Ukraine has no oil. However, Stalin ordered demolition of any and all industrial infrastructure that could be of use to Nazis. That was reasonable: the invaders were denied opportunity to manufacture weapons locally. As result, Hitler had to use railways to send *everything* to his troops, and railways were regularly blown up by guerillas.
No matter how well designed, unless it actually wrecks the gun when tampered with, any security system could potentially be by-passed. I doubt if it would take criminals long to figure out how to do it.
To shoot you only need to apply 100 mA (or something) to the laser diode. Any 1.5V battery will do the trick. This thing is easy to hack.
The worst thing about this gun is the need for a battery. Revolver is the most reliable design so far, IMO.
Are you sure? The only one I can see is this one and it certainly isn't GPL or GPL compatible.
That's what I saw too. It does not have to be GPL, though, as long as you don't plan to hack it, or distribute it - and rarely anybody wants to do any of that. So it is probably OK for a GPL program to use.
But sure, it would be nice if there was a GPL release of a modern (3.1.2) Qt on all platforms. However, Trolltech obviously needs some cash flow to stay in business (and to make other s/w releases available, free or not). They probably get 90% of their revenue from Win32 platform, and anything that might reduce it would be bad for their business.
Opera, VariCAD and Eagle CAD are some of examples. For more case studies you probably should read TrollTech's Web site, I am not going to do their work for them:-)
QT isn't available under GPL on Windows
I just checked, it is available. Not the latest release, though. That might be a concern. You can use an evaluation version on Windows, it is latest, but comes with some strings attached. So GTK looks like a winner in the "freedom" department, and may be a valid option for GPLed software.
But any professionally developed software (free or for money) will be better off with Qt, because it is not a "port" of something UNIX to Win32 - it uses a whole bunch of Win32 things where they are needed (such as threads, semaphores, networking.)
One of BIG advantages of Qt is its guaranteed portability. GTK on other platforms is/was dead, or kludgily ported at best. But Qt is reliably working on X, Windows, Mac and now on Qtopia framebuffer as well. This portability is not a rumor (as it was for GTK when I looked), it is a fact, and there is a solid company behind that fact.
I chose Qt for my work, and I haven't regretted it ever since. It is free for GPLed code, and inexpensive for commercial use - perfect for both worlds - and is well supported.
Re:Improve and go on until a third accident
on
More on Columbia
·
· Score: 1
The goal #1 must be the development of advanced spaceflight science and technologies. Without control of gravity you can not have efficient space travel (unless you live on asteroids). Without higher dimensional jumps you can not get to the stars (in any meaningful time). Without advanced energy sources (fusion, antimatter) you can't use those technologies.
Those are required things, and current experiments on LEO only prove that. Simple physics, in form of laws of reactive motion, tells us how wasteful it must be to launch anything really big, really far and fast. Those laws haven't changed yet. Launches are so expensive simply because so many tons get burned out or thrown away just to put one pound on the low orbit. Any active interplanetary flight is virtually impossible because of the same reasons - you need to have gigatons of reaction mass, even if you can throw it away sufficiently fast. Basically, traditional rockets can not take us to the next level of spaceflight. They are even less than a horse on a highway, limited in speed and range.
All of the required technology will be built on the planet, not in space. If some research has to be in space - fine, launch the experiment. But the majority of the work is done not in space, not by people figuring out how long they can stay weightless before they die. The work is done here.
It is quite similar to cross-atlantic trips in time of Magellan: Yes, you can sail to West India. No, you likely will die on your way there. Technology progressed (not in the oceans, by the way - on the land), and now we can sail and fly across, with little risk and at moderate cost.
Re:Improve and go on until a third accident
on
More on Columbia
·
· Score: 1
it's still a whole lot safer than driving your car on a "lives lost per mile travelled" basis.
Then it is a wrong basis for the comparison, because nobody cares about miles travelled. Number of failures per 100 missions flown, for example, would be the right one.
The Shuttle faces safety comparison to hundreds and hundreds of manned capsule flights - American, Russian and Chinese soon too. It fails to come on top - capsules are inherently safer, and cheaper too. Fact is, modern technology is not ready yet for a fully reusable spacecraft - especially when NASA cancels all the research in that direction!
Ultimately, the choice should belong to those who put their lives on the line.
Yes, as long as the astronauts pay for their flights with their own money. Until then, the taxpayer has the decisive vote.
Probably the most expensive parts are: batteries, motors, custom metal components, and labor (including his extreme factory size!). Microcontrollers are cheap these days, and the industry is well adapted to low volume manufacturing.
sort of environments where people feel they mustn't be the last to come in in the morning [...] aren't necessarily fostering productivity
It's realm of larger companies. As long as the company is 20 people or less this is not a problem. When management can't evaluate the actual productivity then it starts using other, less precise parameters (such as when you come and go).
It is nice that you could tell the PHB to get lost and get away with it. In many companies this is not an option.
Myself, I work about 100-110 hours per week, but I am not an employee. Noblesse oblige, as they say. Our employees work about 8 hours per day, and I don't like when they work much longer (because that rarely contributes to the project, and wrecks the regular schedule).
If PHBs promise to their bosses to be done with such a project in 3 months, then they are stupid, and they will be reprimanded for that. It is obvious that you can not violate laws of nature. Yes, I worked in various teams, but fortunately not under a PHB; managers always understood the basic principles, and listened to coders.
When I worked in a larger company (500 employees) we had a guy who wasn't sure what he is doing, and nobody else had a clue either. So the guy just did some easy paperwork, went to some trade shows, and generally had fun. He was fired after one new manager decided to investigate. Things easily fall through cracks in larger companies...
If the slacker does get paid the same salary then he works more hours to meet the same goal. If he fails to meet a deadline then I will reassign the guy (or fire him if he is worthless.) And I won't pay hourly rate to such a person, because he can't achieve the expected performance.
Of course, if someone is too slow and just ties up resources (cubicle, computer, etc.) then he needs to be replaced, no arguing about that. But if we are talking about 10-15% slower then it is OK, and the guy is free to work 9 hours if he prefers to snooze now and then in between (or post on Slashdot:-) Legally it will be overtime, but factually it is not, as you see.
So in one word, no, I do not keep slackers. We just fired one (a student lost between two worlds.) On the other hand, I try to be reasonable. People cost a lot to hire and bring up to speed on our s/w, so it makes sense to keep them if it is practical.
I guess that the main reason why people want an hourly rate is so that they can estimate how much money they'll have at the end of the week.
It's even easier with salary.
Imagine if you're a young engineer who knows that you can complete a six week $25,000 project in two weeks
Then you should work as a contractor. It is utter naivete to expect your employer to pay you for 4 weeks after you have finished the job. The employer will just think that he miscalculated the complexity of the project. And if the coder insists on collecting the 4 week wage for "improving yourself" he will be thrown out in no time.
If you are that good, you must be your own boss. There is no workaround. Granted, it's a pain to manage your own business, but it can be done; and if it hurts so much, find a partner who is better in these things. This is very common.
Whether you pay both of them salary or an hourly wage, the end result is the same.
We are talking about overtime here. Overtime can be a buffer that accomodates -unusual- decline in productivity. It is also a safety valve that can keep a person employed even when he is not as productive this month as other programmers. Some people are always slow; quite a few of them work slowly but deliver excellent results. You can't measure everyone with the same ruler.
There are many reasons why productivity drops. People have all sorts of personal problems, all the time, and it is unrealistic to expect them to work "top notch" every single day. However, the bureaucratic layer (Catbert) easily can use this fluctuation as a reason for dismissal - especially when there are 100 other guys standing in line to get a job.
And now you stumble upon the unenviable task of estimating time for tasks.
Why is it then that if I finish my job ahead of schedule I'm given more work without more pay
Because you are not skilled in managing your own labor resource:-) You are expected to sell it to the employer. But if you give it away then the employer will gladly take it.
What you can do is to ask for raise, or to relax a bit. Another benefit of relaxation is that you don't want to work significantly better than your colleagues because they then will look bad.
It's all human interaction and workplace politics. You should learn it one day:-)
Sorry, it can't work this way. For example, one guy is lazy and stupid, and it takes him 3 days to code "Hello World" in Perl. And another guy is -normal- (not even genius), and it takes him 3 hours to do a similar job.
Now tell me how can I pay them hourly if the lazy guy just relaxes, while the other one works?
One fair way is to pay per work performed. You estimate some reasonable time needed, you give the assignment, and whenever they finish is up to them. If the lazy guy has to come on weekends, it's his problem.
The only alternative is to fire the lazy guy. But I fail to see how it helps; and as an employer I really don't mind using lazy guy's help even it comes slower than usual. People are different, and something that is obvious to one may require extensive reading to another.
So whoever wants to do a GPL clone will either have to reverse engineer the search and connection code, or develop their own algorithms that perform just as well.
Who chases who will depend on market penetration, and a GPLed client will win even if it not as efficient as the closed source competition. After all, the problem here is quite simple, and GPLed client can be based on any number of established protocols and codebases. It's the proprietary code that can't do that!
Re:Expect fianl report in 6 months
on
Latest Columbia News
·
· Score: 2, Insightful
Last seconds of the flight would be definitely "nice to have", but they are hardly all that important. The wing was mostly destroyed by then, and effects of that are of no relevance -- especially because the original area of the defect was probably falling toward the surface for quite some time already.
Much more valuable is the data about what led to the incident - and that data had been collected normally. So I must concur that in this case a "black box" would be of no use.
Russia continues to waste money on manned spaceflights... If those were cancelled (as some of them will likely be, in today's wake), then funds would be freed, some of which will make it directly into their healthcare system.
It has been already announced that none of Russian fligths will be cancelled (and why should they be, using completely different designs?)
It is also worth noting that Russia has free healthcare for, like, last 80 years or so (along with for-pay service.) If any money would be taken away from space research it would be probably just stolen or mismanaged.
Thay claimed MP3 was evil and now nearly 50% of the DVD players on the market plays MP3...
That's because evil things sell. "Bad" means good business.
This is exactly the problem Sony and others face - one branch of the company wants to block copying of its movies, and another branch wants to sell fastest computers to rip those very movies:-)
Where I worked, they used MS Project to do the project tracking. It has little to do with actual construction work, though. A foreman would report the progress back to the office once a day, and that's it.
The flaw in the original reasoning was that the foreman does not need external info to find out what his men are doing; he is the *source* of that information for everybody else. The foreman only needs a walkie-talkie to be happy. Besides, nobody is going to accept a click on some Palm Pilot device as a substitute for a complete set of paperwork.
There must be a bit of mystery everywhere:-) After all, this is hardly a trade secret. You rarely select such size for a printer (unless you just happen to have an HP plotter), but it's there, in the combo box.
More like unbelievably well.
Well, yes. It is somewhat easier to stretch a room a little in AutoCAD, instead of redoing the whole drawing. Hmm, looks like I did it again:-)
Tell me where these architects work that prefer pen and paper, I'll go steal their job.
They work in a big city in Minnesota that is more like two cities.
You can do infinitely more with a computer rather than pen and paper.
Yes and no. It depends on how much you personally must do, and how much help (CAD operators) you have. Pen in hand is quicker, if you only need to draw a small section here or view there. It all depends, of course. One person was personally attending to floor tiles, but another just took a highlighter and marked rooms according to the floor type. Then I had to put all these markers in all these rooms myself:-) In another case, an architect asked me to draw an elevation of a specific building (we had 46 of them, IIRC), and after that was done he used his favorite red pen to get rid of some stuff and to add some (stucco, as I recall). That was his style; other people had other styles of work.
foremen could fire up their PocketPC handhelds and see what their teams were working on that day.
I worked with architects and construction people, this should be close to shipbuilding. I *guarantee* that no foreman would even *think* of getting any usable info from a Palm Pilot. A foreman has heaps of A0 drawings (if you don't know what A0 size is, check it out), and most of those drawings are already in his head.
A foreman does not *need* a computer. It is too slow, and has too low a resolution. Each drawing has tens of thousands pixels across, and we used all of those pixels - a building is long, and each room and each wall have their dimensions, and these dimensions must be readable.
Also, a foreman does not need to check his Palm Pilot to know what his team is doing. That is because he is right there, with his team, running from one work site to another, checking the work and giving instructions all the time. That's what his job is about - not "checking his Palm Pilot".
Frequently a foreman needs to talk to an engineer who oversees the construction. Then he reaches for his walkie-talkie, or walks to the office, usually with drawings in hand. Then he sits with the architect, who then draws sketches for him to explain this corner, or that insulation layer. A computer here is mostly useless, since pen and drafting paper are much faster. Pen is also easier to use, especially if a foreman does not have a university education.
Computers are widely used as drafting tools, and they do this job reasonably well. But computer manipulation of drawings is not something that even an architect is good with. Most architects prefer pen and paper, and they all draw very well. Maintenance of the drawings is something that only draftsmen do.
Slashdot Mirror
Ukraine has no oil. However, Stalin ordered demolition of any and all industrial infrastructure that could be of use to Nazis. That was reasonable: the invaders were denied opportunity to manufacture weapons locally. As result, Hitler had to use railways to send *everything* to his troops, and railways were regularly blown up by guerillas.
To shoot you only need to apply 100 mA (or something) to the laser diode. Any 1.5V battery will do the trick. This thing is easy to hack.
The worst thing about this gun is the need for a battery. Revolver is the most reliable design so far, IMO.
That's what I saw too. It does not have to be GPL, though, as long as you don't plan to hack it, or distribute it - and rarely anybody wants to do any of that. So it is probably OK for a GPL program to use.
But sure, it would be nice if there was a GPL release of a modern (3.1.2) Qt on all platforms. However, Trolltech obviously needs some cash flow to stay in business (and to make other s/w releases available, free or not). They probably get 90% of their revenue from Win32 platform, and anything that might reduce it would be bad for their business.
Opera, VariCAD and Eagle CAD are some of examples. For more case studies you probably should read TrollTech's Web site, I am not going to do their work for them :-)
QT isn't available under GPL on Windows
I just checked, it is available. Not the latest release, though. That might be a concern. You can use an evaluation version on Windows, it is latest, but comes with some strings attached. So GTK looks like a winner in the "freedom" department, and may be a valid option for GPLed software.
But any professionally developed software (free or for money) will be better off with Qt, because it is not a "port" of something UNIX to Win32 - it uses a whole bunch of Win32 things where they are needed (such as threads, semaphores, networking.)
I chose Qt for my work, and I haven't regretted it ever since. It is free for GPLed code, and inexpensive for commercial use - perfect for both worlds - and is well supported.
Those are required things, and current experiments on LEO only prove that. Simple physics, in form of laws of reactive motion, tells us how wasteful it must be to launch anything really big, really far and fast. Those laws haven't changed yet. Launches are so expensive simply because so many tons get burned out or thrown away just to put one pound on the low orbit. Any active interplanetary flight is virtually impossible because of the same reasons - you need to have gigatons of reaction mass, even if you can throw it away sufficiently fast. Basically, traditional rockets can not take us to the next level of spaceflight. They are even less than a horse on a highway, limited in speed and range.
All of the required technology will be built on the planet, not in space. If some research has to be in space - fine, launch the experiment. But the majority of the work is done not in space, not by people figuring out how long they can stay weightless before they die. The work is done here.
It is quite similar to cross-atlantic trips in time of Magellan: Yes, you can sail to West India. No, you likely will die on your way there. Technology progressed (not in the oceans, by the way - on the land), and now we can sail and fly across, with little risk and at moderate cost.
Then it is a wrong basis for the comparison, because nobody cares about miles travelled. Number of failures per 100 missions flown, for example, would be the right one.
The Shuttle faces safety comparison to hundreds and hundreds of manned capsule flights - American, Russian and Chinese soon too. It fails to come on top - capsules are inherently safer, and cheaper too. Fact is, modern technology is not ready yet for a fully reusable spacecraft - especially when NASA cancels all the research in that direction!
Ultimately, the choice should belong to those who put their lives on the line.
Yes, as long as the astronauts pay for their flights with their own money. Until then, the taxpayer has the decisive vote.
Probably the most expensive parts are: batteries, motors, custom metal components, and labor (including his extreme factory size!). Microcontrollers are cheap these days, and the industry is well adapted to low volume manufacturing.
It's realm of larger companies. As long as the company is 20 people or less this is not a problem. When management can't evaluate the actual productivity then it starts using other, less precise parameters (such as when you come and go).
It is nice that you could tell the PHB to get lost and get away with it. In many companies this is not an option.
Myself, I work about 100-110 hours per week, but I am not an employee. Noblesse oblige, as they say. Our employees work about 8 hours per day, and I don't like when they work much longer (because that rarely contributes to the project, and wrecks the regular schedule).
When I worked in a larger company (500 employees) we had a guy who wasn't sure what he is doing, and nobody else had a clue either. So the guy just did some easy paperwork, went to some trade shows, and generally had fun. He was fired after one new manager decided to investigate. Things easily fall through cracks in larger companies...
Of course, if someone is too slow and just ties up resources (cubicle, computer, etc.) then he needs to be replaced, no arguing about that. But if we are talking about 10-15% slower then it is OK, and the guy is free to work 9 hours if he prefers to snooze now and then in between (or post on Slashdot :-) Legally it will be overtime, but factually it is not, as you see.
So in one word, no, I do not keep slackers. We just fired one (a student lost between two worlds.) On the other hand, I try to be reasonable. People cost a lot to hire and bring up to speed on our s/w, so it makes sense to keep them if it is practical.
It's even easier with salary.
Imagine if you're a young engineer who knows that you can complete a six week $25,000 project in two weeks
Then you should work as a contractor. It is utter naivete to expect your employer to pay you for 4 weeks after you have finished the job. The employer will just think that he miscalculated the complexity of the project. And if the coder insists on collecting the 4 week wage for "improving yourself" he will be thrown out in no time.
If you are that good, you must be your own boss. There is no workaround. Granted, it's a pain to manage your own business, but it can be done; and if it hurts so much, find a partner who is better in these things. This is very common.
We are talking about overtime here. Overtime can be a buffer that accomodates -unusual- decline in productivity. It is also a safety valve that can keep a person employed even when he is not as productive this month as other programmers. Some people are always slow; quite a few of them work slowly but deliver excellent results. You can't measure everyone with the same ruler.
There are many reasons why productivity drops. People have all sorts of personal problems, all the time, and it is unrealistic to expect them to work "top notch" every single day. However, the bureaucratic layer (Catbert) easily can use this fluctuation as a reason for dismissal - especially when there are 100 other guys standing in line to get a job.
And now you stumble upon the unenviable task of estimating time for tasks.
Unenviable it is, but necessary nonetheless.
Because you are not skilled in managing your own labor resource :-) You are expected to sell it to the employer. But if you give it away then the employer will gladly take it.
What you can do is to ask for raise, or to relax a bit. Another benefit of relaxation is that you don't want to work significantly better than your colleagues because they then will look bad.
It's all human interaction and workplace politics. You should learn it one day :-)
Sorry, it can't work this way. For example, one guy is lazy and stupid, and it takes him 3 days to code "Hello World" in Perl. And another guy is -normal- (not even genius), and it takes him 3 hours to do a similar job.
Now tell me how can I pay them hourly if the lazy guy just relaxes, while the other one works?
One fair way is to pay per work performed. You estimate some reasonable time needed, you give the assignment, and whenever they finish is up to them. If the lazy guy has to come on weekends, it's his problem.
The only alternative is to fire the lazy guy. But I fail to see how it helps; and as an employer I really don't mind using lazy guy's help even it comes slower than usual. People are different, and something that is obvious to one may require extensive reading to another.
Who chases who will depend on market penetration, and a GPLed client will win even if it not as efficient as the closed source competition. After all, the problem here is quite simple, and GPLed client can be based on any number of established protocols and codebases. It's the proprietary code that can't do that!
No. The expression is quite common.
Much more valuable is the data about what led to the incident - and that data had been collected normally. So I must concur that in this case a "black box" would be of no use.
It has been already announced that none of Russian fligths will be cancelled (and why should they be, using completely different designs?)
It is also worth noting that Russia has free healthcare for, like, last 80 years or so (along with for-pay service.) If any money would be taken away from space research it would be probably just stolen or mismanaged.
That's because evil things sell. "Bad" means good business.
This is exactly the problem Sony and others face - one branch of the company wants to block copying of its movies, and another branch wants to sell fastest computers to rip those very movies :-)
The flaw in the original reasoning was that the foreman does not need external info to find out what his men are doing; he is the *source* of that information for everybody else. The foreman only needs a walkie-talkie to be happy. Besides, nobody is going to accept a click on some Palm Pilot device as a substitute for a complete set of paperwork.
There must be a bit of mystery everywhere :-) After all, this is hardly a trade secret. You rarely select such size for a printer (unless you just happen to have an HP plotter), but it's there, in the combo box.
More like unbelievably well.
Well, yes. It is somewhat easier to stretch a room a little in AutoCAD, instead of redoing the whole drawing. Hmm, looks like I did it again :-)
Tell me where these architects work that prefer pen and paper, I'll go steal their job.
They work in a big city in Minnesota that is more like two cities.
You can do infinitely more with a computer rather than pen and paper.
Yes and no. It depends on how much you personally must do, and how much help (CAD operators) you have. Pen in hand is quicker, if you only need to draw a small section here or view there. It all depends, of course. One person was personally attending to floor tiles, but another just took a highlighter and marked rooms according to the floor type. Then I had to put all these markers in all these rooms myself :-) In another case, an architect asked me to draw an elevation of a specific building (we had 46 of them, IIRC), and after that was done he used his favorite red pen to get rid of some stuff and to add some (stucco, as I recall). That was his style; other people had other styles of work.
I worked with architects and construction people, this should be close to shipbuilding. I *guarantee* that no foreman would even *think* of getting any usable info from a Palm Pilot. A foreman has heaps of A0 drawings (if you don't know what A0 size is, check it out), and most of those drawings are already in his head.
A foreman does not *need* a computer. It is too slow, and has too low a resolution. Each drawing has tens of thousands pixels across, and we used all of those pixels - a building is long, and each room and each wall have their dimensions, and these dimensions must be readable.
Also, a foreman does not need to check his Palm Pilot to know what his team is doing. That is because he is right there, with his team, running from one work site to another, checking the work and giving instructions all the time. That's what his job is about - not "checking his Palm Pilot".
Frequently a foreman needs to talk to an engineer who oversees the construction. Then he reaches for his walkie-talkie, or walks to the office, usually with drawings in hand. Then he sits with the architect, who then draws sketches for him to explain this corner, or that insulation layer. A computer here is mostly useless, since pen and drafting paper are much faster. Pen is also easier to use, especially if a foreman does not have a university education.
Computers are widely used as drafting tools, and they do this job reasonably well. But computer manipulation of drawings is not something that even an architect is good with. Most architects prefer pen and paper, and they all draw very well. Maintenance of the drawings is something that only draftsmen do.
I had this thing. Worked great, up to 117 km/h on 401, very cheap and stable. Better than 10 segways combined, in terms of speed and energy reserves.
Don't ask developers; ask MS policy makers.