Try getting any of these boutique Linux builders to ship anywhere outside of USA or Canada.
This actually sucks a lot. I'm in the US, so I don't have this problem, but it is legitimately unfortunate for those of you who are.
If you think the market is truly under served, you could always start a non-US Linux Laptop company yourself. The suppliers for whitebox laptops are there (Asus, Quanta, Sager)... you could make some good money.
To start with, we're not even using the same currency here. It looks like the Canadian / US exchange rate doesn't favor US to Canada imports right now.
The discrete graphics thing is weird too. System76 doesn't offer systems with extremely low end discrete graphics like that to compare to (I guess they assume that no-one would want to jank up their system with binary blobs without actually getting solid game performance in exchange).
Obviously System76 is a smaller vendor, and their ~10 different models can't replicate all the different models that Dell offers. I'll stand by my previous statement though - If you compare identical configurations in the same currency you won't see massive price differences.
Did you actually compare similar systems with the same specs? Dell versus http://www.system76.com/ The biggest difference I've found in favor of Dell or Lenovo was around $90 when comparing actually similar systems - the normal difference is more like $40.
Even little things like an inch of screen size or which processor model number *exactly* can mean a $200+ difference.
Most of the time when people make the "few hundred bucks" claim, they're comparing a 15.4" widescreen Dell with a CD-RW drive and a low end Turion processor with a 13.3" System76 mini-notebook with top of the line Core 2 Duo and a dual-layer DVD burner.
If you can give an example where I'm wrong about the $90 maximum difference, please post it. I'm pretty sure I'll find some giant difference between the systems when I look at it. Or I'll find out that you're dumb and are buying high-end gaming desktops pre-built...
As for integrated graphics versus dedicated, if you're going to be running Linux on the thing Intel integrated graphics are *strictly superior* to any other option. They're a little slow for UT2004 and Quake 4, but they have free software drivers and work better for stuff like Compiz / Beryl or even just 2D acceleration for web browsing. Oh, and you get a bunch more battery life.
Most of the companies that specialize in "Linux laptops" that I've ever seen, charge a significant premium.
Most of the people who say this haven't actually looked at the price situation recently.
Configure a system from http://www.system76.com/ and an identical one from Dell. Check what the price difference is. My guess is that the couple bucks difference is worth not having to wipe Windows off the system, being *sure* that Linux supports the hardware, and not wondering if you got special discount "Dell hardware" where Dell demanded that reliability or performance be compromised for cost on the Dell system.
The intent is to confuse people. If you differentiate your product and don't make it 100% obvious as to why, most people will automatically buy whatever the most expensive choice they can afford is. Additionally, those people who care about the differences/want to get a good deal will figure it out and buy what they actually want. There's no downside.
What's going to happen when Dell releases a flavor that can't play MP3s, or some media files, out of the box?
The same thing that happened when they offered any of the versions of Windows that didn't support every media format under the sun by default. Installing codec packs is neither new nor hard - under Linux it's easier than usual, because Linux vendors provide instructions that don't result in getting 45 viruses and 19 pieces of spyware.
Right now Linux notebooks are roughly twice the cost of an identical Windows box.
Wait, what? That's simply not the case.
I specced out identical laptops from System76 (a Linux system retailer), Dell, Lenovo, and HP a couple weeks ago when someone made this claim in another thread, and the prices are about the same.
Make sure you're really getting the same hardware. A "Dell" wireless card isn't the same as a Intel wireless card.
Compare the configuration you would actually buy. For example, Dell starts a little cheaper than System76 but stuff like RAM upgrades may be more expensive.
Don't compare apples to oranges. For example, Laptops with 15.4" widescreen displays are cheaper than other sizes - don't expect a 4:3 screen or a smaller screen to be cheaper.
It's true that for very cheap desktop systems with low-end flatscreen monitors and shitty inkjet printers running Vista Home Basic, Dell currently has better prices than anyone. Don't let that distract you from comparing the prices of stuff that you would actually want to buy.
If you want to have more time in the afternoon, that's easy - arrange to come into work an hour earlier.
Personally, I'd much rather sleep in. There's no reason why you should be promoting a federal law that changes everyone's clocks because you don't like your work schedule.
I feel somewhat safer typing stuff in to my computer than I do typing stuff into a computer at a public library. The chance of someone having the opportunity to install a keylogger on my computer is much smaller than on the Library workstation.
I've been using http://galaxyvoice.com/ - $0/month plus $0.023/minute is pretty damn good, especially with a reliability level of "it's always just worked".
5.5% is a drop in the bucket when overhead is low and you're stealing someone else's R&D.
5.5% is *never* a drop in the bucket. That's a business-killer even if you're selling stolen cars to people who want to over-pay in a world with no cops.
But... back in the real world... 5.5% is a huge hit to a company that did its own R&D and later found out that someone else went and got the government granted monopoly to the obvious ideas they were using first.
A computer is a tool that a user wants to use to get things done.
This is a perfectly valid argument, to a point.
The fact of the matter is, most office workers sit in front of a computer for 8 hours *every day*. That's more time than a carpenter spends working with his cordless drill or a pilot spends in an airplane cockpit. Sure, not everyone needs to know the differences between a semaphore and a spin lock, or a precise vs. conservative garbage collector - but if your *entire job* consists of using a tool, you probably should have some basic grasp of what it is and how to use it.
Almost all types of mainstream software has a Free Software version for GNU/Linux with more than enough functionality for mainstream users.
Sure, a profesional graphics artist with 5 years of Photoshop experience will prefer it to The Gimp or Gimpshop, but 99.8% of people aren't graphics professionals with years of photoshop experience - and for normal people Gimp is great. The same is true with OpenOffice, Evolution, etc.
Yes, it's true that random-CAD-software X doesn't have a Linux port. That's not mainstream, not even 0.2% of people use it - those people an easily keep a Win2k workstation or whatever.
The only exception I can think of is tax preparation software, and there are decent pay-for-use Web apps that do that. The other consideration is games, but most people don't cough up the money for a decent gaming computer anyway.
Basically, the applications barrier to entry that Linux has been facing for years has been largely overcome by the inertia of Free Software development. Yes, there's a one cycle retaining penalty to move to Linux - but that's *exactly* the same penalty as to move to Vista, and it won't necessarily happen again as soon for Linux as it will for Vista + 1.
Are we talking about discrete integers used primarily as sequential labels, or are we talking about decimal floating point numbers. In the first case, 2007 makes obvious sense. In the second case, 2700 (accurate to two places) makes obvious sense.
To divide all programming systems into "compiled" and "interpreted" and put Java in the latter category is strictly inaccurate, and does a giant disservice to both the meaning of the word "compile" and the fact that the topic is reasonably complex. At least say "compiled to native code" if that's what you mean, but even that is oversimplifying.
You definitely want to maintain at least these categories separately, and note that they're linked to programming environments, not languages. You could write an one-line-at-a-time interpreter for assembly language if you wanted, or a native-code compiler for BASIC (actually, there are a couple of each out there).
Compiled to native code, like C normally is.
Compiled to bytecode, with a JIT-enabled VM, like Java usually is.
Compiled to bytecode with a traditional VM, like many "research" languages.
Compiled on startup, like Perl and Python.
Interpreted line-by-line, like a traditional shell script.
And I do believe the system needs some refinement. But I don't think getting rid of it solves anything. It just changes the playing field for the same outcome. Large companies still have an advantage over smaller one. We can change these advantages to ones they truly deserve or we can let them run free with everything and still have a broad advantage.
Patents provide exactly one thing: An excuse to litigate. That's something that will always be better for large companies than it is for the "little guy". No revamping of the patent system will change that fact. In any lawsuit, you can never be sure you're going to win - even if you're in the right. That's a risk that only big companies can afford to take, especially when big companies are free to spend millions on lawyers and then ask the judge to grant them legal fees if they manage to win.
It might be possible to construct some sort of alternate patent system that wasn't just an excuse for legal denial of service attacks, but that's not what we have. What we do have is a horribly unfair system that prevents innovation and strangles the small entrepreneur.
Putting antibiotics into meat animals isn't the least bit harmful... directly. It produces meat that is less likely to have bacteria in it at a lower cost. The antibiotics themselves aren't harmful to humans either. Meat-sellers who use this technique simply are able to sell a better product cheaper.
The problem with overuse of antibiotics are largely a side effect - it results in breeding antibiotic-resistant bacteria. Antibiotics are a shared limited resource - if they're overused they stop working.
As a society, we have two choices:
Let the free market work. It will expend the valuable limited resource as quickly as possible. It could be argued that this is good, because economic development is better now than later.
Introduce regulation that warps the market. In this case, it may make more sense to conserve antibiotics for saving the lives people rather than using them to make steaks cheaper. Further, it looks like "antibiotics as a resource" might have strange properties - if we limit their use we might manage to not kill off the non-resistant strains, which would allow antibiotics to continue working for a very long time.
Perhaps the best way to look at this issue isn't as market problem at all - people who overuse antibiotics are breeding antibiotic-resistant bacteria in the wild. That should probably be covered under some anti-bioterrorism law or something.
It has to be no pattents because of 20 years. It cannot be 5 years instead
So far, I've seen no evidence that patents are beneficial at all, and I've seen the harmful effects they have with my own eyes. Any shortening of the 20 year patent period would be great, but the damage that patents do would still occur with a 5 year duration.
If I take out a patent... you could still use it. All you have to do is satisfy my asking price.
Right. Your asking price and the asking price of everyone else who has a patent that applies to my product. And the large companies competing in the same market also get to use those patented ideas but don't have to pay the licensing fees because of their cross-licensing agreements.
No non-trivial product uses only one patentable idea. For a complex hardware device, hundreds. For a piece of software, thousands. If my product infringes on a hundred patents and each patent holder wants a 5% cut of the profits... I simply don't get to sell my product.
That doesn't even take into account the cases where someone has a patent and decides to use it to exclude all competition from the market. Nothing says they have to license their patent, so they can just take their government granted monopoly and make everyone else wait 20 years (or however long the patent period is) to use "their" idea. It doesn't matter if their idea is absolutely essential to some other important technology, they can just hold progress hostage to their profits (or their sense of entitlement if that decision isn't actually profitable).
You're afraid that big companies will be able to outsell little companies without patents. You're ignoring their giant patent portfolios. The little guy never really gets to say "I have a patent, you can't copy my idea", because the big guy can always come back with "Sure, but you're infringing on these 18 patents. Want to go to court about it?".
Unless the differences are distinguishable in the final product, none of that stuff matters. If that stuff did matter, people would do chemical tests on different brand meats and publish the results - resulting in publicity for the companies with better products. If the products can't be distinguished through testing, then they're not different - regardless of how much you want to say that "range fed cattle are tastier" or whatever.
As for antibiotics and groundwater pollution, those are classic negative externalities most easily resolved through regulation to restore a fair market.
If you used to be able to get through 250 million interations of the loop and now you can get through 247 million iterations of the loop, then you know something is consuming resources on your system.
This would work great, if your system was running *no software*. Unfortunately, the scheduler's job is to run software as quickly as possible - not give you consistent timing numbers. When it comes to actual computers running actual applications, a 1% timing change is well within normal variance.
As for being able to tell *anything* about timing across a network - bullshit. Utter bullshit.
The free market fails in the face of uncompensated externalities.
"A few million lives" for "megabucks" won't produce a strong economy anyway. In an economic analysis, you *can* put a price on human lives - but that price is well over the couple hundred bucks each this statement implies at maximum.
Slashdot Mirror
This actually sucks a lot. I'm in the US, so I don't have this problem, but it is legitimately unfortunate for those of you who are.
If you think the market is truly under served, you could always start a non-US Linux Laptop company yourself. The suppliers for whitebox laptops are there (Asus, Quanta, Sager)... you could make some good money.
To start with, we're not even using the same currency here. It looks like the Canadian / US exchange rate doesn't favor US to Canada imports right now.
The discrete graphics thing is weird too. System76 doesn't offer systems with extremely low end discrete graphics like that to compare to (I guess they assume that no-one would want to jank up their system with binary blobs without actually getting solid game performance in exchange).
Obviously System76 is a smaller vendor, and their ~10 different models can't replicate all the different models that Dell offers. I'll stand by my previous statement though - If you compare identical configurations in the same currency you won't see massive price differences.
If you want a specific brand, you'll have to buy it from the manufacturer. Sorry. That's the way brands work.
Athlon, it's not something that Dell sells.
Wait, no, that was last year.
*All* of those product names sound stupid as a product buyer, but from a marketing perspective they're great.
Did you actually compare similar systems with the same specs? Dell versus http://www.system76.com/ The biggest difference I've found in favor of Dell or Lenovo was around $90 when comparing actually similar systems - the normal difference is more like $40.
Even little things like an inch of screen size or which processor model number *exactly* can mean a $200+ difference.
Most of the time when people make the "few hundred bucks" claim, they're comparing a 15.4" widescreen Dell with a CD-RW drive and a low end Turion processor with a 13.3" System76 mini-notebook with top of the line Core 2 Duo and a dual-layer DVD burner.
If you can give an example where I'm wrong about the $90 maximum difference, please post it. I'm pretty sure I'll find some giant difference between the systems when I look at it. Or I'll find out that you're dumb and are buying high-end gaming desktops pre-built...
As for integrated graphics versus dedicated, if you're going to be running Linux on the thing Intel integrated graphics are *strictly superior* to any other option. They're a little slow for UT2004 and Quake 4, but they have free software drivers and work better for stuff like Compiz / Beryl or even just 2D acceleration for web browsing. Oh, and you get a bunch more battery life.
Emperor Linux is expensive, which is why I didn't mention them.
Other vendors, such as http://www.system76.com/, don't have such ridiculous prices.
Most of the people who say this haven't actually looked at the price situation recently.
Configure a system from http://www.system76.com/ and an identical one from Dell. Check what the price difference is. My guess is that the couple bucks difference is worth not having to wipe Windows off the system, being *sure* that Linux supports the hardware, and not wondering if you got special discount "Dell hardware" where Dell demanded that reliability or performance be compromised for cost on the Dell system.
The intent is to confuse people. If you differentiate your product and don't make it 100% obvious as to why, most people will automatically buy whatever the most expensive choice they can afford is. Additionally, those people who care about the differences/want to get a good deal will figure it out and buy what they actually want. There's no downside.
No way. Screw this 1970's Unix crap. I want a dual boot Plan 9 / Coyotos system.
The same thing that happened when they offered any of the versions of Windows that didn't support every media format under the sun by default. Installing codec packs is neither new nor hard - under Linux it's easier than usual, because Linux vendors provide instructions that don't result in getting 45 viruses and 19 pieces of spyware.
Wait, what? That's simply not the case.
I specced out identical laptops from System76 (a Linux system retailer), Dell, Lenovo, and HP a couple weeks ago when someone made this claim in another thread, and the prices are about the same.
It's true that for very cheap desktop systems with low-end flatscreen monitors and shitty inkjet printers running Vista Home Basic, Dell currently has better prices than anyone. Don't let that distract you from comparing the prices of stuff that you would actually want to buy.
If you want to have more time in the afternoon, that's easy - arrange to come into work an hour earlier.
Personally, I'd much rather sleep in. There's no reason why you should be promoting a federal law that changes everyone's clocks because you don't like your work schedule.
I feel somewhat safer typing stuff in to my computer than I do typing stuff into a computer at a public library. The chance of someone having the opportunity to install a keylogger on my computer is much smaller than on the Library workstation.
I've been using http://galaxyvoice.com/ - $0/month plus $0.023/minute is pretty damn good, especially with a reliability level of "it's always just worked".
5.5% is *never* a drop in the bucket. That's a business-killer even if you're selling stolen cars to people who want to over-pay in a world with no cops.
But... back in the real world... 5.5% is a huge hit to a company that did its own R&D and later found out that someone else went and got the government granted monopoly to the obvious ideas they were using first.
This is a perfectly valid argument, to a point.
The fact of the matter is, most office workers sit in front of a computer for 8 hours *every day*. That's more time than a carpenter spends working with his cordless drill or a pilot spends in an airplane cockpit. Sure, not everyone needs to know the differences between a semaphore and a spin lock, or a precise vs. conservative garbage collector - but if your *entire job* consists of using a tool, you probably should have some basic grasp of what it is and how to use it.
Almost all types of mainstream software has a Free Software version for GNU/Linux with more than enough functionality for mainstream users.
Sure, a profesional graphics artist with 5 years of Photoshop experience will prefer it to The Gimp or Gimpshop, but 99.8% of people aren't graphics professionals with years of photoshop experience - and for normal people Gimp is great. The same is true with OpenOffice, Evolution, etc.
Yes, it's true that random-CAD-software X doesn't have a Linux port. That's not mainstream, not even 0.2% of people use it - those people an easily keep a Win2k workstation or whatever.
The only exception I can think of is tax preparation software, and there are decent pay-for-use Web apps that do that. The other consideration is games, but most people don't cough up the money for a decent gaming computer anyway.
Basically, the applications barrier to entry that Linux has been facing for years has been largely overcome by the inertia of Free Software development. Yes, there's a one cycle retaining penalty to move to Linux - but that's *exactly* the same penalty as to move to Vista, and it won't necessarily happen again as soon for Linux as it will for Vista + 1.
Are we talking about discrete integers used primarily as sequential labels, or are we talking about decimal floating point numbers. In the first case, 2007 makes obvious sense. In the second case, 2700 (accurate to two places) makes obvious sense.
To divide all programming systems into "compiled" and "interpreted" and put Java in the latter category is strictly inaccurate, and does a giant disservice to both the meaning of the word "compile" and the fact that the topic is reasonably complex. At least say "compiled to native code" if that's what you mean, but even that is oversimplifying.
You definitely want to maintain at least these categories separately, and note that they're linked to programming environments, not languages. You could write an one-line-at-a-time interpreter for assembly language if you wanted, or a native-code compiler for BASIC (actually, there are a couple of each out there).
Patents provide exactly one thing: An excuse to litigate. That's something that will always be better for large companies than it is for the "little guy". No revamping of the patent system will change that fact. In any lawsuit, you can never be sure you're going to win - even if you're in the right. That's a risk that only big companies can afford to take, especially when big companies are free to spend millions on lawyers and then ask the judge to grant them legal fees if they manage to win.
It might be possible to construct some sort of alternate patent system that wasn't just an excuse for legal denial of service attacks, but that's not what we have. What we do have is a horribly unfair system that prevents innovation and strangles the small entrepreneur.
Putting antibiotics into meat animals isn't the least bit harmful... directly. It produces meat that is less likely to have bacteria in it at a lower cost. The antibiotics themselves aren't harmful to humans either. Meat-sellers who use this technique simply are able to sell a better product cheaper.
The problem with overuse of antibiotics are largely a side effect - it results in breeding antibiotic-resistant bacteria. Antibiotics are a shared limited resource - if they're overused they stop working.
As a society, we have two choices:
Perhaps the best way to look at this issue isn't as market problem at all - people who overuse antibiotics are breeding antibiotic-resistant bacteria in the wild. That should probably be covered under some anti-bioterrorism law or something.
So far, I've seen no evidence that patents are beneficial at all, and I've seen the harmful effects they have with my own eyes. Any shortening of the 20 year patent period would be great, but the damage that patents do would still occur with a 5 year duration.
Right. Your asking price and the asking price of everyone else who has a patent that applies to my product. And the large companies competing in the same market also get to use those patented ideas but don't have to pay the licensing fees because of their cross-licensing agreements.
No non-trivial product uses only one patentable idea. For a complex hardware device, hundreds. For a piece of software, thousands. If my product infringes on a hundred patents and each patent holder wants a 5% cut of the profits... I simply don't get to sell my product.
That doesn't even take into account the cases where someone has a patent and decides to use it to exclude all competition from the market. Nothing says they have to license their patent, so they can just take their government granted monopoly and make everyone else wait 20 years (or however long the patent period is) to use "their" idea. It doesn't matter if their idea is absolutely essential to some other important technology, they can just hold progress hostage to their profits (or their sense of entitlement if that decision isn't actually profitable).
You're afraid that big companies will be able to outsell little companies without patents. You're ignoring their giant patent portfolios. The little guy never really gets to say "I have a patent, you can't copy my idea", because the big guy can always come back with "Sure, but you're infringing on these 18 patents. Want to go to court about it?".
Unless the differences are distinguishable in the final product, none of that stuff matters. If that stuff did matter, people would do chemical tests on different brand meats and publish the results - resulting in publicity for the companies with better products. If the products can't be distinguished through testing, then they're not different - regardless of how much you want to say that "range fed cattle are tastier" or whatever.
As for antibiotics and groundwater pollution, those are classic negative externalities most easily resolved through regulation to restore a fair market.
This would work great, if your system was running *no software*. Unfortunately, the scheduler's job is to run software as quickly as possible - not give you consistent timing numbers. When it comes to actual computers running actual applications, a 1% timing change is well within normal variance.
As for being able to tell *anything* about timing across a network - bullshit. Utter bullshit.
The free market fails in the face of uncompensated externalities.
"A few million lives" for "megabucks" won't produce a strong economy anyway. In an economic analysis, you *can* put a price on human lives - but that price is well over the couple hundred bucks each this statement implies at maximum.