In a free market price is set by the meeting place of the supply and demand curves. These are over the entire market. What creating a crippled version does is split the single market into two. One market with people willing to pay more for extra features, the other market for the rest. No loss is made on the "el cheapo" market, and a considerable profit is made in the expensive market, in comparison with selling it all at the same price.
Differentiating your market is necessary to sell commodity products (which video cards are). As demonstrated by the major sporting goods manufacturers (nike, adidas,...), which sell differentiated logos with regular clothes attached to them.
That microsoft is forced to differentiate its market like this is good news. It means they're losing their monopoly power. A monopoly can always sell everything at the higher price. Clearly, they can't anymore, not everywhere at least. So three cheers to the downfall of the evil empire.
qmail is broken up into several processes, each running in their own user account. It's very secure. Too bad it has an anal license and so can't be distributed with distro's.
My isp runs it to do the mail account of hundreds of thousands of customers. Mail service has always been smooth.
Any system offering a service to the outside world offers a vector for entry. Yes, running unneeded public services is a bad thing, but if a hacker gains entry through an apache security flaw, there's nothing you can do to stop that. You can only limit the damage they can do once they get on the system.
Hmmm, I don't know, I've always considered plain cotton underwear to be more tempting than any "sexy" lingerie. Victoria's secret just seems tacky to me. This is the kind of gift that might backfire, depending on the guy's tastes.
Okay. Then why not try some penis enlargement pills at xxx.yyy.com.
Don't see that often on slashdot? Hmm... sounds like there must be *some* difference you're missing then because I get this every day in my inbox.
One word: goatse. Slashdot has noise, but slashdot has moderation. Email has moderation too, in the form of distributed spam filters, but it's not built into the medium itself like with slashdot. To me the difference is that slashdot's moderation is on by default, whereas with email it is off by default.
To me those differences aren't very important. You could very well emulate email over slashdot or slashdot over email.
C++ is very good at *hiding* complexity, which is one of the reasons why it can be so much more powerful. But it's arguable that with C, you have a better chance of really understanding your program.
The subtleties of C++ can be understood, though it requires careful reading of stroustrup's book (which does explain them in detail, albeit it reasonably briefly to not have to spend 3000 pages on it). I admit to not having read the standard, I always refer back to stroustrup when in doubt.
I agree C++ is an overly complex language. It is development methodology-agnostic, and to my knowledge is the only language that tries to be so. However you're used to developing, and whatever language you already know, C++ can provide a close match to it. The downside is that every programmer has his own favourite methodology, and in team development projects using C++ you need very strict standards on what parts of C++ you'll use in what way. You can code in C++ as if it is C, so it's not that C++ forces complexity on you. It's just that it offers it, and most people are tempted by the dark side.
Heavy use of templates in large projects can cause an explosion of code size.
Incorrect use of templates balloons code size. It is possible to use templates heavily without having your code expand as if someone shoved an air compressor in an elephant's rear. It's also not easy to figure out how. To me this is a lack of documentation, and a design that leads to incorrect use. It is not a flaw in templates/generics per se for me.
And templates count for one of the biggest reasons to use C++ over C.
To me the biggest reason is having OO features in the language itself. Once you know how C++ OO works, you can start working on any C++ OO project (provided you understand the class structure), whereas every C OO project uses it's very own OO framework.
Operator overloading and references are also nice syntactic sugar. But understanding their intricacies can become bothersome, especially with operator overloading.
Ofcourse, the post was about camino, which doesn't suffer from the described problems. The mozilla suite itself is a dead-end development-wise anyway. All mozilla.org development is going into the birds and the backend. Only outside developers are still contributing to the suite itself. So rest assured, they know it sucks, and they intend to abandon it.
Mozillazine is good wrt getting news on mozilla. Yes, they're pro-mozilla, but if you want to know when the next version of insert-mozilla-product-here will be released, it's the best place to go.
That makes no sense whatsoever. Back and forward ARE tied to a page. If you switch tabs, back and forward will aply to the current tab. Hence, back and forward apply to the currently visible site, the current page, and so should be available in the context menu of the current page.
My guess is the apple developers simply have a rule saying "no more than this many items in the context menu", and adding back, forward and reload would have broken that limit.
First of all, there is no credible difference between holding a discussion over slashdot or holding a discussion over email. Do it through a hotmail account and you're even using the same program to do it. You can come to slashdot and read something you find offensive without warning in advance that it is, just like can happen with email. So trying to draw an arbitrary distinction between anonymous cowards on slashdot and anonymous cowards in email is just that, arbitrary.
One might also argue that shielding yourself from that which you find offensive is bad for the mind. If you shy away from extremes, inevitably your comfort zone shrinks, and you become close-minded. It's only by trying to see the viewpoints of those who disgust you that you can come to truly new realizations about how the world works. Treading the trodden moral paths doesn't take you into uncharted lands, though it does guarantee you a pretty average and "normal" life.
Secondly, the problem is that if a pki system were to take hold to identify senders, eventually it would become required to be identified just for someone to SEE the mail you're sending to them. Although it is possible to devise a system where the net identity of someone is thrustworthy while at the same time not revealing their real life identity, it is ridiculously unlikely that such a system would be promoted by the big isp's. They've already got the riaa and friends breathing down their neck wanting identification of customers, they're not going to back a system that helps people stay anonymous while comitting crime.
Too bad the founding fathers didn't recognize privacy as a right that could be threatened. Until a few decades ago, it wasn't feasible to tie together the knowledge the world has amassed on someone into one large fount of dirty details. Today it is. Most people can have their lives ruined just by the not-so-secrets that are spread around the globe about them (don't believe me? think about everything you've ever purchased with a credit card, now think about everyone in your life knowing about those purchases... unnerving, isn't it?).
There are two ways out of this, force privacy by law, or admit there is no privacy and stop holding people's pasts above their heads. Both are unlikely, and any other system leads to major abuses.
Very admirable. But I've yet to hear from someone signing with a major and then afterwards releasing their music into the open. Ofcourse, a lot of musicians don't own their music, but even those that do generally still follow the old model of clinging hard to their right to copy.
I was unaware that the music industry had been doing much complaining about people making copies of CDs for personal use. I could have sworn they were much more upset about people either A) giving out mix CDs or B) downloading illegal files.
Which is ironic given that most of the loss they're making to copying is NOT caused by mix cd's or mp3 downloads but by good old mass cd pirates.
What they're afraid of is not the loss of income to copied cd's, but the loss of income due to no longer being a monopoly, which is the inevitable outcome of handing over music distribution to filesharing networks (which are very hard to subvert marketingwise, unlike radio/tv/music stores, where you can quite easily market specific musical tastes).
Re:Really? Infamous?
on
Review: KDE 3.2
·
· Score: 4, Informative
Why did they do it in C when there is a object oriented C++ right available?
Several reasons:
- At the time of gnome's creation C++ was slow (wrt compiled code) and unstandardised (wrt source). Well, there were standards, but the popular compilers didn't pay all that much attention to them, and in fact, the MS compilers still don't pay much attention to them. As a result, a C++-based project had an immediate speed and portability hit.
- There was and still is no C++ binary abi. When you upgrade to a new compiler, you have to recompile all your libraries just to compile a new app with it. This is ugly.
And finally and most importantly:
- The gnome programmers were all C fanboys. They didn't know C++, and didn't want to learn it. Better to go with the devil you know than the devil you don't.
It doesn't really matter nowadays. GNOME uses hacks to implement OO in C, KDE/Qt uses hacks (the metacompiler) to implement signalling in C++. Both are a bit of a kludge. And both work well. Though generally I find KDE's architectural design cleaner and easier to get into. But then clearly either can be learnt and learnt well.
Wrt to smooth lines, no bleed, waterproofness and quick dry you should check out fisher space pens. They're expensive, but they're nearly indestructible, and there are some REALLY nice looking fisher pens available. Go with the blue ink though, the other ink colors are "lumpier".
They even sell pens with lifetime guarantees (if the pen ever runs out or breaks in your lifetime, you get a free replacement).
It seems to me most digital watches nowadays are designed not to last. I bought a citizen eco-drive solar-powered watch, because the ads implied that it would run forever. Ofcourse, only after my purchase did I realise that inside was a rechargeable battery, and that this battery would die pretty quickly. So, after three years (which is considerably less than most non-rechargeable watch batteries last) I had the choice: replace the expensive rechargeable battery, or spend slightly more and get me a different watch.
I bought a normal battery-powered watch (laks usb memory watch with citizen internals). Sure, the battery won't last forever, but when it dies it'll be easy and cheap to replace.
Though maybe when the battery in this one runs out I'll just go mechanical. There's something insanely tempting about a watch you can hand down to your grandchildren.
You forget one very important thing. The largest heavy mover of equipment into space is russian space rocket. it could of sent it into space without probs at a fraction of the cost.
But it couldn't have repaired it. Remember the repair mission that was needed to make hubble useful?
why so big? isn't 10 the default standard for most written communication?
I find it kind of ironic that the only people using times new roman 14 will be government employees and high school students trying to fill at least two pages with text.
Check out foneblog. I haven't tried it, since I don't have an mms or wap phone, but there are lots of phoneblogs out there, so clearly it's possible to get it to work.
Not to mention that 68k mac emulation is just about perfect (and fast!) in the shape of basilisk II. I've used it to play around with macos 8. So it's really emulating the ppc that's the problem, not the rom, or the other various bits of hardware (which was a lot more exotic on 68k macs than on ppc macs).
I see this argument over and over and frankly I just don't think it's true. Linux doesn't need native games, it just needs to run games, and run them well. Winex runs a lot of windows games, and runs them well. It runs every windows game I want to run, even those that aren't on the official supported list (like some popcap games).
I admit, I'm not hardcore into games, but linux has better gaming support than the mac, and you never hear people say "you know, macs are just not ready for the desktop because they don't run enough games."
See this list of supported games with a score of 4 or better (which means they're playable).
Even more importantly, a lot of countries (like mine) have arrangements where you can freely choose your eletricity supplier. The electric grid in that case becomes just a common carrier, like the phone network for isp's.
So don't hand out homographically similar domain names. You're not going to be allowed to keep domains like those anyway due to trademark issues, so they might as well make sure nobody can even get them.
I've spent so much time figuring out how to set the belgian be-latin1 keyboard layout on half-installed linux systems that I now know the US keyboard layout by heart. If for some reason the localised keymap doesn't load (which happens surprisingly often given my tendency for tinkering) I can still get work done.
The inability of computers to ask the keyboard for what layout it has is something that baffles me. Nowadays the keyboard, mouse and monitor are pretty much the only parts that don't get automatically recognized by the system, and they happen to be the ones that are most crucial to actually using the system.
Now the lawyers are getting patents. This was to be expected. If they have to wait for some semi-fraudulent IP business to hire them, they might have to wait several weeks or even months. This way, they get immediate and full benefit from their lowlife tactics. Ingenious.
Slashdot Mirror
Because people pay a lot more attention to changes in price - changes in weight or volume go fairly unnoticed.
This is why it is now slowly becoming a legal requirement in EU countries to prominently display the price per weight unit next to the price per unit.
In a free market price is set by the meeting place of the supply and demand curves. These are over the entire market. What creating a crippled version does is split the single market into two. One market with people willing to pay more for extra features, the other market for the rest. No loss is made on the "el cheapo" market, and a considerable profit is made in the expensive market, in comparison with selling it all at the same price.
...), which sell differentiated logos with regular clothes attached to them.
Differentiating your market is necessary to sell commodity products (which video cards are). As demonstrated by the major sporting goods manufacturers (nike, adidas,
That microsoft is forced to differentiate its market like this is good news. It means they're losing their monopoly power. A monopoly can always sell everything at the higher price. Clearly, they can't anymore, not everywhere at least. So three cheers to the downfall of the evil empire.
qmail is broken up into several processes, each running in their own user account. It's very secure. Too bad it has an anal license and so can't be distributed with distro's.
My isp runs it to do the mail account of hundreds of thousands of customers. Mail service has always been smooth.
Any system offering a service to the outside world offers a vector for entry. Yes, running unneeded public services is a bad thing, but if a hacker gains entry through an apache security flaw, there's nothing you can do to stop that. You can only limit the damage they can do once they get on the system.
Hmmm, I don't know, I've always considered plain cotton underwear to be more tempting than any "sexy" lingerie. Victoria's secret just seems tacky to me. This is the kind of gift that might backfire, depending on the guy's tastes.
Okay. Then why not try some penis enlargement pills at xxx.yyy.com.
Don't see that often on slashdot? Hmm... sounds like there must be *some* difference you're missing then because I get this every day in my inbox.
One word: goatse. Slashdot has noise, but slashdot has moderation. Email has moderation too, in the form of distributed spam filters, but it's not built into the medium itself like with slashdot. To me the difference is that slashdot's moderation is on by default, whereas with email it is off by default.
To me those differences aren't very important. You could very well emulate email over slashdot or slashdot over email.
C++ is very good at *hiding* complexity, which is one of the reasons why it can be so much more powerful. But it's arguable that with C, you have a better chance of really understanding your program.
The subtleties of C++ can be understood, though it requires careful reading of stroustrup's book (which does explain them in detail, albeit it reasonably briefly to not have to spend 3000 pages on it). I admit to not having read the standard, I always refer back to stroustrup when in doubt.
I agree C++ is an overly complex language. It is development methodology-agnostic, and to my knowledge is the only language that tries to be so. However you're used to developing, and whatever language you already know, C++ can provide a close match to it. The downside is that every programmer has his own favourite methodology, and in team development projects using C++ you need very strict standards on what parts of C++ you'll use in what way. You can code in C++ as if it is C, so it's not that C++ forces complexity on you. It's just that it offers it, and most people are tempted by the dark side.
Heavy use of templates in large projects can cause an explosion of code size.
Incorrect use of templates balloons code size. It is possible to use templates heavily without having your code expand as if someone shoved an air compressor in an elephant's rear. It's also not easy to figure out how. To me this is a lack of documentation, and a design that leads to incorrect use. It is not a flaw in templates/generics per se for me.
And templates count for one of the biggest reasons to use C++ over C.
To me the biggest reason is having OO features in the language itself. Once you know how C++ OO works, you can start working on any C++ OO project (provided you understand the class structure), whereas every C OO project uses it's very own OO framework.
Operator overloading and references are also nice syntactic sugar. But understanding their intricacies can become bothersome, especially with operator overloading.
Ofcourse, the post was about camino, which doesn't suffer from the described problems. The mozilla suite itself is a dead-end development-wise anyway. All mozilla.org development is going into the birds and the backend. Only outside developers are still contributing to the suite itself. So rest assured, they know it sucks, and they intend to abandon it.
Mozillazine is good wrt getting news on mozilla. Yes, they're pro-mozilla, but if you want to know when the next version of insert-mozilla-product-here will be released, it's the best place to go.
That makes no sense whatsoever. Back and forward ARE tied to a page. If you switch tabs, back and forward will aply to the current tab. Hence, back and forward apply to the currently visible site, the current page, and so should be available in the context menu of the current page.
My guess is the apple developers simply have a rule saying "no more than this many items in the context menu", and adding back, forward and reload would have broken that limit.
First of all, there is no credible difference between holding a discussion over slashdot or holding a discussion over email. Do it through a hotmail account and you're even using the same program to do it. You can come to slashdot and read something you find offensive without warning in advance that it is, just like can happen with email. So trying to draw an arbitrary distinction between anonymous cowards on slashdot and anonymous cowards in email is just that, arbitrary.
One might also argue that shielding yourself from that which you find offensive is bad for the mind. If you shy away from extremes, inevitably your comfort zone shrinks, and you become close-minded. It's only by trying to see the viewpoints of those who disgust you that you can come to truly new realizations about how the world works. Treading the trodden moral paths doesn't take you into uncharted lands, though it does guarantee you a pretty average and "normal" life.
Secondly, the problem is that if a pki system were to take hold to identify senders, eventually it would become required to be identified just for someone to SEE the mail you're sending to them. Although it is possible to devise a system where the net identity of someone is thrustworthy while at the same time not revealing their real life identity, it is ridiculously unlikely that such a system would be promoted by the big isp's. They've already got the riaa and friends breathing down their neck wanting identification of customers, they're not going to back a system that helps people stay anonymous while comitting crime.
Too bad the founding fathers didn't recognize privacy as a right that could be threatened. Until a few decades ago, it wasn't feasible to tie together the knowledge the world has amassed on someone into one large fount of dirty details. Today it is. Most people can have their lives ruined just by the not-so-secrets that are spread around the globe about them (don't believe me? think about everything you've ever purchased with a credit card, now think about everyone in your life knowing about those purchases... unnerving, isn't it?).
There are two ways out of this, force privacy by law, or admit there is no privacy and stop holding people's pasts above their heads. Both are unlikely, and any other system leads to major abuses.
Very admirable. But I've yet to hear from someone signing with a major and then afterwards releasing their music into the open. Ofcourse, a lot of musicians don't own their music, but even those that do generally still follow the old model of clinging hard to their right to copy.
I was unaware that the music industry had been doing much complaining about people making copies of CDs for personal use. I could have sworn they were much more upset about people either A) giving out mix CDs or B) downloading illegal files.
Which is ironic given that most of the loss they're making to copying is NOT caused by mix cd's or mp3 downloads but by good old mass cd pirates.
What they're afraid of is not the loss of income to copied cd's, but the loss of income due to no longer being a monopoly, which is the inevitable outcome of handing over music distribution to filesharing networks (which are very hard to subvert marketingwise, unlike radio/tv/music stores, where you can quite easily market specific musical tastes).
Why did they do it in C when there is a object oriented C++ right available?
Several reasons:
- At the time of gnome's creation C++ was slow (wrt compiled code) and unstandardised (wrt source). Well, there were standards, but the popular compilers didn't pay all that much attention to them, and in fact, the MS compilers still don't pay much attention to them. As a result, a C++-based project had an immediate speed and portability hit.
- There was and still is no C++ binary abi. When you upgrade to a new compiler, you have to recompile all your libraries just to compile a new app with it. This is ugly.
And finally and most importantly:
- The gnome programmers were all C fanboys. They didn't know C++, and didn't want to learn it. Better to go with the devil you know than the devil you don't.
It doesn't really matter nowadays. GNOME uses hacks to implement OO in C, KDE/Qt uses hacks (the metacompiler) to implement signalling in C++. Both are a bit of a kludge. And both work well. Though generally I find KDE's architectural design cleaner and easier to get into. But then clearly either can be learnt and learnt well.
Wrt to smooth lines, no bleed, waterproofness and quick dry you should check out fisher space pens. They're expensive, but they're nearly indestructible, and there are some REALLY nice looking fisher pens available. Go with the blue ink though, the other ink colors are "lumpier".
They even sell pens with lifetime guarantees (if the pen ever runs out or breaks in your lifetime, you get a free replacement).
It seems to me most digital watches nowadays are designed not to last. I bought a citizen eco-drive solar-powered watch, because the ads implied that it would run forever. Ofcourse, only after my purchase did I realise that inside was a rechargeable battery, and that this battery would die pretty quickly. So, after three years (which is considerably less than most non-rechargeable watch batteries last) I had the choice: replace the expensive rechargeable battery, or spend slightly more and get me a different watch.
I bought a normal battery-powered watch (laks usb memory watch with citizen internals). Sure, the battery won't last forever, but when it dies it'll be easy and cheap to replace.
Though maybe when the battery in this one runs out I'll just go mechanical. There's something insanely tempting about a watch you can hand down to your grandchildren.
You forget one very important thing. The largest heavy mover of equipment into space is russian space rocket. it could of sent it into space without probs at a fraction of the cost.
But it couldn't have repaired it. Remember the repair mission that was needed to make hubble useful?
why so big? isn't 10 the default standard for most written communication?
I find it kind of ironic that the only people using times new roman 14 will be government employees and high school students trying to fill at least two pages with text.
Oh, and all White House press conferences must have at least "fifteen pieces of flair".
Not to mention all single syllable words.
It's not like it's hard to say complicated things with words of 4 letters or less.
Check out foneblog. I haven't tried it, since I don't have an mms or wap phone, but there are lots of phoneblogs out there, so clearly it's possible to get it to work.
Not to mention that 68k mac emulation is just about perfect (and fast!) in the shape of basilisk II. I've used it to play around with macos 8. So it's really emulating the ppc that's the problem, not the rom, or the other various bits of hardware (which was a lot more exotic on 68k macs than on ppc macs).
I see this argument over and over and frankly I just don't think it's true. Linux doesn't need native games, it just needs to run games, and run them well. Winex runs a lot of windows games, and runs them well. It runs every windows game I want to run, even those that aren't on the official supported list (like some popcap games).
I admit, I'm not hardcore into games, but linux has better gaming support than the mac, and you never hear people say "you know, macs are just not ready for the desktop because they don't run enough games."
See this list of supported games with a score of 4 or better (which means they're playable).
Even more importantly, a lot of countries (like mine) have arrangements where you can freely choose your eletricity supplier. The electric grid in that case becomes just a common carrier, like the phone network for isp's.
So don't hand out homographically similar domain names. You're not going to be allowed to keep domains like those anyway due to trademark issues, so they might as well make sure nobody can even get them.
I've spent so much time figuring out how to set the belgian be-latin1 keyboard layout on half-installed linux systems that I now know the US keyboard layout by heart. If for some reason the localised keymap doesn't load (which happens surprisingly often given my tendency for tinkering) I can still get work done.
The inability of computers to ask the keyboard for what layout it has is something that baffles me. Nowadays the keyboard, mouse and monitor are pretty much the only parts that don't get automatically recognized by the system, and they happen to be the ones that are most crucial to actually using the system.
Why is that? Are hardware designers such asshats?
Now the lawyers are getting patents. This was to be expected. If they have to wait for some semi-fraudulent IP business to hire them, they might have to wait several weeks or even months. This way, they get immediate and full benefit from their lowlife tactics. Ingenious.