My customer is the next poor slob who has to work on the source code. Software maintenance is a fact of life. Someday, somebody is going to need to read and understand what is in your code.
And that "next poor slob" probably won't even be someone else, it will be yourself.
in the end we're going to have tabbed windows because the future is more likely to be running in a light-weight web-browser interface to the Cloud on any device you can imagine, rather than a resource-heavy hardware-dependent Windows or OS-X environment.
In the future how many operating systems or applications are going to be considered resources-heavy compared to the hardware of the day? These days the browser is probably the most resource intensive app I run regularly. As hardware improves, the apps should become less resouce intensive in general.
And also, it is oft repeated but still true that people like to own their applications and data, even if their backup practices aren't usually as routine as those of a data center.
I would think that correcting Knuth on programming practice has got to be way up there.
You might also consider telling the computer to use some line breaks when you separate your paragraphs, since the aim of slashdot is ostensibly to communicate with other people.
I do work for others at about $55 / hour. Just spend the time I did on researching
Researching components is a sunk cost - unless you want to get mismatched components and pay too much for them in your pre-built computer of choice. The other thing doing your own component research will do is allow you to optimize your computer for low power, silence, speed, reliability and price. Do I want to spend the next 4-6 years putting up with a loud, slow, expensive to run and unreliable machine? I'll spend many more hours regretting my purchase if I don't spend the research hours up front.
OTOH, people who judge others based on their choice of IDE? Those people *are* tools...
Obviously someone who uses Visual Studio isn't necessarily an idiot. But maybe the parent has noticed a correlation (not causation) that extends beyond the anecdote. Someone who uses vim or (and I'll be charitable - it is a powerful editor) emacs almost by definition has to be intelligent enough to see that putting up with the learning curve in the short term is an investment that pays for itself many times over in time saved. And when you use an editor that can edit at the pace of your thinking, you find that you can concentrate better - your train of thought doesn't derail - which will give additional speed and ability. And maybe, a vim/emacs user also needs the intelligence to learn at a fast enough rate that the learning curve does not seem onerous.
You're also going to need some intelligence to even think that a decision on a text editor is something you should research. Someone smart could do a "site:slashdot.org text editor" search on google, find a poll and probably choose vim or emacs or something similar on that basis. If they are that smart, they are probably going to do that for other aspects of coding and it is going to show up in their coding ability. Someone stupid is going to stick with the default, or prioritize immediate ease of use over long term time saved.
In short: dumb users get filtered by the learning curve, smart users have incentive to learn the editor. Because of that the average vim/emacs user is going to be smarter than average, and also a better programmer. However, if someone uses another editor I am not going to make an assumption about their intelligence.
Before embarking on your adventure, you must first determine your own strengths and weaknesses. You have in your possession a physics book and a backpack containing provisions (food and drink) for the trip. You have been preparing your quest by training yourself in physics and doing exercises vigorously to build up your stamina.
To see how effective your preparations have been you must use the dice to determine your INTELLIGENCE, STAMINA and LUCK scores.
Hmmm, ok. Name: Analatoly Bugorski. INTELLIGENCE: 15. STAMINA:18. LUCK: 2...
364 You notice a large steel chamber with what appears to be a Geiger counter and a small tube of poison. You barely have time to register the odor of overripe soft cheese in the air. With a grunt of effort, the pale skinned physicist lunges at you. Pencils clatter to the floor. You quickly side step and extend your leg. He goes down in a heap. You brandish your physics book at him and he cowers in the corner, hissing. You notice that he appears to have two small lumps growing on either side of his forehead. This LHC business is getting stranger by the minute. If you demand to know what is going on in this place, turn to page 152. If you throw him in the metal box in the corner, asking him "How do YOU like it?", turn to page 377.
The only plausible explanation we could think of was that someone was intercepting and decrypting traffic in real time and filtering what they didn't like.
Because in the end, it's all in the mind. If ten million people want to rise up, but believe they are the only ones, then it will be much more unlikely that they really do it.
But if ten people believe that they really can change things, they will rise up, and change things. By showing others that they are not the only ones, and thereby starting the avalanche.
If those ten people get sent to jail, get shot, fall down an elevators shaft, get kneecapped, or even just lose their job, the next hundred people take exceptionally good notes.
Two factors can start off those first ten. If the alternatives are just as bleak, there is no downside to revolution, only upside. The second - if those first ten have wealthy, powerful benefactors bankrolling the "revolution".
Social mammals tend to emulate the alpha individuals of their groups. The alphas, by dint of successfully establishing themselves as alphas, are viewed as successful -- "well, they're doing something right for themselves, guess it'd be smart for me to do the same." When sociopaths lead our companies, the employees themselves will, generally speaking, start behaving more sociopathically. It's basic survival.
It used to be that every farm in North Dakota [where I live] had a windmill powering the farm. Then they disappeared and became an anachronism paying homage to a bygone era. Now windmills are dotting the countryside again. It didn't get windier here.
What changed?
Paid for energy (and associated products) had a more effective business model? My great grandfather had a windmill on his farm, used to pump water from a well. Worked perfectly fine, and for free since the setup cost had been paid for. His boys did him the "favor" of removing the windmill and installing an electric pump. The electric pump was newer and so the thinking went, better. That was just one extra bill after that, and didn't do him any favors.
The perfect business has a perpetually stable or growing income stream (energy, food, consumables, etc). A crappy business sells something that people only need one of, don't have any consumables you can sell them, and don't break down.
The more the business has a stable income stream, the more money it can spend on advertising to make itself the only blip on the radar to the consumer.
It is an unfortunate fact for the environment and those who like better products.
I would be interested in finding out if people with a dependence on both alcohol and tobacco would really be helped by this "vaccine" -- maybe taking the joy out of tobacco could lead to people consuming even more alcohol.
That's pretty much typical with any addiction - the addiction is not given up but rather transferred to something else. Food, alcohol, computer games, reading, work, web surfing, porn, gambling, other drugs, sniffing glue - whatever floats your boat, I'm sure you'll be doing a lot more of it after quitting smoking. But at least a few of those options are less harmful than cigarettes.
I would assume that the Microsoft Excel team did it this way as well, since Joel mentions it in his article. But they also wrote their own compiler because everyone else's was crap, and still managed to ship on time.
Is there anybody who can type as fast as they think? I'm not a great typist, but I can hit 90, and I can think, oh, I dunno, 3x that?
I'm getting at the number of editing passes you make before you metaphorically hit "submit". Sure, maybe I can think 3x faster than I can type as well. But only 10% of the time I spend on writing something is spent on actually typing, so the upper limit of an efficiency increase to the brain-machine interface is 10%. The other 90% of the time is editing - rereading, thinking, correcting, rearranging, final review, preview, etc. None of that activity is limited by the keyboard.
I have never looked at the SELinux code.... have you?
No, I haven't. I overstated with the bit about "all eyes in the world". "Many competent eyes" would be better, and would not change the outcome one bit.
Look at it from NSA's perspective. If I was in their shoes, I would assume Murphy's Law was in full effect. You are putting open source code out there with your name on it. The analogues of NSA in other countries will be looking at it, security researchers will be looking at it. Anyone who found a back door would be famous. As a result of this, the other products for public consumption produced by NSA would be much less trusted. And trust is what you want, right? After all, you are producing cryptographic tools for public use, but primarily you are in the business of reading other people's mail. You'd want people to trust and hence, use, the cryptography you already know how to crack. (For your own use, you use cryptography even you can't break.) Am I wrong?
Interestingly, grandparent dealt with the input side and I raised similar objections for the output side.
That assumes a few things though: that your occipital lobe is the highest bandwidth input possible, that visualizing symbols (words,numbers,etc.) is an efficient means of acquiring knowledge, that the brain couldn't learn faster if it had more efficient inputs, that direct memory creation isn't possible, and that your brain's wiring is optimal.
Personally, I think that we'll have strong AI before we have the answers to any of those. And not because strong AI is easy, but because if we can understand the brain in the level necessary to do much of what you suggest might be possible, we can simulate it, and simulation will be much, much easier than rewiring by hand.
Rewiring the brain to actually improve it - I doubt that would be possible outside of simulation. Direct memory creation - I'd like to see the mechanism proposed for this, as I would think that would be unworkable. More efficient inputs - again, all our ancestors from the time before we were mammals would have had eyes. That's a long time for evolution to come up with a highly efficient visual processing system and also, a visual system to general processing bus. I'd be surprised if a few electrodes are going to beat the existing, highly optimized system. Look at how fast we can visually process already. A good FPS player will notice lags of a few ms.
Slashdot Mirror
And that "next poor slob" probably won't even be someone else, it will be yourself.
In the future how many operating systems or applications are going to be considered resources-heavy compared to the hardware of the day? These days the browser is probably the most resource intensive app I run regularly. As hardware improves, the apps should become less resouce intensive in general.
And also, it is oft repeated but still true that people like to own their applications and data, even if their backup practices aren't usually as routine as those of a data center.
I would think that correcting Knuth on programming practice has got to be way up there.
You might also consider telling the computer to use some line breaks when you separate your paragraphs, since the aim of slashdot is ostensibly to communicate with other people.
does the trick.
I'm still chuckling... nice.
I for one, welcome the arrival of the Pillboxes.
You ever watch C beams glitter in the dark near the Tannhauser Gate... on weed?
Researching components is a sunk cost - unless you want to get mismatched components and pay too much for them in your pre-built computer of choice. The other thing doing your own component research will do is allow you to optimize your computer for low power, silence, speed, reliability and price. Do I want to spend the next 4-6 years putting up with a loud, slow, expensive to run and unreliable machine? I'll spend many more hours regretting my purchase if I don't spend the research hours up front.
Obviously someone who uses Visual Studio isn't necessarily an idiot. But maybe the parent has noticed a correlation (not causation) that extends beyond the anecdote. Someone who uses vim or (and I'll be charitable - it is a powerful editor) emacs almost by definition has to be intelligent enough to see that putting up with the learning curve in the short term is an investment that pays for itself many times over in time saved. And when you use an editor that can edit at the pace of your thinking, you find that you can concentrate better - your train of thought doesn't derail - which will give additional speed and ability. And maybe, a vim/emacs user also needs the intelligence to learn at a fast enough rate that the learning curve does not seem onerous.
You're also going to need some intelligence to even think that a decision on a text editor is something you should research. Someone smart could do a "site:slashdot.org text editor" search on google, find a poll and probably choose vim or emacs or something similar on that basis. If they are that smart, they are probably going to do that for other aspects of coding and it is going to show up in their coding ability. Someone stupid is going to stick with the default, or prioritize immediate ease of use over long term time saved.
In short: dumb users get filtered by the learning curve, smart users have incentive to learn the editor. Because of that the average vim/emacs user is going to be smarter than average, and also a better programmer. However, if someone uses another editor I am not going to make an assumption about their intelligence.
Hmmm, ok. Name: Analatoly Bugorski. INTELLIGENCE: 15. STAMINA:18. LUCK: 2...
Classic (and I use gnome!)
Ah, that seems logical. Thanks.
What would be the motive?
If those ten people get sent to jail, get shot, fall down an elevators shaft, get kneecapped, or even just lose their job, the next hundred people take exceptionally good notes.
Two factors can start off those first ten. If the alternatives are just as bleak, there is no downside to revolution, only upside. The second - if those first ten have wealthy, powerful benefactors bankrolling the "revolution".
No Bud Grace fans here?
It could work, you'd just need a bit more oil.
http://www.virtualschool.edu/mon/SoftwareEngineering/BrooksNoSilverBullet.html
More concisely - a fish rots from the head down.
...this is prima facie evidence that Google's "Don't be evil" policy is working very, very well.
Nice.
Paid for energy (and associated products) had a more effective business model? My great grandfather had a windmill on his farm, used to pump water from a well. Worked perfectly fine, and for free since the setup cost had been paid for. His boys did him the "favor" of removing the windmill and installing an electric pump. The electric pump was newer and so the thinking went, better. That was just one extra bill after that, and didn't do him any favors.
The perfect business has a perpetually stable or growing income stream (energy, food, consumables, etc). A crappy business sells something that people only need one of, don't have any consumables you can sell them, and don't break down. The more the business has a stable income stream, the more money it can spend on advertising to make itself the only blip on the radar to the consumer.
It is an unfortunate fact for the environment and those who like better products.
That's pretty much typical with any addiction - the addiction is not given up but rather transferred to something else. Food, alcohol, computer games, reading, work, web surfing, porn, gambling, other drugs, sniffing glue - whatever floats your boat, I'm sure you'll be doing a lot more of it after quitting smoking. But at least a few of those options are less harmful than cigarettes.
I refer to this article.
I would assume that the Microsoft Excel team did it this way as well, since Joel mentions it in his article. But they also wrote their own compiler because everyone else's was crap, and still managed to ship on time.
I'm getting at the number of editing passes you make before you metaphorically hit "submit". Sure, maybe I can think 3x faster than I can type as well. But only 10% of the time I spend on writing something is spent on actually typing, so the upper limit of an efficiency increase to the brain-machine interface is 10%. The other 90% of the time is editing - rereading, thinking, correcting, rearranging, final review, preview, etc. None of that activity is limited by the keyboard.
No, I haven't. I overstated with the bit about "all eyes in the world". "Many competent eyes" would be better, and would not change the outcome one bit.
Look at it from NSA's perspective. If I was in their shoes, I would assume Murphy's Law was in full effect. You are putting open source code out there with your name on it. The analogues of NSA in other countries will be looking at it, security researchers will be looking at it. Anyone who found a back door would be famous. As a result of this, the other products for public consumption produced by NSA would be much less trusted. And trust is what you want, right? After all, you are producing cryptographic tools for public use, but primarily you are in the business of reading other people's mail. You'd want people to trust and hence, use, the cryptography you already know how to crack. (For your own use, you use cryptography even you can't break.) Am I wrong?
Personally, I think that we'll have strong AI before we have the answers to any of those. And not because strong AI is easy, but because if we can understand the brain in the level necessary to do much of what you suggest might be possible, we can simulate it, and simulation will be much, much easier than rewiring by hand.
Rewiring the brain to actually improve it - I doubt that would be possible outside of simulation. Direct memory creation - I'd like to see the mechanism proposed for this, as I would think that would be unworkable. More efficient inputs - again, all our ancestors from the time before we were mammals would have had eyes. That's a long time for evolution to come up with a highly efficient visual processing system and also, a visual system to general processing bus. I'd be surprised if a few electrodes are going to beat the existing, highly optimized system. Look at how fast we can visually process already. A good FPS player will notice lags of a few ms.