Look, I'm not saying that programmers don't make mistakes...I know that I have made some expensive ones. I also know that there are some fantastic pilots out there too. In fact I would go as far to say that for the most part, most pilots are excellent and I would trust my life with just about all of them. However, My point is this, if and when we get to a point where the software is better than the pilot, I think that the software will still have additional hurdles to over come. Your response demonstrates some of those obstacles.
Your point is taken, and true. One of the most useful and practical examples is with FADEC's, Full-Authority Digital Engine Controls, where the fuel metering to a turbofan engine is completely controlled by a computer reading the thrust level angle from a sensor and then calculating and commanding the needed fuel flow given the conditions rather than a direct link to a highly complex mechanical gadget that meters the fuel. There were objections by many, some pilots as well - what if the computer went haywire, what if it got struck by lightning, what if a sensor started to misbehave and the computer blindly followed it, etc. By and large, these problems were solved rather completely and every jet engine made today (AFAIK) is controlled by FADEC's. So I agree that some opposition to computer controls is irrational - computer controls DO have a place.
But let me use this same example as a cautionary one. In one particular A/C with FADECs, the EMB-145 (it came onto the scene quite recently, 1997), had its FADECs programmed to shut down an engine when the oil pressure went outside of the "safe operating range" for longer than five minutes. This behavior probably seemed quite sensible to the programmers at the time, as the engine would (probably destructively) shut itself down not too long after if the oil pressure stayed out of bounds.
It nearly caused an accident as a brand-new EMB-145 was being flown out of Brazil toward the U.S.A. for delivery.
The seals preventing the high pressure section of the engine from over-pressurizing the oil system were a bit temperamental, and resulted in (what are now determined to be) harmless temporary pressure spikes. The FADEC program to figure out when the oil pressure was too high only sampled once per minute.... perhaps you see what happened when five spikes occurred just at the wrong time, over the Amazon basin. The healthy engine shut itself down. Now that the aircraft was flying on one engine, the one running engine started having the same pressure spikes. There was no logic in the FADEC that would let it talk to the other FADEC and essentially say, "Hey, I have a malfunctioning engine I normally should shut down, but the OTHER engine isn't running at all so I better keep this one going!". In the end, they had to reduce power on the one good engine and go for what was little better than a glide back to Rio. Obviously, this problem in software was fixed, but it is an example of where what seems an obvious thing to do (shutting down an engine that has been malfunctioning for the last 5 minutes before it self-destructs) becomes the WRONG thing to do in the real world.
So while you may worry that when software becomes better than a pilot at a particular task (as it almost inevitably will at some point in time) it will be unfairly (and in a statical sense dangerously) held back, I see it from the other side - immature software and controls will be deployed BEFORE they are better (in order to save money/look good on paper as another "safety feature"). I don't even think "better" is the right word - until all the highly complex side effects of automatic behavior in the real-world are well thought out and worked out.
And on a personal note, thank you for the polite response - more is needed here, I think!
I picked a career in the airlines because I found nothing quite as soul-draining as the prospect of being a cubicle-person 9 to 5 (or as is increasingly the case, 8 to 8 or more), looking forward to my one or two weeks of vacation where I'll see some palm trees in an overpriced tourist trap before resuming my run in the hamster wheel. I mean this as no denigration toward the large majority of people who are in that life by choice or circumstance - I simply point out how it creeps ME out.
To get the lifestyle I currently have, I spent about ten years after graduation in the "apprenticeship" phase, meaning working two jobs for a combined total of (typically) 60 - 80 hours a week to pay the bills, get relevant experience, etc. Ten more years and I'm now working for a major airline.
Now, I average about 16 days off a month - not including vacations. Mind you, there are still costs associated with this. I make less than my peers seniority-wise - but it is by MY choice. I usually work most weekends, and having off holidays is so rare that I tend to forget most holidays anyway. I spend about 10 nights a month in a hotel room away from home - and usually not the nicest of hotels. And lastly, of course, is the fact I might wake up someday, read the headlines, and discover my company/job/career has just vanished.
But what I enjoy most is that there is CHOICE involved. I know many pilots who need to have the large mansion, the multiple racing-car collection, etc., and they work HARD to get it. I chose the opposite extreme and try to minimize my work, so I can tinker with computers, travel to different countries, read and play with math, see my friends, and enjoy my life. I'm fairly frugal with money, so I can do this with less pay. If I wanted to switch to the other extreme, and max out my pay in exchange for minimum time off, all I have to do is bid differently starting next month.
Anyway, my long-winded point is that it is the CHOICE that I think is missing at most American workplaces. It seems that the amount of work you do is quite disconnected from your paycheck, and what is counted as "work" seems to be showing up for long hours at the workplace, somewhat independent of what you do when you get there. And if you say you want to work half as many hours in exchange for half the pay in a month, I suspect the management would laugh.
The current system in most companies where you MUST work a minimum number of hours a week in exchange for two weeks of paid vacation seems geared towards the age of factory assembly lines. I suspect most "mind" workers would be tremendously more productive (not to mention happier, more well-rounded individuals) if they could take a month leave twice a year in exchange for two months without pay, or a week every month for half the year, or some other permutation.
...And folks like you will conviently ignore the thousands of times the computer is programmed poorly, and a human with lives on the line has to fix the mistake, and does so. You don't hear about those cases, because only if tombstones sprout are things investigated. This isn't the media who is worried, it's pilots who have been making up kludgy procedures for years to get around limitations and flaws in the automation who are rightly sceptical about a procedure that says, in effect, "You're about to hit another airplane!!! Do nothing, but watch the autopilot do it, and if IT fucks up, THEN do something with much less time to react while you were waiting for the computer to do the right thing!!!" Does that sound like a sensible emergency procedure to you?
LT: Absolutely. There was a bit of bragging, there was also a bit of, hey, I still, the way I do my work is I sit these days downstairs in my basement alone. And it's nice to just talk to people and a lot of it was probably just social, just saying, hey this is a way to interact with other geeks who are probably also socially inadequate in many ways.
Pretty good insight - it's a way for geeks to socialize other than Star Trek conventions!
(Ducks)
On science and software development:
LT: We shouldn't give credit to Linux per se. There were open source projects and free software before Linux was there. Linux in many ways is one of the more visible and one of the bigger technical projects in this area and it changed how people looked at it because Linux took both the practical and ideological approach. At the same time I don't think this whole "openness" notion is new. In fact I often compare open source to science. To where science took this whole notion of developing ideas in the open and improving on other peoples' ideas and making it into what science is today, and the incredible advances that we have had. And I compare that to witchcraft and alchemy, where openness was something you didn't do. So openness is not something new, it is something that actually has worked for a long time.
Great comparison between open software and science, both of which a lot of people don't get.
On the uselessness of meetings:
KLS: So the face to face thing is a little bit overrated?
LT: I think so. For example I long ago decided I will never go to meetings again because I think face to face meetings are the biggest waste of time you can ever have. I think most people who work at offices must share my opinion on meetings. Nothing ever gets done. When things get done, you usually have someone come into your office to talk about it. But a lot of the time the real work gets done by people sitting, especially in programming, alone in front of their computers doing what they do best.
Dilbert freed from the pointy-haired boss type - Pretty cool. Interesting interview, I may and try and watch it rather than read it.
Yep - Canon's been at the forefront of CMOS technology. I'm trying to figure out what is so special about this particular CMOS sensor that it rated a/.ing. Any idea?
The 5D, with a 13.1 MP full-frame sensor is CMOS. Most camera makers are slowly going over to them because of their much lower power consumption - I presume the reason any one cares about this particular one is because it's cheap.
The main limiter with image quality (unless you're talking medium format or bigger) isn't the sensor any more, it's the lens. And right now, a picture made with a small piece of cheap plastic in front of an 8 MP sensor will reveal exactly all the flaws and distortions in said lens rather than a better image.
I've been using Microsoft DRM with the Napster subscription service for over a year now on an iRiver H10 hard-drive device. You can't beat the convenience and the price -- the cost of a single CD per month for lots of great music.
It's such a good model I even bought four more iRiver devices for others.
To clarify some points in the original comment:
you can download files multiple times (unlike Apple iTunes where you download a file only once and need to copy to other devices)
it's easy to transfer to multiple mobile devices with Windows Media Player
there is a limit on how many total times a file can be downloaded, but when I had to wipe a hard drive and re-install the OS on a particular machine a quick call to Napster got me past that issue -- they'll work with you
The pricing and model beats iTunes. Many, many services will end up using Microsoft DRM. When people wake up and look beyond the fatuous Apple image to practical realities, Microsoft DRM will come out the winner.
Funny, I didn't mention Apple at all in my original post. Got an inferiority complex going there?;)
My point was DRM in general is a pain, and Microsoft has a pretty consistent history of keeping it painful. You are right, I shouldn't pre-judge what they will actually do, but history makes me skeptical. And I like NO DRM, Apple, Microsoft, or otherwise.
Of course that's down to the vendors. I've bought a fair few DRM'd Microsoft Reader ebooks, and the main vendors do allow you to redownload those, for example. It should be a legal requirement for all vendors of DRM'd media files to do that IMHO, but I guess that's about as likely as a very unlikely thing.
Very true - and you make a real good point. If a vendor wants to lock-up a file with DRM, meaning that you have bought the rights to VIEW something instead of OWN something, then they should also be responsible for fixing problems that occur when you can no longer VIEW the content due to technical problems.... sadly, I feel we are dreaming. We need some congresspeople to actually possess a large collection of DRM'ed material and have it all wiped by a technical glitch before they are likely to notice the problem. What this probably requires is new, younger congressfolk! (Apologies to the older folk who are with it - but this is a generational thing)
Boy, they know how to pick a loser. Assuming the carriers go along with this, all Microsoft will have is domination over a standard that nobody will want to use. DRM is annoying enough when it comes to file transfers on computers. Can you imagine how annoying it will be with phones? Will your files survive your phone dropping into the toilet? Or will they be easily transferable to a new phone with the same mobile number assigned to it? You know the answer - and of course, you won't be able to redownload files you've paid for.
It's interesting to see what they waste brainspan and dollars on.
Well, this is/. - Mathematics only to the degree necessary, but rarely appreciated. Nice to know that I wasn't the only one who wondered how you fit l > k mutually orthogonal (or even linearly indendent) vectors into a k-dimensonal space, esp. as many formulations of linear algebra use the maximum number of linearly independent vectors as the DEFINITION of the dimension of that vector space and go from there to define coordinates of vectors.... *sigh*... I need to find a good, active math site for interested amateurs someday. Any suggestions?
"HB2122 passed and was signed by the Governor. This new law allows Permit/License holders to carry onto private parking lots their concealed firearm and store it in their locked vehicle. The Bill States, "No person, property owner, tenant, employer, or business entity shall be permitted to establish any policy or rule that has the effect of prohibiting any person, except a convicted felon, from transporting and storing firearms in a locked vehicle on any property set aside for any vehicle." The law becomes effective 11/1/04." - in fairness, it must be mentioned that this law was set aside (forgive the nonlegalese, IANAL) by a Federal judge, and also the minimum age for possessing a firearm is 21.
But you have to wonder at the logic of a legislature that needs to "protect" kids from videogame violence up until 18, and then at 21 lets them buy REAL guns, carry them around concealed, leave them in their cars (oooh! The car has to be LOCKED - that'll stop a car thief), and so forth. Note this is not an anti-gun post - it is only an anti-hypocrasy post. Don't promote the carrage and use of weapons of deadly force on one hand and then act holier-then-thou and say we're "protecting the children" by not letting them see video-game violence on their own (on the TV it's fine, evidently).
My first thought reading the headline was that this was just called a "hologram" to get some buzz, over what is a very generic, straightforward way of increasing the power delivered to the expensive part, the solar cell. But (for those too lazy to RTFA) this is different for three reasons:
1) It is almost omnidirectional - a Fresnel lens is a flat subsititue for a regular lens, with limited off-axis focusing ability. This seems to use the glass as a lightguide instead, with a broader angular reach (in exchange for limited scalibility - bigger the glass width to thickness ratio, the more light lost because of increased internal reflections & distance from entrance to cell)
2) It uses a hologram to selectively reject useless frequencies like infrared, which is 80% (IIRC) of the energy of sunlight, but generates no electricity from the cell. In fact, infrared is harmful to the cell, because it increases its temperature, which reduces its effeciency!
3) Because of the above features, it does not need a turning mechanism to follow the sun, the solar cell (which is the most expensive part) lasts much longer because it is not heated as much even though it is capturing much more useful light and converting that into electricity, it is flat and relatively easy to handle, unlike traditional solar cells with large, bulky, moving "capture" mechanisms placed in front of them....
In summary, it is cheaper per kilowatt-hr, AND more effecient, AND more practical for installation (no moving parts or seperated pieces). This is pretty neat.
I think that's the logic behind OS X's Keychain app. You don't need remote storage, since the keychain database is encrypted. And to top it off, if your account password is reset by an admin (or someone unfriendly with admin access), the keychain password remains unchanged; if you lose it or forget it.... bye, bye, data. On the whole, a pretty sensible app.
Is there really that much of a difference between using an Intel chip and an AMD chip? I know you need different motherboards and chipsets, but isn't that about it? As far as I understand, there is no difference in the applications, other hardware, etc. So the only thing you would have to maintain (as a boxmaker) is another set of motherboard specs and the BIOSes for them, and in exchange you get (today) better performing chips for the wattage. The fact that a massive organization like Dell has not done so leads me to think that Intel has been doing some arm-twisting.
OTOH, with Apple, which likes working with as small a set of hardware combinations as possible, I can see why they would only want to maintain one microprocessor family, motherboard chipset, etc.
So hard would it really be (financially, organizationally) for a Dell, Gateway or Apple to add the AMD chip to its lineup? Anyone have any concrete knowledge about this?
They still have tremendous inertia in the marketplace. How long has it taken for the general public to have the same perception of Microsoft as a typical/.er? The biggest problem is that Microsoft has gotten people to believe that computers are inherently unreliable, unstable, and buggy - so that people EXPECT such behavior from any computer, not just one with a Microsoft OS or application software on it. As long as a large number of people expect crappily behaving computers, Microsoft's position is secure. In a similar manner, Microsoft has acclimated the business world to the idea that a standalone PC is incomplete without anti-virus, anti-spyware addons and a dedicated IT staff to maintain them.
On the other hand, businesses tend to hold onto computers and custom/favorite apps much longer than individuals do. Apple may have a shot of displacing Windows from household computers, as many people become more savvy about what they need a computer for, which can usually be covered by web browsing, email, document reading, picture viewing, video watching, and music listening, and are willing to pay a bit so they don't have to constantly tinker with their systems. I think replacing Microsoft in the enterprise world will be much harder.
"But even as some on the Mini-Microsoft blog wished for Maria Antoinette-style retribution, other employees defended the decision, if not the people who made it.
"Yes, it's painful. Yes, it's embarrassing," wrote Robert Scoble, a company technical evangelist, on his Scobelizer blog. "But I'd rather have a slipped date than a cruddy product.""
It would have been nice if they had this philosophy a couple of decades ago, rather than trying to transition to a "first in quality rather than first in marketplace" maxim now after all the messes they have institutionalized and all the good, innovative companies that followed the above maxim they have dispatched.
"Maximum landing weight is 230,000 pounds---about twice the maximum landing weight of a Boeing 767 (which, depending on model, ranges from 112,000 to 150,000)---about the same as that of a Boeing 747"
I don't know where you got your 767s from, but our Max Landing Weight is 300,000 lbs. for the -200 and 350,000 lbs. for the -400. Did you mistake kgs. for lbs. or did you forget a main gear strut somewhere?;)
"Here's hoping that the/. effect will spur him on to get this finished in record time!"...or blow his monthly bandwidth limit to the moon. Hope he's ready.
"..absolutely absurd to make [a comparison with Chernobyl]"
No, it's not. While Chernobyl was not designed with safety foremost (as you pointed out), the accident there was also a result of overeager operators going beyond what was authorized in the plant specs, running the plant through a test procedure that had never been operationally been evaluated during the plant's certification, and trying to maintain a higher than rated power output for months at a clip. These are HUMAN failings, not technical ones. Let's take Three Mile Island. Operator error significantly compounded the original problem of a stuck open valve into a major problem that cost a billion dollars to stabilize and fix, although with no loss of human life or health.
This is why I mentioned politics and not planning for the future as reasons why this can go badly. If we approach nuclear power with a long-term view, with a proper safety culture, independent monitoring, no arbitrary "power for no more than this cost by this date" goals, then I think it is as safe as any other power generation technology out there - probably safer when one considers environmental impact (I'd rather have all my toxins in one place than thinly distributed throughout the atmosphere). On the other hand, if we wait until a true energy crisis has developed, and rush through the creation, monitoring, and oversight of these plants, then I DO fear what would happen. While I hope your optimism about our inherent safety culture is true ("No one's going to build and fuel a reactor like that, and they CERTAINLY won't operate it with the absolute idiocy the Soviet crew [did]"), I am a bit more skeptical. In aviation, space travel, nuclear power plants, and other complex endeavors, safely seems to follow a saw-tooth curve; an accident happens, regs are implemented, companies institute safety programs with teeth, and safety improves drastically.... for a while. Then, pressures on cash flow, complacency ("it hasn't happened in years, it must be because we're doing it safely"), industry lobbyists, learning to run the edges rather than the center of the envelope, etc., slowly increase the risk level, until the next accident... and the cycle repeats. Perhaps we can engineer a foolproof nuclear plant, but I wouldn't bet on it.
Nothing is "safe" - but how we judge the odds and risks affects our expectations. The Space Shuttle was planned with an expected hull loss failure rate of about 1 in 10,000, when operational experience and the judgement of the engineers indicated about 1 in 100. If you wish to have a technology that is accepted and utilized properly, the risks have to be judged correctly, not sold like a cure-all. Read the Rodgers Report for more information:
In addition, when you depart the ground in a 747, you may have up to 500,000 lbs. of a highly energetic fuel along for the ride, yet the hull loss rate on 747s is less than 1 per 1,000,000 departures. The presence of explosive material alone does not indicate the risk level. Commercial aviation is the beneficiary of decades of engineering experience, stringent safety regulation, and a culture that encourages disclosure of risk vs. burying the messenger and hoping the probabilities don't catch up on your watch.
So, to be more precise, the political process is more likely to generate safety estimates off by orders of magnitude (in order to sell the product), and then when those fail to come true (not through the fault of the technology, but the way it was used), the whole technology could be needlessly scrapped for decades.... like nuclear power was in the U.S. after the Three Mile Island incident. I was expressing the hope that we would be slightly wiser this time around in deploying a very costly but potentially very beneficial technology, and the most likely obstacles to it.
Nuclear power plants are a reasonable option, if we can do two things:
a) Depoliticize the running of them - the first thing that suffers when politics overtakes reality is safety, as NASA is a perfect illustration of. Having a good, long-term safety policy built into an organization isn't something that can be done overnight, and building an agency to replace the DOE is impossible in the current polarized political environment.
b) Figure out how to prevent proliferation of high-grade fissionable materals; the technology for generating and seperating such is inherent with breeder reactors, if I understand them correctly.
Both of the above are certainly technically possible; but with our current disdain for effective governmental organizations, I think it unlikely we will achieve the above soon.
So, nuclear plants will work, and be built, but I wonder how they will be built. Slowly, carefully and sited & managed rationally? Or be built hurriedly, in a panic to dampen swings in energy prices, with no coherent oversight, policy or judgement?
If it is the latter, we will have our own Chernobyl in time.
Humph... when we were doing comedy, we didn't have any o' this namby-pamby television and camera and writer malarkey. We had to go up on a stage (well, it was more of a pit, really), no costumes, no lights, no dialog, make at least a dozen people die from laughing their guts up with jokes that we had just thought up that second in sign-language, and get violated by the club manager, IF we were lucky!
But, the young comedians today, they just won't believe you....
(If having it tougher than your kids isn't a timeless joke, I don't know what is, and they did it great)
Quite familiar with them, Eddie Murphy before he "changed", John Stewart, Mark Russell, Keith Robinson, Sherrod Small, etc., etc. - I like good comedy. But Python has something timeless and brilliant, particularly in sketch comedy - I don't think (with the exception of Cleese) they would have been so good at stand-up, or Adam Sandler style movies.
No arguing taste! I wish I found another group that captured that wonderful sense of tweaked properness, of just barely concealed rage, or pure self-loathing - I haven't seen any other group get at these quite the same way.
...It is still amazing how well these shows have held up over 30+ years. Perhaps it is just sentimentality, but Monty Python is still one of the best comedy troupes, if not THE best, I have seen in my life.
Are we getting less funny? Or is familiarity part of the enjoyment? Or are they really just uniquely brilliant?
So when will the War on Terrorism be done? Let me know what the criteria are so I can prod my local bureaucrats into restoring a few freedoms that have been lost this round. Ditto for the War on Drugs, running for decades now, with no clear winners or losers or end.
It is a bit more insidious in modern times, I think...
Slashdot Mirror
Your point is taken, and true. One of the most useful and practical examples is with FADEC's, Full-Authority Digital Engine Controls, where the fuel metering to a turbofan engine is completely controlled by a computer reading the thrust level angle from a sensor and then calculating and commanding the needed fuel flow given the conditions rather than a direct link to a highly complex mechanical gadget that meters the fuel. There were objections by many, some pilots as well - what if the computer went haywire, what if it got struck by lightning, what if a sensor started to misbehave and the computer blindly followed it, etc. By and large, these problems were solved rather completely and every jet engine made today (AFAIK) is controlled by FADEC's. So I agree that some opposition to computer controls is irrational - computer controls DO have a place.
But let me use this same example as a cautionary one. In one particular A/C with FADECs, the EMB-145 (it came onto the scene quite recently, 1997), had its FADECs programmed to shut down an engine when the oil pressure went outside of the "safe operating range" for longer than five minutes. This behavior probably seemed quite sensible to the programmers at the time, as the engine would (probably destructively) shut itself down not too long after if the oil pressure stayed out of bounds.
It nearly caused an accident as a brand-new EMB-145 was being flown out of Brazil toward the U.S.A. for delivery.
The seals preventing the high pressure section of the engine from over-pressurizing the oil system were a bit temperamental, and resulted in (what are now determined to be) harmless temporary pressure spikes. The FADEC program to figure out when the oil pressure was too high only sampled once per minute.... perhaps you see what happened when five spikes occurred just at the wrong time, over the Amazon basin. The healthy engine shut itself down. Now that the aircraft was flying on one engine, the one running engine started having the same pressure spikes. There was no logic in the FADEC that would let it talk to the other FADEC and essentially say, "Hey, I have a malfunctioning engine I normally should shut down, but the OTHER engine isn't running at all so I better keep this one going!". In the end, they had to reduce power on the one good engine and go for what was little better than a glide back to Rio. Obviously, this problem in software was fixed, but it is an example of where what seems an obvious thing to do (shutting down an engine that has been malfunctioning for the last 5 minutes before it self-destructs) becomes the WRONG thing to do in the real world.
So while you may worry that when software becomes better than a pilot at a particular task (as it almost inevitably will at some point in time) it will be unfairly (and in a statical sense dangerously) held back, I see it from the other side - immature software and controls will be deployed BEFORE they are better (in order to save money/look good on paper as another "safety feature"). I don't even think "better" is the right word - until all the highly complex side effects of automatic behavior in the real-world are well thought out and worked out.
And on a personal note, thank you for the polite response - more is needed here, I think!
I picked a career in the airlines because I found nothing quite as soul-draining as the prospect of being a cubicle-person 9 to 5 (or as is increasingly the case, 8 to 8 or more), looking forward to my one or two weeks of vacation where I'll see some palm trees in an overpriced tourist trap before resuming my run in the hamster wheel. I mean this as no denigration toward the large majority of people who are in that life by choice or circumstance - I simply point out how it creeps ME out.
To get the lifestyle I currently have, I spent about ten years after graduation in the "apprenticeship" phase, meaning working two jobs for a combined total of (typically) 60 - 80 hours a week to pay the bills, get relevant experience, etc. Ten more years and I'm now working for a major airline.
Now, I average about 16 days off a month - not including vacations. Mind you, there are still costs associated with this. I make less than my peers seniority-wise - but it is by MY choice. I usually work most weekends, and having off holidays is so rare that I tend to forget most holidays anyway. I spend about 10 nights a month in a hotel room away from home - and usually not the nicest of hotels. And lastly, of course, is the fact I might wake up someday, read the headlines, and discover my company/job/career has just vanished.
But what I enjoy most is that there is CHOICE involved. I know many pilots who need to have the large mansion, the multiple racing-car collection, etc., and they work HARD to get it. I chose the opposite extreme and try to minimize my work, so I can tinker with computers, travel to different countries, read and play with math, see my friends, and enjoy my life. I'm fairly frugal with money, so I can do this with less pay. If I wanted to switch to the other extreme, and max out my pay in exchange for minimum time off, all I have to do is bid differently starting next month.
Anyway, my long-winded point is that it is the CHOICE that I think is missing at most American workplaces. It seems that the amount of work you do is quite disconnected from your paycheck, and what is counted as "work" seems to be showing up for long hours at the workplace, somewhat independent of what you do when you get there. And if you say you want to work half as many hours in exchange for half the pay in a month, I suspect the management would laugh.
The current system in most companies where you MUST work a minimum number of hours a week in exchange for two weeks of paid vacation seems geared towards the age of factory assembly lines. I suspect most "mind" workers would be tremendously more productive (not to mention happier, more well-rounded individuals) if they could take a month leave twice a year in exchange for two months without pay, or a week every month for half the year, or some other permutation.
Thoughts?
...And folks like you will conviently ignore the thousands of times the computer is programmed poorly, and a human with lives on the line has to fix the mistake, and does so. You don't hear about those cases, because only if tombstones sprout are things investigated. This isn't the media who is worried, it's pilots who have been making up kludgy procedures for years to get around limitations and flaws in the automation who are rightly sceptical about a procedure that says, in effect, "You're about to hit another airplane!!! Do nothing, but watch the autopilot do it, and if IT fucks up, THEN do something with much less time to react while you were waiting for the computer to do the right thing!!!" Does that sound like a sensible emergency procedure to you?
LT: Absolutely. There was a bit of bragging, there was also a bit of, hey, I still, the way I do my work is I sit these days downstairs in my basement alone. And it's nice to just talk to people and a lot of it was probably just social, just saying, hey this is a way to interact with other geeks who are probably also socially inadequate in many ways.
Pretty good insight - it's a way for geeks to socialize other than Star Trek conventions!
(Ducks)
On science and software development:
LT: We shouldn't give credit to Linux per se. There were open source projects and free software before Linux was there. Linux in many ways is one of the more visible and one of the bigger technical projects in this area and it changed how people looked at it because Linux took both the practical and ideological approach. At the same time I don't think this whole "openness" notion is new. In fact I often compare open source to science. To where science took this whole notion of developing ideas in the open and improving on other peoples' ideas and making it into what science is today, and the incredible advances that we have had. And I compare that to witchcraft and alchemy, where openness was something you didn't do. So openness is not something new, it is something that actually has worked for a long time.
Great comparison between open software and science, both of which a lot of people don't get.
On the uselessness of meetings:
KLS: So the face to face thing is a little bit overrated?
LT: I think so. For example I long ago decided I will never go to meetings again because I think face to face meetings are the biggest waste of time you can ever have. I think most people who work at offices must share my opinion on meetings. Nothing ever gets done. When things get done, you usually have someone come into your office to talk about it. But a lot of the time the real work gets done by people sitting, especially in programming, alone in front of their computers doing what they do best.
Dilbert freed from the pointy-haired boss type - Pretty cool. Interesting interview, I may and try and watch it rather than read it.
Yep - Canon's been at the forefront of CMOS technology. I'm trying to figure out what is so special about this particular CMOS sensor that it rated a /.ing. Any idea?
The 5D, with a 13.1 MP full-frame sensor is CMOS. Most camera makers are slowly going over to them because of their much lower power consumption - I presume the reason any one cares about this particular one is because it's cheap.
The main limiter with image quality (unless you're talking medium format or bigger) isn't the sensor any more, it's the lens. And right now, a picture made with a small piece of cheap plastic in front of an 8 MP sensor will reveal exactly all the flaws and distortions in said lens rather than a better image.
I've been using Microsoft DRM with the Napster subscription service for over a year now on an iRiver H10 hard-drive device. You can't beat the convenience and the price -- the cost of a single CD per month for lots of great music. It's such a good model I even bought four more iRiver devices for others. To clarify some points in the original comment: you can download files multiple times (unlike Apple iTunes where you download a file only once and need to copy to other devices) it's easy to transfer to multiple mobile devices with Windows Media Player there is a limit on how many total times a file can be downloaded, but when I had to wipe a hard drive and re-install the OS on a particular machine a quick call to Napster got me past that issue -- they'll work with you The pricing and model beats iTunes. Many, many services will end up using Microsoft DRM. When people wake up and look beyond the fatuous Apple image to practical realities, Microsoft DRM will come out the winner. Funny, I didn't mention Apple at all in my original post. Got an inferiority complex going there? ;)
My point was DRM in general is a pain, and Microsoft has a pretty consistent history of keeping it painful. You are right, I shouldn't pre-judge what they will actually do, but history makes me skeptical. And I like NO DRM, Apple, Microsoft, or otherwise.
Of course that's down to the vendors. I've bought a fair few DRM'd Microsoft Reader ebooks, and the main vendors do allow you to redownload those, for example. It should be a legal requirement for all vendors of DRM'd media files to do that IMHO, but I guess that's about as likely as a very unlikely thing. Very true - and you make a real good point. If a vendor wants to lock-up a file with DRM, meaning that you have bought the rights to VIEW something instead of OWN something, then they should also be responsible for fixing problems that occur when you can no longer VIEW the content due to technical problems.... sadly, I feel we are dreaming. We need some congresspeople to actually possess a large collection of DRM'ed material and have it all wiped by a technical glitch before they are likely to notice the problem. What this probably requires is new, younger congressfolk! (Apologies to the older folk who are with it - but this is a generational thing)
Boy, they know how to pick a loser. Assuming the carriers go along with this, all Microsoft will have is domination over a standard that nobody will want to use. DRM is annoying enough when it comes to file transfers on computers. Can you imagine how annoying it will be with phones? Will your files survive your phone dropping into the toilet? Or will they be easily transferable to a new phone with the same mobile number assigned to it? You know the answer - and of course, you won't be able to redownload files you've paid for.
It's interesting to see what they waste brainspan and dollars on.
Well, this is /. - Mathematics only to the degree necessary, but rarely appreciated. Nice to know that I wasn't the only one who wondered how you fit l > k mutually orthogonal (or even linearly indendent) vectors into a k-dimensonal space, esp. as many formulations of linear algebra use the maximum number of linearly independent vectors as the DEFINITION of the dimension of that vector space and go from there to define coordinates of vectors.... *sigh* ... I need to find a good, active math site for interested amateurs someday. Any suggestions?
"HB2122 passed and was signed by the Governor. This new law allows Permit/License holders to carry onto private parking lots their concealed firearm and store it in their locked vehicle. The Bill States, "No person, property owner, tenant, employer, or business entity shall be permitted to establish any policy or rule that has the effect of prohibiting any person, except a convicted felon, from transporting and storing firearms in a locked vehicle on any property set aside for any vehicle." The law becomes effective 11/1/04." - in fairness, it must be mentioned that this law was set aside (forgive the nonlegalese, IANAL) by a Federal judge, and also the minimum age for possessing a firearm is 21.
But you have to wonder at the logic of a legislature that needs to "protect" kids from videogame violence up until 18, and then at 21 lets them buy REAL guns, carry them around concealed, leave them in their cars (oooh! The car has to be LOCKED - that'll stop a car thief), and so forth. Note this is not an anti-gun post - it is only an anti-hypocrasy post. Don't promote the carrage and use of weapons of deadly force on one hand and then act holier-then-thou and say we're "protecting the children" by not letting them see video-game violence on their own (on the TV it's fine, evidently).
My first thought reading the headline was that this was just called a "hologram" to get some buzz, over what is a very generic, straightforward way of increasing the power delivered to the expensive part, the solar cell. But (for those too lazy to RTFA) this is different for three reasons:
1) It is almost omnidirectional - a Fresnel lens is a flat subsititue for a regular lens, with limited off-axis focusing ability. This seems to use the glass as a lightguide instead, with a broader angular reach (in exchange for limited scalibility - bigger the glass width to thickness ratio, the more light lost because of increased internal reflections & distance from entrance to cell)
2) It uses a hologram to selectively reject useless frequencies like infrared, which is 80% (IIRC) of the energy of sunlight, but generates no electricity from the cell. In fact, infrared is harmful to the cell, because it increases its temperature, which reduces its effeciency!
3) Because of the above features, it does not need a turning mechanism to follow the sun, the solar cell (which is the most expensive part) lasts much longer because it is not heated as much even though it is capturing much more useful light and converting that into electricity, it is flat and relatively easy to handle, unlike traditional solar cells with large, bulky, moving "capture" mechanisms placed in front of them....
In summary, it is cheaper per kilowatt-hr, AND more effecient, AND more practical for installation (no moving parts or seperated pieces). This is pretty neat.
I think that's the logic behind OS X's Keychain app. You don't need remote storage, since the keychain database is encrypted. And to top it off, if your account password is reset by an admin (or someone unfriendly with admin access), the keychain password remains unchanged; if you lose it or forget it.... bye, bye, data. On the whole, a pretty sensible app.
Is there really that much of a difference between using an Intel chip and an AMD chip? I know you need different motherboards and chipsets, but isn't that about it? As far as I understand, there is no difference in the applications, other hardware, etc. So the only thing you would have to maintain (as a boxmaker) is another set of motherboard specs and the BIOSes for them, and in exchange you get (today) better performing chips for the wattage. The fact that a massive organization like Dell has not done so leads me to think that Intel has been doing some arm-twisting.
OTOH, with Apple, which likes working with as small a set of hardware combinations as possible, I can see why they would only want to maintain one microprocessor family, motherboard chipset, etc.
So hard would it really be (financially, organizationally) for a Dell, Gateway or Apple to add the AMD chip to its lineup? Anyone have any concrete knowledge about this?
They still have tremendous inertia in the marketplace. How long has it taken for the general public to have the same perception of Microsoft as a typical /.er? The biggest problem is that Microsoft has gotten people to believe that computers are inherently unreliable, unstable, and buggy - so that people EXPECT such behavior from any computer, not just one with a Microsoft OS or application software on it. As long as a large number of people expect crappily behaving computers, Microsoft's position is secure. In a similar manner, Microsoft has acclimated the business world to the idea that a standalone PC is incomplete without anti-virus, anti-spyware addons and a dedicated IT staff to maintain them.
On the other hand, businesses tend to hold onto computers and custom/favorite apps much longer than individuals do. Apple may have a shot of displacing Windows from household computers, as many people become more savvy about what they need a computer for, which can usually be covered by web browsing, email, document reading, picture viewing, video watching, and music listening, and are willing to pay a bit so they don't have to constantly tinker with their systems. I think replacing Microsoft in the enterprise world will be much harder.
From the article:
"But even as some on the Mini-Microsoft blog wished for Maria Antoinette-style retribution, other employees defended the decision, if not the people who made it.
"Yes, it's painful. Yes, it's embarrassing," wrote Robert Scoble, a company technical evangelist, on his Scobelizer blog. "But I'd rather have a slipped date than a cruddy product.""
It would have been nice if they had this philosophy a couple of decades ago, rather than trying to transition to a "first in quality rather than first in marketplace" maxim now after all the messes they have institutionalized and all the good, innovative companies that followed the above maxim they have dispatched.
"Maximum landing weight is 230,000 pounds---about twice the maximum landing weight of a Boeing 767 (which, depending on model, ranges from 112,000 to 150,000)---about the same as that of a Boeing 747"
;)
I don't know where you got your 767s from, but our Max Landing Weight is 300,000 lbs. for the -200 and 350,000 lbs. for the -400. Did you mistake kgs. for lbs. or did you forget a main gear strut somewhere?
"Here's hoping that the /. effect will spur him on to get this finished in record time!" ...or blow his monthly bandwidth limit to the moon. Hope he's ready.
"..absolutely absurd to make [a comparison with Chernobyl]"
No, it's not. While Chernobyl was not designed with safety foremost (as you pointed out), the accident there was also a result of overeager operators going beyond what was authorized in the plant specs, running the plant through a test procedure that had never been operationally been evaluated during the plant's certification, and trying to maintain a higher than rated power output for months at a clip. These are HUMAN failings, not technical ones. Let's take Three Mile Island. Operator error significantly compounded the original problem of a stuck open valve into a major problem that cost a billion dollars to stabilize and fix, although with no loss of human life or health.
This is why I mentioned politics and not planning for the future as reasons why this can go badly. If we approach nuclear power with a long-term view, with a proper safety culture, independent monitoring, no arbitrary "power for no more than this cost by this date" goals, then I think it is as safe as any other power generation technology out there - probably safer when one considers environmental impact (I'd rather have all my toxins in one place than thinly distributed throughout the atmosphere). On the other hand, if we wait until a true energy crisis has developed, and rush through the creation, monitoring, and oversight of these plants, then I DO fear what would happen. While I hope your optimism about our inherent safety culture is true ("No one's going to build and fuel a reactor like that, and they CERTAINLY won't operate it with the absolute idiocy the Soviet crew [did]"), I am a bit more skeptical. In aviation, space travel, nuclear power plants, and other complex endeavors, safely seems to follow a saw-tooth curve; an accident happens, regs are implemented, companies institute safety programs with teeth, and safety improves drastically.... for a while. Then, pressures on cash flow, complacency ("it hasn't happened in years, it must be because we're doing it safely"), industry lobbyists, learning to run the edges rather than the center of the envelope, etc., slowly increase the risk level, until the next accident... and the cycle repeats. Perhaps we can engineer a foolproof nuclear plant, but I wouldn't bet on it.
Nothing is "safe" - but how we judge the odds and risks affects our expectations. The Space Shuttle was planned with an expected hull loss failure rate of about 1 in 10,000, when operational experience and the judgement of the engineers indicated about 1 in 100. If you wish to have a technology that is accepted and utilized properly, the risks have to be judged correctly, not sold like a cure-all. Read the Rodgers Report for more information:
t ml
http://www.ralentz.com/old/space/feynman-report.h
In addition, when you depart the ground in a 747, you may have up to 500,000 lbs. of a highly energetic fuel along for the ride, yet the hull loss rate on 747s is less than 1 per 1,000,000 departures. The presence of explosive material alone does not indicate the risk level. Commercial aviation is the beneficiary of decades of engineering experience, stringent safety regulation, and a culture that encourages disclosure of risk vs. burying the messenger and hoping the probabilities don't catch up on your watch.
So, to be more precise, the political process is more likely to generate safety estimates off by orders of magnitude (in order to sell the product), and then when those fail to come true (not through the fault of the technology, but the way it was used), the whole technology could be needlessly scrapped for decades.... like nuclear power was in the U.S. after the Three Mile Island incident. I was expressing the hope that we would be slightly wiser this time around in deploying a very costly but potentially very beneficial technology, and the most likely obstacles to it.
Nuclear power plants are a reasonable option, if we can do two things:
a) Depoliticize the running of them - the first thing that suffers when politics overtakes reality is safety, as NASA is a perfect illustration of. Having a good, long-term safety policy built into an organization isn't something that can be done overnight, and building an agency to replace the DOE is impossible in the current polarized political environment.
b) Figure out how to prevent proliferation of high-grade fissionable materals; the technology for generating and seperating such is inherent with breeder reactors, if I understand them correctly.
Both of the above are certainly technically possible; but with our current disdain for effective governmental organizations, I think it unlikely we will achieve the above soon.
So, nuclear plants will work, and be built, but I wonder how they will be built. Slowly, carefully and sited & managed rationally? Or be built hurriedly, in a panic to dampen swings in energy prices, with no coherent oversight, policy or judgement?
If it is the latter, we will have our own Chernobyl in time.
Humph... when we were doing comedy, we didn't have any o' this namby-pamby television and camera and writer malarkey. We had to go up on a stage (well, it was more of a pit, really), no costumes, no lights, no dialog, make at least a dozen people die from laughing their guts up with jokes that we had just thought up that second in sign-language, and get violated by the club manager, IF we were lucky!
But, the young comedians today, they just won't believe you....
(If having it tougher than your kids isn't a timeless joke, I don't know what is, and they did it great)
Quite familiar with them, Eddie Murphy before he "changed", John Stewart, Mark Russell, Keith Robinson, Sherrod Small, etc., etc. - I like good comedy. But Python has something timeless and brilliant, particularly in sketch comedy - I don't think (with the exception of Cleese) they would have been so good at stand-up, or Adam Sandler style movies.
No arguing taste! I wish I found another group that captured that wonderful sense of tweaked properness, of just barely concealed rage, or pure self-loathing - I haven't seen any other group get at these quite the same way.
...It is still amazing how well these shows have held up over 30+ years. Perhaps it is just sentimentality, but Monty Python is still one of the best comedy troupes, if not THE best, I have seen in my life.
Are we getting less funny? Or is familiarity part of the enjoyment? Or are they really just uniquely brilliant?
So when will the War on Terrorism be done? Let me know what the criteria are so I can prod my local bureaucrats into restoring a few freedoms that have been lost this round. Ditto for the War on Drugs, running for decades now, with no clear winners or losers or end.
It is a bit more insidious in modern times, I think...