the sequence to allow for missed button presses. It will never open for the same code twice in succession. That defeats simple replay attacks, though if the attacker knows the code sequence alorithm that does allow attackers a toe-hold.
If you use two fobs, there is a separate code sequence for each one. If you run out of retries or otherwise get out of synch with the vehicle, you must (shudder!) use the actual metal key to open the door.
The 900MHz band (and 2.4GHz, 27Mhz CB, and more) are allocated to Part 18 ISM equipment on a primary basis in the US. These "Industrial, Scientific, and Medical" devices include diathermy heating machines, RF induction heaters, microwave ovens, and similar high-power devices that aren't for communication. Any other device operating on allocated ISM bands must accept (and expect) interference from the ISM equipment. Ironically the ISM bands are good places for unlicensed users, because the extremely high power commercial and military users aren't authorized to use the ISM "no-man's land" frequencies. ISM equipment has maximum field strength limits, while broadcasters must meet required minimum signal levels specified for their class of license.
No matter what frequency they're on, Part 15 devices are the lowest class users. Basically the Part 15 rules say "Buyer beware!" (though in many more words, as is the habit of government agencies). So it's good engineering for the Part 15 designers to consider all other uses of the spectrum avaialble for their product, and select the frequencies where they will be most compatible.
Somehow, I can't see how military radar bands would meet that criteria!
Negative video has it right. Coatings make rework a pain. Over-spray on contact pins is a major source of "NTF" failures. That's why conformal coatings are used where they're needed, not on evrything under the sun.
It's called "conformal coating". It protects circuits against humidity, conductive contamination, fungus, and other environmental hazards. There are two big reasons PC manufacturers don't use it:
Anything that adds cost and isn't absolutely essential doesn't go into a product with merciless cost pressure. Consumer-grade PCs will last for years in home use without the conformal coating, so they don't have it.
Anything that increases production time and isn't essential doesn't go into a high volume product. Applying and curing the coating takes additional time, space, and equipment. A fume hood or spray booth is needed to control fumes during application, and a UV curing "oven" is desirable for high volume products (air drying takes much longer).
Conformal coating is widely used in military products, and also is in some industrial products that must survive severe environments. But they are rarely used in office equipment, and I've never seen it in a mass market PC.
The Hummers (both original and macho-'burb H2) have one characteristic that makes them poor choices for serious primitive-road(*) use: they're too big! Too wide to fit established tracks, too long to negotiate tight corners, and too tall to clear low-hanging tree limbs. Like the jacked-up monster truck wannabes, they're at home striking a pose on Main Street--not in the boondocks. There's a reason hunters prefer old Jeep CJs, Blazers, Broncos, and Land Cruisers: they get you to the back country, and back home again, without a three-figure "retrieval" operation.
(*)The ads like to talk about "off-road", but the powers-that-be will skin you alive for doing it. "Beyond this point, vehicles are permitted only on marked routes." Tread lightly: it's not just a good idea, it's the law!
You hit it right on the head! The perimeter should be sealed up tight. The operators/hosts of the compromised sites should have been notified immediately, and they should immediately shut down the infected servers; rip out the network and power cables if that's what it takes! It is inexcusably negligent to leave a system on-line when it is causing this much harm.
If they continue to operate their web sites, knowing that it is distributing malicious code, how is that any different from deliberate sabotage? Mitnick went to prison for causing less damage; why should anyone who doesn't take prompt and decisive action to halt their infected server (upon receiving credible reports that it is aiding a criminal act) be treated any differently?
If the servers were properly shut down, identifying them poses no risk to the public. The real danger is if these zombies are allowed to remain in operation!
If you must use cubicles, at least include...
on
Building a Better Office
·
· Score: 2, Interesting
Storage space, and lots of it. You can't RTFM if there's no place to keep TFM. Modular cubicles with built-in "efficient furniture" typically have one small (often non-locking) book shelf. That is utterly inadequate for technical staff.
Glare-free lighting. For me it's not how much or little light, but how it hits the screen and desk. A hot spot on the screen or a overhead light aimed at the eyeglasses is a migraine waiting to happen.
Flexible arrangement. What works for my body may not suit yours. Monitor and keyboard positions change depending on eye height, length of arms, bifocals vs. 20/20 vision, etc. Frequently used items will be arranged differently for the sin vs. dex vs. amidextrous workers. Another strike against one-size-fits-all modular furnishings.
QUIET telephone ringers, perhaps completely muted (use a "ring" light instead). You don't need to hear it from down the hall, that's why you have voice mail and a cell phone. How I miss the days of Western Electric phones, when you could stuff tissue into your neighbor's ringer and kill the stinking noise!
Single cubicles only; no double bunking. Even if you practice ExP, allow everybody some own personal space. When it's team programming time, pull up an extra chair (another real chair, not some broken-down reject from the lunch room).
Dilbert clippings hanging on every available wall. It's not a geek department without them!
The suggestions omit these important requirements, mostly for mailing list operations:
All email lists must be created by confirmed opt-in, without exception.
All subscriptions are for one mailing list only. They may never be used for any other purpose, including "related" mailings from the same organization and automatic subscription transfers to a "replacement" list if the list folds. If the subscriber didn't explicitly ask for it, it's junk mail!
Unsubscribe requests must be processed through plain-text email to an address in the originating domain.
Users should be taught to never use any URLs in unsolicited or suspect email, and that all such links should be considered malicious. In particular, "click here to remove" is to be considered synonymous with "attack my PC".
Legitimate mailing lists messages never include scripts, attachments, or executable content of any description. If executables are part of the list's charter, they will be accessed only by user-initiated download, never by attachment or automatic invokation.
I could go on much longer, but that covers my most common gripes!
Slashdot Mirror
The continuous roll paper makes it simple to determine how each elector voted. There is a record of who voted in what order (signatures on a ledger, for example) and the paper record is in the same order. Line them up, and all traitors to the party can be identified and scheduled for "re-education".
The Subaru Justy was available with a Van Doorne CVT in the early '90s. This transmission did hold constant RPM: 2500 for standard economy/cruise operation, or 4000 in the "sport" mode/passing gear. This set-up was so effective you would never know the car had a tiny (1.2L?) 3-cyliner engine.
It works great until the CFOs see fourth-quarter projections below their MBO bonus targets. Then you'll get laid off from all those jobs and need to find another. About your resume, sir. How did you get laid off six times in the last three months?... We'll be in touch.
It's much more likely a plot to gather creditinformation, bank account numbers, SSN, income, etc. In other words, a phishing scam.
Spam will exist as long as those who outfit the spammers can make money doing it. Mailing lists, bulk mail software, web hosts who will ignore any AUP for money, email laundering through cracked PCs: these guys make money from spam, lots of it. They don't care if the spammer makes money or not, as long as they get paid. If you want a real solution to spam go after the wholesale suppliers, not the small fry dealers.
People prefer going back to the main page to find the link to the next part of the article, instead of having the obvious "next" button.
Dye color is a different phenomenon. The energy carried by one photon of visible light (h*nu) is sufficient to move electrons from one orbital to another. The energy in one photon of 1.5MHz RF is far less; there is eight and a half orders of magnitude difference: 5.3e-20J for light at 600nm (480THz), 1.6e-28J for RF at 200m (1.5MHz). Compare this to Boltzmann thermal energy: kT = 4.1e-21J at 300K. A photon of light exceeds thermal energy by an order of magnitude; the 1.5MHz RF photon is seven and a half orders of magnitude less than thermal energy. One visible light photon is sufficient to change chemical bonds stable at room temperature (thus extended exposure to bright light fades the dye). But one RF photon has negligible probability of causing such a change (a 1K temperature increase is much more significant). Thus a large flux of RF photons (high power density) is needed to accumulate enough energy to have a measurable effect, whether by thermal or postulated non-thermal means.
The reason wavelength matters is because the human body is a more effective "receiving antenna" for wavelengths where body features are near 1/4 wavelength, 1/2 wavelength, etc. That's why the FCC exposure limits http://a257.g.akamaitech.net/7/257/2422/14mar20010 800/edocket.access.gpo.gov/cfr_2002/octqtr/pdf/47c fr1.1310.pdf are lowest between 30MHz (10m) and 300MHz (1m). Higher exposure is allowed at frequencies where the body is a poorer antenna (549V/m at 1.5MHz vs. 27.5V/m at 30-300MHz). Current, voltage, and resulting heat within the body are higher (for a given field strength) at wavelengths where body parts form resonant antennas; bulk absorption is not the only mechanism at work at those wavelengths.
I don't think that standing next to a 50 kW transmitter is very healthy.
Agreed! The usual textbook formula (R=sqrt(30*EIRP)/E, MKS units) isn't valid in the near field of the transmitting antenna, and may provide a false sense of security. For AM broadcast transmitters, the near field is about 30-100m radius. 100m also is about as close as you should get to a 50KW (EIRP) FM broadcast antenna.
In any case, the amount of power the human body absorbs from a 1500kHz AM signal is phenomenally small. The body is small compared to the signal wavelength (2m/200m=0.01 wavelength), which means it absorbs almost none of the radiated power. The only way it is likely to be a hazard is if you touch a conductor with considerable RF voltage on it. That could give you an RF burn.
The re-allocation of spectrum world wide is constrained by treaty. All nations follow the agreements of the International Telecommunications Union (ITU), a UN agency. What flexibility the nations have is limited to signals that stay within their borders. In practice, this means that any spectrum use that overlaps an authorized satellite link or that could propogate over the horizon (i.e., all below 30MHz) is subject to the ITU plans. Until and unless the ITU adopts the "commons" principle, it will see limited use.
Ask those who were coding a decade or so earlier. More than 80 characters won't fit on a punch-card. That leaves you 72 usable columns, after taking out the continuation character and reserved line number columns. Teletypes and early VDTs had the same limitation. I'm not saying there's anything sacred about 80 columns, just that it originally was a hardware limitation.
Security is a good reason for using the minimum connection range, but there's another that's just as important. The available spectrum is limited, prospects for increased spectrum are long range at best (not to mention that your existing system can't use it), and the user base is growing. For broadband wireless to remain useful, all users must copperate so channels can be re-used a short distance away, just as cell phones do. The alternative is a chaotic war zone where productive activity is impossible: observe what happened to the 27MHz CB radio spectrum. Making a long distance link is an impressive achievement, but comes at the expense of many others who also need to use the channel. Make your range sufficient for the job, no more. Don't hog shared resources!
The late change in patentability also destroys the most affordable protection available to financially limited small businesses: expired patents. The least risky path is to select a good, expired patent and follow its claims exactly. This strategy is not perfect, but it is strong: find prior art in the public domain, then copy it faithfully!
Gasoline engines for street cars already run at cooler combustion temperatures than they can withstand. This is to reduce nitrogen oxide (Nox) emissions, which increase greatly at high combustion temperatures. The emissions control system uses combinations of intake dilution (exhuast gas recirculation), reduced compression ratios, and retarded ignition timing to limit peak temperatures, and thus control Nox emissions. So while ceramic would allow higher temperatures, that isn't an issue for emissions-controlled engines.
The most common solid state designs employ feedback around the entire amplifier. But some use only local feedback, and some a combination of both methods (I recall the SWTP "Universal Tiger" being one of these).
In any case, arguing too long and loud over amplifier distortion distracts you from the goal, which is getting the best performing system for your money. The big sonic variables are the electromechanical elements: loudspeakers, microphones, and their interactions with the environment you put them in. Today's amplifiers range from decent to superb; there is much more variation in loudspeaker quality. Speakers are your most important audio purchase.
Now we have a comparatively stagnant economy, where companies are postponing the computer and software upgrades. The ROI on computer systems must be demonstrated; these days bean counters don't automatically approve any purchase request with "computer" written on it.
Microsoft has had a longer winning streak than any other PC business. But once you take the top spot in the market, it's difficult to find enough new market opportunity to maintain that momentum. Passive, defensive business practices (cost cutting) gain favor over active, productive ones ("betting the company" on an all-new product). Microsoft probably won't go away, but they're already not the company we once knew. People grow old; companies do too. The hard part is doing so gracefully.
It also can negate the call-the-cops option. In some places, using "fighting words" meets the legal definition of "disorderly conduct". If you call in the law at that point, both you and the employee may be needing bail bonds!
For an X-Box or standard TV set, it makes very little sense. The service contract is a high percentage of the replacement cost, you can live without one for a while, and the out-of-warranty failure rate is very low.
For automobiles, it may be worth it (depending on complexity and that make's durability). In the case of my wife's FWD V6 mini-van, one look under the hood convinced me. There's no way I would attempt anything beyond routine maintenance at home. Access to major engine components (heads, manifolds, etc.) requires dropping the entire power train, which means dismantling the front suspension. It took roughly 8 hours of shop time to replace a leaky intake gasket. Required tools included a coil spring compressor and a wheel alignment rack. The insurance company lost money on that bet.
But for my 4 cylinder RWD pick-up, it would be silly. Everything is easily accessible for DIY repairs, the parts don't cost much, and it has none of the expensive-to-fix options. I can afford to assume the foreseeable repair costs, so I won't pay the insurance company to assume that risk.
The key fob and auto receiver use "rolling codes". Once the vehicle accepts a code as valid, it increments to the next code in the sequence. Likewise, each time the fob is pressed it increments to the next code in the sequence. The vehicle allows, say, +5/-0 codes in the sequence to allow for missed button presses. It will never open for the same code twice in succession. That defeats simple replay attacks, though if the attacker knows the code sequence alorithm that does allow attackers a toe-hold.
If you use two fobs, there is a separate code sequence for each one. If you run out of retries or otherwise get out of synch with the vehicle, you must (shudder!) use the actual metal key to open the door.
No matter what frequency they're on, Part 15 devices are the lowest class users. Basically the Part 15 rules say "Buyer beware!" (though in many more words, as is the habit of government agencies). So it's good engineering for the Part 15 designers to consider all other uses of the spectrum avaialble for their product, and select the frequencies where they will be most compatible. Somehow, I can't see how military radar bands would meet that criteria!
Negative video has it right. Coatings make rework a pain. Over-spray on contact pins is a major source of "NTF" failures. That's why conformal coatings are used where they're needed, not on evrything under the sun.
Anything that adds cost and isn't absolutely essential doesn't go into a product with merciless cost pressure. Consumer-grade PCs will last for years in home use without the conformal coating, so they don't have it.
Anything that increases production time and isn't essential doesn't go into a high volume product. Applying and curing the coating takes additional time, space, and equipment. A fume hood or spray booth is needed to control fumes during application, and a UV curing "oven" is desirable for high volume products (air drying takes much longer).
Conformal coating is widely used in military products, and also is in some industrial products that must survive severe environments. But they are rarely used in office equipment, and I've never seen it in a mass market PC.
(*)The ads like to talk about "off-road", but the powers-that-be will skin you alive for doing it. "Beyond this point, vehicles are permitted only on marked routes." Tread lightly: it's not just a good idea, it's the law!
If they continue to operate their web sites, knowing that it is distributing malicious code, how is that any different from deliberate sabotage? Mitnick went to prison for causing less damage; why should anyone who doesn't take prompt and decisive action to halt their infected server (upon receiving credible reports that it is aiding a criminal act) be treated any differently?
If the servers were properly shut down, identifying them poses no risk to the public. The real danger is if these zombies are allowed to remain in operation!
Storage space, and lots of it. You can't RTFM if there's no place to keep TFM. Modular cubicles with built-in "efficient furniture" typically have one small (often non-locking) book shelf. That is utterly inadequate for technical staff.
Glare-free lighting. For me it's not how much or little light, but how it hits the screen and desk. A hot spot on the screen or a overhead light aimed at the eyeglasses is a migraine waiting to happen.
Flexible arrangement. What works for my body may not suit yours. Monitor and keyboard positions change depending on eye height, length of arms, bifocals vs. 20/20 vision, etc. Frequently used items will be arranged differently for the sin vs. dex vs. amidextrous workers. Another strike against one-size-fits-all modular furnishings.
QUIET telephone ringers, perhaps completely muted (use a "ring" light instead). You don't need to hear it from down the hall, that's why you have voice mail and a cell phone. How I miss the days of Western Electric phones, when you could stuff tissue into your neighbor's ringer and kill the stinking noise!
Single cubicles only; no double bunking. Even if you practice ExP, allow everybody some own personal space. When it's team programming time, pull up an extra chair (another real chair, not some broken-down reject from the lunch room).
Dilbert clippings hanging on every available wall. It's not a geek department without them!
All email lists must be created by confirmed opt-in, without exception.
All subscriptions are for one mailing list only. They may never be used for any other purpose, including "related" mailings from the same organization and automatic subscription transfers to a "replacement" list if the list folds. If the subscriber didn't explicitly ask for it, it's junk mail!
Unsubscribe requests must be processed through plain-text email to an address in the originating domain.
Users should be taught to never use any URLs in unsolicited or suspect email, and that all such links should be considered malicious. In particular, "click here to remove" is to be considered synonymous with "attack my PC".
Legitimate mailing lists messages never include scripts, attachments, or executable content of any description. If executables are part of the list's charter, they will be accessed only by user-initiated download, never by attachment or automatic invokation. I could go on much longer, but that covers my most common gripes!