So, I propably should try to find intelligent life on another planet.
I don't know if IRC has lost most of its 'meaning' - whatever that meaning is/was, but I meet my friends (and co-workers) every day in IRC.
I also meet many interesting people in IRC every day, not everyone there is an idiot. If meeting intelligent life in IRC seems hard for you, don't try any #teensex channels.
>Who wants to upgrade to more bloatware when >previous versions of the browser worked better. >Fix bugs, don't add more annoying features that >slow things down.
Nobody wants to upgrade to something that doesn't work, but believe it or not, there are some of us who actually like progress. I don't seen anything wrong with adding features (_if_ they can also be turned off) while fixing the bugs already found.
If everyone would just be happy with what we have and fixing bugs in existing software, Linux would not be what it is.
I don't understand this root thingy. Why do people think that virus should get root access to be effective. Of course that would make it a lot more dangerous.But even if a newbie would never log in as root, the virus would have access to all his data. When people still used dos and win3.x, viruses were spread in binaries, copying programs. Virus exploiting a security hole would live only as long as that hole. A lot more dangerous, in a long run, would be a hiding virus, stealing secret data and sending it somewhere. Of course, this only becomes issue, when we have many clueless users, downloading and running binaries right off the net.
So true. When there are bunch of people downloading and running binaries, life is lot easier for a virus.
Average windows-user doesn't know anything that is actually going on inside his/her computer. This is not going to change anytime soon, even if they all started using "most secure OS ever" (whatever that could be).
Problem with moving away from record companies is, if we like it or not, that they still are necessary evil. They are still handling the distribution and MARKETING. It's pretty hard to sell something if no-one knows anything about it. People have to be told that there's new good album out. Right now the big companies handle this, for a price. This is good for artists and consumers as well as companies. Maybe this will change in the future as the internet grows. But I fail to see how it would happen. Majority of people would still check only few places for records and pay for those sites to download them.
Did that make any sense? Point was that marketing is still necessary, expensive and risky. There are only few who can do it globally.
I know I would love to pay for the music I listen. I just don't want to by the whole cd when I only want one song.
It would be great if it was possible to: 1) listen to music (radio etc.) 2) decide if I want it for myself, to listen whenever I want 2) pay for it 3) the money would go to the musician who made it 4) this would be really easy, one click to select a song and pay for it
I don't know if this is possible right now, but I would think so. I just don't know where I could do it.
The limit for maximum size of perl-written application comes from the computing power and memory available. You propably don't want to completely drop c++ for any language.
I'm not one, who could define maximum size of application in any language, but I use following as a rule of thumb: If you want development speed, use perl. If you want run-time speed, use c++. If you want both, or something in between, use python (maybe combined with c++). And don't forget java for web-applications.
"Good argument": following is true in almost all cases: If you can do something in some language, you can do the same thing in perl, but faster.
Perl is a language, where it's possible to make things in more than one way. Main object is to get things done. If you want to write bigger applications in perl, that works too, but the language itself doesn't force you to write good, elegant code. Many use this as an argument against perl, but I think it should be the programmers responsibility. The language should just provide a way to make things happen. That's what perl does.
And for examples where perl is used, I suggest you do your web-search yourself.
I just wonder what operating system it will have. Linux is out of question, but I hope they don't install standard win2000 in it. Probably it will have a special version of W2K, or what do you think?
Why is it shame that IBM has abandoned OS/2 Warp? I'd rather see IBM starting to support (and maybe develop) Linux than to try to do anything to/with dying/dead operating system (OS/2).
(sorry, but I just have to say this) This doesn't have anything to do with UCITA. UCITA belongs to the USA. Software licences affect the whole world. If the worst (for you) would happen, and UCITA is passes, I welcame you all to europe.
It's the same thing with words and letters. If I put some letters after another forming a book or a novel, that's my text and no-one has the right to use it without my permission. Should I 'own' this comment. Bottom of this page says: "Comments are owned by the Poster". Interesting thing is, how short a story or piece of music can be 'owned'. Surely I can't have "rights" for a story only a few letters long, or a piece of music, that has only few notes.
(traditional) patter recognition doesn't work, because there are awful lot of patterns that could have 1 or 2 stones that could mean difference between good and awful move. Traditional neural networks would be far too slow and inaccurate for this kind of work. As long as we don't have the NN-hardware for this (I've got a good idea for this), I don't think we have the actual computational strenght for this.
(I'm drunk, but try to hold on) Just my opinion, but so far there have been weak programs written by strong players and really bad programs written by decent programmers. What we really need for go is a program, written by a pro, that knows the fundamentals and is able to actually _learn_ from that and evolve to something really sophisticated.
That's surely the way. I'm not a great player, but the shapes are extremely important, at least for human. One of the first thing a go player learns is to look for shapes, are they good or bad and are they 'alive' or 'dead'. This brings down the number of possible 'good' moves. The problem as I see it, is that there are so many shapes where only one stone is the difference between great move and lost game. Currently it seems to be out of our reach to figure out these automatically with current computing power.
There have been arguments that maybe the best person to make a go program would be one who is not very strong at go. If the programmer just knew the basic game and made a program to do all the fancy thinking. Or maybe if would be good if a weak player made a program and as the player gets better, he would always make sure the program can beat him. I have spent some time trying to figure out a way to make a strong go-program, but this is beyond my capability and/or interest.
I agree on the profit part. I think that amazon is going down, no matter what. Their only chance of survival is to start making money. That's not going to happen by filing and/or enforcing patents.
Did I misunderstand something, or are you saying that perl is not a good language to start with, because one can do 'bad code' easily with perl? Perl encourages people to do things the easy way. If these students have a good teacher, I don't know better language to start with. In perl, there are very few rules to learn and you can use (teach) same language for scripting and 'serious' programming.
hmm... you work for AMD, right? I don't think that any company in this industry could (or would) just sit and wait for the competitors to catch up (2.). When they could go on and practically dominate the market. It would be impossible (or very stupid) for AMD to hold back those "well over 1GHz" chips if they had any.
Slashdot Mirror
>If you're making an email killerapp, first make
>sure that the email part works, then add other
>functionality.
Why? What advantage you would get by developing
only one part of the program at a time? Assuming
you have many developers.
>When you get down to it what is so earth
>shattering about Linux? Nothing. It's just
>another OS for your apps to run on. Big deal.
Hmm... It's just another OS, but it happens to
be open-source. Really nothing earth-shattering,
but I think it's the best OS out there.
So, I propably should try to find intelligent life on another planet.
I don't know if IRC has lost most of its 'meaning' - whatever that meaning is/was, but I meet my friends (and co-workers) every day in IRC.
I also meet many interesting people in IRC every day, not everyone there is an idiot. If meeting intelligent life in IRC seems hard for you, don't try any #teensex channels.
>Who wants to upgrade to more bloatware when >previous versions of the browser worked better. >Fix bugs, don't add more annoying features that
>slow things down.
Nobody wants to upgrade to something that doesn't
work, but believe it or not, there are some of us
who actually like progress. I don't seen anything
wrong with adding features (_if_ they can also be
turned off) while fixing the bugs already found.
If everyone would just be happy with what we
have and fixing bugs in existing software, Linux
would not be what it is.
I don't understand this root thingy. Why do people think that virus should get root access to be effective. Of course that would make it a lot more dangerous.But even if a newbie would never log in as root, the virus would have access to all his data.
When people still used dos and win3.x, viruses were spread in binaries, copying programs. Virus exploiting a security hole would live only as long as that hole. A lot more dangerous, in a long run, would be a hiding virus, stealing secret data and sending it somewhere. Of course, this only becomes issue, when we have many clueless users, downloading and running binaries right off the net.
So true. When there are bunch of people downloading and running binaries, life is lot easier for a virus.
Average windows-user doesn't know anything that is actually going on inside his/her computer. This is not going to change anytime soon, even if they all started using "most secure OS ever" (whatever that could be).
Problem with moving away from record companies is, if we like it or not, that they still are necessary evil. They are still handling the distribution and MARKETING.
It's pretty hard to sell something if no-one knows anything about it. People have to be told that there's new good album out. Right now the big companies handle this, for a price. This is good for artists and consumers as well as companies. Maybe this will change in the future as the internet grows. But I fail to see how it would happen. Majority of people would still check only few places for records and pay for those sites to download them.
Did that make any sense? Point was that marketing is still necessary, expensive and risky. There are only few who can do it globally.
Or if no-one asks if it runs Linux, someone is going to ask if one could build a beowulf cluster out of them.
You can't underestimate stupidity.
I know I would love to pay for the music I listen.
I just don't want to by the whole cd when I only want one song.
It would be great if it was possible to:
1) listen to music (radio etc.)
2) decide if I want it for myself, to listen whenever I want
2) pay for it
3) the money would go to the musician who made it
4) this would be really easy, one click to select a song and pay for it
I don't know if this is possible right now, but I would think so. I just don't know where I could do it.
The limit for maximum size of perl-written application comes from the computing power and memory available. You propably don't want to completely drop c++ for any language.
I'm not one, who could define maximum size of application in any language, but I use following as a rule of thumb:
If you want development speed, use perl. If you want run-time speed, use c++. If you want both, or something in between, use python (maybe combined with c++). And don't forget java for web-applications.
"Good argument":
following is true in almost all cases:
If you can do something in some language, you can do the same thing in perl, but faster.
Perl is a language, where it's possible to make things in more than one way. Main object is to get things done.
If you want to write bigger applications in perl, that works too, but the language itself doesn't force you to write good, elegant code. Many use this as an argument against perl, but I think it should be the programmers responsibility. The language should just provide a way to make things happen. That's what perl does.
And for examples where perl is used, I suggest you do your web-search yourself.
I just wonder what operating system it will have. Linux is out of question, but I hope they don't install standard win2000 in it. Probably it will have a special version of W2K, or what do you think?
You are absolutely rigth. That's why general docs, reference manuals and tutorials are written. They are meant for people who want different things.
Why is it shame that IBM has abandoned OS/2 Warp? I'd rather see IBM starting to support (and maybe develop) Linux than to try to do anything to/with dying/dead operating system (OS/2).
Obviously IBM is just counting on customers to be insane enough to by systems running linux. They could be right, too.
(sorry, but I just have to say this)
This doesn't have anything to do with UCITA. UCITA belongs to the USA. Software licences affect the whole world. If the worst (for you) would happen, and UCITA is passes, I welcame you all to europe.
It's the same thing with words and letters. If I put some letters after another forming a book or a novel, that's my text and no-one has the right to use it without my permission.
Should I 'own' this comment. Bottom of this page says: "Comments are owned by the Poster". Interesting thing is, how short a story or piece of music can be 'owned'. Surely I can't have "rights" for a story only a few letters long, or a piece of music, that has only few notes.
(traditional) patter recognition doesn't work, because there are awful lot of patterns that could have 1 or 2 stones that could mean difference between good and awful move. Traditional neural networks would be far too slow and inaccurate for this kind of work. As long as we don't have the NN-hardware for this (I've got a good idea for this), I don't think we have the actual computational strenght for this.
(I'm drunk, but try to hold on)
Just my opinion, but so far there have been weak programs written by strong players and really bad programs written by decent programmers. What we really need for go is a program, written by a pro, that knows the fundamentals and is able to actually _learn_ from that and evolve to something really sophisticated.
That's surely the way. I'm not a great player, but the shapes are extremely important, at least for human. One of the first thing a go player learns is to look for shapes, are they good or bad and are they 'alive' or 'dead'. This brings down the number of possible 'good' moves. The problem as I see it, is that there are so many shapes where only one stone is the difference between great move and lost game.
Currently it seems to be out of our reach to figure out these automatically with current computing power.
There have been arguments that maybe the best person to make a go program would be one who is not very strong at go. If the programmer just knew the basic game and made a program to do all the fancy thinking. Or maybe if would be good if a weak player made a program and as the player gets better, he would always make sure the program can beat him.
I have spent some time trying to figure out a way to make a strong go-program, but this is beyond my capability and/or interest.
So I can actually own a shape! This is great news. I have these design ideas I call 'cube', 'triangle' and 'ball'.
I agree on the profit part. I think that amazon is going down, no matter what. Their only chance of survival is to start making money. That's not going to happen by filing and/or enforcing patents.
Did I misunderstand something, or are you saying that perl is not a good language to start with, because one can do 'bad code' easily with perl? Perl encourages people to do things the easy way. If these students have a good teacher, I don't know better language to start with. In perl, there are very few rules to learn and you can use (teach) same language for scripting and 'serious' programming.
hmm... you work for AMD, right? I don't think that any company in this industry could (or would) just sit and wait for the competitors to catch up (2.).
When they could go on and practically dominate the market. It would be impossible (or very stupid) for AMD to hold back those "well over 1GHz" chips if they had any.