Am I going to get calls from irate people telling me all the bad things my OS has been doing, or am I going to have to keep a careful eye on what all those other similar aged OS's are doing, to make sure they don't get my Linux system into trouble?
I agree - In my opinion, by the time someone reaches high school, they either are interested in science & maths, or they aren't.
For me though, my love of math's and science came about because my father got me interested in it from an early age, and I do wonder whether or not I would have discovered it to the same extent if I had been left to my own devices. I somehow get a feeling that I wouldn't have...
I mean, they can't really be relying on countries as poor as Brazil for their revenue, that would be plain wrong. On the otherhand, developed countries whose people do not live in such deep poverty, should pay for the drugs.
As long as Brazil doesn't try to profit from these drugs i.e. sell them to other countries, then I don't see a problem.
However, I cannot help wondering if a black market will arise, where the drug is trafficked from Brazil and those in need who can't afford it, to those in developed countries who are simply too tight to pay for it.
I had another reason for writing for QT as well - the fact that it has full support on both Windows, Linux, and (coming soon), the Mac (though I am unusure as to whether trolltech intend to release a free version as they have for Windows & Linux). Now, I'm only writing free software, but if I was to start writing code that I wished to sell, then I would want it to be cross platform (to capture as much of the market as possible), and I would want it to be easy to program (and QT is very easy to program).
IMHO, this means that kde will in the future, when Linux gains more popularity on the desktop, have the advantage over GNOME for commercial developers, and so more "big" apps (and closed source, and non-distributable...) might appear for KDE than GNOME.
If you consider this a good thing or a bad things I suppose depends on exactly where you stand on the Free software/open source argument.
Well, lot of GUI programs do offer quite advanced scripting ( MS Office being prime example) but the problem is that going from simple GUI Word functionality to VB scripting is not one step up.
For most people it is like 10 stories up and that is why hardly anybody bothers with learning this stuff.
OK, I've never used it, but I assume that VB scripting is still basically a "type commands down in a sort-of-programming type of way".
Or in other words, it's just a normal GUI with a shell/command line/programming language tacked on for those who know how to use it.
My initial idea is something like a flowchart design program, where the elements of the flow chart would be simple wrappers for the various standard unix commands, the idea being that you simply plug them together to do what you want. Each element would have an associated dialog box so that you could set specific switches, etc.
However, this would basically just be a wrapper for the shell, and although it would remove some of the "i can't use that!" factor, it would not bring the power to the everyday user.
So the next thing that needs to be done, would be to change it so that the user basically just specifies what goes in to the "GUI script", and what they expect to come out. The GUI would go away, decide which components (i.e. unix commands, etc.) could be used to do the job, construct the flowchart itself, and then give the user a nice dialog box asking for specific details on how to do it.
I haven't really worked out the details recently (and it's been a while since I last worked any details out) but a limited implementation at least seems feasable.
I'll try a quick code-up with a few limited commands and see how it goes if I have a chance
Who uses hotmail as their primary email anyway?
on
Hotmail Hacked
·
· Score: 1
Oops - lot's of my friends do actually...
Hmm....
Nah, (dismisses it with a wave of his hand), I'm too nice...
hmm...
Re:So what? (Humour, may offend but i doubt it)
on
Hotmail Hacked
·
· Score: 1
Actually it's very easy, as long as you obey one simple rule:
Subsitute "building and exploding atomic bombs in major cities for fun" or "disemboweling children at random" for "creating a free AI" and then see if you would still dismiss anyone who uses phrases like "monstrous and evil" and argues that perpetrators and advocates alike should be killed.
Actually, I would. Please hear me out.
By using phrases such as "monstrous and evil" it suggests that you have already written these people off as such, and are not prepared to discuss the situation any further.
In particular, it suggests that any question posed to you will be answered with the same phrase.
Also, you have to be very careful deciding what is evil and what is not, because it changes as society changes.
For example, contradicting the Bible used to be heresy, a monstrous and evil act since you were seen as speaking out against the will of God, and what could be more monstrous and evil than that? So they used to burn alive the people who did it.
Nowadays, burning people alive would probably be considered as a monstrous and evil act, and depending on where in the world the perpetrators lived they would either be locked away for life, electrocuted, hung, killed in a gas chamber, shot, etc. etc. etc.
Umm... which in themselves are (IMHO) fairly monstrous and evil acts, yet considered perfectly fine in the circumstances because "the perpetrator was monstrous and evil, and thus deserved it"
(incidentally, I never suggested that "people opposed to [my] argument should be killed," only that people attempting or supporting attempts to destroy humanity should)
But your argument here suggests that creating AI will destroy humanity. Fair enough, that's your opinion, but not everyone shares that opinion. That means that there is a chance that you are wrong. So you cannot (safely, without any doubt of becoming monstrous and evil yourself) say that any particular group of people is trying to destroy humanity unless you know absolutely that it is there intention to do so.
If it isn't their intention to do so, then the best that you can say is that they are misguided, in which case you should be aiming to educate them, but you can't to this by imposing your view on them and forcing them to accept it (after all, they might be right and you might be wrong - you can't be sure), so the only way to do the education is to discuss the situation with the people involved and try to convince them that your argument is best, whilst being prepared to take the risk that their reasoning might be better than yours.
To conclude, that's the reason I don't take people who use "monstrous and evil" in their posts seriously:-)
When I say that software should be free, I mean that once it has been written, it should be distributable for everyone for the cost of distribution (i.e, essentially free), since there is relatively no cost in distribution, compared to the cost of writing the software in the first place.
(And I've suddenly realised this argument can be extended to basically any media that can be copied over the internet, such as music, books, etc.)
The idea that I'm trying to get across is that there is only one copy of software that can be sold - the first one which was created through long hours of work, sweat and toil. Every other copy of the software costs the exact amount it does to, well, copy it.
Which means that there is a _big_ difference in price between the first copy and every other copy.
So the question that I am searching to answer so that this type of utopia could possibly exist is, who buys that first piece of software?
I agree that most everyday tasks of computers are best achieved with a GUI (which is why I specifically mentioned them as things you can do with a GUI).
Unfortunately, the GUI "dumbs people down", because it makes anything even slightly more complex (such as involving 2 seperate progams) very hard to do, or at least, very hard to do quickly. If you've ever tried doing something that the GUI doesn't specifically support then you know what I mean - you automatically feel that you have chosen the wrong way to do it, or at least think that this half hour of your time could have been better served doing something more productive.
I specifically picked an over-the-top example for what shell/command line can do for this very reason, but I agree that command lines are not for the majority of users.
Let us consider though, if someone could type the instruction that I did in English to the computer, and the computer could understand and carry out the task. Would that be a "dumbing people down" interface? Of course not, because it allows the user to specify exactly what they want to happen, and get it done.
Let me clarify my thoughts - the "dumbing down" that people suffer from is that they DO NOT REALISE that the computer can do something, and for this reason, they do not try to do it.
This is a problem with today's GUI interfaces. I'm not suggesting that I'm an expert, but I think that an improvement (an attempt of which I am contemplating) would be to somehow meld the simple concept of the GUI, with the incredible power offered by the "plug one program into another into another" power of the command line and scripting languages in such a way that is accessible to the majority of people.
A tall order, I know, but hell - I like impossible challenges...:-)
"Oh look, I want to write a letter, check my email, and check out slashdot at the same time.
Terminals and shells are great for:
I want to search this directory for bmps, take them and convert them into jpegs. Oh, I also want to count up how many of them there are, and make a nice sorted list of them showing me where they all are. Incidentally, I know there's over a gig's worth of images to process so please don't do it till 3 in the morning, and only if nobody else is using the system.
I challenge someone to do the example above with a GUI.
Never mind that, like it or not, Windows is a corporate standard that people need to know to get "standard" jobs.
Hmmm, if you mean secretarial stuff, then maybe but the majority of software I see running when I'm out at shops, or at the dentists, are THE MOST ARCANE, HIDEOUS, TEXT-BASED DATABASE PROGRAMS I never imagined where still in use.
If people can be trained to use them, then something as simple as MS Word should be a breeze.
But I found one that taught me the basics, how to use different colorspaces to your advantage, details that were independant of the graphics program.
Agreed. Technique, the "art" of what you are trying to do is independant of the interface you use, as long as you are familiar with that interface. To be honest, it took me no time at all to transition between Windows and KDE, and even my mother (who used Windows to browse the web on -gasp- 2 occasions) was able to start browsing the web on KDE as well after about 2 minutes prompting. (I wouldn't trust her to startup/shutdown either OS BTW...)
I guess it means that you get a bundle of shite software with it that you can do without - that's what happened when I first started using Blueyonder & surf unlimited.
Also, I'm pretty sure there's nothing stopping the author from relicensing the software under a different license. Isn't this what trolltech is doing with the QT toolkit?
I have a problem with closed source for a very simple reason - it can force the user to pay for work that they didn't want developers to do in the first place.
Say I'm happy with a proprietary piece of software "A version 1", except that it has a few bugs that need fixing, and could do with a few extra things. But - oh dear, version 2 just came out. Now version 2 has the functionality I want, but has a lot of extra stuff that I don't want, and they have stopped working on version 1. I can either a) buy version 2 and pay for all that stuff that I didn't want done, or b) live with the bugs and deficiencies in my current version.
Now consider that the same software was open source. I still paid the original developers for the work (assume for the moment that the license prevents redistribution - bare with me), but now they have released version 2, which I don't want. If it's cheaper for me, I could pay a different developer to patch up version one for me.
In other words, I pay for what I want done, and I have freedom to choose who I pay to do the work. Developers have the right to get paid for what they develop, but not the right to prevent other developers from getting paid for extending their product.
So how does this fit in with the GPL license? Obviously, the major difference is that GPL'd software is redistributable by anyone. What does this mean? That the price of software is in the development, and not the distribution, or in other words, developers get paid to write software, they don't get paid to distribute it.
Software should be free, development of it shouldn't (unless the developer so wishes)
Of course, the problem with this comes when we have a large scale project that takes great effort to write, with several developers, and which has several thousands of customers who have an interest in it being written. How do we make sure that the developers get paid the fairly hefty sum that they will be asking, without anyone losing out?
Um, d00d, one hour of C programmer time costs a lot more than a new sound card. What did you save again? Negative a hundred dollars?
I'm a student, and so don't have the luxury of earning hundred dollars an hour yet.
But ok, if I bought a new sound card, I would still have to install it into my computer, there could be side effects (though I haven't had any such problems any time recently) It would still take several hours of my time, plus the time for it to arrive. Ok, it's not the most mission-critical piece of hardware ever invented (in fact I don't think I _have_ any mission critical hardware), and so really it's more of a "I can do this, so I'm going to" than anything else - it's all good practice.
And anyhow, I'm perfectly happy with the soundcard I've got, it still works perfectly now I fixed the driver, I don't see why I should have to throw it away and buy a new one.
Slashdot Mirror
Am I going to get calls from irate people telling me all the bad things my OS has been doing, or am I going to have to keep a careful eye on what all those other similar aged OS's are doing, to make sure they don't get my Linux system into trouble?
I agree - In my opinion, by the time someone reaches high school, they either are interested in science & maths, or they aren't.
For me though, my love of math's and science came about because my father got me interested in it from an early age, and I do wonder whether or not I would have discovered it to the same extent if I had been left to my own devices. I somehow get a feeling that I wouldn't have...
I mean, they can't really be relying on countries as poor as Brazil for their revenue, that would be plain wrong. On the otherhand, developed countries whose people do not live in such deep poverty, should pay for the drugs.
As long as Brazil doesn't try to profit from these drugs i.e. sell them to other countries, then I don't see a problem.
However, I cannot help wondering if a black market will arise, where the drug is trafficked from Brazil and those in need who can't afford it, to those in developed countries who are simply too tight to pay for it.
PS "way FEWER words", and don't split your infinitives. Illiteracy is quite relative sometimes.
"To Boldy go where no man has gone before"
or should that be,
"To go boldly where no man has gone before"?
or even :
"To go where no man has gone before boldly"?
I think Shakespeare's vocabulary was about 10 times that of a normal person.
:-)
Though I have to wonder sometimes if he just used to make a lot of words up
I had another reason for writing for QT as well - the fact that it has full support on both Windows, Linux, and (coming soon), the Mac (though I am unusure as to whether trolltech intend to release a free version as they have for Windows & Linux). Now, I'm only writing free software, but if I was to start writing code that I wished to sell, then I would want it to be cross platform (to capture as much of the market as possible), and I would want it to be easy to program (and QT is very easy to program).
IMHO, this means that kde will in the future, when Linux gains more popularity on the desktop, have the advantage over GNOME for commercial developers, and so more "big" apps (and closed source, and non-distributable...) might appear for KDE than GNOME.
If you consider this a good thing or a bad things I suppose depends on exactly where you stand on the Free software/open source argument.
Well, lot of GUI programs do offer quite advanced scripting ( MS Office being prime example) but the problem is that going from simple GUI Word functionality to VB scripting is not one step up.
For most people it is like 10 stories up and that is why hardly anybody bothers with learning this stuff.
OK, I've never used it, but I assume that VB scripting is still basically a "type commands down in a sort-of-programming type of way".
Or in other words, it's just a normal GUI with a shell/command line/programming language tacked on for those who know how to use it.
My initial idea is something like a flowchart design program, where the elements of the flow chart would be simple wrappers for the various standard unix commands, the idea being that you simply plug them together to do what you want. Each element would have an associated dialog box so that you could set specific switches, etc.
However, this would basically just be a wrapper for the shell, and although it would remove some of the "i can't use that!" factor, it would not bring the power to the everyday user.
So the next thing that needs to be done, would be to change it so that the user basically just specifies what goes in to the "GUI script", and what they expect to come out. The GUI would go away, decide which components (i.e. unix commands, etc.) could be used to do the job, construct the flowchart itself, and then give the user a nice dialog box asking for specific details on how to do it.
I haven't really worked out the details recently (and it's been a while since I last worked any details out) but a limited implementation at least seems feasable.
I'll try a quick code-up with a few limited commands and see how it goes if I have a chance
Oops - lot's of my friends do actually...
Hmm....
Nah, (dismisses it with a wave of his hand), I'm too nice...
hmm...
Actually it's very easy, as long as you obey one simple rule:
Start soft...
Subsitute "building and exploding atomic bombs in major cities for fun" or "disemboweling children at random" for "creating a free AI" and then see if you would still dismiss anyone who uses phrases like "monstrous and evil" and argues that perpetrators and advocates alike should be killed.
:-)
Actually, I would. Please hear me out.
By using phrases such as "monstrous and evil" it suggests that you have already written these people off as such, and are not prepared to discuss the situation any further.
In particular, it suggests that any question posed to you will be answered with the same phrase.
Also, you have to be very careful deciding what is evil and what is not, because it changes as society changes.
For example, contradicting the Bible used to be heresy, a monstrous and evil act since you were seen as speaking out against the will of God, and what could be more monstrous and evil than that? So they used to burn alive the people who did it.
Nowadays, burning people alive would probably be considered as a monstrous and evil act, and depending on where in the world the perpetrators lived they would either be locked away for life, electrocuted, hung, killed in a gas chamber, shot, etc. etc. etc.
Umm... which in themselves are (IMHO) fairly monstrous and evil acts, yet considered perfectly fine in the circumstances because "the perpetrator was monstrous and evil, and thus deserved it"
(incidentally, I never suggested that "people opposed to [my] argument should be killed," only that people attempting or supporting attempts to destroy humanity should)
But your argument here suggests that creating AI will destroy humanity. Fair enough, that's your opinion, but not everyone shares that opinion. That means that there is a chance that you are wrong. So you cannot (safely, without any doubt of becoming monstrous and evil yourself) say that any particular group of people is trying to destroy humanity unless you know absolutely that it is there intention to do so.
If it isn't their intention to do so, then the best that you can say is that they are misguided, in which case you should be aiming to educate them, but you can't to this by imposing your view on them and forcing them to accept it (after all, they might be right and you might be wrong - you can't be sure), so the only way to do the education is to discuss the situation with the people involved and try to convince them that your argument is best, whilst being prepared to take the risk that their reasoning might be better than yours.
To conclude, that's the reason I don't take people who use "monstrous and evil" in their posts seriously
Let me clairfy my thoughts.
When I say that software should be free, I mean that once it has been written, it should be distributable for everyone for the cost of distribution (i.e, essentially free), since there is relatively no cost in distribution, compared to the cost of writing the software in the first place.
(And I've suddenly realised this argument can be extended to basically any media that can be copied over the internet, such as music, books, etc.)
The idea that I'm trying to get across is that there is only one copy of software that can be sold - the first one which was created through long hours of work, sweat and toil. Every other copy of the software costs the exact amount it does to, well, copy it.
Which means that there is a _big_ difference in price between the first copy and every other copy.
So the question that I am searching to answer so that this type of utopia could possibly exist is, who buys that first piece of software?
I'm still working on the answer...
Ok, I've calmed down a bit now :-)
:-)
I agree that most everyday tasks of computers are best achieved with a GUI (which is why I specifically mentioned them as things you can do with a GUI).
Unfortunately, the GUI "dumbs people down", because it makes anything even slightly more complex (such as involving 2 seperate progams) very hard to do, or at least, very hard to do quickly. If you've ever tried doing something that the GUI doesn't specifically support then you know what I mean - you automatically feel that you have chosen the wrong way to do it, or at least think that this half hour of your time could have been better served doing something more productive.
I specifically picked an over-the-top example for what shell/command line can do for this very reason, but I agree that command lines are not for the majority of users.
Let us consider though, if someone could type the instruction that I did in English to the computer, and the computer could understand and carry out the task. Would that be a "dumbing people down" interface? Of course not, because it allows the user to specify exactly what they want to happen, and get it done.
Let me clarify my thoughts - the "dumbing down" that people suffer from is that they DO NOT REALISE that the computer can do something, and for this reason, they do not try to do it.
This is a problem with today's GUI interfaces. I'm not suggesting that I'm an expert, but I think that an improvement (an attempt of which I am contemplating) would be to somehow meld the simple concept of the GUI, with the incredible power offered by the "plug one program into another into another" power of the command line and scripting languages in such a way that is accessible to the majority of people.
A tall order, I know, but hell - I like impossible challenges...
Hey, now they'll get force-fed Linux propaganda instead :-)
Someone mentioned that people using Apple 2's now use Windows ME - I'm not suprised. I prefered DOS to using those pieces of shite.
(Recently converted to Linux by the gorgeous KDE 2 desktop)
Oh, make that "directory tree" - my Bmps are all sorted by things such as "Computer Pics", "Fruit", "3D Graphics", etc.
And this is a made up example. Thought I should point that out...
GUI is great for:
"Oh look, I want to write a letter, check my email, and check out slashdot at the same time.
Terminals and shells are great for:
I want to search this directory for bmps, take them and convert them into jpegs. Oh, I also want to count up how many of them there are, and make a nice sorted list of them showing me where they all are. Incidentally, I know there's over a gig's worth of images to process so please don't do it till 3 in the morning, and only if nobody else is using the system.
I challenge someone to do the example above with a GUI.
Hmmm, if you mean secretarial stuff, then maybe but the majority of software I see running when I'm out at shops, or at the dentists, are THE MOST ARCANE, HIDEOUS, TEXT-BASED DATABASE PROGRAMS I never imagined where still in use.
If people can be trained to use them, then something as simple as MS Word should be a breeze.
But I found one that taught me the basics, how to use different colorspaces to your advantage, details that were independant of the graphics program.
Agreed. Technique, the "art" of what you are trying to do is independant of the interface you use, as long as you are familiar with that interface. To be honest, it took me no time at all to transition between Windows and KDE, and even my mother (who used Windows to browse the web on -gasp- 2 occasions) was able to start browsing the web on KDE as well after about 2 minutes prompting. (I wouldn't trust her to startup/shutdown either OS BTW...)
And Neopaint was (in my eyes) the _ultimate_ paint package.
Erm...
The reasons that I like to be able to browse books online are
a)It's easy to flick between relevant pages (assuming that keywords are hyperlinked)
b)It's incredibly easy to search them for specific phrases/words (as long as they have a search function, of course)
c)Um, that's it.
I don't see how flash would improve any of these criteria, in what circumstances do you deem it necessary?
I guess it means that you get a bundle of shite software with it that you can do without - that's what happened when I first started using Blueyonder & surf unlimited.
Also, I'm pretty sure there's nothing stopping the author from relicensing the software under a different license. Isn't this what trolltech is doing with the QT toolkit?
Isn't gcc 3 still unstable in certain respects too? Pretty much any software I've checked out so far tells you to use gcc 3 at your own risk.
Why don't you try writing one and find out?
Was you using a 370Kb page off slashdot, or the link on the site (which just leads to the homepage of slashdot)?
OK, this is my view/opinion on the issue.
I have a problem with closed source for a very simple reason - it can force the user to pay for work that they didn't want developers to do in the first place.
Say I'm happy with a proprietary piece of software "A version 1", except that it has a few bugs that need fixing, and could do with a few extra things. But - oh dear, version 2 just came out. Now version 2 has the functionality I want, but has a lot of extra stuff that I don't want, and they have stopped working on version 1. I can either a) buy version 2 and pay for all that stuff that I didn't want done, or b) live with the bugs and deficiencies in my current version.
Now consider that the same software was open source. I still paid the original developers for the work (assume for the moment that the license prevents redistribution - bare with me), but now they have released version 2, which I don't want. If it's cheaper for me, I could pay a different developer to patch up version one for me.
In other words, I pay for what I want done, and I have freedom to choose who I pay to do the work. Developers have the right to get paid for what they develop, but not the right to prevent other developers from getting paid for extending their product.
So how does this fit in with the GPL license? Obviously, the major difference is that GPL'd software is redistributable by anyone. What does this mean? That the price of software is in the development, and not the distribution, or in other words, developers get paid to write software, they don't get paid to distribute it.
Software should be free, development of it shouldn't (unless the developer so wishes)
Of course, the problem with this comes when we have a large scale project that takes great effort to write, with several developers, and which has several thousands of customers who have an interest in it being written. How do we make sure that the developers get paid the fairly hefty sum that they will be asking, without anyone losing out?
I'm still working on that one...
I'm a student, and so don't have the luxury of earning hundred dollars an hour yet.
But ok, if I bought a new sound card, I would still have to install it into my computer, there could be side effects (though I haven't had any such problems any time recently) It would still take several hours of my time, plus the time for it to arrive. Ok, it's not the most mission-critical piece of hardware ever invented (in fact I don't think I _have_ any mission critical hardware), and so really it's more of a "I can do this, so I'm going to" than anything else - it's all good practice.
And anyhow, I'm perfectly happy with the soundcard I've got, it still works perfectly now I fixed the driver, I don't see why I should have to throw it away and buy a new one.