The bill bans the sale (and, I think, import for sale) rather than the manufacture of 100W general purpose bulbs that don't meet the required efficiency standards. So, I guess this would spark a "War On Light Bulbs" - perhaps we will eventually have a Light Bulb Enforcement Agency to squash the illicit selling of illegal light bulbs by profiteering hoarders.
(So, since the sale is the illegal part, I assume that manufacture will stop well before the deadline so the supply chain can be drained by 1/1/2012. Better stock up EARLY (for personal use only of course)).
CFLs do contain mercury, but it is only a problem if the bulbs are broken or disposed of improperly. The amount of mercury in an average CFL is about 1/100 the amount contained in one of those old mercury thermometers. Also, the largest man-made source of mercury pollution is coal-fired power plants, which CFL usage will reduce. In addition, newer CFLs are being released that contain much less mercury than current ones AFAIK, this (and other bullet points) are fairly accurate, but this Hg comparison is interesting.
Home-use Hg fever thermometers are a relic now due to the Hg issue (not sure if this was by legislation or consumer interest). However, if the Hg in thermometers was a risk and CFLs have 1% of that amount, there seems to be cause for concern. Was the risk of Hg in thermometers over-hyped, or is the risk of Hg in CFLs being glossed over at the alter of Global Warming?
I'm quite certain I will consume more than 100x as many CFLs in my life as I would have Hg fever thermometers. Also, a broken fever thermometer is small (I could store all the ones I have or would have broken in my life in the space of a pencil) so they are easy to store for proper disposal while CFLs are larger and therefore more difficult (read, less likely) to be stored for proper disposal. Also, it's very obvious that a broken Hg thermometer has Hg in it (the silvery blobs is a pretty good hint) and that maybe special care is appropriate but the same can't be said of CFLs.
Ah, so we will see a sharp increase in sales of vibration resistant bulbs around, oh, perhaps 2012. I predict a technology race to reduce the manufacturing cost of vibration resistant bulbs that can be built on old bulb manufacturing lines. The law of unintended consequences will make the U.S. a leading innovator in vibration resistant bulbs.
(And, here in California, the earth moves enough we need vibration resistant bulbs)
I'm not a Ron Paul supporter, but it seems he would cut back on military so that at least would give him a source of savings to make room for some tax cuts w/o increasing the deficit. Also, he seems like a cantankerous guy who might be willing to just shut the government down using his veto power if needed to control spending.
Or, he could just sell out like most do within minutes of taking the oath of office...
A new CEO of RH using four versions of Linux at home is interesting. But, if I were an investor, I would prefer that he ran one or one version of Linux, a recent (preferably a server version) version of Windows, a recent version of Mac (ideally Leopard), a version of Open Solaris, and perhaps something out of the mainstream. RH's competitor is not other Linux version, it's Windows, Solaris, and perhaps AIX (and, on the desktop, perhaps Mac).
I was responding more generally to the notion that tazers are inappropriate because officers should respond in teams (apparently in teams of six) and wrestle knife wielding "fools" to the ground instead of using a tazer.
As I noted, there are tazer abuses and, perhaps, the Vancouver case is an example of this. I would not presume to have an opinion on this specific case without seeing at least a detailed impartial analysis and the evidence (and maybe even then I would conclude that there was insufficient reliable information to form a firm opinion). Certainly, however, I could not form an opinion based solely on the fact that (1) the weapon was a table and (2) the tazer was used less then 30 seconds after the initial threat. For example, if the middle aged man was in good shape (or appeared to be anything but feeble - police actually don't have a chance to administer strength tests and EKGs to a suspect before responding to a physical threat by the suspect), the attack was in a narrow hallway where only one officer could access the suspect, and the suspect was swinging the table vigorously while approaching the officer rapidly, the use of a tazer might be completely appropriate.
I never claimed that police officers are "likely to die" on the job. I asserted that police should be able to pursue a lifelong career "on the street" with an expectation that they will have a very small chance of dying on the job and that expecting police to engage in hand-to-hand combat with a knife wielding "fool" instead of using a tazer (or, I suppose, just killing the person since if a tazer is unavailable, that's the realistic option) is inconsistent with that expectation.
And, yes, I would believe that swinging a table at police in an threatening manner in an enclosed area at close range is working pretty hard to evoke a response from the police that is sufficient to stop that behavior without putting a police officer at substantial risk of personal injury.
"one fool with a knife" (especially one who turns out to be on meth) may eventually be subdued by six officers. However, one of those police officers may not go home to their family that night (or, perhaps ever again). Knife wounds can be very lethal and can be inflicted in the initial stage of combat before the suspect is under control. Being a police officer is a job (and, not a great paying one at that). A police officer has no obligation to put themselves at substantial risk of dying just to make life more comfortable for a "fool with a knife". A police officer who stays "on the street" in a high crime area in the United States for 25 years can't take much risk on each encounter or the odds are that they will die "on the job" since they may encounter a situation where, without a tazer, they would need to engage in physical combat with "fools with knives" a few times each year (esp. as the bad guys figured out that the risk of being tazered or shot if they threatened officers with knives was minimal). Deciding to be a police officer should not be a "death sentence". We spend a lot of effort to protect workers in other hazardous jobs, police officers deserve the same.
As well, even if six officers attempt to subdue a suspect with a knife instead of tazering them, some percentage of those suspects will be injured or killed by accident also (ranging from broken bones, paralysis, or death by various means - esp. if the suspect has a medical weakness of some sort).
The notion that police can travel in "packs" of six in case it's necessary to detain a belligernet individual is absurd. Note that when someone is pulled over for a traffic stop and pulls out a knife, the officer in the car doesn't have time to call for backup - (s)he's got to deal with the knife now if the suspect is coming closer to them because backup is minutes away, the suspect is one or two seconds away at most. The notion that, for example, the California Highway Patrol can/should stop having one-person cars and instead have (perhaps?) minivans full of six officers is not cost effective and will result in a reduced ability to respond to situations such as a report of a drunk driver OR about six times the cost of labor.
It seems fairly rare that someone gets tazered when they are following an officer's instructions. Also, I suspect it's rare that being belligerent or threatening an officer actually makes one less likely to be arrested or detained. Hence, it's just stupid to be belligerent or threatening to an officer. There are obviously excesses here and there, but in the vast majority of the tazer cases I've heard of, the recipient worked pretty hard to get tazered.
Note that I'm not defending all uses of tazers (it's hard to imagine, for example, why a handcuffed suspect should be tazered) and policies, training, and disciplinary action should control the use of tazers a bit more than they are now.
Loosening the tinfoil a bit... ah, there, feels much better... crawling out of basement... ah, there, the view is much clearer from up here... (but, what is that big glowing yellow/orange thing the sky - that is truly terrifying looking...)
Shutting down the ability to withdraw funds for six months for this reason would also require preventing transfers and check payments for the same supposed reason. Doing this would, by itself, probably destroy the entire economy of any modern commerce based society so it would make no sense. It would be like committing suicide to prevent getting a cold.
What you describe isn't a problem with welfare, it's a problem with some people who abuse the system.
Perhaps, but that means that "welfare" programs are structured incorrectly - there will always be greedy lazy people and we can't hope to change that, so we need to change the program to exclude those people from it.
If able bodied welfare recipients had to work 40 hours a week for the government (community service, picking up trash in parks, cleaning public restrooms, performing janitorial services in city/county buildings, weeding by the side of the road, providing daycare for children of other welfare recipients who are off working for their checks etc) in order to get their check, I suspect there would be less requests for checks. And who knows, for those that still choose to participate, perhaps the work experience would help them get a job in the private sector (such as working for a janitorial service). Obviously some flexibility would be appropriate to accommodate time off for bona fide job interviews.
If the existing USPTO surplus is adequate (or fairly close to it) to fund a sufficiently fully qualified and larger examiner force, I'm happy that's the case and retract the notion that a significant increase in patent fees might be appropriate to fund such an effort. Although I was aware of the surplus, I don't have a good grasp of its degree vs. the cost of increasing the qualifications and size of the examiner force. I would of course want to refund the past fees that patent seekers paid w/o receiving corresponding services from the USPTO, but that's another issue for an ethical Congress to address -- and I'm not holding my breath that such a Congress will exist or choose to refund such fees.
My point about installment payments was just that if substantial fee increases were necessary, perhaps the patent seeker would not need to pay the increased cost of the examination up front and if, in the end, they decided to abandon the patent to the public domain because they couldn't figure out how to exploit it, the cost to the USPTO for approving the patent might be eaten by either the taxpayers or others who file for new patents.
However, I see no reason that a "time and materials" based fee schedule should not be implemented. If I want a public record, I pay based on research and reproduction costs -- and that's just for me to get access to "public records" - why should commercially oriented services offered by the government be any less?
(BTW, I work in software development in the Valley and some years ago seriously considered switching to law, but realized I would, by nature, end up representing the completely right, but outmatched, "little guy" against big corp/govt interests and wasn't willing to take a BIG pay cut to do that (esp. since if all I got were "actual damages" for my client as opposed to absurd "punitive" or "pain and suffering" damages, I don't think I would be able in good conscious to collect my contingency) -- yep, I'm not as principled as I'd like to be. Also, I was perplexed by why the LSAT was difficult and was concerned that such an easy test would be used as ANY sort of filter to weed out the other future Lawyers I'd be working with/against - I would assume that engineers ace the LSAT normally but have never seen any stats on that.)
This would in turn put up the tax bill for the average american
Or, raise patent fees - perhaps including a "time and materials" component which would vary depending on the skills and effort required by the patent office. Those patents that required an uberexaminer with doctorates in three seemingly unrelated fields and 20 years of commercial experience would be expensive - but that seems okay. It seems likely that some of the most novel patents (slapping head saying "dang, that's really clever and useful, why didn't I think of that?") actually would require the least effort/cost to grant -- and these are the sort of patents that seem to be the best for society as they reflect true innovation and insight.
Perhaps part (1/2?) of the fees above some limit could be paid on an installment basis over the life of the patent (possibly with interest) - and the patent holder could decide to stop making the payments and thereby irrevocably release the patent into the Public Domain. On the positive side, this would help the "little guy" a bit as some of the expenses could be deferred or, even never incurred. It would also open up patents more quickly to the Public Domain if the inventor ('s company) hadn't figured out how to commercially exploit the patent. Also, on the positive side it would tend to reduce, over time, the number of active patents and thereby reduce the search effort (for both the applicant and the patent office) and also remove some of the mines from the minefield that technology companies walk every day. On the negative side, some of the expenses of patent applications would eventually be borne by the taxpayer or by amortizing the ultimately unrecovered costs across other patent applications.
And claming that a HD STB costs $1200 is bullshit too. If it was true, the cable would loose money for every customer they signed on. By the time they made back the $1200 (16 years at $6/mo), the STB would need to be replaced. And don't forget the $5/month in interest (assuming about 5% per year) for that $1200. So at a $6/month charge, only about $1 from the $6 is left for depreciation in order to break even. So, the STB would have to last for 1200/(1*12) = 100 years for this to happen. Only 60 years ago the first practical transistor is demonstrated so
if history repeats itself, that STB might be a bit long in the tooth by the time the cable company broke even in 100 years at $6/month.
On the bright side, they are testing out a pre-grant peer review process so that the public can submit prior art before any claims are allowed. Still, it requires affirmative action on the publics part. That's pretty neat - anyone have a link to a good summary of this?
But, won't all those FOSS developers who get distracted by searching for prior art result in a reduction of the amount/quality/functionality of FOSS?
Perhaps it's already been done - and we are looking at the result.
Maybe the police will get 100 reports of "hey, that's Mr. Smith, the third grade teacher down at my kid's school" so they pick Smith up and make his life hell for six months......until they can't find any collaborating evidence and realize Chester is really much smarter than they thought. Because, Chester is a really smart image processing expert who has developed breakthrough algorithms to use pictures of Smith gathered from public sites (using insecure wireless etc. to maintain anonymity) from multiple angles, built a 3d model of both Smith's and his own head and then to locate his head in every pic/frame and substitute Smith's head for his (making alterations to account for angle, lighting, other objects etc,). Oh, and then twirling the result to amuse himself watching the police salivate over their own brilliance and buy some time to slip away while police are focusing on Mr. Smith.
Although, given the misinformation on this thread, I think it's safe to say that Chester isn't on/. (or is he and he expects the police will therefore rule him out as insufficiently clever to pull this off even if they start to get close to him?). However, maybe all the top image processing experts should be put on the "possible suspect" list.
In the unlikely case that Chester gets the last laugh on this one, let's hope it's really his last laugh because he dies laughing (and when his mom is going through his stuff in the basement, she finds and reports his stash of originals so there is some closure).
The internet is so new here, this county only got cable modems within the last two years. Are they selling well? I would think people would return them when they can't find anything sticking out of the wall that fits the connector on the back (and that attempts to hook it to the mains seems to just create smoke and sparks)?
2) [...] hash the information. [...] Should you wish to confirm your votes were accuratly recorded, go to www.x y z.com and enter this number. I've been an advocate of this sort of solution for some time. However, it won't work if the voter can verify their vote w/o insuring that others are not permitted to view the "counted" vote (else, we are right back to enhancing vote buying and/or coercion).
Although with more thought I'm sure a better system could be found, one approach would be to encrypt the vote selection (and database record number to later match with the actual recorded vote) on the receipt. One pass of encryption would be done with a voter supplied key/passphrase (not appearing on the receipt and to be retained only by the voter) and probably some salt (which may need to appear on the receipt in cleartext). A second pass would be done with an 'n-of-m' scheme using all the individual public keys of a panel of m "trusted" individuals (probably judges). The voter would always have the option of excluding the encrypted vote from their receipt (making it impossible for them to ever verify how they voted, but making voters more confident of the anonymity and confidentiality of their vote). The voter's registration number would also have to be on the receipt in cleartext.
If a voter desired to validate the recording of their vote, they would go to a "vote validation" site at publicly announced times after the count is announced. At these sites, at least n of the m panel members would be present (where n isn't so close to m that death or illness or forgotten keys of a few panel members would make it impossible to verify votes - but n is large enough to give confidence that collusion among panel members would be difficult). The vote validation sites would be constructed (physically and electronically) to insure privacy of the voter (so only they could see how their vote was recorded on their receipt and in the counting database) and the privacy of the panel members' keys. Upon showing appropriate picture id which (along with the registration list) shows that the person requesting verification and the receipt match, they would use a secure terminal to scan their receipt (which would verify it was the same receipt that was just matched to their id), enter their private key or passphrase. Simultaneously, n of the panel members would enter their private keys. Then, the system would decrypt the the vote recorded on their receipt and show it to the voter (only). If voter then wanted to (which they presumably usually would), the "recorded vote" would be retrieved from the database (since the record number is now available) and displayed to them. If any inconsistencies were found the computer would give the voter the OPTION (but not requirement) to initiate a review process at a higher level (high level state and/or Federal level judges probably) as election fraud is obvious (since the review process would necessarily result in some of those involved in the review process being aware of the actual votes recorded/cast by the contesting voter, the review must be optional).
There would probably have to be a procedure to limit the number of vote validations in order to keep the cost and delay down. Perhaps if more than some small percentage of the voters request the validation of their vote, some would be selected randomly to be validated and each party would be allowed to select a number (based on the percentage of total voters registered with the party) of "validation" requesters whose requests would be honored.
If I'm the mightiest tech employer [MTE] in the U.S. and I say I'm paying too much for you and you are earning $30k/yr, now what?
Umm... The MTE will end up staffed with workers that feel they are worth $30K/yr - the rest will work for competitors looking for higher quality employees. If it turns out that the MTE can figure out how to effectively harness 400 inferior $30K/yr employees to produce more/better work than 100 higher quality $120K/yr employees, they will remain the MTE, if not, they will end up only being the answer to a trivia question. (Yes, I simplified by assuming overhead per employee for office space, equipment, health care etc scales linearly with salary even though it rarely would).
Actually, this test is not as good as it may seem because the specification is ambiguous - there are several materially different outputs (and hence code) that are valid depending on how the candidate resolves these ambiguities.
I would consider a program producing any one of the valid outputs to be acceptable after verifying that the output matched the candidate's understanding of the problem. The only "good" answer is for the candidate to notice the ambiguity and request clarification (or, if the test is not interactive, include some comments that indicate there are ambiguities and which way the candidate choose to resolve them) before starting to write code. I'd much rather hire someone who makes a minor error in the coding after noticing the ambiguity and explicitly resolving it than someone who writes "correct" code without noticing the ambiguity. The first candidate will probably get the code correct before checking it in, the second candidate would likely check in the code never realizing they hadn't implemented what was really required. Undetected ambiguities and incorrect assumptions lead to big problems while coding errors are easy to catch (via compiler, desk check, unit test, peer code review) and much easier to fix once discovered. IMHO, the person who notices the ambiguity up front may qualify as an "Engineer", the one who codes a solution without noticing it is likely only a mechanic (i.e., coder).
In case there are any/. readers who are not engineers, perhaps additional clarification will be helpful...
The problem statement from codinghorror is: Write a program that prints the numbers from 1 to 100. But for multiples of three print "Fizz" instead of the number and for the multiples of five print "Buzz". For numbers which are multiples of both three and five print "FizzBuzz".
The output that seems "most popular" (and, based on the comments on the codinghorror site, the "intended" one) is of the following form:
1
2
Fizz
4
Buzz
[...]
14
FizzBuzz
16
[...]
However, an equally valid sequence would start something like:
1
2
Fizz
4 5 Buzz
[...]
The preceding is quite reasonable because the scope of the phrase "instead of the number" in the second sentence of the specification is unclear but it seems to apply only to those numbers that are multiples of 3 and NOT to those that are multiples of 5. If the number '5' was not supposed to be printed out (along with 'Buzz'), the second sentence would have better read (although even this is ambiguous for reasons that will become obvious when considering what to output for '15' -- unfortunately fixing the specification sentence by sentence is difficult): But for multiples of three or five, print "Fizz" or "Buzz" respectively instead of the number.
This reworded sentence is shorter and more precise so it's quite reasonable to assume the author of the specification probably didn't mean this - else they would have written the shorter and more precise form (or some equivalent).
The third sentence is even more problematic than the second. Is it a request to add a string to the output for numbers that are multiples of 5 and 3 or to replace the previously specified output? If one assumes replacement, one gets the output identified as "most popular" above, but this seems incorrect given that the author of the specification carefully included the word instead in the second sentence but not in the third sentence and arbitrarily introducing such a construct where it does not exist is a pure flight of fantasy.
A more defensible output fragment would be as follows (if one make
US traffic lights vary - seemingly on a regional basis. In urban areas in California, for example, traffic signals usually have black or rather dark green cases - and virtually never yellow ones. In some other parts of the country (and in some more rural areas of California) yellow cases seem to be more common. I don't know why this is - perhaps the yellow cases make the signals more visible in some adverse weather conditions?
(The signals I was specifically discussing happen to have dark cases rather than yellow ones.)
However, there are times when the sunlight reflected off the moon is insufficient to provide meaningful illumination. Without street lighting in these cases, you get no advance warning of anything beyond the range of your headlights -- which can increase risk over a situation where there are some fixed sources of artificial light to provide some visibility beyond your headlights.
In the area I live, there are some stretches of freeway where there are no street lights and virtually no human presence (business or residential) near the freeway to cast artificial light on the freeway. When deer are (for reasons only a deer would understand - and I'm not sure even they do) standing in the middle of the freeway at 3AM just because a car hasn't driven by for five minutes, it's quite a bit more exciting on the "moonless" nights because by the time your headlights pick them up, you need to react very quickly to avoid hitting them (especially when, for reasons known only to themselves, they are distributed across all three or four lanes). Obviously one needs to drive at speeds within the limits of one's headlights, but a minor distraction or moment of inattention becomes substantially riskier if you have nearly zero time to react due to a lack of lighting beyond your headlights.
Not to say that there should be street lights on these stretches of road (it would be absurdly wasteful actually), but it seems likely it would be "safer" in at least some situations (sorry, but I'm skeptical that limited properly designed street street lighting on these stretches would reduce safety in any way that would offset the safety benefits).
Slightly off topic, but... Some of those horrid yellow lights have another little problem which could probably be dealt with but isn't in some places. That problem is that they are so close to the color and intensity of a yellow traffic light that they "camouflage" yellow traffic lights.
On one major street that I drive on at nights sometimes, you have to learn to remember where you saw a green light and, when you glance back (from scanning the sides of the road and your mirrors like one should), remember where it was - because if it's turned yellow, you can't quickly pick it out from all the yellow street lights - it just looks like suddenly a signal that you recall having seen is no longer there... that is until it "suddenly" turns red and results in people jamming on their brakes.
Once you learn to remember where the signals are because you drive the street regularly, it's not so bad - but it's quite disconcerting the first few times and certainly detracts attention from other elements of your driving even once you learn to compensate for it.
It's critically important that a manager establishes a culture and expectations. These expectations must be fair, they must be clear, they must be consistently applied in hiring, work assignments, resource planning, firing, transferring, and performance evaluation decisions and be exemplified by the manager's own actions on a day-to-day, minute-by-minute basis. The level of flexibility an individual manager has in this regard is of course related to their level (counted from the TOP, not from the BOTTOM, of the org chart!) - but if you can't do what is right, reject the position and keep coding - far too many technical people who SHOULDN'T become managers do, and it's painful for all around them.
It's a manager's job to not only insulate, but to stand up (to their manager, their manager's manager, the CEO etc) when, for example, it's important to keep from launching on an impossible task. An effective development manager must always put their team before themselves. An effective manager must be willing to sacrifice their own job and career if the alternative is to sacrifice the team (it turns out, if done responsibly, rarely does one have to actually sacrifice their job - although their career path is a completely different question).
Rarely will you go wrong by your team if you first ask "will my team members follow me to another company because they respect, although not always agree with, me" before asking "what is the impact of this on MY career/job".
Also, a first level manager must always remember that everyone is unique. In reality, you're not going to make big changes in the behavior of senior developers -- either figure out how to exploit their strengths or help advance their career with a transfer or dismissal if they are never going to be happy working for you. Also, mentor the heck out of junior developers -- that's part of your job and your responsibility and they should appreciate it.
Help your people from getting into trouble... Increase their estimates if they are notoriously optimistic, add time for additional desk checking/code review if they are quick but a bit sloppy, step between them and YOUR management without hesitation to keep them out of a nasty situation.
A manager must embrace and USE the process (until they are successful at getting the stupid parts of the process eliminated if necessary). It's there, it's sometimes in your way -- so you might as well exploit it -- if a Senior PHB wants grossly detailed breakdowns of estimates to justify your rollups, start tossing in the time for every little thing so the schedule only swells - that will usually shut upper management up.
Put everything you communicate (especially upwards) in writing and review this stream periodically to both understand yourself better and to use as future ammunition if needed. If you reach a verbal agreement, follow up with an email -- such as: "To recap what we discussed earlier today, it's my understanding that the performance of function X in standby mode is unimportant so our team will not spend time on performance analysis of X in standby mode. Please let me know if this decision changes in the future.".
Although a component of the job is technical, your FIRST job is as a manager.
Oh, and stop reading/. -- you won't have time anymore!
If you make this market truly competitive, then there will be interest. Now, granted the price will necessarily be higher, and that's where the main objection from people living out in the rural area comes from. But at least there will be service for a price.
Without subsidies, it will not make economic sense to run "wired" (truly high speed) broadband infrastructure to some rural areas due to the high cost per subscriber.
If the cost just to finance the infrastructure build-out divided by the number of eligible subscribers exceeds some value, the build-out isn't feasible because as the minimum possible price (something higher than the cost of servicing the debt on the infrastructure build-out) increases, a smaller and smaller percentage of the potential subscribers will sign up, thereby driving the per subscriber cost yet higher, causing fewer willing subscribers and yet higher prices and so on... In addition, the risk of future competition (such as from a wireless solution which is not yet practical, but may be feasible in ten years) drawing away some customers makes it even less likely that a business would invest in such a build-out since it would require charging yet more for service to factor in the risk that the infrastructure's life will be shortened by future competition and/or obsolescence.
It may be that controlled monopolies, possibly with subsidies from "urban" subscribers (much as the telephone system did), will be required to get broadband everywhere in the U.S. (This isn't to say we should do this - I would probably be against all or most such plans).
Although I agree with your sentiments about the statue, I don't think that most people viewing a statue (which is really a work of art) would expect it to be an authoritative and precise rendition of the situation depicted. This seems a bit different than altering a photo in a substantive fashion (as opposed to, say, altering the contrast or "brightness" to attempt to compensate for the limited dynamic range of the original media). I kinda liked the idea of diversity in the statue but, since it was based on a specific photo (as opposed to an artist's creation), to modify it just to "adjust the diversity" seemed inappropriate.
(But don't get me started on those "false color" photos from the Hubble - an entire generation of children is going to grow up thinking space is very colorful and be very disappointed when they are jettisoned into it:) Of course, an alarming number of Americans think that dinosaurs and "cave men" roamed the earth together - perhaps because of cartoons and comics?)
Slashdot Mirror
The bill bans the sale (and, I think, import for sale) rather than the manufacture of 100W general purpose bulbs that don't meet the required efficiency standards. So, I guess this would spark a "War On Light Bulbs" - perhaps we will eventually have a Light Bulb Enforcement Agency to squash the illicit selling of illegal light bulbs by profiteering hoarders.
(So, since the sale is the illegal part, I assume that manufacture will stop well before the deadline so the supply chain can be drained by 1/1/2012. Better stock up EARLY (for personal use only of course)).
Home-use Hg fever thermometers are a relic now due to the Hg issue (not sure if this was by legislation or consumer interest). However, if the Hg in thermometers was a risk and CFLs have 1% of that amount, there seems to be cause for concern. Was the risk of Hg in thermometers over-hyped, or is the risk of Hg in CFLs being glossed over at the alter of Global Warming?
I'm quite certain I will consume more than 100x as many CFLs in my life as I would have Hg fever thermometers. Also, a broken fever thermometer is small (I could store all the ones I have or would have broken in my life in the space of a pencil) so they are easy to store for proper disposal while CFLs are larger and therefore more difficult (read, less likely) to be stored for proper disposal. Also, it's very obvious that a broken Hg thermometer has Hg in it (the silvery blobs is a pretty good hint) and that maybe special care is appropriate but the same can't be said of CFLs.
Ah, so we will see a sharp increase in sales of vibration resistant bulbs around, oh, perhaps 2012. I predict a technology race to reduce the manufacturing cost of vibration resistant bulbs that can be built on old bulb manufacturing lines. The law of unintended consequences will make the U.S. a leading innovator in vibration resistant bulbs.
(And, here in California, the earth moves enough we need vibration resistant bulbs)
Or, he could just sell out like most do within minutes of taking the oath of office...
A new CEO of RH using four versions of Linux at home is interesting. But, if I were an investor, I would prefer that he ran one or one version of Linux, a recent (preferably a server version) version of Windows, a recent version of Mac (ideally Leopard), a version of Open Solaris, and perhaps something out of the mainstream. RH's competitor is not other Linux version, it's Windows, Solaris, and perhaps AIX (and, on the desktop, perhaps Mac).
Not to worry... They will be assembled with Security TORX screws - absolutely no chance an unauthorized person could open one up.
I was responding more generally to the notion that tazers are inappropriate because officers should respond in teams (apparently in teams of six) and wrestle knife wielding "fools" to the ground instead of using a tazer.
As I noted, there are tazer abuses and, perhaps, the Vancouver case is an example of this. I would not presume to have an opinion on this specific case without seeing at least a detailed impartial analysis and the evidence (and maybe even then I would conclude that there was insufficient reliable information to form a firm opinion). Certainly, however, I could not form an opinion based solely on the fact that (1) the weapon was a table and (2) the tazer was used less then 30 seconds after the initial threat. For example, if the middle aged man was in good shape (or appeared to be anything but feeble - police actually don't have a chance to administer strength tests and EKGs to a suspect before responding to a physical threat by the suspect), the attack was in a narrow hallway where only one officer could access the suspect, and the suspect was swinging the table vigorously while approaching the officer rapidly, the use of a tazer might be completely appropriate.
I never claimed that police officers are "likely to die" on the job. I asserted that police should be able to pursue a lifelong career "on the street" with an expectation that they will have a very small chance of dying on the job and that expecting police to engage in hand-to-hand combat with a knife wielding "fool" instead of using a tazer (or, I suppose, just killing the person since if a tazer is unavailable, that's the realistic option) is inconsistent with that expectation. And, yes, I would believe that swinging a table at police in an threatening manner in an enclosed area at close range is working pretty hard to evoke a response from the police that is sufficient to stop that behavior without putting a police officer at substantial risk of personal injury.
"one fool with a knife" (especially one who turns out to be on meth) may eventually be subdued by six officers. However, one of those police officers may not go home to their family that night (or, perhaps ever again). Knife wounds can be very lethal and can be inflicted in the initial stage of combat before the suspect is under control. Being a police officer is a job (and, not a great paying one at that). A police officer has no obligation to put themselves at substantial risk of dying just to make life more comfortable for a "fool with a knife". A police officer who stays "on the street" in a high crime area in the United States for 25 years can't take much risk on each encounter or the odds are that they will die "on the job" since they may encounter a situation where, without a tazer, they would need to engage in physical combat with "fools with knives" a few times each year (esp. as the bad guys figured out that the risk of being tazered or shot if they threatened officers with knives was minimal). Deciding to be a police officer should not be a "death sentence". We spend a lot of effort to protect workers in other hazardous jobs, police officers deserve the same.
As well, even if six officers attempt to subdue a suspect with a knife instead of tazering them, some percentage of those suspects will be injured or killed by accident also (ranging from broken bones, paralysis, or death by various means - esp. if the suspect has a medical weakness of some sort).
The notion that police can travel in "packs" of six in case it's necessary to detain a belligernet individual is absurd. Note that when someone is pulled over for a traffic stop and pulls out a knife, the officer in the car doesn't have time to call for backup - (s)he's got to deal with the knife now if the suspect is coming closer to them because backup is minutes away, the suspect is one or two seconds away at most. The notion that, for example, the California Highway Patrol can/should stop having one-person cars and instead have (perhaps?) minivans full of six officers is not cost effective and will result in a reduced ability to respond to situations such as a report of a drunk driver OR about six times the cost of labor.
It seems fairly rare that someone gets tazered when they are following an officer's instructions. Also, I suspect it's rare that being belligerent or threatening an officer actually makes one less likely to be arrested or detained. Hence, it's just stupid to be belligerent or threatening to an officer. There are obviously excesses here and there, but in the vast majority of the tazer cases I've heard of, the recipient worked pretty hard to get tazered.
Note that I'm not defending all uses of tazers (it's hard to imagine, for example, why a handcuffed suspect should be tazered) and policies, training, and disciplinary action should control the use of tazers a bit more than they are now.
Loosening the tinfoil a bit... ah, there, feels much better... crawling out of basement... ah, there, the view is much clearer from up here... (but, what is that big glowing yellow/orange thing the sky - that is truly terrifying looking...)
Shutting down the ability to withdraw funds for six months for this reason would also require preventing transfers and check payments for the same supposed reason. Doing this would, by itself, probably destroy the entire economy of any modern commerce based society so it would make no sense. It would be like committing suicide to prevent getting a cold.
If able bodied welfare recipients had to work 40 hours a week for the government (community service, picking up trash in parks, cleaning public restrooms, performing janitorial services in city/county buildings, weeding by the side of the road, providing daycare for children of other welfare recipients who are off working for their checks etc) in order to get their check, I suspect there would be less requests for checks. And who knows, for those that still choose to participate, perhaps the work experience would help them get a job in the private sector (such as working for a janitorial service). Obviously some flexibility would be appropriate to accommodate time off for bona fide job interviews.
If the existing USPTO surplus is adequate (or fairly close to it) to fund a sufficiently fully qualified and larger examiner force, I'm happy that's the case and retract the notion that a significant increase in patent fees might be appropriate to fund such an effort. Although I was aware of the surplus, I don't have a good grasp of its degree vs. the cost of increasing the qualifications and size of the examiner force. I would of course want to refund the past fees that patent seekers paid w/o receiving corresponding services from the USPTO, but that's another issue for an ethical Congress to address -- and I'm not holding my breath that such a Congress will exist or choose to refund such fees.
My point about installment payments was just that if substantial fee increases were necessary, perhaps the patent seeker would not need to pay the increased cost of the examination up front and if, in the end, they decided to abandon the patent to the public domain because they couldn't figure out how to exploit it, the cost to the USPTO for approving the patent might be eaten by either the taxpayers or others who file for new patents.
However, I see no reason that a "time and materials" based fee schedule should not be implemented. If I want a public record, I pay based on research and reproduction costs -- and that's just for me to get access to "public records" - why should commercially oriented services offered by the government be any less?
(BTW, I work in software development in the Valley and some years ago seriously considered switching to law, but realized I would, by nature, end up representing the completely right, but outmatched, "little guy" against big corp/govt interests and wasn't willing to take a BIG pay cut to do that (esp. since if all I got were "actual damages" for my client as opposed to absurd "punitive" or "pain and suffering" damages, I don't think I would be able in good conscious to collect my contingency) -- yep, I'm not as principled as I'd like to be. Also, I was perplexed by why the LSAT was difficult and was concerned that such an easy test would be used as ANY sort of filter to weed out the other future Lawyers I'd be working with/against - I would assume that engineers ace the LSAT normally but have never seen any stats on that.)
Perhaps part (1/2?) of the fees above some limit could be paid on an installment basis over the life of the patent (possibly with interest) - and the patent holder could decide to stop making the payments and thereby irrevocably release the patent into the Public Domain. On the positive side, this would help the "little guy" a bit as some of the expenses could be deferred or, even never incurred. It would also open up patents more quickly to the Public Domain if the inventor ('s company) hadn't figured out how to commercially exploit the patent. Also, on the positive side it would tend to reduce, over time, the number of active patents and thereby reduce the search effort (for both the applicant and the patent office) and also remove some of the mines from the minefield that technology companies walk every day. On the negative side, some of the expenses of patent applications would eventually be borne by the taxpayer or by amortizing the ultimately unrecovered costs across other patent applications.
But, won't all those FOSS developers who get distracted by searching for prior art result in a reduction of the amount/quality/functionality of FOSS?
Perhaps it's already been done - and we are looking at the result.
...until they can't find any collaborating evidence and realize Chester is really much smarter than they thought. Because, Chester is a really smart image processing expert who has developed breakthrough algorithms to use pictures of Smith gathered from public sites (using insecure wireless etc. to maintain anonymity) from multiple angles, built a 3d model of both Smith's and his own head and then to locate his head in every pic/frame and substitute Smith's head for his (making alterations to account for angle, lighting, other objects etc,). Oh, and then twirling the result to amuse himself watching the police salivate over their own brilliance and buy some time to slip away while police are focusing on Mr. Smith.
/. (or is he and he expects the police will therefore rule him out as insufficiently clever to pull this off even if they start to get close to him?). However, maybe all the top image processing experts should be put on the "possible suspect" list.
Maybe the police will get 100 reports of "hey, that's Mr. Smith, the third grade teacher down at my kid's school" so they pick Smith up and make his life hell for six months...
Although, given the misinformation on this thread, I think it's safe to say that Chester isn't on
In the unlikely case that Chester gets the last laugh on this one, let's hope it's really his last laugh because he dies laughing (and when his mom is going through his stuff in the basement, she finds and reports his stash of originals so there is some closure).
[...]
Should you wish to confirm your votes were accuratly recorded, go to www.x y z.com and enter this number. I've been an advocate of this sort of solution for some time. However, it won't work if the voter can verify their vote w/o insuring that others are not permitted to view the "counted" vote (else, we are right back to enhancing vote buying and/or coercion).
Although with more thought I'm sure a better system could be found, one approach would be to encrypt the vote selection (and database record number to later match with the actual recorded vote) on the receipt. One pass of encryption would be done with a voter supplied key/passphrase (not appearing on the receipt and to be retained only by the voter) and probably some salt (which may need to appear on the receipt in cleartext). A second pass would be done with an 'n-of-m' scheme using all the individual public keys of a panel of m "trusted" individuals (probably judges). The voter would always have the option of excluding the encrypted vote from their receipt (making it impossible for them to ever verify how they voted, but making voters more confident of the anonymity and confidentiality of their vote). The voter's registration number would also have to be on the receipt in cleartext.
If a voter desired to validate the recording of their vote, they would go to a "vote validation" site at publicly announced times after the count is announced. At these sites, at least n of the m panel members would be present (where n isn't so close to m that death or illness or forgotten keys of a few panel members would make it impossible to verify votes - but n is large enough to give confidence that collusion among panel members would be difficult). The vote validation sites would be constructed (physically and electronically) to insure privacy of the voter (so only they could see how their vote was recorded on their receipt and in the counting database) and the privacy of the panel members' keys. Upon showing appropriate picture id which (along with the registration list) shows that the person requesting verification and the receipt match, they would use a secure terminal to scan their receipt (which would verify it was the same receipt that was just matched to their id), enter their private key or passphrase. Simultaneously, n of the panel members would enter their private keys. Then, the system would decrypt the the vote recorded on their receipt and show it to the voter (only). If voter then wanted to (which they presumably usually would), the "recorded vote" would be retrieved from the database (since the record number is now available) and displayed to them. If any inconsistencies were found the computer would give the voter the OPTION (but not requirement) to initiate a review process at a higher level (high level state and/or Federal level judges probably) as election fraud is obvious (since the review process would necessarily result in some of those involved in the review process being aware of the actual votes recorded/cast by the contesting voter, the review must be optional).
There would probably have to be a procedure to limit the number of vote validations in order to keep the cost and delay down. Perhaps if more than some small percentage of the voters request the validation of their vote, some would be selected randomly to be validated and each party would be allowed to select a number (based on the percentage of total voters registered with the party) of "validation" requesters whose requests would be honored.
Umm... The MTE will end up staffed with workers that feel they are worth $30K/yr - the rest will work for competitors looking for higher quality employees. If it turns out that the MTE can figure out how to effectively harness 400 inferior $30K/yr employees to produce more/better work than 100 higher quality $120K/yr employees, they will remain the MTE, if not, they will end up only being the answer to a trivia question. (Yes, I simplified by assuming overhead per employee for office space, equipment, health care etc scales linearly with salary even though it rarely would).
Actually, this test is not as good as it may seem because the specification is ambiguous - there are several materially different outputs (and hence code) that are valid depending on how the candidate resolves these ambiguities.
/. readers who are not engineers, perhaps additional clarification will be helpful...
I would consider a program producing any one of the valid outputs to be acceptable after verifying that the output matched the candidate's understanding of the problem. The only "good" answer is for the candidate to notice the ambiguity and request clarification (or, if the test is not interactive, include some comments that indicate there are ambiguities and which way the candidate choose to resolve them) before starting to write code. I'd much rather hire someone who makes a minor error in the coding after noticing the ambiguity and explicitly resolving it than someone who writes "correct" code without noticing the ambiguity. The first candidate will probably get the code correct before checking it in, the second candidate would likely check in the code never realizing they hadn't implemented what was really required. Undetected ambiguities and incorrect assumptions lead to big problems while coding errors are easy to catch (via compiler, desk check, unit test, peer code review) and much easier to fix once discovered. IMHO, the person who notices the ambiguity up front may qualify as an "Engineer", the one who codes a solution without noticing it is likely only a mechanic (i.e., coder).
In case there are any
The problem statement from codinghorror is:
Write a program that prints the numbers from 1 to 100. But for multiples of three print "Fizz" instead of the number and for the multiples of five print "Buzz". For numbers which are multiples of both three and five print "FizzBuzz".
The output that seems "most popular" (and, based on the comments on the codinghorror site, the "intended" one) is of the following form:
1
2
Fizz
4
Buzz
[...]
14
FizzBuzz
16
[...]
However, an equally valid sequence would start something like:
1
2
Fizz
4
5 Buzz
[...]
The preceding is quite reasonable because the scope of the phrase "instead of the number" in the second sentence of the specification is unclear but it seems to apply only to those numbers that are multiples of 3 and NOT to those that are multiples of 5. If the number '5' was not supposed to be printed out (along with 'Buzz'), the second sentence would have better read (although even this is ambiguous for reasons that will become obvious when considering what to output for '15' -- unfortunately fixing the specification sentence by sentence is difficult):
But for multiples of three or five, print "Fizz" or "Buzz" respectively instead of the number.
This reworded sentence is shorter and more precise so it's quite reasonable to assume the author of the specification probably didn't mean this - else they would have written the shorter and more precise form (or some equivalent).
The third sentence is even more problematic than the second. Is it a request to add a string to the output for numbers that are multiples of 5 and 3 or to replace the previously specified output? If one assumes replacement, one gets the output identified as "most popular" above, but this seems incorrect given that the author of the specification carefully included the word instead in the second sentence but not in the third sentence and arbitrarily introducing such a construct where it does not exist is a pure flight of fantasy.
A more defensible output fragment would be as follows (if one make
US traffic lights vary - seemingly on a regional basis. In urban areas in California, for example, traffic signals usually have black or rather dark green cases - and virtually never yellow ones. In some other parts of the country (and in some more rural areas of California) yellow cases seem to be more common. I don't know why this is - perhaps the yellow cases make the signals more visible in some adverse weather conditions?
(The signals I was specifically discussing happen to have dark cases rather than yellow ones.)
However, there are times when the sunlight reflected off the moon is insufficient to provide meaningful illumination. Without street lighting in these cases, you get no advance warning of anything beyond the range of your headlights -- which can increase risk over a situation where there are some fixed sources of artificial light to provide some visibility beyond your headlights.
In the area I live, there are some stretches of freeway where there are no street lights and virtually no human presence (business or residential) near the freeway to cast artificial light on the freeway. When deer are (for reasons only a deer would understand - and I'm not sure even they do) standing in the middle of the freeway at 3AM just because a car hasn't driven by for five minutes, it's quite a bit more exciting on the "moonless" nights because by the time your headlights pick them up, you need to react very quickly to avoid hitting them (especially when, for reasons known only to themselves, they are distributed across all three or four lanes). Obviously one needs to drive at speeds within the limits of one's headlights, but a minor distraction or moment of inattention becomes substantially riskier if you have nearly zero time to react due to a lack of lighting beyond your headlights.
Not to say that there should be street lights on these stretches of road (it would be absurdly wasteful actually), but it seems likely it would be "safer" in at least some situations (sorry, but I'm skeptical that limited properly designed street street lighting on these stretches would reduce safety in any way that would offset the safety benefits).
Slightly off topic, but... Some of those horrid yellow lights have another little problem which could probably be dealt with but isn't in some places. That problem is that they are so close to the color and intensity of a yellow traffic light that they "camouflage" yellow traffic lights.
... that is until it "suddenly" turns red and results in people jamming on their brakes.
On one major street that I drive on at nights sometimes, you have to learn to remember where you saw a green light and, when you glance back (from scanning the sides of the road and your mirrors like one should), remember where it was - because if it's turned yellow, you can't quickly pick it out from all the yellow street lights - it just looks like suddenly a signal that you recall having seen is no longer there
Once you learn to remember where the signals are because you drive the street regularly, it's not so bad - but it's quite disconcerting the first few times and certainly detracts attention from other elements of your driving even once you learn to compensate for it.
It's critically important that a manager establishes a culture and expectations. These expectations must be fair, they must be clear, they must be consistently applied in hiring, work assignments, resource planning, firing, transferring, and performance evaluation decisions and be exemplified by the manager's own actions on a day-to-day, minute-by-minute basis. The level of flexibility an individual manager has in this regard is of course related to their level (counted from the TOP, not from the BOTTOM, of the org chart!) - but if you can't do what is right, reject the position and keep coding - far too many technical people who SHOULDN'T become managers do, and it's painful for all around them.
/. -- you won't have time anymore!
It's a manager's job to not only insulate, but to stand up (to their manager, their manager's manager, the CEO etc) when, for example, it's important to keep from launching on an impossible task. An effective development manager must always put their team before themselves. An effective manager must be willing to sacrifice their own job and career if the alternative is to sacrifice the team (it turns out, if done responsibly, rarely does one have to actually sacrifice their job - although their career path is a completely different question).
Rarely will you go wrong by your team if you first ask "will my team members follow me to another company because they respect, although not always agree with, me" before asking "what is the impact of this on MY career/job".
Also, a first level manager must always remember that everyone is unique. In reality, you're not going to make big changes in the behavior of senior developers -- either figure out how to exploit their strengths or help advance their career with a transfer or dismissal if they are never going to be happy working for you. Also, mentor the heck out of junior developers -- that's part of your job and your responsibility and they should appreciate it.
Help your people from getting into trouble... Increase their estimates if they are notoriously optimistic, add time for additional desk checking/code review if they are quick but a bit sloppy, step between them and YOUR management without hesitation to keep them out of a nasty situation.
A manager must embrace and USE the process (until they are successful at getting the stupid parts of the process eliminated if necessary). It's there, it's sometimes in your way -- so you might as well exploit it -- if a Senior PHB wants grossly detailed breakdowns of estimates to justify your rollups, start tossing in the time for every little thing so the schedule only swells - that will usually shut upper management up.
Put everything you communicate (especially upwards) in writing and review this stream periodically to both understand yourself better and to use as future ammunition if needed. If you reach a verbal agreement, follow up with an email -- such as: "To recap what we discussed earlier today, it's my understanding that the performance of function X in standby mode is unimportant so our team will not spend time on performance analysis of X in standby mode. Please let me know if this decision changes in the future.".
Although a component of the job is technical, your FIRST job is as a manager.
Oh, and stop reading
Without subsidies, it will not make economic sense to run "wired" (truly high speed) broadband infrastructure to some rural areas due to the high cost per subscriber.
If the cost just to finance the infrastructure build-out divided by the number of eligible subscribers exceeds some value, the build-out isn't feasible because as the minimum possible price (something higher than the cost of servicing the debt on the infrastructure build-out) increases, a smaller and smaller percentage of the potential subscribers will sign up, thereby driving the per subscriber cost yet higher, causing fewer willing subscribers and yet higher prices and so on... In addition, the risk of future competition (such as from a wireless solution which is not yet practical, but may be feasible in ten years) drawing away some customers makes it even less likely that a business would invest in such a build-out since it would require charging yet more for service to factor in the risk that the infrastructure's life will be shortened by future competition and/or obsolescence.
It may be that controlled monopolies, possibly with subsidies from "urban" subscribers (much as the telephone system did), will be required to get broadband everywhere in the U.S. (This isn't to say we should do this - I would probably be against all or most such plans).
Although I agree with your sentiments about the statue, I don't think that most people viewing a statue (which is really a work of art) would expect it to be an authoritative and precise rendition of the situation depicted. This seems a bit different than altering a photo in a substantive fashion (as opposed to, say, altering the contrast or "brightness" to attempt to compensate for the limited dynamic range of the original media). I kinda liked the idea of diversity in the statue but, since it was based on a specific photo (as opposed to an artist's creation), to modify it just to "adjust the diversity" seemed inappropriate.
:) Of course, an alarming number of Americans think that dinosaurs and "cave men" roamed the earth together - perhaps because of cartoons and comics?)
(But don't get me started on those "false color" photos from the Hubble - an entire generation of children is going to grow up thinking space is very colorful and be very disappointed when they are jettisoned into it