First, it doesn't really matter what people prefer. In a decent market, there's variety, whether having to choose one manufacturer from many inconveniences some people or not. For example, 99% of cars aren't made by Ford. There are several companies that produce cars, and none of them has a mononpoly.
Second, the reason why there's no standard UI is because it's not needed. TCP/IP needs to be a standard in order to find some way for all computers to talk to each other. However, there's no such problem with user interfaces. Logging on a computer and seeing Gnome instead of KDE might be a minor inconvenience to me, but it doesn't stop me from going online and using programs. KMail will start in a Gnome environment just fine.
Third, you're confusing something here. GTK and Qt use X as if it was a framebuffer. Using the Linux fb device doesn't change much, because all that changes is the canvas they paint on. X, or the framebuffer know nothing of the appearance of Qt widgets, Qt dictates it.
Besides, I don't want to see Qt widgets with whatever look the Y developers like. I'm very happy with how Qt widgets look, thank you very much.
Heh, some people definitely have too much free time. So much time uselessly wasted on talking about something I wrote without giving it much importance.
I calculated a while ago that assuming that RAM was 5 cm away from the CPU, at 5 GHz a clock cycle would be lost on waiting for the signal to travel the 5 cm to the RAM and back.
If the speed of light is not far from being a limit at this point, then clock speed improvements can't continue working for long.
Besides, there's the question of whether it will "fly" or not. Clock speed doesn't measure performance. It especially says nothing of the performance of a new chip.
Yup, much happier. "Copyright infringement" would be a more correct term, but that's fine.
Corruption of the meaning of language is a powerful and quite evil tactic some people use. Copying software is not the same as stealing somebody's hard disk. It's also ridiculous to say that software piracy, which is essentially illegal sharing of data has something to do with taking over a ship, killing people and raping women.
The same way, the term "terrorist" has been redefined to almost mean "somebody who does anything the government doesn't like", making it easy to apply it to protesters in a manifestation.
Of course, if you use the exact terms it doesn't sound so impressive, does it?
I've said it before, and I'll say it again. Linux will never have one unique interface. It's simply impossible.
Windows has only one because it's made by one company that's got a monopoly. Linux has no company behind it, it's all made by volunteers. Try to convince Gnome people that they're losing their time and should be working on KDE. Or the other way.
Besides, the uniqueness of the Windows desktop is an artificial situation, not a natural one! If we had a decent market, there would be no MS monopoly, and instead we'd have 4-5 competing desktop OSes, probably based on standards.
Instead of having just Windows, if there was any justice, now we'd have a variety of Windows, OS/2, BeOS, Linux desktops, and who knows what else. That's how things are in other markets, and how they really should be.
Nope. You steal something when you remove something that isn't yours from its original position.
The whole concept of stealing was made long before computers appeared and only makes sense when applied for physical things. You can steal a computer, but not the software on it, unless you take the whole disk.
I got tired of it a long time ago. Last time I used ICQ (years ago) it was a huge program with a really long menu that seemed to include everything.
Didn't take me long to switch to licq, and later to Psi. So much better, just IM and no other crap. If I want to play games then I will just start a game.
Um, bandwidth adds up on the backbone. You don't need much processing power to handle a 2 Mbps connection. However, in order for you to have that there have to be places with traffic measured in hundreds of Gbps, and those will definitely benefit from faster processors.
Depends really. You *can* update a driver, recompile it, and insert without changing the kernel. How do you think people write modules, recompiling the whole kernel every time? No, they vi, gcc, insmod, rmmod, vi, gcc, insmod...
Besides, Windows is static, and Linux is getting better.
Yeah, I guess that Linux firewall at work that boots from a floppy and has no hard disk is a figment of my imagination.
There you have one place where Linux competes on tecnical merit. Try to come up with a way of putting Win2K firewall on a 1.44MB floppy. Heck, or even a 8MB flash card.
Why? It's pretty easy to understand. I'm pretty sure that most kids at the age of 10 and their parents can understand pretty well that HalfLife is nothing more but a glorified shooter, not that different from those old games ambiented in the west where you controlled the cursor and had to hit only the bad guys.
Swearing of course is much more realistic and somewhat more likely to have RL effects than pointing at 3D graphics and clicking
Question: Why do you use a dual Athlon MP for gaming? I'm tying this from one, but it doesn't help much with games. It's great for compiling code, and Gentoo, but doesn't improve games much. So, why aren't you using the XP2400 for games?
Oh yeah, WinRAR is pretty good. But few people use it, and I don't know anybody who registered it. And now that I found the GPLd 7-zip that compresses better I don't need or recommend it anymore.
Which is what I was pointing at. Yes, you can earn some money with your software. But you've got to be really really good. WinRAR, WinZIP are some of the very few shareware programs that got any money. For every one of those there's a thousand that will never pay for the developer's pizza.
That makes it a really crappy business model. I mean, I can easily buy fruit, potatoes, or whatever, and sell it without too much effort. I might not get a huge profit, but it's nothing compared to spending half a year on working on a program to then see that of 5000 people that downloaded only 5 bothered to register it.
Ah, but you're only taking into account the batteries. What about the rest?
From what I've seen, cars with gasoline motors break very often. An electric car should be very simple in comparison, and require less maintenance. Does anybody have numbers of how often has an electric car to be repaired, and about how difficult it is and how much it costs?
Nope, selling software is a very risky business. It works for some, but it's a very risky business these days.
Games and Maya sell just for one reason: They provide something not available anywhere else. There are very few programs that provide unique funcionality out there: 3DS Max/Maya/etc, vmware, big databases, etc. Nearly everything else, as operating systems, compression programs, text editors is available for free.
For Microsoft the reason they earn so much is that they sell to OEMs and businesses. Most normal people just get a copy from their friend. MS would be bankrupt long ago if it didn't have companies buying massive amounts of Office licenses.
To try to sell your own software these days is a pretty bad idea. Come on, just try to sell your own tool. Ask the thousands of shareware authors if they ever got more than $100 from all their users. And now Linux is starting to provide for free the only two product they have that are profitable. Yes, MS released a statement that said that only Windows and Office are profitable. Nearly everything else loses money.
These days it seems much easier to earn money by consulting, writing custom programs, support, maintenance, etc. And it looks like it's only going to go even further that way with time.
People will indeed pay if they have specific needs. That's why the Freenet project is managing to pay a programmer while having a small user base. But there are really few unique projects like that.
The load average says very little about how busy is the system, it only shows the last 1, 5 and 15 minutes of activity. Now if the output of procinfo was there you could judge if the system has been busy or not.
I've got a dual Athlon 2000+. The boot time improvement is next to inexistent. Overall it boots slower, due to the initial delay (about half a minute) to initialize ECC RAM, and the second or two Linux needs to initialize SMP.
Now, it's definitely a really big improvement over a 1 CPU system.
It's really smooth, and I can:
Burn CDs at 24x and play Quake 3
Compile programs using both CPUs and play videos at the same time.
Kill high priority programs (like sound daemons) that went mad for some reason and got into an infinite loop. This happened me with KDE a few months ago. With 1 CPU my computer froze for minutes until it could react to my request to kill the daemon. With SMP, no problem, I still have an idle CPU.
It's really nice for Gentoo. KDE compiles in several hours, and my computer isn't slowed down noticeably by the compilation. The basic Gentoo installation is done in a day, I can get all the necessary programs compiled during the second day.
The hardware is really stable too. Never locks up, never crashes. ECC RAM gives great peace of mind, too.
That probably won't work too well either. Several problems:
First, nobody got micropayments working yet. One simple reason is that it's probably costs more than $ 0.01 to process a $0.01 payment. The other reason is that many companies are trying it, but micropayments only work well when everybody uses the same company.
Now, let's ignore that. There are much larger problems:
First, nobody will want to enforce that. Just take a look at the discussion about TMDA, the amount of people who find it so rude that they'll never reply to it, and its conflicts with automated responses, tech support (reply coming from another address), etc.
Second, people upgrade very slowly. Forcing people to pay you'll cut yourself from 99% of the world, including most of your computer illiterate friends who will not know if it's a joke or what when they get the "payment required" email.
Third, ISPs will need to establish communication with every other ISP. I don't see why some small ISP in Russia would care much about some computer in Afghanistan claiming it wants a payment. In what currency, and what amount btw?
Fourth, it doesn't remove spam. Spamming will be simply done via worms that will infect the largest amount of computers possible and mail everybody in the address book. Given that currently we see that millions of computers are infected with Windows worms, this is going to be pretty effective.
And finally, nobody is going to like this system much anyway. Email is so popular because it's so simple. Making it like this will only make people switch to instant messengers or something else that's not so complicated, which is what's going to be spammed massively then.
Crypto or no crypto, the big problem here is the payment, which you ignore too easily. Ok, let's say grandma wants to send me an email. Let's also suppose that all this stuff has been integrated into Outlook already as well. Let's also assume that she just got a computer, so she's not in my list yet.
1. Grandma types her email address 2. Grandma sends it 3. Email arrives somewhere.
Now, how does the payment work? Clearly it has to work somehow. One option is that the payment is handled by the receiving server. I like running my own server, so I would do the charge. This means that grandma would have to send me her credit card number. Mistype the email address, and the wrong person gets the credit card number. Great opportunity for abuse.
Okay, suppose I can't run my mail server because I need a certification of that I'm an ISP, etc, etc. So my ISP's mail server does the charge instead. Now they end with a huge database of credit card numbers just waiting for being hacked. And of course some bright guy will manage to get the appropiate certs without being an ISP...
Ok, let's try without credit cards. We just make grandma's ISP pay me money. Her ISPs charges her, and then does a transfer to my ISP, that should be quite safe. There's also some kind of cryptographic authentication to make sure everybody is who they claim to be. But this is not good either. Ok, in the US maybe there could be a law mandating this implementation. But how do you get your mail from grandma when she lives in Russia and you moved to the US? Surely the russian ISP would have to agree to pay, and I don't see why would they bother.
Your ISP then drops the charge from Russia, but then the russian spammers can do pretty much as they please.
Additionally, even supposing the system is perfect it's still easy to abuse. A spammer writes a worm, infects my grandma's computer with it, and uses that to send 10000 emails at $0.01 each. Grandma can handle paying $100 you say? But she lives in Russia where an university teacher's wage is pretty close to $100. Seriously.
Charge a fair ammount for the country's economy? No, that won't work either. A spammer in the US would get an account in Russia, and pay almost nothing. Spammers are profitable with very little. If 1 in 10000 people pays $200, it's profitable. If mailing in Russia costs $0.0001, even more. I'm pretty sure some ISP in Russia would be very happy with such a customer.
Slashdot Mirror
Okay, several things.
First, it doesn't really matter what people prefer. In a decent market, there's variety, whether having to choose one manufacturer from many inconveniences some people or not. For example, 99% of cars aren't made by Ford. There are several companies that produce cars, and none of them has a mononpoly.
Second, the reason why there's no standard UI is because it's not needed. TCP/IP needs to be a standard in order to find some way for all computers to talk to each other. However, there's no such problem with user interfaces. Logging on a computer and seeing Gnome instead of KDE might be a minor inconvenience to me, but it doesn't stop me from going online and using programs. KMail will start in a Gnome environment just fine.
Third, you're confusing something here. GTK and Qt use X as if it was a framebuffer. Using the Linux fb device doesn't change much, because all that changes is the canvas they paint on. X, or the framebuffer know nothing of the appearance of Qt widgets, Qt dictates it.
Besides, I don't want to see Qt widgets with whatever look the Y developers like. I'm very happy with how Qt widgets look, thank you very much.
Heh, some people definitely have too much free time. So much time uselessly wasted on talking about something I wrote without giving it much importance.
I calculated a while ago that assuming that RAM was 5 cm away from the CPU, at 5 GHz a clock cycle would be lost on waiting for the signal to travel the 5 cm to the RAM and back.
If the speed of light is not far from being a limit at this point, then clock speed improvements can't continue working for long.
Besides, there's the question of whether it will "fly" or not. Clock speed doesn't measure performance. It especially says nothing of the performance of a new chip.
The ones everything uses, of course. ASCII, POSIX, TCP/IP, XML... it's not that hard to build several OSes that can interoperate.
Things would be much easier if MS just stopped reinventing the wheel. There was no need for SMB, NFS already existed.
As a geek I like arguing about semantics ;-)
:-P
Besides, my arguing about semantics can't be a sign of me losing the argument, because the whole argument was about semantics in the first place
What precognition? I said pretty much the same thing in the previous discussion about this. I'm just saying that it's not a new thought I had.
Yup, much happier. "Copyright infringement" would be a more correct term, but that's fine.
Corruption of the meaning of language is a powerful and quite evil tactic some people use. Copying software is not the same as stealing somebody's hard disk. It's also ridiculous to say that software piracy, which is essentially illegal sharing of data has something to do with taking over a ship, killing people and raping women.
The same way, the term "terrorist" has been redefined to almost mean "somebody who does anything the government doesn't like", making it easy to apply it to protesters in a manifestation.
Of course, if you use the exact terms it doesn't sound so impressive, does it?
I agree completely.
I've said it before, and I'll say it again. Linux will never have one unique interface. It's simply impossible.
Windows has only one because it's made by one company that's got a monopoly. Linux has no company behind it, it's all made by volunteers. Try to convince Gnome people that they're losing their time and should be working on KDE. Or the other way.
Besides, the uniqueness of the Windows desktop is an artificial situation, not a natural one! If we had a decent market, there would be no MS monopoly, and instead we'd have 4-5 competing desktop OSes, probably based on standards.
Instead of having just Windows, if there was any justice, now we'd have a variety of Windows, OS/2, BeOS, Linux desktops, and who knows what else. That's how things are in other markets, and how they really should be.
Nope. You steal something when you remove something that isn't yours from its original position.
The whole concept of stealing was made long before computers appeared and only makes sense when applied for physical things. You can steal a computer, but not the software on it, unless you take the whole disk.
I got tired of it a long time ago. Last time I used ICQ (years ago) it was a huge program with a really long menu that seemed to include everything.
Didn't take me long to switch to licq, and later to Psi. So much better, just IM and no other crap. If I want to play games then I will just start a game.
Well, PCI-X can certainly handle half a gigabyte per second, so I suppose that there's got some really expensive card that can do it too.
Um, bandwidth adds up on the backbone. You don't need much processing power to handle a 2 Mbps connection. However, in order for you to have that there have to be places with traffic measured in hundreds of Gbps, and those will definitely benefit from faster processors.
Depends really. You *can* update a driver, recompile it, and insert without changing the kernel. How do you think people write modules, recompiling the whole kernel every time? No, they vi, gcc, insmod, rmmod, vi, gcc, insmod...
Besides, Windows is static, and Linux is getting better.
You should try Perl.
You set up a socket, via either IO::Socket::INET, or IO::Socket::SSL (which of course requires a bit more mess to pass the keys, etc)
Whatever method you use, you can use IO::Select on it.
If this works in Perl, it can't be impossible to make some wrapper to make SSL as transparent in C as it is in Perl.
Yeah, I guess that Linux firewall at work that boots from a floppy and has no hard disk is a figment of my imagination.
There you have one place where Linux competes on tecnical merit. Try to come up with a way of putting Win2K firewall on a 1.44MB floppy. Heck, or even a 8MB flash card.
Why? It's pretty easy to understand. I'm pretty sure that most kids at the age of 10 and their parents can understand pretty well that HalfLife is nothing more but a glorified shooter, not that different from those old games ambiented in the west where you controlled the cursor and had to hit only the bad guys.
Swearing of course is much more realistic and somewhat more likely to have RL effects than pointing at 3D graphics and clicking
S/he is referring to the real life player dying, not the in-game character.
Question: Why do you use a dual Athlon MP for gaming? I'm tying this from one, but it doesn't help much with games. It's great for compiling code, and Gentoo, but doesn't improve games much. So, why aren't you using the XP2400 for games?
Oh yeah, WinRAR is pretty good. But few people use it, and I don't know anybody who registered it. And now that I found the GPLd 7-zip that compresses better I don't need or recommend it anymore.
Which is what I was pointing at. Yes, you can earn some money with your software. But you've got to be really really good. WinRAR, WinZIP are some of the very few shareware programs that got any money. For every one of those there's a thousand that will never pay for the developer's pizza.
That makes it a really crappy business model. I mean, I can easily buy fruit, potatoes, or whatever, and sell it without too much effort. I might not get a huge profit, but it's nothing compared to spending half a year on working on a program to then see that of 5000 people that downloaded only 5 bothered to register it.
Ah, but you're only taking into account the batteries. What about the rest?
From what I've seen, cars with gasoline motors break very often. An electric car should be very simple in comparison, and require less maintenance. Does anybody have numbers of how often has an electric car to be repaired, and about how difficult it is and how much it costs?
Nope, selling software is a very risky business. It works for some, but it's a very risky business these days.
Games and Maya sell just for one reason: They provide something not available anywhere else. There are very few programs that provide unique funcionality out there: 3DS Max/Maya/etc, vmware, big databases, etc. Nearly everything else, as operating systems, compression programs, text editors is available for free.
For Microsoft the reason they earn so much is that they sell to OEMs and businesses. Most normal people just get a copy from their friend. MS would be bankrupt long ago if it didn't have companies buying massive amounts of Office licenses.
To try to sell your own software these days is a pretty bad idea. Come on, just try to sell your own tool. Ask the thousands of shareware authors if they ever got more than $100 from all their users. And now Linux is starting to provide for free the only two product they have that are profitable. Yes, MS released a statement that said that only Windows and Office are profitable. Nearly everything else loses money.
These days it seems much easier to earn money by consulting, writing custom programs, support, maintenance, etc. And it looks like it's only going to go even further that way with time.
People will indeed pay if they have specific needs. That's why the Freenet project is managing to pay a programmer while having a small user base. But there are really few unique projects like that.
The load average says very little about how busy is the system, it only shows the last 1, 5 and 15 minutes of activity. Now if the output of procinfo was there you could judge if the system has been busy or not.
I've got a dual Athlon 2000+. The boot time improvement is next to inexistent. Overall it boots slower, due to the initial delay (about half a minute) to initialize ECC RAM, and the second or two Linux needs to initialize SMP.
Now, it's definitely a really big improvement over a 1 CPU system.
It's really smooth, and I can:
Burn CDs at 24x and play Quake 3
Compile programs using both CPUs and play videos at the same time.
Kill high priority programs (like sound daemons) that went mad for some reason and got into an infinite loop. This happened me with KDE a few months ago. With 1 CPU my computer froze for minutes until it could react to my request to kill the daemon. With SMP, no problem, I still have an idle CPU.
It's really nice for Gentoo. KDE compiles in several hours, and my computer isn't slowed down noticeably by the compilation. The basic Gentoo installation is done in a day, I can get all the necessary programs compiled during the second day.
The hardware is really stable too. Never locks up, never crashes. ECC RAM gives great peace of mind, too.
That probably won't work too well either. Several problems:
First, nobody got micropayments working yet. One simple reason is that it's probably costs more than $ 0.01 to process a $0.01 payment. The other reason is that many companies are trying it, but micropayments only work well when everybody uses the same company.
Now, let's ignore that. There are much larger problems:
First, nobody will want to enforce that. Just take a look at the discussion about TMDA, the amount of people who find it so rude that they'll never reply to it, and its conflicts with automated responses, tech support (reply coming from another address), etc.
Second, people upgrade very slowly. Forcing people to pay you'll cut yourself from 99% of the world, including most of your computer illiterate friends who will not know if it's a joke or what when they get the "payment required" email.
Third, ISPs will need to establish communication with every other ISP. I don't see why some small ISP in Russia would care much about some computer in Afghanistan claiming it wants a payment. In what currency, and what amount btw?
Fourth, it doesn't remove spam. Spamming will be simply done via worms that will infect the largest amount of computers possible and mail everybody in the address book. Given that currently we see that millions of computers are infected with Windows worms, this is going to be pretty effective.
And finally, nobody is going to like this system much anyway. Email is so popular because it's so simple. Making it like this will only make people switch to instant messengers or something else that's not so complicated, which is what's going to be spammed massively then.
Crypto or no crypto, the big problem here is the payment, which you ignore too easily. Ok, let's say grandma wants to send me an email. Let's also suppose that all this stuff has been integrated into Outlook already as well. Let's also assume that she just got a computer, so she's not in my list yet.
1. Grandma types her email address
2. Grandma sends it
3. Email arrives somewhere.
Now, how does the payment work? Clearly it has to work somehow. One option is that the payment is handled by the receiving server. I like running my own server, so I would do the charge. This means that grandma would have to send me her credit card number. Mistype the email address, and the wrong person gets the credit card number. Great opportunity for abuse.
Okay, suppose I can't run my mail server because I need a certification of that I'm an ISP, etc, etc. So my ISP's mail server does the charge instead. Now they end with a huge database of credit card numbers just waiting for being hacked. And of course some bright guy will manage to get the appropiate certs without being an ISP...
Ok, let's try without credit cards. We just make grandma's ISP pay me money. Her ISPs charges her, and then does a transfer to my ISP, that should be quite safe. There's also some kind of cryptographic authentication to make sure everybody is who they claim to be. But this is not good either. Ok, in the US maybe there could be a law mandating this implementation. But how do you get your mail from grandma when she lives in Russia and you moved to the US? Surely the russian ISP would have to agree to pay, and I don't see why would they bother.
Your ISP then drops the charge from Russia, but then the russian spammers can do pretty much as they please.
Additionally, even supposing the system is perfect it's still easy to abuse. A spammer writes a worm, infects my grandma's computer with it, and uses that to send 10000 emails at $0.01 each. Grandma can handle paying $100 you say? But she lives in Russia where an university teacher's wage is pretty close to $100. Seriously.
Charge a fair ammount for the country's economy? No, that won't work either. A spammer in the US would get an account in Russia, and pay almost nothing. Spammers are profitable with very little. If 1 in 10000 people pays $200, it's profitable. If mailing in Russia costs $0.0001, even more. I'm pretty sure some ISP in Russia would be very happy with such a customer.