#1 I did the TurboTax software on the web. Doesn't it work under a Linux browser?
Yes, but there are other reasons it isn't right for everybody. It all depends on your own tax circumstances. For me, in the past few years with an S corporation, various depreciating assets, etc., the dumbed-down web-based version just didn't hack it. The full offline "Turbo-Tax Business" version probably would have, but I ended up using an accountant anyway, since it was too complex.
But I no longer own a business, so next year I'll revert to doing it myself. I probably will use the online TurboTax.
#2 Wouldn't it be easier just to have a tax preparer do them for $40, or is your time worth that little?
More like $350 or so. Once again, it all depends upon your circumstances. If your taxes are real simple -- salary income, a few simple investments, and a handful of itemized deductions, you might be able to find a $40 preparer. But, if your taxes are that simple, it would be pretty trivial to just complete the forms yourself, too.
However, lead-free solder isn't much more difficult to work with (at least as an elecronics hobbiest).
For factory production, basic soldering, e.g. simple wave soldering or single-sided reflow SMT is probably not much more difficult with unleaded than with leaded solder. Where it gets exciting is when you do fancy stuff.
Some soldering processes are pretty sophisticated. Consider a board with reflow SMT components on both sides, some of them using solderball or solder-column attachment, areas of selective through-hole wave solder, and areas with layered (interposer) board attachment.
The manufacturing engineers do some pretty amazing stuff. By careful selection of which solder alloys are used where (i.e. different melting points), and very tight control of thermal profiles, they can get seemingly impossible things to happen. To me, it's black magic that they can reflow the top side of a PCB without the components already on the bottom falling off.
So, they depend upon not just one solder, but a variety of different alloys with different properties. Trying to re-engineer a complex process to switch to all-new (lead free) alloys is far from trivial.
Another issue is that the higher temperatures may stress parts more during attachment. This can lead to higher fallout, shorter component life, etc. Not my specialty, but some QA folks tell me this can cause problems.
Despite all these difficulties, I'm a fan of eliminating lead from the process, and am glad to see it finally happening. It's the right thing to do.
Does your state [context: New Hampshire] have property tax? It is possible that your property tax is far higher than it is in California.
I lived in NH for several years. So, I'll have a crack at answering.
Yes, NH does have property taxes, but at the time I lived there (I left 10 years ago), they weren't excessive. Certainly nothing nearly as high as California, Colorado, or Massachusetts
NH really has no state income tax. NH does have sales tax, but on a very restricted basis -- it applies only to restaurant meals, hotel rooms, ski tickests, and perhaps a few other tourism-related things. It has high gas (petrol) taxes, about comparable to most other states in the US. Everything else in ones daily lives -- clothing, (non-restaurant) food, cars, etc., truly are sales-tax free. NH also have two major toll highways, but the only portion subject to toll is the first few miles beyond the Massachusetts border - once again, it's tourism focussed.
When I lived in New Hampshire, I paid about $1000/yr in property tax (I owned a small house). I paid, say, $500 a year in restaurant sales tax, because I ate out a lot. Maybe $50 a year in tolls. And I gas (petrol) taxes which I won't enumerate becasue they're about the same in all states. This totals, say, $1550/yr in taxes.
For comparison, when I lived in Massachusetts before that, I paid almost $5000/yr in state income tax alone, even before you get to the higher property taxes, and almost across-the-board sales taxes.
Massachusettes residents who defended all the monster Massachusetts taxes (yes, some people actually like them!) would always either (a) vaguely cite alleged benefits associated with living in a more "progressive" state, and (b) alledge that New Hampshire taxes were "just as high", because property taxes made up for the difference. The numbers above belie this claim.
I certainly never saw a single shred of evidence of truth in either of these two arguements. In terms of percentage of property values, New Hampshire property taxes were about comparable with those in a typical Massachusetts town. But in terms of absolute dollars, they were _much_ lower, because New Hampshire didn't suffer from Massachusetts's outrageous housing prices. I don't know how that's changed in recent years?
The bottom line, NH gets by without income and sales taxes, in two ways: By taxing tourism, and by keeping it's spending under control. I lived there enough to know that it REALLY does work.
I never saw any downside to the moderation in spending. The state legislature was part-time and unpaid. Essential services all seemed to work just fine. Road maintenence, snow clearing, hospitals, courts, etc., all seemed excellent. I can't comment on schools, since I didn't have kids.
Now if only I could live in New Hampshire, without having to be on the east coast!
I like the idea in principle, but see plenty of problems in it's practical impelementation.
As described, the proposed law would hard-code the concept of using icons to disclose this information. What about fundamentally non-graphical programs (drivers, daemons)? What about overall non-graphical environments (servers, embedded)?
I fear this scheme would further what is already an increasing problem: that everybody wants to attach a GUI to every program, even if it's totally inappropriate (e.g. printer drivers). The proliferation of spurious GUI interfaces leads to the proliferation of inappropriate design choices in exception reporting (pop-ups instead of log files), configuration methods, etc.
I'm not anti-GUI, by the way. I'm anti-inappropriate-GUI, and I fear hard-coding icon requirements into every piece of software makes this trend even worse. Immagine if every.deb or.rpm package in your Linux system had a spurious GUI component, just to comply with a well-intentioned but poorly-considered law!
On the other hand, I would definitely like to see these icons displayed on the labels of software packages and disks, or on the web pages that software is downloaded from.
Oh, and something the article didn't mention, but I'd propose this ammendment to the act: Make it hard to add any additional icons (i.e. to make the program behavior worse) in upgrades. If any icons are added, the vendor must either (1) continue to support the old version for future bug fixes, security patches, etc., or (2) refund the purchase price to buyers who choose not to continue using the product. (Obviously, there'd have to be a time limit, but long enough to prevent the use of "incrimental-spyware" as a bait-and-switch technique.)
I'm not denying that the lead reduction is real. It is real.
But this isn't anything unique to Intel, and it isn't done out of the goodness of their green little hearts.
Every IC manufacturer, in fact practically every manufacturer of anything electronic, is already investigating lead reduction or elimination at some level or other. Not all are making a public hoopla about it, though.
Lead free solder requires the development of new alloys and new processes. The changeover isn't trivial, but some promissing candidates exist. Typically they have very high tin content, plus some mix of Silver, Copper, and Antimony.
There are several reasons for this trend: Regulatory changes (pending in the US, and I think already passed in Europe?), Liability/Insurance cost (employee lawsuits), and waste treatment cost, including waste water.
My opinion: I don't beleive lead in electronics will ever be totally eliminated, nor outright outlawed. I'm no solder/process expert, but those I know tell me that leadless soldering presents many challenges. More likely in my opinion, regulations will take the form of taxes and fees on lead content, driving manufacturers to use it only where no good alternative exists.
You've got to have some system to identify which employees get what share of the loot. I'm not egalitarian enough to advocate the same raises and bonusses for top performers as for useless slugs. And I'm certainly not too thrilled with my own company's policy (all the loot goes to the CEO and the CEO's closest henchpeople).
When done right, a ranking system works just fine. And, in all but the smallest companies, it's pretty much unavoidable.
Notice that I refer to ranking in the context of pay and rewards, not layoffs. Once you use the ranking system to select layoff candidates, things go very badly awry.
Bad things companies do with ranking systems:
1. Firing all the people below some threshold.
2. Continuing to assume a gaussian curve, even after several rounds of layoffs.
3. Applying gaussian distribution to small populations, e.g. individual project teams, rather than large populations like the entire large corporation.
4. Applying the rank-and-fire methodology only to the rank-and-file. Too often the real problems are in upper management or exectutive levels, and those folks are invariably excluded from the system.
5. Letting prior year's ranking affect this year's ranking. i.e. judging whether to raise or lower a person's ranking vs. last year, rather than starting with a blank sheet of paper each year.
An inflexible "fire (at least) the bottom 5%" rule, as adopted by my employer, is just plain dumb. Apart from the obvious morale impact on the survivors, it just doesn't make sense. Do it once, you may get rid of dead wood. Do it twice, there's no dead wood left for the second round. After a while, you're firing really good people, just to meet a quota. (And using up prodigous amounts of toner, as your remaining good people print their resumes looking for a more rational company to employ them.)
The quota is completely fake, based on the mistaken assumption that when you lop off the tail of a gaussian curve, what's left behind is still a gaussian curve.
In my opinion, the root cause of this problem, as with so many corporate problems today, is executives and upper management that have only a short-term outlook. Short term, layoffs save money, and it really doesn't matter who gets the chop, or what effect it has on the morale, motivations, and behavior of the survivors. Long term, it may be disasterous, but corporate governence these days seems to care little about long term consequences of their activities.
A very weak rationalization by Google, in my opinion. It seems to be based upon the "method of dimensions" arguement, i.e. "if the dimensions come out right, the answer must be right". I don't buy it. The method of dimensions is a helpful but non-rigorous technique, and doesn't always correctly handle dimensionless constants.
Hertz can't just mean "anything" per second.
For starters, consider that minimally it must be restricted to mean "anything dimensionless per second". You can't measure velocity or power in Herz, even though both have units of "anything per second" (metres or joules respectively).
In fact, either by convention or definition, I suspect the former, the "anything" is further restricted to be a cycle of some periodically repeating event. One rotation of a roundabout. One cycle of a sine-wave. etc.
Travel one radian around the roundabout, and you don't end up back where you started. Travel 2*pi radians, and you do -- you've completed a cycle.
If Herz really meant "anything per second", why don't Intel just claim their Xeons are 12.4GHz instead of 2GHz? I hope Intel's marketing folks aren't listening.:-)
Google may have a rationalization for why their answer is "right", but any engineer who has ever worked with filters knows better.
A barrel roll, unlike a loop, can be performed without noticably exceeding the normal 1G.
Quite true, in many aircraft types.
slow rolls being done without the knowledge of passengers
Not quite true.
Don't confuse barrel rolls and slow rolls. A slow roll, properly executed, involves minus one G (a.k.a. hanging upside down from your seatbelt). It also involves significant sideways G during the knife-edge portions.
By the way, "slow" in "slow roll" isn't really a measure of the speed of the roll, only the type of roll. (And, roll-rate isn't a judging criterion in aerobatic competition.) The name came about because slow rolls are inherently slower than snap/flick rolls.
In a nutshell:
Barrel roll: Corkscrew-shaped flight path. Curvature of flight path during the inverted portion keeps the pilot feeling positive G force.
Slow roll: maintain a constant, level, undeviating flight path, while rotating around the longitudinal axis.
Snap/flick roll: Snap=US term, flick=UK term. Combined use of yaw and rapid pich change to cause one wing to stall (stop producing much lift). Rotation results from the unbalanced lift generated by the other wing. Rotation is usually quite rapid.
Aileron roll: Lazy pilot's slow roll. Like a slow-roll, but the flight path is a freefall-like parabola rather than a straight, level line, which makes it much easier to coordinate.
Rolling turn: Like a slow roll, but the flight path is a curved, turning, level path.
Only snap/flick, slow, and rolling turns are used in IAC sanctioned competition, other than in freestyle where anything goes,
Re:Not harassment
on
Beyond Pay?
·
· Score: 2, Insightful
Telling you to work overtime or you'll be fired is not harassment
Classic dilemma -- we have two vocabularies. Common usage, and legal definitions.
Obviously, this can be indeed be a form of harassment in the common usage of the term, but it's not harrasment by any legal definition of the word (in US law, anyway).
Reminds me of a discussion I had with an aviation liability lawyer. He told me that there's no such thing as a "frivolous lawsuit". I cited many examples, but he wouldn't budge on his claim that they simply do not exist. That's because I was using the English word "frivolous" (consult dictionary), wheras he was using the Legal word "frivolous" (consult definitions in laws and statutes).
I can't speak for the original poster, but I'll describe why this solution doesn't work for me.
I live in Colorado. Not sure if it's the same in other places, but here basic home phone service is nearly $40 a month. That's even before you pay for long distance, call time, anything. That's just flat monthly fees and taxes for having a dial tone.
No big deal, if you're amortizing that $40 over zillions of hours of calls. But, in my case I make very few calls. And I need a cell phone too. My call volume is so low, it makes sense to just use the cell for everything, and disconnect the home phone. I did this a few months ago, and it works great: the little extra airtime is a drop in the bucket compared to the $40/month saved.
The one problem I haven't solved yet: What to do about overseas calls? Most (all?) cell phone plans available here _either_ have totally outrageous charges for international calls, _or_ have high monthly fees, incompatible with my low call volume. And they all block the use of calling cards for international.
I have a cable modem, and my hope is that some VOIP solution will eventually solve my international calling dilemma. I'd be happy with outgoing-only VOIP service. (Incoming international calls go to my cell, and I keep the call short and tell the caller I'll call back on the VOIP.)
Some have suggested trading in cable modem for DSL plus land line. But that costs more where it is available, and isn't available where I live, anyway.
If anybody has any experience with low- or no- monthly fee VOIP that solves this problem, I'd love to hear about it.
... because they both (as typically implemented, anyway) start/stop each servide one at a time. Way too much unnecessary serialization!
A good system would explicitly know about dependencies, and would then concurrently start/stop everything for which the dependencies are already satisfied. That's what multi-tasking is for. Each time a daemon reports a successful start, all the other services whose dependencies are satisfied by it, would immediately be concurrently launched.
Problem is, no distribution seems to have standardized an implementation of this, although I've heard of experimental examples.
There's no good reason for computers to take forever to start and stop. We accept it because we're used to it. Aside from the fscking, the rest of the bootup should take only a few seconds.
They want to know what financial value people place on their privacy. Why?
I'd love to beleive that this is a purely academic social science study, but I'm not that naive. The rationale behind it must be to make money. Someone, somewhere is dreaming up a business model to make you and I pay for our privacy.
Imagine... Opt-outs, where you pay a fee not to have your personal details sold.
Sure, it's extortion. But, so are a lot of very common business practices today, unfortuntely. And governments pay lip service to privacy concerns, but big business can always buy the loopholes they want.
Don't think it will happen? Consider unlisted telephone numbers: you already pay a monthly fee for you privacy!
This study is all about putting a dollar value on privacy -- to help develop new ways to extort profit by selling, or more likely renting, our privacy back to us. Just wait and see.
I've interviewed and hired a lot of people. And, my choices have always turned out well. Admitedly, my experience may not be directly applicable, since I'm in hardware engineering, but I'd be surprised if it's that different in your area.
I seldom bother to ask a candidate what degrees and certificates they hold. And I never let the answer affect my final choice. I talk about engineering, problem solving, and experience. As for experience, I don't ask questions like "how many years doing X?" . I ask questions like "give a specific example of your experience of X".
I can name many very smart and talented people I've hired over the years, and I still don't even know whether they have a BS, MS, or PhD, nor from what school. On the rare occasions when the subject comes up (I don't bring it up), I see essentially no correlation between degree of formal education and the value of the engineer.
One caveat, here. I select candidates from the pool who already have a foot in the door. Perhaps lots of degrees and certificates are helpful in getting the foot in the door in the first place. In my current organization, some pre-screening is done by other interviewers before they get to me. So, I know the candidates all meet a certain baseline, including a BSEE degree. But the pre-screening lets an awful lot goofballs, fast-talkers, shmoozers, empty-shells and slugs through. And, if anything, the worst of the hopeless candidates frequently turn out to be (on paper) the best educated ones.
OK. But I don't grok the connection in your original post between encrypting and "inability to change stuff without a trail". Encryption without signature is deniable -- you can't prove the statement is authentic, so it's useless as a trail. With a signed document, you have proof that the trail is authentic, they can't deny it.
Original: Any company offering encrypted emails as a paperless billing solution? That way I have a copy of everything and they can't change stuff without a trail. I've got a PGP public key around somewhere...;)
Yeah, I know it's lousy writing, and pedantic, and repetitive, but... it's core theme is valid. So slog though it, and take it to heart.
Communicate with your customers, partners, suppliers, and employees in clear, simple, unambiguous, TRUE language. Don't mislead by omission. Don't obfuscate. Don't hide the truth in-amongst tons of verbiage. Make sure the letter of your message is the same as the spirit of your message. Don't hide behind legalities and technicalities. Say what you mean, and mean what you say.
Even if these numbers are too large, this still makes you think about how inefficient our cars are.
Several posters have countered with the suggestion that mother nature is inefficient, using so much plant material to make so little fuel.
But, both the "cars are inefficeint" and "nature is inefficient" arguements miss one important point: That the huge amount of biomass was spread out over millions of years of growth, with the vast majority of the material being recycled from one growth generation to the next. Obviously, just by virtue of the fact that a gallon of petrol weighs a lot less than a small forrest, we must conclude that most of the material didn't become fuel. Most of it became fertilizer/compost, and fueled the next generation of growth.
Adding up the mass of all these generations of plant growth is really just repeatedly counting the same material over and over.
1. USB subsystem. Maybe it's just me, but I've never been able to make it work reliably.
2. ATAPI support for CD burners. The whole passthrough driver concept is a mess. Especially when something goes wrong, and error recovery can't clean up properly because it is an application program like cdrecord tracking the drive state, a function which should reside in an OS driver. Yuck.
3. Sound card support. Probably works fine if you know how to set it up, but I've always found it to be the hardest thing to get right, and the least automated in terms of hardware discovery.
4. Distros that are a little "too helpful", and install stuff that doesn't work, but just gets in the way. Nautilus and autofs, for example.
5. Dependency nightmares. OK, it's probably an intractible problem, since I would reather run a very conservative stable Linux, yet install one or two bleeding edge packages. (Why don't bleeding edge projects build statically linked archive binaries by default, until they get mature enough to be inculded in stable distros?)
I'm not knocking Linux, though. Numbers 2, 4, and 5, especially, are far worse in windows. Still, I'd love to see these fixed.
I'm far from an expert on the subject, so someone please stop me if I've got my facts wrong. But I think this is accurate...
I beleive it's smarter not to use 128bit, but rather use 64 bit (which is really 40 bit encryption, by the way, since 14 of them are not random).
Here's why.
WEP provides no real privacy. The algorithm has been cracked wide open, and there are readily available exploits. Also, the known exploits scale linearly, so using twice as many bits only gives a 2x increase in crack time. In short, forget WEP as a privacy measure -- it isn't one.
So, why use WEP at all? It's still worth using, in my opinion, but for different reasons. Despite the lack of protection against deliberate attack, WEP does provide protection against accidental use through mistaken identification of your access point. (The same "protection" is equally well achieved with arp filtering 'though, perhaps a better alternative?)
When viewed this way, WEP still has it's uses, but 128 bit WEP is no better or worse than 64 bit. So, why not pick 64 to reduce the overhead on your poor laptop CPU?
If you value your security, you should NOT rely on WEP for privacy or authentication. Only use it on networks that are adequately firewalled from your sensitive network, and layer some application or socket level encryption over it for all sensitive data.
Nothing mysterious about it, just a little strange. The clock used a single-pole synchronous motor, so it could run equally well in either direction. It had a sweep second hand, so it looked pretty funny going backwards. There's a starting circuit that's supposed to give it a kick in the right direction when the power is applied, but it occasionally got it wrong anyway. Once it was running backwards, it would continue indefinitely. I liked to leave it that way, just to see who noticed.
Slashdot Mirror
#1 I did the TurboTax software on the web. Doesn't it work under a Linux browser?
Yes, but there are other reasons it isn't right for everybody. It all depends on your own tax circumstances. For me, in the past few years with an S corporation, various depreciating assets, etc., the dumbed-down web-based version just didn't hack it. The full offline "Turbo-Tax Business" version probably would have, but I ended up using an accountant anyway, since it was too complex.
But I no longer own a business, so next year I'll revert to doing it myself. I probably will use the online TurboTax.
#2 Wouldn't it be easier just to have a tax preparer do them for $40, or is your time worth that little?
More like $350 or so. Once again, it all depends upon your circumstances. If your taxes are real simple -- salary income, a few simple investments, and a handful of itemized deductions, you might be able to find a $40 preparer. But, if your taxes are that simple, it would be pretty trivial to just complete the forms yourself, too.
However, lead-free solder isn't much more difficult to work with (at least as an elecronics hobbiest).
For factory production, basic soldering, e.g. simple wave soldering or single-sided reflow SMT is probably not much more difficult with unleaded than with leaded solder. Where it gets exciting is when you do fancy stuff.
Some soldering processes are pretty sophisticated. Consider a board with reflow SMT components on both sides, some of them using solderball or solder-column attachment, areas of selective through-hole wave solder, and areas with layered (interposer) board attachment.
The manufacturing engineers do some pretty amazing stuff. By careful selection of which solder alloys are used where (i.e. different melting points), and very tight control of thermal profiles, they can get seemingly impossible things to happen. To me, it's black magic that they can reflow the top side of a PCB without the components already on the bottom falling off.
So, they depend upon not just one solder, but a variety of different alloys with different properties. Trying to re-engineer a complex process to switch to all-new (lead free) alloys is far from trivial.
Another issue is that the higher temperatures may stress parts more during attachment. This can lead to higher fallout, shorter component life, etc. Not my specialty, but some QA folks tell me this can cause problems.
Despite all these difficulties, I'm a fan of eliminating lead from the process, and am glad to see it finally happening. It's the right thing to do.
Does your state [context: New Hampshire] have property tax? It is possible that your property tax is far higher than it is in California.
I lived in NH for several years. So, I'll have a crack at answering.
Yes, NH does have property taxes, but at the time I lived there (I left 10 years ago), they weren't excessive. Certainly nothing nearly as high as California, Colorado, or Massachusetts
NH really has no state income tax. NH does have sales tax, but on a very restricted basis -- it applies only to restaurant meals, hotel rooms, ski tickests, and perhaps a few other tourism-related things. It has high gas (petrol) taxes, about comparable to most other states in the US. Everything else in ones daily lives -- clothing, (non-restaurant) food, cars, etc., truly are sales-tax free. NH also have two major toll highways, but the only portion subject to toll is the first few miles beyond the Massachusetts border - once again, it's tourism focussed.
When I lived in New Hampshire, I paid about $1000/yr in property tax (I owned a small house). I paid, say, $500 a year in restaurant sales tax, because I ate out a lot. Maybe $50 a year in tolls. And I gas (petrol) taxes which I won't enumerate becasue they're about the same in all states. This totals, say, $1550/yr in taxes.
For comparison, when I lived in Massachusetts before that, I paid almost $5000/yr in state income tax alone, even before you get to the higher property taxes, and almost across-the-board sales taxes.
Massachusettes residents who defended all the monster Massachusetts taxes (yes, some people actually like them!) would always either (a) vaguely cite alleged benefits associated with living in a more "progressive" state, and (b) alledge that New Hampshire taxes were "just as high", because property taxes made up for the difference. The numbers above belie this claim.
I certainly never saw a single shred of evidence of truth in either of these two arguements. In terms of percentage of property values, New Hampshire property taxes were about comparable with those in a typical Massachusetts town. But in terms of absolute dollars, they were _much_ lower, because New Hampshire didn't suffer from Massachusetts's outrageous housing prices. I don't know how that's changed in recent years?
The bottom line, NH gets by without income and sales taxes, in two ways: By taxing tourism, and by keeping it's spending under control. I lived there enough to know that it REALLY does work.
I never saw any downside to the moderation in spending. The state legislature was part-time and unpaid. Essential services all seemed to work just fine. Road maintenence, snow clearing, hospitals, courts, etc., all seemed excellent. I can't comment on schools, since I didn't have kids.
Now if only I could live in New Hampshire, without having to be on the east coast!
I like the idea in principle, but see plenty of problems in it's practical impelementation.
.deb or .rpm package in your Linux system had a spurious GUI component, just to comply with a well-intentioned but poorly-considered law!
As described, the proposed law would hard-code the concept of using icons to disclose this information. What about fundamentally non-graphical programs (drivers, daemons)? What about overall non-graphical environments (servers, embedded)?
I fear this scheme would further what is already an increasing problem: that everybody wants to attach a GUI to every program, even if it's totally inappropriate (e.g. printer drivers). The proliferation of spurious GUI interfaces leads to the proliferation of inappropriate design choices in exception reporting (pop-ups instead of log files), configuration methods, etc.
I'm not anti-GUI, by the way. I'm anti-inappropriate-GUI, and I fear hard-coding icon requirements into every piece of software makes this trend even worse. Immagine if every
On the other hand, I would definitely like to see these icons displayed on the labels of software packages and disks, or on the web pages that software is downloaded from.
Oh, and something the article didn't mention, but I'd propose this ammendment to the act: Make it hard to add any additional icons (i.e. to make the program behavior worse) in upgrades. If any icons are added, the vendor must either (1) continue to support the old version for future bug fixes, security patches, etc., or (2) refund the purchase price to buyers who choose not to continue using the product. (Obviously, there'd have to be a time limit, but long enough to prevent the use of "incrimental-spyware" as a bait-and-switch technique.)
The announcement is just PR!
I'm not denying that the lead reduction is real. It is real.
But this isn't anything unique to Intel, and it isn't done out of the goodness of their green little hearts.
Every IC manufacturer, in fact practically every manufacturer of anything electronic, is already investigating lead reduction or elimination at some level or other. Not all are making a public hoopla about it, though.
Lead free solder requires the development of new alloys and new processes. The changeover isn't trivial, but some promissing candidates exist. Typically they have very high tin content, plus some mix of Silver, Copper, and Antimony.
There are several reasons for this trend: Regulatory changes (pending in the US, and I think already passed in Europe?), Liability/Insurance cost (employee lawsuits), and waste treatment cost, including waste water.
My opinion: I don't beleive lead in electronics will ever be totally eliminated, nor outright outlawed. I'm no solder/process expert, but those I know tell me that leadless soldering presents many challenges. More likely in my opinion, regulations will take the form of taxes and fees on lead content, driving manufacturers to use it only where no good alternative exists.
You've got to have some system to identify which employees get what share of the loot. I'm not egalitarian enough to advocate the same raises and bonusses for top performers as for useless slugs. And I'm certainly not too thrilled with my own company's policy (all the loot goes to the CEO and the CEO's closest henchpeople).
When done right, a ranking system works just fine. And, in all but the smallest companies, it's pretty much unavoidable.
Notice that I refer to ranking in the context of pay and rewards, not layoffs. Once you use the ranking system to select layoff candidates, things go very badly awry.
Bad things companies do with ranking systems:
1. Firing all the people below some threshold.
2. Continuing to assume a gaussian curve, even after several rounds of layoffs.
3. Applying gaussian distribution to small populations, e.g. individual project teams, rather than large populations like the entire large corporation.
4. Applying the rank-and-fire methodology only to the rank-and-file. Too often the real problems are in upper management or exectutive levels, and those folks are invariably excluded from the system.
5. Letting prior year's ranking affect this year's ranking. i.e. judging whether to raise or lower a person's ranking vs. last year, rather than starting with a blank sheet of paper each year.
An inflexible "fire (at least) the bottom 5%" rule, as adopted by my employer, is just plain dumb. Apart from the obvious morale impact on the survivors, it just doesn't make sense. Do it once, you may get rid of dead wood. Do it twice, there's no dead wood left for the second round. After a while, you're firing really good people, just to meet a quota. (And using up prodigous amounts of toner, as your remaining good people print their resumes looking for a more rational company to employ them.)
The quota is completely fake, based on the mistaken assumption that when you lop off the tail of a gaussian curve, what's left behind is still a gaussian curve.
In my opinion, the root cause of this problem, as with so many corporate problems today, is executives and upper management that have only a short-term outlook. Short term, layoffs save money, and it really doesn't matter who gets the chop, or what effect it has on the morale, motivations, and behavior of the survivors. Long term, it may be disasterous, but corporate governence these days seems to care little about long term consequences of their activities.
A very weak rationalization by Google, in my opinion. It seems to be based upon the "method of dimensions" arguement, i.e. "if the dimensions come out right, the answer must be right". I don't buy it. The method of dimensions is a helpful but non-rigorous technique, and doesn't always correctly handle dimensionless constants.
:-)
Hertz can't just mean "anything" per second.
For starters, consider that minimally it must be restricted to mean "anything dimensionless per second". You can't measure velocity or power in Herz, even though both have units of "anything per second" (metres or joules respectively).
In fact, either by convention or definition, I suspect the former, the "anything" is further restricted to be a cycle of some periodically repeating event. One rotation of a roundabout. One cycle of a sine-wave. etc.
Travel one radian around the roundabout, and you don't end up back where you started. Travel 2*pi radians, and you do -- you've completed a cycle.
If Herz really meant "anything per second", why don't Intel just claim their Xeons are 12.4GHz instead of 2GHz? I hope Intel's marketing folks aren't listening.
Google may have a rationalization for why their answer is "right", but any engineer who has ever worked with filters knows better.
Yeah, but don't ask it for 80 hertz in radians per second. It gives the wrong answer (off by a factor of 2*pi).
Pity it's buggy,:
Try this...
16 hertz in radians per second
Notice that the answer is off by a factor of 2*pi.
I reported it to them months ago, but they haven't fixed it.
Quite true, in many aircraft types.
slow rolls being done without the knowledge of passengers
Not quite true.
Don't confuse barrel rolls and slow rolls. A slow roll, properly executed, involves minus one G (a.k.a. hanging upside down from your seatbelt). It also involves significant sideways G during the knife-edge portions.
By the way, "slow" in "slow roll" isn't really a measure of the speed of the roll, only the type of roll. (And, roll-rate isn't a judging criterion in aerobatic competition.) The name came about because slow rolls are inherently slower than snap/flick rolls.
In a nutshell:
Barrel roll: Corkscrew-shaped flight path. Curvature of flight path during the inverted portion keeps the pilot feeling positive G force.
Slow roll: maintain a constant, level, undeviating flight path, while rotating around the longitudinal axis.
Snap/flick roll: Snap=US term, flick=UK term. Combined use of yaw and rapid pich change to cause one wing to stall (stop producing much lift). Rotation results from the unbalanced lift generated by the other wing. Rotation is usually quite rapid.
Aileron roll: Lazy pilot's slow roll. Like a slow-roll, but the flight path is a freefall-like parabola rather than a straight, level line, which makes it much easier to coordinate.
Rolling turn: Like a slow roll, but the flight path is a curved, turning, level path.
Only snap/flick, slow, and rolling turns are used in IAC sanctioned competition, other than in freestyle where anything goes,
RockyMountain.
(Competition aerobatic pilot, Pitts Special).
Telling you to work overtime or you'll be fired is not harassment
Classic dilemma -- we have two vocabularies. Common usage, and legal definitions.
Obviously, this can be indeed be a form of harassment in the common usage of the term, but it's not harrasment by any legal definition of the word (in US law, anyway).
Reminds me of a discussion I had with an aviation liability lawyer. He told me that there's no such thing as a "frivolous lawsuit". I cited many examples, but he wouldn't budge on his claim that they simply do not exist. That's because I was using the English word "frivolous" (consult dictionary), wheras he was using the Legal word "frivolous" (consult definitions in laws and statutes).
Thanks for the info.
Gentoo's one of the few distro's I've never taken a look at. Mybe it's time for me to take a closer look.
I can't speak for the original poster, but I'll describe why this solution doesn't work for me.
I live in Colorado. Not sure if it's the same in other places, but here basic home phone service is nearly $40 a month. That's even before you pay for long distance, call time, anything. That's just flat monthly fees and taxes for having a dial tone.
No big deal, if you're amortizing that $40 over zillions of hours of calls. But, in my case I make very few calls. And I need a cell phone too. My call volume is so low, it makes sense to just use the cell for everything, and disconnect the home phone. I did this a few months ago, and it works great: the little extra airtime is a drop in the bucket compared to the $40/month saved.
The one problem I haven't solved yet: What to do about overseas calls? Most (all?) cell phone plans available here _either_ have totally outrageous charges for international calls, _or_ have high monthly fees, incompatible with my low call volume. And they all block the use of calling cards for international.
I have a cable modem, and my hope is that some VOIP solution will eventually solve my international calling dilemma. I'd be happy with outgoing-only VOIP service. (Incoming international calls go to my cell, and I keep the call short and tell the caller I'll call back on the VOIP.)
Some have suggested trading in cable modem for DSL plus land line. But that costs more where it is available, and isn't available where I live, anyway.
If anybody has any experience with low- or no- monthly fee VOIP that solves this problem, I'd love to hear about it.
... because they both (as typically implemented, anyway) start/stop each servide one at a time. Way too much unnecessary serialization!
A good system would explicitly know about dependencies, and would then concurrently start/stop everything for which the dependencies are already satisfied. That's what multi-tasking is for. Each time a daemon reports a successful start, all the other services whose dependencies are satisfied by it, would immediately be concurrently launched.
Problem is, no distribution seems to have standardized an implementation of this, although I've heard of experimental examples.
There's no good reason for computers to take forever to start and stop. We accept it because we're used to it. Aside from the fscking, the rest of the bootup should take only a few seconds.
They want to know what financial value people place on their privacy. Why?
I'd love to beleive that this is a purely academic social science study, but I'm not that naive. The rationale behind it must be to make money. Someone, somewhere is dreaming up a business model to make you and I pay for our privacy.
Imagine... Opt-outs, where you pay a fee not to have your personal details sold.
Sure, it's extortion. But, so are a lot of very common business practices today, unfortuntely. And governments pay lip service to privacy concerns, but big business can always buy the loopholes they want.
Don't think it will happen? Consider unlisted telephone numbers: you already pay a monthly fee for you privacy!
This study is all about putting a dollar value on privacy -- to help develop new ways to extort profit by selling, or more likely renting, our privacy back to us. Just wait and see.
I've interviewed and hired a lot of people. And, my choices have always turned out well. Admitedly, my experience may not be directly applicable, since I'm in hardware engineering, but I'd be surprised if it's that different in your area.
I seldom bother to ask a candidate what degrees and certificates they hold. And I never let the answer affect my final choice. I talk about engineering, problem solving, and experience. As for experience, I don't ask questions like "how many years doing X?" . I ask questions like "give a specific example of your experience of X".
I can name many very smart and talented people I've hired over the years, and I still don't even know whether they have a BS, MS, or PhD, nor from what school. On the rare occasions when the subject comes up (I don't bring it up), I see essentially no correlation between degree of formal education and the value of the engineer.
One caveat, here. I select candidates from the pool who already have a foot in the door. Perhaps lots of degrees and certificates are helpful in getting the foot in the door in the first place. In my current organization, some pre-screening is done by other interviewers before they get to me. So, I know the candidates all meet a certain baseline, including a BSEE degree. But the pre-screening lets an awful lot goofballs, fast-talkers, shmoozers, empty-shells and slugs through. And, if anything, the worst of the hopeless candidates frequently turn out to be (on paper) the best educated ones.
#define MIN_CHEST 25
size_t chest;
Minimum diameter 25! Whoah! Are you sure you don't mean circumference?
(BTW, I'm assuming inches. Maybe that's where I'm going wrong?)
OK. But I don't grok the connection in your original post between encrypting and "inability to change stuff without a trail". Encryption without signature is deniable -- you can't prove the statement is authentic, so it's useless as a trail. With a signed document, you have proof that the trail is authentic, they can't deny it.
;)
Original: Any company offering encrypted emails as a paperless billing solution? That way I have a copy of everything and they can't change stuff without a trail. I've got a PGP public key around somewhere...
I think you mean "digitally signed", rather than "encrypted".
(Just being a nitpicker. Sorry.)
Yeah, I know it's lousy writing, and pedantic, and repetitive, but... it's core theme is valid. So slog though it, and take it to heart.
Communicate with your customers, partners, suppliers, and employees in clear, simple, unambiguous, TRUE language. Don't mislead by omission. Don't obfuscate. Don't hide the truth in-amongst tons of verbiage. Make sure the letter of your message is the same as the spirit of your message. Don't hide behind legalities and technicalities. Say what you mean, and mean what you say.
Several posters have countered with the suggestion that mother nature is inefficient, using so much plant material to make so little fuel.
But, both the "cars are inefficeint" and "nature is inefficient" arguements miss one important point: That the huge amount of biomass was spread out over millions of years of growth, with the vast majority of the material being recycled from one growth generation to the next. Obviously, just by virtue of the fact that a gallon of petrol weighs a lot less than a small forrest, we must conclude that most of the material didn't become fuel. Most of it became fertilizer/compost, and fueled the next generation of growth.
Adding up the mass of all these generations of plant growth is really just repeatedly counting the same material over and over.
1. USB subsystem. Maybe it's just me, but I've never been able to make it work reliably.
2. ATAPI support for CD burners. The whole passthrough driver concept is a mess. Especially when something goes wrong, and error recovery can't clean up properly because it is an application program like cdrecord tracking the drive state, a function which should reside in an OS driver. Yuck.
3. Sound card support. Probably works fine if you know how to set it up, but I've always found it to be the hardest thing to get right, and the least automated in terms of hardware discovery.
4. Distros that are a little "too helpful", and install stuff that doesn't work, but just gets in the way. Nautilus and autofs, for example.
5. Dependency nightmares. OK, it's probably an intractible problem, since I would reather run a very conservative stable Linux, yet install one or two bleeding edge packages. (Why don't bleeding edge projects build statically linked archive binaries by default, until they get mature enough to be inculded in stable distros?)
I'm not knocking Linux, though. Numbers 2, 4, and 5, especially, are far worse in windows. Still, I'd love to see these fixed.
I'm sure (well, I hope anyway) JN had his tongue in his cheek when he proposed a perpetual motion application for this technology.
Remember, they're not just the "good ideas" of thermodynamics, they're the law.
I'm far from an expert on the subject, so someone please stop me if I've got my facts wrong. But I think this is accurate...
I beleive it's smarter not to use 128bit, but rather use 64 bit (which is really 40 bit encryption, by the way, since 14 of them are not random).
Here's why.
WEP provides no real privacy. The algorithm has been cracked wide open, and there are readily available exploits. Also, the known exploits scale linearly, so using twice as many bits only gives a 2x increase in crack time. In short, forget WEP as a privacy measure -- it isn't one.
So, why use WEP at all? It's still worth using, in my opinion, but for different reasons. Despite the lack of protection against deliberate attack, WEP does provide protection against accidental use through mistaken identification of your access point. (The same "protection" is equally well achieved with arp filtering 'though, perhaps a better alternative?)
When viewed this way, WEP still has it's uses, but 128 bit WEP is no better or worse than 64 bit. So, why not pick 64 to reduce the overhead on your poor laptop CPU?
If you value your security, you should NOT rely on WEP for privacy or authentication. Only use it on networks that are adequately firewalled from your sensitive network, and layer some application or socket level encryption over it for all sensitive data.
Nothing mysterious about it, just a little strange. The clock used a single-pole synchronous motor, so it could run equally well in either direction. It had a sweep second hand, so it looked pretty funny going backwards. There's a starting circuit that's supposed to give it a kick in the right direction when the power is applied, but it occasionally got it wrong anyway. Once it was running backwards, it would continue indefinitely. I liked to leave it that way, just to see who noticed.